PHENOMENA

REFERENCE DATA Applicant hereby incorporates by reference Disclosure Document No. 343952, filed December

6, 1993, for Holographic Photonic Transistor , and Disclosure Document No. 352533, filed April 19, 1994, for Photonic Amplifier, Boolean AND, Power Amplifier, Photonic Controller, Avalanche Amplifier. Filed in the U.S. Patent office.

TECHNICAL FIELD

The present invention relates to energy beam control of wave-type energy beams including electromagnetic waves, acoustical waves, and moving particles, optical computers, optical signal processing, optical signal amplification, and basic optical logic functions.

BACKGROUND ART

The "Optical Computing Method Using Interference Fringe Components Regions" (U.S. patent No. 5,093,802, Hait, 1992) provides amplification by using constructive interference to remove energy from an energy-supplying beam which remains on, diverting it into an output along with energy from a niodulated input beam. As a result, the amount of energy that appears in the modulated portion of the output is greater than the amount of energy in the modulated input beam.

The disadvantage of the prior method is that a portion of the energy supplying beam appears in the output all the time. This occurs because the prior art uses the most popular and well studied interference effects, such as Young's fringes, where energy appears at the location of fringe component separation, whenever any one of die input beams is on by itself. See, U.S. Pat. 5,093,802, State 2 of Figs. 1 and 2, and Col. 6, lines 7 - 45, especially lines 36 - 40. However, there exist other interference phenomena which may be used to alleviate the prior art problems.

These special interference phenomena are produced whenever the geometry of the apparatus is such that energy from a plurality of beams causes destructive interference at the first location(s) where energy from the input beams appears when any one of die input beams is on by itself. Since the law of conservation of energy requires mat the energy in the beams not be destroyed by the destructive interference, when an out-of-phase beam is on, the energy must appear somewhere else. Depending on die geometry of beam superposition, the energy will be reflected, or diverted to a position adjacent to die first locations), or at some angle in between. The important result is that energy from die plurality of beams is actually diverted away from die first location(s) where destructive i-αterference occurs and on to a second location where constructive interference occurs, outside of the area where at least one input beam appears in die absence of interference.

In the most elementary examples, having only two input beams, two types of special interference are manifest. With the first type, neither of the input beams contribute energy to the second location when either one is on by itself. When both input beams are on, interference causes

energy from both beams to appear at the second location.

With the second type of special interference, die first input beam contributes no energy to die second location when it is on by itself. When die second input beam comes on, interference causes energy from both input beams to appear at die second location. However, energy from the second beam does appear at die second location when it is on by itself.

Some embodiments and applications of die present invention is able to use either type of special interference. There are some dungs, however, that require one type or die other, but will not work for both types; e.g., the logical AND, discussed below.

The individual beams, in either type of special interference, actually produce images at the locations where interference takes place, even if these images are just simple spots. These images then interfere with each other.

In complex images, one or more input beams are able to produce image component area(s) that correspond to die simple examples above. The inputs are subsets of a plurality of input beams diat form images. When only one beam set is on, and as a result its image is on, the energy pattern defines a set of "first" locations by d e presence of energy. When at least two of die subsets are on, interference occurs between die two images, and energy from both images is removed from the first locations by destructive interference. That energy men appears at die second locations) because of constructive interference. The second locations lie outside of die area where die first locations are.

Holograms, especially but not exclusively computer-generated holograms, like other pictures, are made up of individual pixels. From each pixel comes a group of rays that eventually combine to produce die wave-front reconstructed holographic image. As a result, each spot on the image is produced by a group of rays from die hologram. The rays constitute a set of beams. When a whole set of beams are modulated in concert, the image it produces, and the complex interference that occurs between it and outer images is also modulated. Interference between such images, made by subsets of all input beams, are also able to be used to produce die special interference phenomena used by die present invention.

The important difference between these special interference phenomena and Young's fringes used in die prior art is that energy from at least one of die input beam sets, which appears at the second locations), appears while interfeience is occurring, and does not appear at mat locations) in the absence of interference. On die other hand, die input beams used in Young's fringes do appear at that second locations) in die absence of interference, when any of those beams are on by themselves.

These special phenomena are analog in nature, in tiiat die amount of energy diat appears at the second locations) is proportional to die amount of energy in die two input beams or images. The energy appearing at die second location(s) has been diverted from die first location(s). If one input is held constant, and a second input(s) is increased, die amount of energy contributed to die second locations) from the first input(s) reaches a limit where die addition of more energy in the second input(s) is unable to cause more energy from die first input(s) to appear at die second locations).

The phenomena may be utilized in digital energy circuits through die use of discrete levels for

modulating die input beams, to establish discrete states of die interference images, having discrete amounts of energy in their component parts.

The present invention uses d-ese heretofore-unused phenomena to produce a basic means and method of energy beam control that has direct application in optical computing, photonic signal processing, acoustic imaging, and moving particle imaging.

An additional difference between die present invention and nearly all of the prior art is diat die present invention operates at die full speed of whatever energy form is being used. For example, if light is the energy form used, die present invention operates completely at the speed of light, as does die invention disclosed in U.S. Patent 5,093,802. Any introduction of electronic, mechanical, or acousto-niechanical components only serves to limit processing speed to diat of the slowest component. U.S. patent 5,239,173, by Yang, is an excellent example of an attempt to amalgamate light with slow components, while also using Young's fringes. Yang states, in col. 2 lines 17 and 18, "A mechanical or electro-optical shutter is provided at each slit so as to turn die light ON and OFF," and restates diis in his claim 1, col. 5 line 23. Optical sensors or detectors are also used as noted at col. 3, lines 58 & 58, and col. 5 lines 61 & 62. These also limit die speed of operation to that of the slow electronic sensors.

Electrons are simply too slow. Photons are much faster, and diat is why die present invention should be practiced without utilizing any components diat require changes in die energy type used, although die present invention is able to use one embodiment for acoustical waves for detecting or generating acoustical images, while another embodiment is able to use light for processing (hose images after tiiey have been converted to optical signals.

Yang also uses Young's fringes. This is evident by die description of his first three figures at col. 2 line 58 to col. 3 line 58 which describes double slit diffraction, a common term of art for Young-type interference fringes. The fact diat his AND device requires "two detectors working cooperatively" (col. 3 lines 54 to 57) shows tiiis to be die case. The "null" as well as die "constructive interference" at two different positions must be sensed in order to detect a state where bom input beams are on so diat his AND is actually the result of ANDing two sensor outputs, and is not a direct result of using interference alone.

As a result of tiiese inherent problems of the prior art, the whole conceptual process of the present invention is geared to die interaction of components without requiring conversion from one energy form to another, and to die use of special interference, which is able to produce effects diat Young's fringes do not.

DISCLOSURE OF THE INVENTION

The present invention is a basic means and method for controlling one set of beams of energy widi another set of beams by using die above-described special interference phenomena. Outputs are positioned at locations where die interference images interact to produce energy components in accordance widi die interference states.

Because die use of special interference is completely new to die field, it necessitates die disclosure of a large number of components and component interconnections diat are used in new ways, requiring new organizational and interconnectional methods for accomplishing familiar tasks. As a result, this basic means and method is able to accomplish a variety of energy control and signal processing tasks, including active filtering, gated limiting amplifying, multiple bit binary information storage, energy beam oscillation, computer logic, and signal processing, as well as a multitude of otiier tasks and functions, which will become more apparent from a full reading of this specification. 1. Definition of some terms

Certain terms are herein defined in order to make die disclosure more clear. "Energy. " The special interference phenomena used by die present invention are able to be produced widi any type of energy diat exhibits a wave nature including, but not limited to, sound, moving particles, electrons, light, X-rays, microwaves, or other electromagnetic energy. While die present invention will operate using any wave-type energy, die description is presented herein in optical terms for clarity and consistency of die description. The apparatus used to implement die invention include any energy-directing elements) or optical element(s), including holograms, diat are compatible widi die type of energy in use.

"Beams and beam sets. " Because images as well as individual beams are able to be used in die present invention, die term "beam set" includes beams diat have been projected to produce complex images. In fact, a simple spot produced by a single beam is just a subset of die possible images that are able to be produced by a beam (or set of beams,) depending upon die optics used. As a result, a

"beam" or "beam set" that is on or off, or modulated widi analog information, is considered to include die production of both simple spots and complex images, wherein die entire image or spot is modulated in unison widi die same information. As a result, images are die same as beams and beam sets having multiple similar acting locations. The first beam set generally refers to a power or constant beam which is normally held on, while die second beam set generally refers to a control and/or modulation beam. Other beams and/or beam sets are as defined below in tiiis specification, claims and/or abstract.

"Interference. " The present invention uses multiple beam interference of wave-type energy. As a result, die terms "interference" and "special interference", unless otherwise stated, refer to multiple beam interference, and not to die projection of die type of images defined as "beam sets ^* above. A complex example of diat would be holographic images diat are actually produced by interference as individual images, but die images, in turn, "interfere" widi each otiier when multiple images are present. In tiiis case, the type of "interference" referred to would be between two or more holographic images.

"Diversion, divert, and diverted" are used herein to describe die phenomenon diat occurs when

destructive interference is present. Normally, energy from a beam set diat is on by itself contributes to die total amount of energy at a first locations), but does not contribute to die amoum of energy appearing at a second location(s) at that time. When an out-of-phase beam is superimposed upon die first one, destructive interference occurs, causing energy from bo beams to appear at die second location(s). As a result, the energy is said to be "diverted" away from die first location(s) and to die second location(s) by the process of interference.

Some scientists have a different viewpoint of how interference works; however, the concept of diversion is very useful for explaining the steps and components of die present invention, even if more complex processes are physically involved. The important point is diat energy from a certain beam appears at a location only when interference is occurring and does not appear there in die absence of die second interfering beam.

Input and output "levels" refer to energy levels rather tiian amplitudes per se, even though it may be obvious from die context that amplitude behavior is important; after all, it does form die basis for understanding interference effects. However, beams - even tiny beams - do have a cross-sectional area, which brings die concept of energy levels into die picture. Amplitudes and intensities may change depending upon die apparatus used. A given amount of energy is able to be focused onto a small area to increase the amplitude and die intensity, or be spread out onto a larger area, reducing die amplitude and intensity, while die total amount of energy remains die same. However, while adding energy to the modulated portion of an information arrying beam produces -unplification, reducing that energy is attenuation. The actual amplitude and area covered is able to be adjusted as needed by die proper selection of die optics used.

"Phase, inverted and uninverted. " The energy beams and beam sets function as carrier waves for die operation of die present invention. As a result, some confusion can arise over the use of die term "phase, " since it could refer to die phase of die carrier wave itself, or of any modulated waveform envelope impressed upon die carrier wave. In this disclosure, die term "phase" always refers to the carrier wave phase diat is able to determine if constructive or destructive interference will occur at a location when it is superimposed upon another carrier wave. Such terms as "inverted" and "uninverted" refer to the modulated waveform envelope, and not to the carrier wave phase. When an amplitude- modulated carrier wave is said to be "inverted," it is turned off during die time die "uninverted" carrier wave is turned on, as is the case with its analog equivalent, "ci" is constructive interference, "di" is destructive interference.

"controller" is die basic unit of die present invention as explained in sections 2 and 3 below. 2. The basic means and method using the first type of special interference. The present invention is a basic means and method of controlling a plurality of energy beam sets widi at least one set of the plurality of energy beam sets, using tiiis heretofore-unused special interference, comprising die steps of: a. producing a plurality of input beam sets --v -ng a first beam set having at least one first input beam of energy, and a second beam set having at least one second input beam of energy

modulated with controlling information, wherein die first and second beams sets are directed to at least one first location; b. producing interference widi diat plurality of input beam sets at die at least one first location, diverting energy proportional to die second beam set from die plurality of input beam sets to at least one second location; whereby energy from the plurality of input beam sets is absent from diat at least one second location when eitiier or both of die first and second beam sets are off and when die first beam set is in phase widi said second beam set,

(this is die first type of special interference where none of die input beams contributes energy to die second location^) in die absence of die others) and c. separating energy from the at least one second location to provide at least one output, thereby producing an energy beam controller of die first type.

In each case, all of die rays in a beam set operate in concert widi each other to produce images at die locations) where die various image components are separated, directing energy into die outputs. When at least two beam sets are on (and out of phase widi each otiier), die images interfere to produce a composite image having an energy distribution which is different from die images produced by die individual beam sets. Separation is tiien able to take place because die special locations of die distributed energy either conform to output positions (producing an "on" or "high" condition having a level proportional to the relative levels and phases of die input beam sets), or they do not, producing an "off" or "low" or reduced output condition).

"Separation," in eitiier case, is able to be accomplished by die strategic positioning of any element (optical or odierwise, depending upon the energy form used), diat permits energy from one location to exit into an output while blocking energy from another location from getting into die output. This separator is able to be as simple as a mask with a hole in it (as illustrated), as complex as a hologram diat directs energy from one set of locations in a different manner than it does when die energy comes from some other set of locations, or a device such as a strategically placed end of an optical fiber. The important point is that die energy is separated by die apparatus so diat it appears, or does not appear, in a certain output according to die teachings of die present invention.

The present invention is also able to use more than two beam sets as inputs. The complex combination of die group will determine die output. To produce an output, any two of die inputs must be on and out of phase widi each odier. Beams that are in phase will produce constructive rather tiian destructive interference at die first locations). The combinations of two beams of different phases are also able to produce a combined signal diat is completely out of phase widi a third beam set and so on. As a result, many complex means, methods, and devices are able to be produced from die present invention.

The above list of steps describes die use of die first type of special interference where neither of die inputs contributes energy to die output when only one beam set is on. The second list of steps, below, uses die second type of special interference, where at least one of die input beam sets does not contribute energy to die output when on by itself.

3. The basic means and method using of die second type of special interference.

A means and method of controlling one set of beams of energy widi another set of beams of energy, comprising die following steps: a. Producing a first beam set having at least one first input beam of energy directed toward at least one first location. Whether die "beam set" is only one little beam, or a whole group of rays diat form an image, tiiey are directed to where interference is going to be produced at certain times. The "at least one first location" is able to be a simple point, or an image made up of many locations that will function in unison. This beam set is die one diat does not show up at a second location in the absence of interference; b. Producing a second beam set having at least one second input beam of energy modulated with controlling information. This is die beam set that does contribute some of its energy to die second location when the first beam set is off, and is die primary difference between die second type of special interference and die first type explained above; c. Producing interference between die first and die second beam sets at diat at least one first location, when both die beam sets are on, diverting energy from bodi die beam sets to at least one second location in proportion to the second beam set, whereby energy from the first beam set is absent from die at least one second location when the second beam set is off or in phase widi die first beam set and is present at the at least one second location when the second beam set is on and out of phase widi die first beam set, (die interfere-re-controlling beam set is able to be amplitude- or phase-modulated. At die time when both beam sets have an equal level, die maximum amount of energy occurs at the second locations) because complete destructive interference occurs at die first locations). At input levels, and phases in between, die level at die second location(s) is proportional to the input beam sets, depending on die bandwiddi of the implementation, explained below.) and d. Separating energy from the at least one second location to provide at least one output, thereby producing an energy beam controller of die second type.

4. Foreview of more complex embodiments.

A suitable name for this basic amngement, using either type of special interference, is a "controller. " One could even refer to it as a type one controller or a type two controller. The common name is "photonic transistor;" however, die word controller is used herein, because die invention is able to use non-photonic energy.

This basic invention is able to be reproduced and interconnected with itself and other devices to produce a great variety of useful functions. To accomplish this interconnection, the precise locations, phases, timing, and relationships to various component types are required.

To provide an understanding of these requirements, a limited number of interconnected processes are herein disclosed diat use selected principles that are not obvious in die prior art. Once taught, these basic components and interconnecting methods are able to be rearranged and interconnected for producing an even greater variety of functions.

To begin widi, an explanation of some very basic processes diat explain die operation of the basic invention and die differences that occur depending on which type of special interference is being employed is in order. Such items as the logical AND, amplifier, gated amplifier, limiter, phase demodulator, phase comparer, active filter, cascading one into another, and die use of feedback are defined. Each of these items results from defining die input types to produce various kinds of outputs.

A further explanation of some of die more complex components diat use several of these principles to accomplish more complex tasks, such as time division multiplexing and demultiplexing, and frequency division demultiplexing, along widi logic devices such as a bistable flip flop and a multi- bit flip flop diat is able to store frequency multiplexed bits, will men be undertaken.

The principle of input beam summing and how it affects an amplifier with and without feedback makes it possible for die present invention to act as a threshold detector. Selection of input levels within certain ranges results in a variety of useful processes, including computer logic such as a set reset bistable flip flop, a multiple input AND, and a multiple input OR. Special timing and phase relationships are needed to clock a set/reset flip flop. The process of energy beam differentiation is a new one. It is needed in order to satisfy die tuning needs of a clocked flip flop so diat it is able to be used as a binary counter.

Using special interference opens the way to an entirely new field of invention. Some of the principles taught herein have no counterpart in electronics or any otiier prior art. Others have a counterpan, but must be used in unique ways. Consequendy, these many items have not been included just for d e fun of making a big application. Each of these many requirements for using the present invention needs to be taught. And each one has been selected so as to teach a particular aspect of using die basic invention.

5. The logical AND. The primary difference between die two types of special interference and Young's fringes is easily illustrated by die logical AND. Using die first type of special interference, a two-input AND gate is able to be made, because when eitiier of die two inputs is on by itself die output remains off.

(Note: "off is able to be anything between full "off" and a low state in comparison to an "on" or high state, because some implementations of die present invention may leak energy. So even if such implementations are not perfect, tiiey may still be useful.) Only when botii inputs are on does energy from both beams appear at the second locations), where die output of the AND is taken from.

Thus, in order to produce die logical AND using die present invention with die first type of special interference, the only additional step necessary is to modulate die input beams widi binary information, and die invention will accomplish die AND function. The second type of interference will not work to produce a logical AND, because energy from one of die beams does appear in die output in die absence of two-beam interference, which of course is contrary to die definition of a logical AND.

Young's interference, as used in the prior art, also fails to provide die logical AND function.

Young's interference provides energy to die output when eitiier or botii of die inputs are on. In the

prior art, that is described as an OR.

There are other ways to produce a logical AND; one of the most interesting is with the use of a threshold detector. This will be discussed below, as dwre are some fundamental concepts diat need to be taught first. The following sections explain bow the present invention responds to changes in input levels.

6. Basic amplifier and gated amplifier.

Amplification is able to be accomplished using either of the basic embodiments described in sections 2 and 3 above.

Even tiiough the first beam set (power beam) remains on as a substantially constant energy level, the output will be off in the absence of the second beam set (the control beam). Eitiier type of special interference is able to be used, because at least one of die inputs functions this way.

When die second beam set is modulated widi analog or binary information, the output will be an amplified version of the modulated input because energy from botii of the input beam sets is diverted into the output in proportion to die level of the second beam set and its phase relationship to the first beam set.

If the second beam set has a m- nmm level equal to diat of the first beam set and is exactly 180 degrees out of phase with die first beam set, and provided the optics are not sloppy, destructive interference will occur at the first locations). In tiiis case, all of die energy will be diverted into die areas of constructive interference at the second location-^) which lie(s) outside of the area where the first beam set appears in the absence of interference. As a result, the energy level of the combined output is able to be up to twice the value of the second beam set when die levels of both beam sets are equal. If the second beam set is smaller than the first beam set, the output will contain a contribution of energy from both beams in proportion to the level (and phase relationship) of the second beam set. The procedure used in making an amplifier is: a. Use the basic invention in utilizing either type of special interference, and b. Maintain die first beam set at a substantially constant above-zero level, diereby providing an amplifier by producing an amplitude modulated uninverted output having more energy than the second beam set has.

If die first type of special interference is used, die amplifier is able to be gated on and off by turning die first beam set on and off, thereby providing a gated amplifier by producing an amplitude modulated output having more energy than said second beam set has, and gating the amplitude modulated output off and on by switching die first beam set off and on.

The reason the second type of interference will amplify but is unable to be gated off by shutting off die first beam set is that a residual output from the second beam set appears in die absence of the first beam set. 7. Inverter.

Energy diverted into the output by interference produces an "uninverted" output. That is, when die control beam is on, so is the uninverted output. That diverted energy is removed from die power beam and die control beam so that their contribution to the first locations) is diminished.

Directing die remaining energy from the first location^) to a separate output produces an inverted waveform envelope. This inverted output operates differential to die uninverted output.

If the modulated input is modulated widi binary information, and die inverted output used, then die logic element produced is a logical NOT using eitiier type of special interference. When both input beam sets are modulated widi binary information, and die first type of special interference is used, die result is the EXCLUSIVE OR. These functions are die same as those of die prior art; however, here tiiey result from using special interference rather tiian Young's interference.

8. The limiter.

The present invention is also able to function as a limiting amplifier using eitiier type of special interference. When die first beam set is held at a -nibstantially constant level, it establishes a saturation level (depending on die exact optics used). When die second beam set is below that saturation level, die output will increase, as described above, producing an amplified output. However, as die level of die second beam set increases, more and more of die first beam set is diverted into die output. Eventually, a saturation level is reached wherein a further increase in die input level of the second beam set fails to divert more energy from the first beam set into die output because there is no more energy available in die first beam set to be diverted.

When all of die energy from die first beam set has been diverted into die output, amplification has reached its maximum value. Depending on die optical arrangement used, increasing die second beam set even more may or may not increase die output level. Amplification is curtailed, however, because die first beam set simply has no more power to contribute.

The procedure used in making the limiter is:

Modulate die second beam set to a sufficiently high level so as to divert all energy available from die first beam set into the amplitude modulated output, diereby providing a limiter by producing an amplified output when the second beam set is below saturation level, and limiting amplification of the amplitude modulated output when die second beam set exceeds die saturation level.

More information about limiters using die two different types of special interference is covered in item 38 below, "Basic theory of operation. "

9. Time division multiplexer. Among die many processes and devices that are able to be made widi gated amplifiers is a time division multiplexer. The procedure used in building a time division multiplexer is: a. Provide a plurality of gated amplifiers; b. Direct the outputs of the gated amplifiers to at least one third location; c. Direct energy from diat tiiird location to provide at least one common output; d. Modulate each of die second beam sets of die gated amplifiers widi information to be time division multiplexed, and e. Sequentially gate the amplifiers on by sequentially pulsing on the power beam sets of the individual amplifiers during each sequential time division, diereby providing a time division multiplexer by sequentially gating digital information from

each of the second beam sets of the plurality of gated amplifiers into a common output.

There are several ways of providing sequential pulsed beam sets. One simple way is to provide an initial pulse, direct a portion of die pulse beam into several delay paths, each having a different lengtii, then direct each delay path to a separate gated amplifier. 10. Time division demultiplexer.

A time division demultiplexer is very similar to die time division multiplexer, except that the inputs are common and die outputs are separate. The procedure used in building a time division demultiplexer is: a. Provide a plurality of gated amplifiers. b. Provide a multiplexed input beam modulated widi time division multiplexed information. c. Direct a portion of diat multiplexed input beam to die second beam set of each gated amplifier, and d. Sequentially gate each amplifier by sequentially pulsing on the first beam set of each successive amplifier during each successive sequential time division, thereby providing a time division demultiplexer by sequentially gating time division multiplexed information during each time division into a separate output of each of a plurality of gated amplifiers.

Time division multiplexing is an important way of directing a variety of information sources into a common transmission system. One example is the multiplexing of telephone calls into a fiberoptic cable. The present invention has a tremendous speed and bandwidth advantage over existing technology. Using the present invention for direct photon switching of fiberoptic signals will increase the capacity of fiber trunk lines.

Until the slow electronic interfaces have been replaced with faster photonic ones of the present invention, time division multiplexing utilizing the present invention will allow a large number of slower electronic systems to be multiplexed into a high bandwiddi optical link by modulating each of die separate input lines of the multiplexer with a separate electro-optical modulator, and each separate output of die demultiplexer with a photonic bistable arrangement and an electro-optical sensor.

Essentially, die multiplexer and demultiplexer are a parallel-to-serial converter and a serial-to- parallel converter. The following sections explain how die present invention responds to changes in phase and frequency. 11. Phase demodulator.

If die control input of die above-described amplifier is phase modulated, an amplitude modulated output will result diereby providing a phase demodulator wherein die anφUtude-modulated output is at a high level when the first and second beam sets are of opposite phase, at a low level when die first and second beam sets are in phase, and proportional to the phase of the second beam set in between those two phase extremes. (See die "Basic theory of operation" below, and the discusion of bandwidth.)

Depending on die optics used, the output signal provides a clean, amplitude-modulated signal

diat is free of any phase-modulated component because constructive interference occurs at the second locations), which is in phase widi die first beam set. However, if die output locations) are not located directiy at the position where complete constructive interference occurs, die ampUtude-modulated ouφut will contain some phase-modulated component. 12. Active filter.

The present invention is able to be used as a phase and frequency sensitive, precision active filter using die first type of special interference. If either of die input beams contains energy diat is not of die same frequency and opposite phase of die otiier input, no uninverted output will occur. As a result, die present invention is able to be used to demultiplex frequency multiplexed signals, distinguish colors, and demodulate frequency-modulated and phase-modulated signals.

If more than one color (wavelength) is supplied to both beam sets, a single device will operate independently and simultaneously at each wavelength. As a result, the present invention is able to be used to switch, separate, and organize broad-band signals.

By supplying multiple-wavelength energy of a substantially constant (above-zero) level(s) as the first beam set of a gated amplifier, along widi a multiple-wavelength second beam set, an amplified signal that matches each wavelength diat occurs simultaneously in both inputs will appear at die output. By switching die individual wavelengths of the power beam set on and off, die filtering process is able to be gated for selecting and demultiplexing die matching signals.

A plurality of diese active filters are able to be used eitiier in parallel, or in a tree structure to demultiplex frequency multiplexed signals of all kinds, including those used in optical fiber transmission, microwave and even radio.

The active filter uses the present, basic invention by adding these steps to die means and method: a. Providing die first beam set widi energy at a constant above-zero-level having at least one wavelength, and often several wavelengths; b. Switching wavelengths of the first beam set off and on to gate filtering of those individual wavelenguis off and on; c. Providing die second beam set widi energy at multiple wavelengths to be filtered, and d. Producing special interference widi a subset of the multiple wavelenguis matching die first beam set wavelenguis and rejecting all other wavelenguis, diereby providing a means and method of gated active filtering by producing an output only at wavelenguis diat exist simultaneously in both input beam sets. 13. Removing signals using an active filter.

It should be noted that either type of special interference is able to be used for filtering, but die relationships between die input signals and output signals in die second type of interference differ somewhat from those in die first type of interference.

Widi die second type of interference, die filtered, uninverted output will contain a contribution from die second beam set unless die second beam set is equal to and in phase widi die first beam set's wavelenguis, in which case constructive interference will occur at die first locations) at those

wavelengths removing that energy from die second locations) and the uninverted ouφut.

Adding an inverted ouφut, as with die inverter above, produces an ouφut which is differential to the uninverted ouφut at every wavelength present in the power beam, but is not differential at otiier wavelengths. The procedure for producing a differential active filter, using either type of interference, begins widi an amplifier with an inverted ouφut and continues widi the following steps: a. Providing the first beam set with its substantially constant above-zero level energy having at least one wavelengdi; b. Providing die second beam set widi multiple wavelengths to be filtered, and c. Producing interference widi a subset of those multiple wavelengths that match the at least one wavelength in die first beam set to divert energy of matching wavelengths away from die first locations) and into die second locations), diereby providing an inverted active filter by producing an inverted ouφut deficient of wavelengths which exist simultaneously in both input beam sets. This inverted ouφut is differential to the uninverted ouφut, just as widi die inverter above, only in tiiis instance, inputs having a variety of wavelengths are provided for die purpose of filtering, removing, and separating one wavelength from another while preserving any information present in the wavelength(s) being filtered. 14. Frequency demultiplexer.

Frequency multiplexing is easily performed by combining individually modulated signals of different frequencies into a common beam path. Demultiplexing is more complex. The procedure used in building a frequency demultiplexer is: a. Provide a plurality of active filers; b. Provide a frequency multiplexed beam set having a plurality of modulated wavelengths; c. Direct a portion of the frequency multiplexed beam set into die second (control) beam set of each filter, and d. Provide die first beam set of each filter with a different frequency of energy matching each of die plurality of modulated wavelengths, diereby providing a frequency demultiplexer by producing a separate modulated output, from each filter, matching each different frequency. The second type of special interference is not used, because frequencies not having a matching power beam will pass through into the output.

If die second beam sets of die active filters use die same input and first locations, step c. above happens simultaneously as die energy is being directed to that first location. Each frequency will produce a ci at a different location, where the separate outputs are taken. IS. Phase comparison.

When die first type of special interference is used widi two inputs of die same level, the ouφut level will be proportional to the phase difference between die two when multiple-wavelength sized locations are used, which produces an averaged, broadband device. If a narrow band device, having a small number of location sets, or even just one, is used, then die phase will have to be more precise,

and die output will not be proportional through the entire range of phases from 0 to 180. Precision engineering is able to narrow the phase bandwiddi to cover a smaller range. (See die discussion in "Basic theory of operation" below.)

In eidier case, die process used in building a phase comparer is: a. Provide die first and second beam sets widi energy having phases to be compared, and b. Use die special interference to produce an output diat is at a high level when die first and second beam sets are of opposite phase, and at a lower level at other phase differences, diereby providing a phase comparer. If both input phases are variable widi respect to diat used in odier components, additional stages will be needed to lock die combination phase produced from a single stage to diat of die odier stages.

16. Single stage and double stage bistable flip flops.

The principle of feedback has several important applications in die present invention. For purposes of building a flip flop, a portion of die uninverted ouφut is redirected, eidier directly or through an intermediate process, and fed back into die second beam set. Initially the ouφut is off, and so is the feedback signal.

When a set pulse also enters die second beam set, it is amplified in die ouφut. A portion of that ouφut is fed back into die second beam set, which is again amplified, making die feedback signal larger, etc. This is regenerative feedback.

The percentage of output which is fed back and die organization of the optics determines how the process operates. If the feedback signal is large, die amplifier will be driven quickly into saturation; if it is small, the amplifier will regenerate to produce a greater ouφut, but will not saturate.

Pulsing off eidier die first beam set or die feedback signal will reset the bistable arrangement. Either type of special interference is able to be used because die energy needed to maintain die set state comes from die first beam set. Because the control beam is off during reset, shutting off die power causes die output to shut off, turning off die feedback beam also.

In addition to its normal binary information storage function, tiiis type of flip flop is able to be made very sensitive by providing a large feedback signal. This sensitivity allows die flip flop to be set by even a single properly phased photon in die second beam set having a wavelengdi which matches the first beam set's wavelength. As a result, it is very color and level sensitive.

Moreover, by powering it widi multiple-wavelength energy, multiple bits are able to be frequency multiplexed using die same components. The procedure used in building such a flip flop is: a. Use eitiier type of controller; b. Maintain die first beam set at a substantially constant above-zero level widi energy of multiple wav c. Pulse the second beam set on widi energy having at least one wavelengdi which matches at least one of those multiple wavelengths; d. Direct a portion of die ouφut into die second beam set as a feedback signal to hold die output on at each wavelengdi pulsed on, and

d. Pulse either beam set off as a reset signal to terminate the output, and turn off the feedback signal to hold die ouφut off for each wavelength pulsed off, thereby providing a multi-bit bistable function by holding die ouφut on for each wavelengdi set by the set signal, and holding the ouφut off for each wavelengdi reset by the reset signal. One way of providing a flip flop diat is not so sensitive to changes is to use two inverters that drive each other die way static RAM cells are made in electronics. Such a flip flop was described in U.S. patent 5,093,802. However, diat invention uses Young's type interference.

Because the present invention also functions as a phase-sensitive active filter, multibit operation is also feasible, producing a clean, uninverted, frequency multiplexed ouφut. The procedure used in building such a phase-sensitive active filter is: a. Provide first and second inverters; b. Maintain die first beam set of botii inverters at a substantially constant above-zero level with energy of multiple wavelengths; c. Direct die inverted ouφut of die first inverter into die second beam set of the second inverter to hold die inverted ouφut of die second inverter off at each of the multiple wavelenguis present in die inverted output of die first inverter; d. Direct the inverted output of die second inverter into the second beam set of die first inverter to hold die inverted output of the first inverter off at each of die multiple wavelengths present in the inverted ouφut of the second inverter; e. Provide a set signal by pulsing on the second beam set of die first inverter widi energy having at least one wavelength which matches at least one of die multiple wavelengths, and f . Provide a reset signal by pulsing on the second beam set of the second inverter widi energy having at least one wavelengdi which matches at least one of tiiose multiple wavelengths, thereby providing a multi-bit set/reset bistable function by holding die uninverted output of the first inverter on for each wavelengdi set by the set signal, and holding the uninverted ouφut of die first inverter off for each said wavelengdi reset by the reset signal.

Because the two inverters balance against each other, the set and reset pulses have to be large enough to overcome die balance to drive die flip flop into die new state. Pulses diat are too small to cause a state change will not affect die state. As a result, tiiis arrangemem is less sensitive to incoming noise.

Another way to reduce die sensitivity is by using threshold detection. 17. Threshold detector.

Threshold detection is able to be accomplished using the present invention so that a schmitt trigger, a neuro circuit detector, a fuzzy logic element, an AND, an OR, and a less sensitive set/reset flip flop are able to be produced by simply supplying the needed number of inputs and adjusting die direshold level so that the process responds as needed.

The prior optical art has no counterpart using Young's interference, and so die method of interconnecting die various signals and die needed relationships between diem are herein disclosed. An analogy is able to be made widi die electronic world, but die architecture of the present invention

differs because of die requirements of beam phasing and setting die relative levels of multiple inputs, not to mention die critical timing needed to operate energy wave interactions.

The basic principle of multiple-beam summing provides die needed input signals for threshold detection. Beam summing is able to be accomplished by directing multiple beams into die first and second locations of an amplifier/limiter/phase filter in a controller having a substantially constant first beam set of at least one wavelength.

However, separating die summing locations) from die amplifying locations) allows multiple inputs to be summed prior to amplification, while providing a much clearer u---fersta-----ing of the basic principles involved. Therefore, a third locations) is used where a number of input signals are summed.

After die input signals are summed at die third location(s), they are separated and directed into die second beam set of die amplifier. As a phase filter, die basic amplifier, using eidier type of special interference, responds by producing an uninverted ouφut only when die modulated input has die proper phase. The uninverted output will then depend upon die total sum of die inputs relative to die first beam set.

The summing (third) locations) operate in a special way that makes direshold detection possible. There are two types of input beam sets to the summing locations). The first type are "trigger inputs" , and die second are "dueshold-controUing inputs. " These two types are 180 degrees out of phase widi each odier. In complex arrangements, some of these input beams have special assignments, and are given special names such as "set" or "reset" inputs, yet they provide energy diat is in phase widi one type or die odier.

Under die principle of superposition, the amplitudes of die superimposed beams add algebraically. The sum of all trigger beams balances die sum of all direshold-controlling beams. The total algebraic sum of die two sums has the interesting and useful quality of being in phase widi the threshold-controlling beams whenever die sum of all d-reshold-controlling beams is greater tiian die sum of all trigger beams. That total is out of phase widi die t---reshold-controlling beams whenever die sum of all trigger beams is the greater. If the two sums are equal, die total sum is zero.

If any of die input beams were not zero or 180 degrees, a combination phase would result (widi broad band optics). However, because die inputs are of one phase or die odier, die sum will be only one phase or die odier (or off if they balance).

If at least one of die threshold-controlling beams is held at a substantially constant level, and die level of die sum of die trigger beam(s) is increased from zero, the amplitude of die total sum will decrease, but its phase will remain die same as diat of die threshold-controlling beam.

Energy from die summing locations) is separated and directed into die control input of die amplifier, having a phase upon arriving diat holds die amplifier at cutoff (uninverted ouφut off). Because any sum diat produces this same phase produces constructive interference at die first locations), die input level to die amplifier has no effect. The amplifier remains in cutoff regardless of fluctuations in die input levels.

When die trigger input sum equals the threshold-co-urolling sum, die total sum is zero; as a

result, the output of the amplifier remains cut off.

As die trigger sum rises above die level of die direshold-controlling sum, the total amplitude rises. However, its phase has switched 180 degrees, and is now in phase with the trigger sum. If die trigger sum raises rapidly, the phase of the total sum will not pass through all of the phases from zero to 180; rather, the phase jumps from zero to 180. This sudden phase change principle is used for threshold detection by detecting this phase change. As soon as the control input to the amplifier sees die new phase, destructive interference takes place at die first locations), and energy appears in the uninverted ouφut.

As long as die direshold-controlling input is held constant, it will establish die level at which this phase cross-over takes place. Without this balancing input, the amplifier will be sensitive to the slightest input; its direshold level is zero. However, adding die summing locations) ahead of and outside of die amplifier allows a direshold to be established at some point above zero.

By changing die threshold-controlling sum, eitiier by changing the level of the main direshold- controlling beam or by adding beams of eidier phase, from various sources, widi various timings, a considerable variety of functions are able to be accomplished from this basic threshold detector.

Starting with die basic amplifier or limiter, the procedure for tiiis basic means and method of direshold detection is: a. Provide at least one trigger beam set having at least one beam of energy directed to at least one third (summing) location, modulated widi information to be threshold-detected; (When tiiis arrangemem is used as a schmitt trigger, tiiis trigger input is analog-modulated with the information diat is to be digitized by turning on the ouφut whenever this input is above the threshold. When it is used as a neuro detector, fuzzy logic detector, AND or OR, this arrangement has a multitude of trigger inputs diat are all in phase widi each odier.) b. Provide a direshold-controlling beam set having at least one beam of energy at a substantially constant level directed to die tiiird (summing) locations);

(This input determines the direshold level that die arrangemem will respond to. It is out of phase with the trigger beams.)

c. Produce destructive interfeience with die trigger beam set and the direshold-controlling beam set at the uϋid location(s) when die trigger and direshold-controlling beam sets are botii on, so that the combined phase of energy at the third locations) is in phase with the threshold- controlling beam set when die trigger beam set is smaller than die direshold-controlling beam set, and is out of phase with the threshold-controlling beam set when die trigger beam set is greater than the direshold-controlling beam set, (This "diird" summing location(s) is used to sum die various inputs. When die inputs have only one of two opposing phases, separating die sum of all inputs to this location(s) will produce a signal diat is of eidier one phase or the otiier, but not in between;

(The threshold-controlling input is kept constant during the time when threshold detection is to take place. If die trigger input(s) are off, the ouφut from die third locations) is of a phase that does

not pass through die limiter.) d. Direct energy from die third locations) into die second (control) beam set of die phase demodulator, diereby providing a direshold detector by demodulating phase changes in the combined energy at die tiiird locations).

The phase demodulator of die present invention produces an ouφut only when its two input beam sets are of opposite phase. When they are in phase, constructive interference rather than die destructive interference needed to divert energy into die uninverted ouφut occurs at the "first location^)." At this point, die direshold has been detected, because die amplifier will produce an output only when die sum of the trigger input(s) is greater than die direshold. When die trigger input fells below die direshold, die anφlifier goes again into cutoff. Thus, if die particular use for threshold detection requires that die "turning on" direshold be die same as die "turning off" direshold, along with a binary output, then die power beam of the amplifier is adjusted so that saturation occurs at low control beam levels. The uninverted ouφut is then able to be cascaded into another amplifier, if more power is needed.

In many binary circuits, tiiere is a range of levels available so diat die turn-on and turn-off diresholds need not be die same. By rapidly changing die direshold, die amplifier is able to be driven quickly into eidier saturation or cutoff upon reaching die direshold. Two interrelated principles, feedback and schmitt trigger snap action, are explained above in the discussion of flip flops. Next a feedback signal is added to die threshold detector.

The difference here is diat die feedback signal, which is taken from a portion of die uninverted ouφut, is directed into die summing locations) as an additional trigger input out of phase widi the direshold-controlling input rather than being fed directiy into die control input as before. When die direshold is reached by a slowly rising trigger input, die uninverted ouφut comes on, directing energy along die feedback patii. If diat path is short, the time delay will be negligible in comparison to die rise time of die trigger input. If die delay time is not negligible, then it must be accounted for in die design of die rest of the components in a complex system. For this discussion, however, we will assume that the feedback delay is negligible. Upon reaching die summing locations), die feedback signal immediately lowers die direshold level. If die trigger plus die feedback input is sufficient, die amplifier will go immediately into saturation. This is a "snap" toggle-like action. Once die turn-on threshold has been reached by die trigger input, die amplifier snaps directiy into saturation. As long as die amplifier remains saturated, fluctuations in die trigger input will have littie or no effect on die ouφut because of die limiting action. What happens when die trigger input goes down depends upon die level of die feedback signal relative to the odier summed inputs. If die feedback signal is smaller tiian die direshold-controlling beam, botii the trigger and die feedback beams will be required in order to overcome die uireshold- controlling beam.

As die trigger input falls, the amplifier comes out of saturation and die ouφut begins to drop.

As the ouφut drops, so does die feedback signal, raising die threshold. The rapidly changing, regenerative driven fall of the feedback signal causes the ouφut to drop just as rapidly as it rose before. Because die regenerative nature of the snap action is quite rapid in comparison to the rise and fall times of die trigger pulse, it is just as easy to snap die amplifier on as it is to snap it off. As a result, the turn-on direshold will be very nearly die same as the turn-off threshold. When the trigger input is above die threshold, the ouφut snaps on, and when it drops below die direshold, die ouφut snaps off. It is a schmitt trigger.

The additional step needed to produce die schmitt trigger action is: e. Direct a portion of die amplitude modulated ouφut, as a feedback signal, smaller than the threshold-controlling beam set, into die at least one tiiird location, out of phase with die direshold-controlling beam set, diereby providing a schmitt trigger by driving die phase demodulator to a greater ouφut than occurs when the regenerative feedback caused by die feedback signal is absent. 18. Bistable flip flop. Changing die maximum level of the feedback signal changes die way die direshold detector operates after it has first been turned on.

If die feedback signal is larger than the threshold-controlling input and large enough to maintain saturation, reducing die trigger input has little or no effect on die ouφut. The amplifier has toggled on, and will remain on. It becomes bistable. The trigger input becomes a "set" input, and there are at least two ways to reset it.

The power beam is able to be pulsed off, but diat would require an additional component. A better way is to provide anodier threshold-controlling input which is pulsed to reset the arrangemem. When the reset pulse is combined widi the feedback signal and die main threshold-∞ntrolling input, the direshold is raised above the level of die feedback signal. The phase of die total sum reverts to die phase of the direshold-controlling input, and die amplifier goes into cutoff.

However, there is also a snap action that occurs here. As the reset pulse rises, the total sum falls. At the saturation point, the amplifier comes out of saturation, and tite output (and as a result, die feedback signal) also begins to fall. This causes a regenerative amplification of the fell time. That is, the loss of feedback signal, in turn, causes die output to drop some more, which causes die feedback signal to fall, and on and on. The combination of rising reset pulse and falling feedback causes die amplifier to snap quickly into cutoff.

The reset pulse must be at least large enough to bring die amplifier out of saturation. As a result, tiiere is a "turn-off" threshold, below which die reset pulse will be unable to bring the amplifier out of saturation to toggle it off.

The bistable arrangemem described above produces a set/reset flip flop that has an advantage over die previously discussed flip flops, in diat it is turned on and off by pulses to separate inputs, whereas die previous flip flops required diat one of die beams be shut off.

Starting with the threshold detector above, the procedure for producing a complete set/reset

bistable flip flop is: a. Provide at least one reset beam set, directed to die at least one tiiird (summing) location in die threshold detector, in phase widi die direshold-controlling beam set; b. Direct a feedback signal from a portion of die amputude-modulated ou ut into die at least one third locations), the feedback signal at die at least one tiiird locations) being out of phase widi and greater tiian the direshold-conrro-ling beam set (at the summing location); c. Pulse the trigger beam as a set pulse, and d. Pulse die reset beam set as a reset pulse, diereby providing a bistable function by turning the ainpUtude-modulated output on with the set pulse, holding die ampule-modulated ouφut on widi die feedback signal, then turning die amplitude-modulated ouφut off widi die reset pulse, and holding die ampUtude-modulated ouφut off due to die absence of die feedback signal.

Multi-bit operation also works with this bistable arrangement by using a power beam having multiple wavelengths, by using broad band optics, and then setting and resetting die arrangement widi die individual wavelengths.

19. Neuro detector, fuzzy detector, and logical AND.

Providing a multitude of trigger inputs to die direshold detector produces a total sum diat is at its maximum when all of die trigger inputs are on. By setting die threshold level just below diat sum, and above die level where all but one of die trigger inputs are on, threshold detection will occur only when all of die trigger inputs are on at die same time, producing a saturated ouφut from the limiter. If this arrangemem is connected in a neuro circuit, tiien it will function as a neuro direshold detector, widi die direshold-controlling input being die "weight" signal. Likewise widi fuzzy logic. Analog inputs of eidier phase are able to be summed to provide a "crisp" ouφut from a multitude of "fuzzy" information sources. If this arrangement is connected in a binary digital circuit, tiien it functions as a multi-input logical AND. The process used in producing a multi-input logical AND is:

Provide a plurality of trigger beam sets as AND inputs modulated widi binary information, die AND inputs having an energy sum, die energy sum of die AND inputs being greater than the substantially constant level of the threshold-controlling beam set when all die AND inputs are on and less than the substantially constant level of die direshold-controlling beam set when one of die AND inputs is off and die remainder of die AND inputs are on, diereby providing a multi-input AND by producing an on output only when all die AND inputs are on.

This process will work widi or without the feedback circuit of die schmitt trigger. The schmitt trigger action is especially useful when the input is analog, as is die case widi die neuro logic applications.

20. Multi-input logical OR.

If die direshold-ccmtrolling signal is smaller than die smallest of die trigger inputs in die

threshold detection arrangement, tiien the limiter will turn on whenever any one of the trigger inputs is on. As a result, the function is a n-nlti-i-αput logical OR. The steps used in producing die multi-input logical OR are:

Provide a plurality of die trigger beam sets as OR inputs niødulated widi binary information, 5 said OR inputs having an energy sum, the energy sum of the OR inputs being greater than the substantially constant level of the direshold-controlling beam set when only one of die OR inputs is on, diereby providing a multi-input OR by producing an on output when at least one of the OR inputs is oa 10 The multi-input logical OR is able to be used botii widi and without feedback. If no feedback is used, die sum signal which is larger than die threshold will be amplified. 21. Multiple input bistable a-rrangements.

Interesting things happen when the feedback signal is large enough to make the arrangement bistable. Setting the threshold low is die same as placing a multi-input OR on multiple set and reset 15 signals. Setting the threshold high is the same as placing a multi-input AND on multiple input set and reset signals.

Setting the threshold in between causes the threshold to be reached when two out of three, or three out of five, or some similar proportion of inputs has been reached, a function diat is very useful in fuzzy logic environments. 20 22. One shot

The one-shot function is able to be provided using any of die bistable arrangements by adding a delayed feedback beam, die easiest being those diat have direct set and reset inputs such as die set/reset bistable device above that uses the direshold detector. The procedure used in providing die one-shot function is to: 25 a. Direct a portion of die anφfitude-modulated output along a delay path to provide a delay period and a delayed-feedback beam set, and b. Direct die delayed-feedback beam set to provide die reset pulse, thereby providing a one-shot function wherein the amphtude-i∞dulated ouφut is pulsed on by die set pulse, held on during the delay period by the feedback signal (already a part of die set/reset flip 30 flop), and tiien turned off after the delay period by the delayed-feedback beam set.

The delayed-feedback signal is in addition to any feedback diat is used to make die original arrangemem bistable. Essentially, the one-shot trigger turns die bistable arrangement on, and die delayed-feedback signal turns it off. The length of die delay also affects die duty cycle, because die arrangemem is unable to be set again until all energy has exited die delay path. In order to allow a 5 second one shot to begin immediately upon die ending of another, the delayed-feedback signal is able to be differentiated and made into a shorter pulse. Differentiation is discussed below in greater detail. 23. Cascaded amplifiers

To accomplish many of die tasks set out above, many embodiments will require more energy for die many outputs. As a result, a number of amplifiers will be needed in place of the single

amplifiers described. Cascaded amplifiers have die qualities of die individual amplifiers plus some important new features.

Cascading one anφlifier into another increases die amoum of energy in the modulated envelope at each stage. When all of die stages are on, the last ouφut in die cascade series has the accumulated energy of die entire series in the modulated envelope. Since die ouφut of each stage provides more energy to die next stage, each stage is able to be powered by larger and larger beams. The control input to die first amplifier in the series then becomes the controlling input to die whole cascade. As a result, a low level input to die cascade will control die much larger (uninverted) ouφut of die series, thus making the small beam the controller of a larger beam. The amoum of energy in die ouφut of the series has a maTiimnri level of

E ₀ - E. 2" where n — the number of stages in die series, and E, is die level of die small input beam. However, E ₀ is also limited by die input level to die individual amplifiers in die series. As a result, the maximum level of the ouφut will be limited by die level of die energy-supplying beams. When die cascaded amplifier reaches tiiis limit, it will go into saturation. The addition of more energy imo the small input will not cause any more energy from the energy-supplying beams to be diverted to die output because all diat is able to be diverted into die output has been.

The procedure used in providing tiiis means and method of cascading amplifiers of die present invention is: a. Provide a plurality of amplifiers connected in a cascade series; b. Cascade die ouφut of each amplifier into die second (control) beam set of the next amplifier in die series, and c. Provide at least one large beam, greater than the second (control) beam set of die first amplifier in die series, wherein die large beam is die first (power) beam set of another amplifier in die series, thereby providing control of die large beam by a smaller beam by diverting energy from die large beam through the cascade series and into die amplitude modulated ouφut of die last amplifier of die cascade series in response to die smaller beam.

Since die level of die modulated signal is substantially doubled at each stage, a considerable amount of amplification is able to be accomplished. Such amplifiers have die advantage of not introducing noise into die signal as long as there is no noise in die power beams and no vibrations in die optics. 24. Gated amplifiers used to produce a multi-input AND.

Cascaded amplifiers are able to be used in all of die energy beam circuits described in die present invention. If die first type of special interference is used, die amplifiers are able to be gated. In fact, turning off any one of die power beams turns off the entire series. As a result, cascading is an excellent way of providing a multi-input AND function. The steps are:

1. Provide a plurality of logical ANDs in a cascade series, and

2. Cascade die output of each logical AND imo the second beam set of die next logical AND in

die cascade series, diereby providing a multi-input AND wherein all of the first beam sets of the plurality of logical ANDs and the second beam set of the first logical AND in the cascade series must be on in order to turn on the at least one ouφut of the last logical AND in the cascade series. If die power beams are modulated with analog information, die cascaded series of amplifiers are able to be used as an amplifying mixer. 25. An Oscillator.

The four main items needed to sustain oscillation are: an amplifier, a power source, a frequency determining device, and a feedback path. The present invention provides die amplification. The constam input beam set provides die power source. By directing die output imo a delay path, an ouφut pulse from the amplifier will be delayed for a certain period as it travels along the delay path. This provides die frequency determining device. By directing the ouφut of die delay path back imo the input of die amplifier a feedback patii is established.

One or two other things, though, are also needed. Because the amplified ouφut of a simple amplifier is not inverted, an initiation pulse must be provided. Once oscillation is started, die above arrangemem is able to be designed so as to continue oscillating, or it is able to be designed so diat die oscillations die out, as in a ring oscillator.

To make a self-starting oscillator, an inverter must be added to die feedback circuit. Together the feedback path tiien provides an inverted, delayed signal to drive die amplifier. Initially the amplifier is off. The inverter provides a signal to turn the amplifier on. A portion of the output is delayed and inverted before it reaches the amplifier's input. As a result, the amplifier turns off, waits for the delay period, turns on, waits for the delay period, and turns off again in response to die delayed, inverted signal.

Because die feedback signal is the sole source of energy for the control input, and the arrangemem is self-starting, shutting die power beam off and on gates the oscillator off and on when using eidier type of special interference.

As a result, a means and method of energy beam oscillation begins with the basic amplifier and is produced by die following procedure: a. Provide an inverted delayed signal to the amplifier by directing a first portion of energy from the amphtuάe-modulated output along a delay path and tiirough an inverter means. The inverter means is able to be the inverter described above, or the Young's interference inverter described in die prior art; b. Direct die inverted delayed signal to the second beam set to turn the amplitude-modulated ouφut on and off, widi die second beam set remaining on during die period when die inverted delayed signal is on, and off when die inverted delayed signal is off to produce oscillation, and c. Switch die (power) first beam set off during the time when oscillation is to be gated off, diereby producing energy beam oscillation by successively turning the uninverted ouφut on or off during each delay period, and gating off energy beam oscillation by turning off the first (power) beam set.

Energy beam oscillation is able to be accomplished using any of die amplifying components, including die bistable and direshold processes. Components are connected in circuits just as their electronic counterparts are; however, each interconnection must have the correct phase and timing relationships in order to work. 26. Phase locking and demultiplexing.

A phase-fluctuating beam, such as is die case when beams from two differem lasers or odier sources need to be phase matched, is directed in parallel into a group of phase filters/demodulators of die present invention. The quality of die optics used will have a great bearing on die actual bandwiddi passed by each phase filter/demodulator. Sufficient filter/demodulators are provided in order to cover die entire spectrum used by die phase-fluctuating input. Each demodulator is provided widi a differem, but constam, phase signal. The outputs will be a demultiplexed set of signals diat produce their individual ouφuts as die phase-fluctuating signal fluctuates into die bandpass region of each individual filter/demodulator.

The phase filter/demodulators produce a maximum ouφut only during die time when die phase- fluctuating signal and die individual power beams are 180 degrees out of phase. As a result, die ouφuts have a substantially constam phase output during die time diat they are at maximum. Phase adjusting each of die filter/demodulator ouφuts and tiien recombining diem into a single beam again results in an output beam diat has a substantially constam phase, relative to the power source of the controllers, having energy from botii the power and phase-fluctuating beams. As a result, the two sources have been phase-locked togedwr.

If die phase-fluctuating beam is at a substantially constam level, tiien die phase-locked output will also have a substantially constam level, even though die individual controllers will be on individually in accordance widi die phase at any given moment. Because at least one will be on at any given instant, however, die ouφut will remain on, and if die input is amplitude-modulated, die ou ut will likewise be amplitude-modulated.

If narrow band controllers are used, die first locations) are able to be used for all of die controllers. The ouφut locations are able to be close enough together to form a continuous band. An optical phase changing element is able to be positioned so diat energy at each point along die band will receive a differem phase change. Then, directing all of the phase changed outputs to die phase-locked ouφut location brings diem to one location with the same phase.

The procedure used in producing the means and method of phase locking is: a. Use die phase demodulators of the present invention; b. Provide a phase-fluctuating beam set; c. Direct a portion of the phase-fluctuating beam set into each second (control) beam set of each of a plurality of the demodulators; d. Provide die first (power) beam set of each demodulator with a differem phase of energy; e. Direct the amplitude modulated ouφuts of each demodulator to at least one tiiird location, such diat energy from all demodulator ouφuts arrives at that tiiird location mutually in phase, and f. Direct energy from that tiiird location to provide a phase-locked ouφut,

thereby phase-locking energy from the phase-fluctuating beam set to a substantially constam phase ouφut having a substantially constant amplitude.

An important use of the phase-locked ouφut is phase-locking one wave train widi another wave train in order to produce a substantially constant phase ouφut from multiple wave trains so as not to lose information being stored and processed in an optical computer because of pluse slήfring in the power source.

The principle of operation is to provide enough phase demodulators so the sum of their band widths covers the range of fluctuation. The fluctuating phase signal is viewed as a phase multiplexed signal wherein only one phase is present at any one given moment. Because the phase demodulators produce constructive interference at tiieir ouφut locations), the phase of that signal will be of a known phase, namely die phase of die power beam. When die control input is not of die correct phase, the output will be off (within the bandwidth range determined by the precision of the optics).

The output from the demodulator having the matching (and known) phase is directed imo the phase-locked ouφut after the phase is first adjusted to match the phase of all the other demodulator ouφuts. The demodulators take turns firing, depending on the phase of the fluctuating phase input. However, die combined output has a substantially stable phase made up of energy from both beams.

The stability of the output phase depends on die bandwiddi of die individual demodulators, and die number of them being used, for diat determines die resolution of die process.

Why would anyone warn to combine two cw signals to produce a constam phase signal? One application is to lock wave trains together. Even die best lasers produce finite wave trains. That means diat abrupt phase changes sometimes occur in the laser ouφuts, which amounts to a form of phase modulation.

When the tail end of one wave train is locked to die start of die next wave train, die phase- locked ouφut is able to m--im---n a substantially stable phase, or at least one that changes very slowly. Substantially constam phase is necessary for the operation of high-speed optical computers.

These processes are able to be incorporated right into the feedback path of the laser itself so that it will ouφut wave trains that are phase-locked together.

Multiple sources, even some non-laser sources, are able to be phase-locked together using tiiis means. The phase locking and filtering qualities of die present invention allow a seed signal, such as a low-power laser, or even a high-quality color filter, to be used as a locking standard for extracting and phase-locking energy from a thermal source, sunlight, or white light. Certainly, the present invention will be very useful in spectroscopy. Information extraction:

If the phase-fluctuating beam has been amplitude-modulated, die phase-locked output will also be anφUtude-modulated by the phase-fluctuating beam. As a result, information is able to be extracted from one source, such as light from an optical fiber, and prepared for processing in a phase-locked system such an optical computer.

When diat beam is frequency-modulated, all of the phase demodulators turn off together when the wavelength changes so that it is out of the combined bandwiddi of die group of demodulators. As a

result, information transmitted using one energy source is able to be transferred over to devices driven from another energy source by using eidier frequency or amplitude modulation.

Phase locking is essential for producing high-speed communications links between optical computers, including fiberoptic telephone transmission systems. The additional step needed to extract modulated information from a phase fluctuating signal is: g. Amplitude- or frequency-modulate die phase-fluctuating beam widi information to be extracted, diereby providing an anφtitude-modulated ouφut containing diat information and having a substantially constam phase at die phase-locked ouφut.

27. Phase encoder/modulator. Phase encoding and phase modulating from an amplitude-modulated signal are able to be accomplished by producing a differential anφUtude-modulated signal pair and recomputing the two signals having opposite phases. When one of the differential pairs is on, the combined output has one phase; when the otiier one is on, the combined signal has die opposite phase. In between it has a combined phase. The procedure for tiiis means and meuiod of phase modulating or encoding is: a. Use eitiier type of controller; b. Maintain the first (power) beam set at a substantially constam above-zero level; c. Amplitude-modulate die second (control) beam set; d. Direct a portion of the at least one output, having a first phase, to at least one tiiird location; e. Separate a portion of energy from the first location to produce an inverted signal having a phase opposite the first phase; f . Direct die inverted signal to diat dύrd location, and g. Direct energy from diat third location to produce a phase-modulated output, thereby producing a phase modulator by causing energy of die first phase to output when the second inpm beam set is on, causing energy of die second phase to output when die second input beam set is off, and causing the phase of energy at the phase-modulated ouφm to change proportional to the amplitude of die second (control) beam set.

28. NAND.

Boolean algebra provides for die production of various logic gates by combi-αing logic functions. When combining functions that use special interference, certain tilings (such as the carrier wave phase) need attention, as they have no counterpart in electronics. As a result, tiiis disclosure also contains some examples, so diat process interconnection is substantially and clearly taught.

In Boolean logic, an inverted AND output produces a NAND. This is die procedure: a. Use any of the previously described AND functions, and b. Direct die at least one output of the logical AND into a logical NOT means, thereby providing a logical NAND.

The NOT is able to be eidier of die present invention or of the prior art.

29. NOR.

The logical NOR is produced by inverting the two inputs to an AND. The procedure used in producing a logical NOR is:

a. Use the AND of the presem invention; b. Provide a first logical NOT means having an energy beam ouφut directed into die first beam set of the logical AND, and c. Provide a second logical NOT means having an energy beam output directed imo the second 5 beam set of die logical AND, diereby producing a logical NOR.

As common as these iiuerconnections are in the electronic world, they have not been done before using energy beam ANDs that use special interference. Again, the energy beams must interface while meeting all of the phasing and directing requirements for the individual parts disclosed herein. 10 30. EXCLUSIVE OR.

The Exclusive OR function of the prior art produces an ouφut having a phase-modulated component. When one input beam is on by itself, the ouφut is of one phase. When the other beam is on by itself, the ouφut is of the other phase. The presem invention corrects that problem by separating the phase-modulated ouφut, correcting the phase and recombining diem back together. The process 15 used in building an Exclusive OR is: a. Use die phase demodulators of die presem invention; b. Produce a first Exclusive OR input beam set having at least one beam of energy modulated with binary information directed toward at least one tiiird location; c. Produce a second Exclusive OR input beam set having at least one beam of energy modulated 0 widi binary information directed toward that third location; d. Produce destructive interference at that third location when die first and second Exclusive OR input beam sets are on; e. Direct energy from that third location into first and second said phase demodulators, the first beam sets of the first and second phase demodulators being of opposite phase so diat the 5 amplitude-modulated ouφut of the first phase demodulator is on when die first Exclusive OR input beam set is on by itself, and the second phase demodulator is on when the second Exclusive OR input beam set is on by itself; f. Direct energy from said first phase demodulator to provide at least one Exclusive OR output, and 0 g. Direct energy from the second phase demodulator to the Exclusive OR ouφut while providing a 180-degree phase shift so that the energy from the first and second phase demodulators have matching phases at the Exclusive OR ou ut, diereby providing an Exclusive OR having a substantially constam phase ouφut. 31. A Binary half adder. 5 The Binary half adder function is produced by providing an AND output as a carry signal, and an Exclusive OR as a sum signal, from a common input. The procedure used in making a binary half adder is: a. Use die Exclusive OR of the presem invention; b. Direct a portion of energy from the first Exclusive OR input beam set to a first input of a

logical AND means, and c. Direct a portion of energy from the second Exclusive OR input beam set to a second input of die logical AND means, thereby providing a binary half adder by providing die Exclusive OR ouφut as a sum ouφut and an ouφut of die logical AND as a carry output. 32. A clock signal for the bistable flip flop.

Adding a clock signal to a bistable flip flop of die presem invention involves more than what is needed in electronics. The high-speed nature of the carrier waves used in die presem invention will cause false signals to appear at times, and in places where tiiey would not in slower media, such as conventional electronics.

A bistable flip flop of the present invention is used that has separate set and reset inputs. An amplitude-modulated clock pulse stream must be directed first to die set input and tiien to die reset input. If die clock signal is directed to botii inputs simultaneously, unpredictable results will occur. As a result, two AND gates are used to direct die clock signals to die appropriate inputs by enabling the gates individually for a short time slightly longer than the length of the clock pulses. This insures good solid set and reset pulses. The procedure is: a. Use eitiier of die set/reset binary flip flops of the presem invention that have an on-pulse reset input as well as an on-pulse set input; b. Provide first and second logical AND means; c. Provide a clock beam set having at least one pulsed energy beam of alternating first and second pulses; d. Direct a first portion of the clock beam set imo a first input of the first logical AND means; e. Direct an ouφut of the first logical AND means to provide die set pulse; f . Direct a second portion of the clock beam set into a first input of the second logical AND means; g. Direct an output of the second logical AND means to provide die reset pulse; h. Direct a portion of the amphtiide-modulated ouφut of the bistable function along a first delay padi, providing a delay time, and a delayed bistable beam set; i. Direct a first portion of the delayed bistable beam set imo a logical NOT means; j. Direct an output of the logical NOT means to a second input of the first logical AND means, and k. Direct a second portion of die delayed bistable beam set into a second input of die second logical AND means, diereby providing a clocked bistable function by setting the bistable function using the first pulses that pass through the first logical AND means when the second input of the first logical AND means is held on by the logical NOT means in the absence of die delayed bistable beam set and die first pulses are prevented from passing through the second logical AND means by die absence of the delayed bistable beam set, tiien resetting the bistable function by using the second pulses diat pass through the second logical AND means when the second input of die second logical AND means is held on by die

delayed bistable beam set and the second pulses are prevented from passing through the first logical AND means by the presence of the delayed bistable beam set, which is inverted by the logical NOT means to hold the first logical AND means off.

Delaying the bistable beam before it is able to turn on the NOT, and turn off the first AND, allows the first clock pulse to pass through to set the flip flop solidly on before that opportunity expires after the delay period. Delaying die bistable beam before it is able to turn off the second AND allows die second clock pulse to solidly reset the flip flop before the time period expires. In each case, the clock pulse must be shorter than the delay period; otherwise, unpredictable things happen. 33. Binary digit counter, energy beam differentiation, and integration. Because the clock pulses of the clocked bistable arrangemem set forth above must be shorter than the delay period, additional carrier wave circuitry must be added so that one bistable device is able to be directed into another bistable device. To accomplish that task, a clock signal having a long pulse width must be differentiated.

Differentiating a carrier wave pulse is a little differem from differentiating an electronic pulse. Differentiating a positive electronic pulse results in a short positive pulse, followed by a delay period equal to the length of the input pulse minus the positive pulse, followed by a negative pulse. The analogous differentiation of a long carrier wave pulse results in a short carrier wave pulse, followed by a long off period diat equals the length of the input pulse minus the short pulse, followed by a short pulse having phase opposite that of the first short pulse. Energy beam pulse differentiation is accomplished by dividing die pulsed beam into two portions. One portion is directed along a delay path having a delay equal to the differentiated pulse width. The delayed portion is then recombined with an undelayed portion at a locations) for performing the differentiation. An output is provided from that differentiation location(s). During that short delay period, energy is ouφut as the leading short pulse. When the delayed energy arrives from the delay path, destructive interference is produced at die differentiation locations) which shuts off the output. As long as the input continues, the differentiated ouφut remains off.

When the input shuts off, destructive interference stops. However, energy is still coming from the delay path. This energy is ouφut as the trailing short pulse. However, this pulse is 180 degrees out of phase widi the leading pulse. The result is a differentiated pulse pair having opposite phases, which is produced from a single input pulse. The advantage of such an arrangemem is that the off period between die two pulses is able to be any length without affecting down stream components, while the delay period that establishes the differentiated pulse widths is able to be standardized to match the requirements of other components. An excellent example is in the production of binary counters. In order to produce a counting stage, die bistable device must be clocked on or off on every other binary pulse ouφut from a bistable component that is able to have pulse widths of various lengths. Therefore, die input clock is first differentiated so that standard-lengdi short pulses will be presented to die following binary stage. Because the energy beam AND means of the presem invention respond only to inpm signals of the proper phase, either of the differentiated pulses are able to be used

by simply adjusting the phase of die differentiated signal to match that needed by the ANDs. The ANDs will respond to only one of die pulses.

The steps used in producing this means and method of producing a binary counting stage are: a. Providing a binary input beam set having pulses longer than the delay time; b. Directing a portion of the binary input beam set to at least one fourth location; c. Directing another portion of the binary inpm beam set along a second delay path and then to the at least one fourth location as a delayed beam set; d. Producing destructive interference using the binary input beam set and die delayed beam set at the at least one fourth location when both the beam sets are on; e. Separating energy from the at least one fourth location to provide differentiated pulses, and f . Directing the differentiated pulses to provide the clock beam set of the clocked bistable function, thereby providing a binary digit counter which allows use of binary inpm beam pulses longer than the delay time to clock die clocked bistable function by differentiating the binary input beam set to produce a leading pulse having a constam pulse length, and a trailing pulse having a constam pulse length, the leading pulse being out of phase widi die trailing pulse and die clocked bistable function responding to at least one of die pulses.

Differentiation is a very effective way to build a sequencer. A long pulse is differentiated. If only one of die pulses is needed, then an amplifier is used to remove one of diem. The ouφut is apportioned out imo a group of delay lines having differem lengths. This approach is able to be used to produce quadrature pulses or sequence pulses for an optical processor. Integration

The energy beam equivalent of an electronic capacitor is the delay line. Energy is directed imo a delay path. At some later time the energy exits the delay path. This process is used to differentiate a pulse by turning it off after a delay period, and then retrieving die energy stored in die delay patii at the end of die pulse.

Integration is the accumulation of energy over time. To accomplish integration with energy beams, the beam to be integrated is portioned out into one or more delay lines of differem lengths. The ouφut of the delay lines is accumulated at a summing location. To differentiate using tiiis arrangemem, the delayed pulse is brought back out of phase. To integrate, the delayed energy is brought back in phase.

The length of the input pulse relative to the delay periods determines what kind of waveform will result. If the pulse is longer than the delay lines, then an amplitude increase win occur as energy from the various paths arrives simultaneously at the summing location. If the several delay lines have delay times that increase by one pulse time each, the integrated output will be a series of pulses that are able to be close enough together so that they will function as a single long pulse. In practice that may be difficult, because even die slightest break in energy flow may disrupt die process. The presem invention cures these problems by using a threshold detector, a schmitt trigger, a limiter or a bistable arrangemem to provide a constant-level ouφut from the varying-level

integrated signal.

Interestingly, a frequency multiplier is able to be made by first differentiating a long pulse and delaying portions of it for successively longer times, so that the final waveform turns on and off at a frequency greater than the main input pulse. 34. Binary counter.

The next step is to produce a binary counter by interconnecting a number of variable-length dockable flip flops in a cascade series. The procedure used in making such a binary counter is:

a. Use the means and method of variable pulse length clocking; b. Provide a plurality of die binary digit counters in a cascade series; c. Direct the anφUtude-modulated ouφut of each binary digit counter to provide die binary inpm beam set of the next binary digit counter in the cascade series, and d. Provide die binary input beam set of the first of the binary digit counters in die cascade series with pulses to be counted, thereby providing a binary counter by connecting a plurality of the binary digit counters in a cascade series that produces binary ouφuts representing pulse count.

35. Square wave oscillator.

The oscillator described above is capable of producing a variety of wave forms including a sine wave, modulated waveform envelope. A square wave results from connecting a bistable arrangemem to produce oscillation. The procedure used in making the square wave oscillator is: a. Use any of the above bistable arrangements which has a reset input; b. Hold die set pulse on during the time when square wave oscillation is to take place and off when die square wave oscillation is to be gated off; c. Direct a portion of the amplitude modulated output along a delay path which has a delay period and through a logical NOT means providing a delayed bistable signal which is inverted, and d. Direct the delayed bistable signal to provide the reset pulses, each of the reset pulses being larger than the sum of the set pulse and die feedback signal, thereby providing a gated square wave energy beam oscillator by repeatedly turning the bistable function on or off at least once for each delay period, and gating oscillation off by turning off the set pulse.

36. D type flip flop.

A D type flip flop is a fundamental circuit that stores a binary bit inpm on one beam set upon turning on an enabling beam set. The procedure used in making a D type flip flop is: a. Use either of the set/reset binary flip flops having pulse set, and reset inputs; b. Provide first and second logical AND means; c. Provide a data beam set having at least one beam of energy modulated widi binary information; d. Provide an enabling beam set having at least one beam of energy modulated widi data storage enabling information; e. Direct a first portion of the data beam set into a first input of the first logical AND means;

f Direct a second portion of the data beam set into a logical NOT means; g. Direct an output of the logical NOT means to a first input of the second logical AND means; h. Direct a first portion of the enabling beam set into a second input of the first logical AND means, and a second portion of the enabling beam set imo a second inpm of die second logical S AND means; i. Direct an ouφut of die first logical AND means to provide die set pulse, and j. Directing an ouφut of the second logical AND means to provide die reset pulse, diereby providing a D-type bistable function by setting or resetting the bistable function depending upon die state of the data beam set during the time when the enabling beam set is on. 10 37. Multiple uses, multiple energy forms.

The principles for interconnecting energy beam components, described herein, are able to be used in a countless variety of complex arrangements, just as the basic building blocks of electronics are. Energy beam differentiation, inpm summing, threshold detection, frequency and phase filtering, energy beam phasing for component interconnections, and using multiple rays to form interacting images are 15 just some of the important features that allow the presem invention to be interconnected and used in complex arrangements.

All of these things have been disclosed so as to provide a broad foundation for using special interference. In view of the many transistor-like functions described herein, and its ability to use electromagnetic energy, the presem invention is able to be rightfully referred to as a "photonic 0 transistor."

The endless variety of complex devices built from standard electromc building blocks are able to be duplicated using die presem invention. If electromagnetic energy is used, die speed of such devices is phenomenal in comparison to tiieir electromc counterparts.

When electromagnetic energy is used in digital photonic circuits, botii the power amplifying 5 and power limiting ability of the presem invention provide a wide range of uses that are directiy analogous to the use of amplifiers and limiters in electronic circuits, and are able to perform die same types of tasks at a higher speed than conventional electronics.

When the wave properties of electron beams are used in die presem invention, free space electronic structures that perform the functions outlined herein are able to be produced. There are many 0 advantages to the use of free space electrons, or electrons confined to spaces wherein their wave properties are able to be exploited.

Very sensitive instruments are able to be made using the presem invention. For example, small numbers of certain subatomic particles are difficult to detect. By using a stream of such particles as the power beam set in the presem invention, small numbers of input particles to the second beam set 5 are able to be turned imo proportionately larger numbers of such particles which are then more easily examined.

There are many embodiments diat are able to be made using various energy forms. If the presem invention is incorporated into a radar antenna or X-ray machine, phase-modulated reflected or transmitted signals are able to be detected. If an array is set up whereby each pixel is an amplifier of

the presem invention, the output will be a composite, anφUtude- -xlulated image of the reflected or transmitted radar or X-ray phase-modulated image. From such information, full tiiree-dimensional radar images are able to be constructed by providing a matching array of frequency converters of the prior art that convert the non-visible pixel amplitudes to visible pixels. The same is true for sonar when using acoustical energy in the presem invention and converting each pixel to visible energy.

The present invention is also able to operate using multiple differing energy forms in the same locations. For example, one embodiniem is able to include both acoustical waves and light by orienting die input components for each one so that they have a common point for output separation, and given die proper media at these locations, the various energy forms are able to be made to interact with each other. Likewise, information carried by differem frequencies of the same energy form are able to be transferred from one frequency to another.

Reflected ultrasonic sonar waves are essentially modulated beam sets, capable of producing constructive and destructive interference at the "first" location of this combination embodiment. Constructive and destructive interference of the acoustical wave will cause the medium in the emtodiment to compress and decompress. Changing the density of a medium changes its index of refraction and, as a result, will phase-modulate the light beam going through it. This phase modulation is able to be used directly or in cooperation with a second beam so that the effect is able to be amplified right at die point of energy form conversion.

The result is a super-sensitive microphone which is able to detect phase changes in the acoustical wave and direct die gathered infoπnation into a photonic computer made of many interconnected components of the presem invention.

Certain types of materials react to light to produce beat frequencies. Using such materials in an embodi-ment of die presem invention allows light to be heterodyned, frequency multiplied, and divided, even while die signals are being amplified. Other types of materials are photosensitive to light of one frequency, while being able to cause a reaction with light of another frequency. When such materials are placed in an embodiment of the presem invention, information contained on a beam of one frequency is able to be transferred to a beam of another frequency.

Any two energy types or frequencies of the same wave-type energy is able to be used sunultaneously to interact with the presem invention by providing the proper medium for promoting the interaction. The presem invention is then used to transfer information from the one energy type to die odier by providing a sensitive, amplifying environment in which to make the transfer. 38. Basic theory of operation.

Applicant theorizes that the amplitude and intensity of energy at the purely constructive interference locations, using the first type of special interference, are able to be calculated using an adaptation of the standard vector sum of amplitudes method used with otiier interference phenomena.

The basic formula for intensity has been derived from the law of cosines and considers just two incoming rays. That formula is:

A « amplitude of the first beam.

B — amplitude of die second beam.

Theta- phase difference between die two beams.

Intensity = I - A ² + B ² +2AB Cos(Theta)

The Total amplitude T _d « square root of I, just as A ² - the intensity of amplitude A.

At die center of die constructive interference (ci) area, Theu - 0, and die Cos(Theta)- + 1. At the center of the destructive interference (di) area, Theta - 180 degrees, and die Cos(Theta)«— 1. As a result, the vector sum of two amplitudes at these two locations is also the algebraic sum of the amplitudes.

The two rays are in phase in the ci area, so the sum has that same phase. As a result, the ci intensity formula is, Lj - A ² + B ² +2AB - (A + B) ² In the di area: The two rays are out of phase in the di area, so diat the vector sum is the difference of die two amplitudes, which takes on die phase of die largest of the two. If they are equal, the algebraic sum is zero. The di formula for intensity becomes: I* - A ² + B ² -2AB - (A - B) ²

These two conditions are also able to be viewed as die vector sums of three differem rays, which will be labeled B„ B^, and U. In die di area, B <= -Bi «= B _2> so that U is the difference between A and B, and A - B+U

When A is on by itself, die amplitude at location 1 is the vector sum of B, and U. The intensity is (B, + U) ² .

When beam B _j comes on it combines with die first two. Since it is 180 degrees out of phase widi B, and U, die totals of amplitude and intensity are as shown in formulae 1. Formulae 1, di location for all of the interference types: I, - L _j - (B, + U - B,) ² - U ² also, by substitution we get:

I, - L _j - A ² + B ² -2AB = (B+U) ² + B ² -2B(B+U)

- B ² + 2BU + U ² + B ² -2B ² -2BU - U» This is exactly what is expected because the amplitudes add algebraically, and die intensity is the square of die amplitude.

This indicates diat die addition of an out-of-phase beam smaller than the first beam leaves energy having an amplitude equal to the difference of the two. If it is viewed as die sum of three

beams, two of which are equal in amplitude but of opposite sign, ύie third beam is equal to the amplitude of ύie energy remaining at tiiis location after all three have been summed.

The process of interference relocates energy within a fringe image. The equivalent amoum of energy diat is missing from the di areas appears in the ci areas. As shown above, when two unequal beams interfere destructively, not all of the energy in the di areas is relocated into the ci areas. The remainder is exactly equal to the difference between the two unequal beams. This remainder has not been relocated; it continues to arrive at the di location. As a result, this remaining energy is able to be called "undiverted" energy, because it has not been diverted into ύie ci areas by the interference.

As a result, one is able to describe die energy which is apparently missing from ύie di area as "diverted" energy.

In Young's ci area:

In die case of Young's type interference, the amplitude of energy arriving at a second location, namely ύie ci area, when only one beam is on is A. A is able to be considered as die sum of two amplitudes B, and U.

Again, when beam B ₂ comes on, it combines with ύie first two. Since it is in phase widi B, and U,

B ^« B ^{1 «} B ² , and die totals of amplitude and intensity are as shown in formulae 2. Formulae 2, Young's interference type in either amplification or saturation: T ₂ - T _d - B, + U + B ₂ - 2B + U

I ₂ - j - (B, + U + Bj) ² - (2B + U) ² also, by substitution we get:

I ₂ . j _ A ² + B ² +2AB - (B+U) ² + B ² +2B(B+U)

- B ² + 2BU + U ² + B ² +2B ² +2BU - 4B ² + 4BU + U ²

- (2B + U) ²

This is also exactly what is expected because the amplitudes add algebraically, and ύie intensify is ύie square of ύie amplitude.

In tiiis case, Young's type interference has energy directed to tiiis ci location when only one beam is on. It is able to be viewed as having two components. When die second beam comes on, energy from the di area is diverted into the ci area. As shown above, the amoum added to die ci area by interference exactly equals ύie amoum removed from die di area.

As a result, two equal parts exist, B, and B ₂ . One came from beam A and die odier from beam B. The difference between ύie two is U. In both the ci case and the di case, U remains unchanged. It has been called "undiverted" energy. Apparently, it remains unaffected by die interference that is taking place between B, and B _∑ , even in die ci area.

If B rises to become equal widi A, U drops to zero at bot places. The resulting interference image goes completely dark at ύie di location, and the intensity at the ci location goes to 4A ² - 4B ² .

All of the energy contributes to ύie interference image.

When A and B are not equal, die image formed is able to be viewed as being die sum of two images. One image is the interference image formed by portions B, and B ₂ in die familiar interference fringe pattern. The other image is a consistent spot, having no contrast change from one part to S another; its amplitude equals U, and its intensity is U ² .

As a result, U, die difference between two unequal beams, can rightly be called "undiverted," for it arrives at the same locations and in die same pattern as when B, and B _? are off.

B, and B _? are rightly called "diverted" energy, because this energy has been rearranged, or "diverted," in order to form ύie interference image. In that image, the energy from the di location is 0 diverted into die ci location to combine widi an equal contribution from the other beam that will arrive there anyway in the absence of interference. In special interference:

Next, we examine special interference. Special interference has no contribution to location 2, the ci location, when only one beam is on. This occurs because the beams are small in comparison to the di location, and are directed only toward die di location, and are not spread out to cover die location where ci will eventually take place.

The di area functions exactly as described above, as having two in phase beams from A, with the out-of-phase B.

The ci area has no energy in die absence of interference. Most importantly, it has no "undiverted" energy (that is, U-0).

When the second beam (Bj) comes on, interference occurs producing an interference image that removes energy from the di location, (B, - B ₂ ), leaving U as residual energy.

The energy removed from the di location is diverted mto die ci location as B, + Bj. It has an intensity of (B, + B^ . Again by substitution we get:

I ₂ ■ _j _ A ² + B ² +2AB - (B+U) ² + B ² +2B(B+U)

= B ² + 2BU + U ² + B ² +2B ² +2BU

- 4B ² + 4BU + U ²

- (2B + Uf However, U - 0 at tiiis location, producing die important relation as shown in formulae 3.

Formulae 3, first interference type in eitiier amplification or saturation: T ₂ - 2B

I ₂ - (2B + 0) ² - 4B ² As a result, a formula for the first type of special interference has been derived for botii die amplitude and die intensity.

The total amoum of energy in any one application depends upon die area of ci and die area of di, because they are able to be made up of many rays, even thousands or billions of rays. The total energy is able to be expanded to cover large areas, or focused to --mall areas. The ouφut characteristics will be a function of die size, locations, and die ratio of ouφm area to image area of the image

component separator relative to ύie image. Contributions of energy from the other parts of the image mat are not pure ci or di also contribute to the overall operation of ύie invention.

The importance of these formulae to the process of amplification and limiting cannot be overstated. As an example, a substantially constam power beam A that is directed to location 1 and a control beam B (which is smaller than A) produce an interference image at locations 1 and 2, widi di at 1, and ci at 2.

The ouφut intensity is 4B ² , and ύie amplitude is 2B. It does not matter how much larger A is than B, within the limits of the breakdown of die optics or other factors that would physically change the arrangement. Energy diverted into die output is directly proportional to the control beam B. When ύie control beam is amplitude modulated, the ouφut is also --πφUtude-modulated, having twice the amplitude of the control beam. The energy in the information carrying portion of the ou ut waveform has been doubled. Unlike die amplifier of die prior art that uses Young's interference, ύie presem invention does not produce the residual output U, the undiverted leftover energy diat does not contribute to die interference image. As long as ύie modulated beam is smaller than die constam beam, the output will be amplified.

The ouφut amplitude is always double die smaller of ύie two.

Next consider what happens when ύie modulated control beam rises above the level of the constam power beam. With B > A, for any given instant the output will be twice ύie smaller of ύie two. It is ύie same as switching the beam names in the formulae above. Because ύie smaller one is also ύie constam one, the output will be a constam 2A no matter how highly B is modulated, again within ύie realm of not destroying or modifying the optical arrangemem. This condition is called "saturation. " All of the energy from beam A that is able to be has been diverted into die ouφut.

As a result, the amplification curve of the presem invention is NON-LINEAR. Non-linear optics that operate at the speed of light is able to accomplish many tasks that are otherwise impossible. A modulated waveform will be limited and clipped at die point where die two input beams are equal. Second type of special interference:

The second type of special interference is also able to be viewed as having three component amplitudes. The power beam (A) is directed to die di location; none of it is directed to die ci location, just as with ύie first type of special interference. The control beam (B) is directed to both locations. For that reason, tiiis type of interference will not produce a logical AND in a single stage; however, it makes an excellent anφlifier. When the control beam is off, I ₂ — 0, and I, — B, +- U. When ύie control beam is less than ύie substantially constam power beam, A « B, + U, and B - Bj. The amplitude at location 1 will be B, + U. Formulae 4, 2nd interference type in amplification:

Amplitude - T ₂ - B- + B, - 2B

Intensity - I ₂ - (B, + Bj) ² - 4B ²

This is the same as with ύie first type of special interference. The difference appears when ύie arrangemem goes into saturation. When that occurs, the undiverted energy (U), which equals B-A

(because B is larger) does not come from tiie power beam. In tins case the residual energy comes from die control beam which is directed straight into die ouφut. As a result, die ouφut during saturation is as shown in formulae 5. Formulae S, 2nd interference type in saturation: Amplitude - T ₂ - B, + Bz + U - 2B + U - 2A + U

Intensity - I ₂ - (B, + Bz + U) ² - 4A ² + 4AU + U ² Amplification is reduced because A is constam. All of the available energy of die power beam has been diverted into die ouφut. Further increases in B only increase ύie size of U, which is not doubled. When squaring to produce die intensity, die 4AU factor indicates that there exists some interaction with energy from otiier parts of the interference image, but U remains tiie same.

As a result, mis second type of special interference behaves like die first type of special interference when B < A. However, it behaves like Young's interference when B > A. Amplification is still limited somewhat, but it is not clipped. Broad band and narrow band arrangements. The above-described process is phase-dependent. The energy removed from the di location is relocated in die ci position. But what if die signals arrive at the first location at some odier phase? In tins case, ύie ci position is at some otiier location, resulting in near binary operation of a phase- modulated signal. The inputs would have to be exactly out of phase in order for die ci location to be the same as the outpm location. In practice, die optics used will have to be engineered in wavelengdi units and wavelengdi sizes. Most optical arrangements rely on an averaging of energy from multiple points of the cross- section of an input beam. Averaging of energy from these multiple points produces die familiar sinusoidal interference fringe.

If die amplifier is engineered to include a large number of such points so as to use the averaging principle, then it will have a wide bandwiddi and will be able to function using a number of input frequencies. The ouφut locations function as if a group of controllers were placed side by side, each one using an individual ray set.

In tins case, the ouφut hole includes a large number of wavelengdi size locations. For slightly differem phases and slightly differem frequencies, die ci location from each pair of input locations will be at slightly differem output locations. If those ouφut locations happen to be within the area of the hole, the energy will ouφut. If they are not, it will not.

Modern optics is capable of operations at wavelength sizes. Wavelengdi size inpm beams and wavelength size ouφut holes will produce processes that operate considerably differently from die multiple-location averaging style of optics. The more precise tiie optics are, tiie more precisely phases and frequencies must be in order for die ci area to hit die outpm hole.

Wavelength size precision will cause a phase-modulated signal to ouφut only when die phase is close enough to 180, at ύie first location, in order for die ci area to hit the wavelengdi size ouφut hole. The output from an analog phase-modulated signal would be a binary outpm that occurs only when the two inputs are exactly out of phase.

If multiple frequencies are used, tiie only ones that will be able to hit tiie ouφut hole will be those that match the wavelength geometry so that ύie ci location is where ύie tiny hole is.

As a result, each method and each device must be engineered to produce tiie type of amplifier needed. If a phase demodulator is to operate widi an analog input, it will have to be of the averaged, multiple location (broad band) type. If it is to be used in a binary circuit, then die single wavelength- size location (narrow band) type will work quite well.

It is possible to produce a considerable number of composite operations by using a number of wavelength-size controllers having a common first location, but separate output locations; thus, a variety of signals are able to be handled all at once. A frequency division demultiplexer is able to be produced by inputing ύie beams from differem locations directed to a common location. Each differem frequency will produce its ci at a differem output location. If each ouφut location has its own ouφut hole in die image component separator, a complex group of frequencies in ύie input will be separated into separate outputs. Meanwhile, it will filter out any frequencies in between, because no output hole is provided for those frequencies, and no matching input frequency is provided.

If die control inpm is directed to a common location and a .--umber of power inputs are used, each having a differem frequency and a differem location, die geometry is able to be arranged so that the ci locations all match, producing a very accurate frequency-selectable filter. All frequencies that match a power beam will have their ci at ύie common ouφut hole. All other frequencies will not. The difference between this arrangemem and the broad band averaging arrangemem is mat each of ύie frequencies that pass through die filter must match precisely the frequency and phase of the power beam. At wavelength sizes, filters are capable of providing die best selectivity of any known means, especially at light wave frequencies and beyond.

These basic principles of operation produce functions similar to the way electronic transistors perform similar functions. As a result, tiie presem invention warrants the common name "photonic transistors. " Even though the presem invention is quite capable of using non-photonic wave-type energy, photonic embodiments are expected to become the most common in operation. 39. The presem invention provides die following advances in die art of photonics: a. A complete technological base for energy beam (especially photonic) computing using special interference. b. Means, method and apparatus for performing: i. Basic amplification, limiting, and clipping, ϋ. Basic Boolean logic, iii. Basic threshold detection. iv. Digital information storage. v. Frequency and phase-sensitive filtering. c. Interconnecting means and method for interfacing energy beam components. d. A clear understanding of energy beam operational principles, including: i. Multiple beam summing for function interfacing.

ii. Simultaneous, multi-frequency operation. ϋi. Energy beam differentiation. iv. Energy beam integration. v. Regenerative feedback. vi. Phase requirements. vii. Effect of maintaining one beam constam. viii. Amplification. ix. Saturation. x. Cutoff. xi. Phase relationships among multiple input beams. e. Establish a basic mathematical theory of operation for special interference, including its non-linear amplification characteristics. f . Establish die technological and conceptual basis for die simultaneous operation of multiple energy forms. g. More firmly establish a conceptual basis for the use of energy beams to accomplish tasks heretofore reserved for electronics, and tasks far beyond the capability of conventional electronics, h. Establish die operational similarities and differences between electronic transistor circuits and energy beam circuits so diat the common term "photonic transistors" will be rightly applied to die presem invention. i. Establish die technological and conceptual basis for producing electron-wave electronic functions.

The foregoing benefits of the presem invention will become clearer through an examination of the drawings, description of die drawings, best mode(s) for carrying out die invention, claims and abstract which follow.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will become more understandable upon the examination of ύie following drawings along widi ύie detailed description below. Please note that ύie energy beam angles and sizes are exaggerated so as to provide clarity of unrderstanding. Fig. 1. Controller using special interference.

Figs. 2A through 2E. Controlling Locations in five states.

Fig. 2A. Controller using botii types of special interference, having botii inpm beams off.

Fig. 2B. Controller using both types of special interference, having one input beam on.

Fig. 2C. Controller using first type of special interference, having tiie 2nd inpm beam on. Fig. 2D. Controller using both types of special interfeience, having botii inpm beams on.

Fig. 2E. Controller using second type of special interference, having the 2nd inpm on.

Fig. 3. Inverter having one input beam oa

Fig. 3A. Inverter having both input beams on.

Fig. 4. Frequency Spectrum Filtering. Fig. 5. Cascaded Amplifiers.

Fig. 6. Bistable Configuration with Feedback.

Fig. 7. Threshold Detector.

Fig. 8. Summation Vectors.

Fig. 8A. Threshold Detector input ouφut waveforms. Fig. 9. Feedback in Threshold Detector.

Fig. 10. Schmitt Trigger input/ouφut waveforms.

Fig. 11. AND and OR Threshold Detection input/output waveforms.

Fig. 12. NAND Logic.

Fig. 13. NOR Logic. Fig. 14. Exclusive OR Logic.

Fig. 15. Binary Half Adder Logic.

Fig. 16. Clocked Bistable Logic.

Fig. 17. Binary Digit Counter input/ouφut waveforms.

Fig. 18. Binary Digit Counter. Fig. 19. Binary Counter and input, intermediate and ouφut waveforms.

Fig. 20. Oscillator Logic.

Fig. 21. Square Wave Osci-Uator/Schmitt Trigger Logic.

Fig. 22. D-type Bistable Logic and input/ouφut waveforms.

Fig. 23. Demultiplexers and Logic. Fig. 24. Multiplexer Logic.

Fig. 24A. Multiplexer inpm waveforms.

Fig. 25. Phase Locking Logic.

BEST MODE(S) FOR CARRYING OUT INVENTION 1. The basic energy beam controller.

The perspective view of Fig. 1 illustrates die preferred embodiment of the basic unit of die presem invention, an energy beam controller utilizing interference of eitiier type as discussed in die Disclosure of the Invention above. It consists of a first beam set (1) of at least one beam of energy and a second beam set (2) of at least one other beam of energy that are superimposed at at least one location (3). In this illustration, one of each is shown, ahhough "at least one" means that in practice there are able to be many such locations and beams diat behave as those illustrated.

When both beam sets are on, they interfere to produce an interference image having a di area at location (3), and a ci area at at least one otiier location (4). In this illustration two ci areas (4) are shown. Energy behaves the same in all areas of locations (4). Energy from the ci areas continues on to provide an oυφut (S). Four rays or beams of die ouφut beam set are shown in tiiis illustration. They behave the same and operate in unison.

An image component separator (6), in tiiis case an opaque mask, prevents energy from location (3) from becoming a part of tiie ouφut (5).

When tiie optics are designed to take advantage of die first type of special interference, all of die energy from eidier beam (1) or beam (2), when on by itself, is directed to location (3) and stopped by mask (6), and ouφut (5) is off. The optics do not direct energy from eidier beam towards location (4) or imo ouφut (5). The inpm beams are oriented so that destructive interference covers the entire area where the beams arrive individually, including location (3), when both beams (1) and (2) are on. If beam (1) is on first, and then beam <2) comes on, less than or equal to die level of beam (1), the total amoum of energy in the di area, including location (3), is reduced by an amoum equal to the level of beam (2). More than di occurs. The energy does not just go away. The law of conservation of energy requires that the energy which disappears from the di area must appear at a ci area, so diat an entire interference image is formed. The ci areas are labeled as locations (4).

The energy which is diverted away from die di areas, such as locations (3) and into die ci areas, such as locations (4), includes energy from both beams (1) and (2). If die beams are of equal amplitude, tiien the amplitude at die locations (3) will be zero, and all of die energy from botii beams will appear at locations (4). Whenever energy appears at locations (4), it continues on into ouφut (S). If beam (1) is larger than beam (2), energy with an amplitude equal to the difference between (1) and (2) will appear at locations (3) and be stopped by image component separator (6). The amplitude of die energy diverted into locations (4) will be equal to twice the amplitude of die smaller of die two, in tiiis case beam (2). All of die energy from both beams is diverted away from die di locations (3) and imo die ci locations (4).

If die amplitude of beam (2) increases so diat its amplitude is greater than that of beam (1), tiie amplitude of die energy diverted by interference imo ouφut (5) is equal to twice die amplitude of die smaller of the two, namely, beam (1). The intensity of die energy, or power per unit area, is approximately equal to the amplitude squared. As a result, the intensity is four times as great in ouφut

(5), as eitiier of die inpm beams by themselves at location (3).

To calculate the total amoum of energy in ouφut (5), one must also take into account surface area and time. Since ύie two inpm beams are ύie only source of energy, ύie total amoum of energy in ouφut (5) does not exceed twice die amoum of energy in die smaller of ύie two beams. The fact diat ύie intensity is four times as great merely indicates that the energy has been concentrated into a smaller surface area.

If beam (1) is held at a substantially constant level, and ύie level of beam (2) fluctuates, or is modulated below ύie level of beam (1), ouφut (5) will contain twice tiie amoum of energy that there is in the modulated inpm beam (2), assuming perfect optics. The additional energy in ouφut (5) is removed from beam (1) by interference. Since the level of ouφut (5) is a function of the level of the modulated input at beam (2), ouφut (5) contains an amplified modulation envelope just like the modulation envelope of beam (2), only having up to twice the amoum of energy.

Since ύie power for this amplifier comes from beam (1), it is called ύie "power beam" when the presem invention is used as an amplifier. The modulated input beam (2) is called the "control beam" because it controls how much energy will be diverted from die power beam by interference.

If ύie level of control beam (2) rises above ύie level of ύie power beam (1), ouφut (5) will hold steady at twice the level of the power beam. Amplification stops, and die modulated waveform is clipped off. This occurs because ύie ouφut is equal to double ύie level of tiie smaller of the two beams, and in tiiis case ύie level of ύie smaller power beam is constam, producing a constam ouφut.

How these component locations and beams interact to produce a variety of functions are able to be understood more fully by examining the critical locations more closely. Figs. 2A - 2D are an energy beam controller of the first type; that is, using the first type of special interference.

Four states of the most basic operation are shown in Figs. 2A - 2E. Angles in ύie figures are exaggerated for clarity. In this discussion, "beams" are sets of beams or rays diat operate in unison as if they were single beams. Such beam sets are able to have multiple locations that behave in unison as described herein, but they all function as depicted in this close-up view.

Two input beams (1) and (2), directed to a common location (3), along widi location (4), ouφut (5) and mask (6), are depicted for four states. Fig. 2A state shows the state when both input beams (1) and (2) are off, (as indicated by die cross-hatch). Energy does not appear at any of the locations. This null state is important and occurs in many applications of die presem invention, especially digital applications. Output (5) is off.

Fig. 2B state depicts ύie state when beam (1) is on and beam (2) is off. Energy does not appear at location (4), because die energy is purposely not directed there. Mask (6) prevents energy from reaching output (5), so it is off.

Fig. 2C state depicts die state when beam (2) is on and beam (1) is off. In this state, energy does not appear at location (4) because it is not directed there by ύie optics, so ouφut (5) is off.

Fig. 2D state depicts die state when both beams (1) and (2) are on. The beams are superimposed so as to produce destructive imerference at their common location (3). Mathematically

the two beams cancel; however, die energy in those beams is not destroyed. Rather, ύie energy from both beams is diverted away from location (3) and into location (4), where constructive interference takes place. As a result, die interference image in state Fig. 2D covers an area diat is differem from die area covered in state Figs. 2A through 2C. Energy diverted to location (4) continues on, past mask (6), to produce an on ouφut at (5).

Beams (1) and (2) are able to be very small, even single rays as those that come from single pixels of a hologram. Beams (1) and (2) are also able to be projected images produced by sets of beams or rays at location (3) during state Figs. 2B and 2C, including those produced by holograms. During state Fig. 2D the images interfere to produce a new image which is able to have multiple component parts separated so as to provide multiple ouφuts such as location (4).

Beams (1) and (2) are also able to be component parts of complex, multi-input dynamic images (images diat change continuously as computation proceeds, as within a photonic computer.) As components of more complex images, then Figs. 2A - 2E represem the operation of one set of locations among many multiple sets of locations diat operate simultaneously as differem componem parts of multiple complex images functioning in parallel.

The image componem separator at location (6) is able to be any optical element positioned so as to separate energy in one area of the image from energy in another area of ύie image. For simplicity, a mask is illustrated. It blocks energy diat appears at location (3), but passes energy at location (4) to become die output (5) in state Fig. 2D when botii inputs (1) and (2) are on. To produce a maximum ouφut at ouφut (S), beams (1) and (2) must be 180 degrees out of phase at location (3) (which is, of course, a requiremem for producing di) and of equal strength. When these conditions are met, all of the energy from both beams will be diverted into die constructive imerference areas at locations (4), in phase, and on imo ouφut (5). When these energy beam controllers are i-nterconnected widi otiier components, die carrier wave phases are compared to these two locations (3) and (4).

The phase of an inpm beam is defined as being its phase at location (3) when on by itself (unless otherwise stated.) As a result, beams (1) and (2) are considered to be 180 degrees out of phase widi each odier in order to produce an output at ouφut (5).

The phase of an ouφut beam is defined as being die same as die combined phase of energy at location (4). When two or more controllers, or other phase-dependem components, are interconnected, their positions and optics are engineered so diat die needed phase relationships will occur. For example, if die ouφut of one controller is directed into die inpm of another controller, the two are located so that the ouφut measured from location (4) will provide die proper phase of energy at location (3) of the succeeding controller in order to accomplish its design task. The precision of operation of Figs. 2A - 2E depends on die tolerances used in producing ύie optics used to produce it. If die tolerance is larger than wavelength sizes, then operation will generally be broad band. If die tolerance is in wavelength sizes, narrow band operation is facilitated. Broad band applications function tiie same as when many narrow band controllers, widi slightly differem band passes, are operated in parallel, diat is, when multiple precise locations (4) exit energy imo output (5).

This is because die precise (4) locations will be slightly differem for different frequencies and phase differences. The more a design is able to single out precise (4) locations, the narrower will be its bandwidth.

2. Using die second type of special interference. The second type of interference allows energy from one of die beams to appear in ύie ouφut when it is on by itself, but not otherwise. In this case, also shown in Figs. 2A, 2B and 2D, state Fig. 2C does not occur; instead, state Fig. 2E occurs. This type of interference is able to be used to provide an amplifier by using beam (1) as die energy-supplying power beam. However, such interference cannot be used to produce die Boolean AND function or die gated amplifier. This is an energy beam controller of the second type. Botii broad band and narrow band arrangements are able to be produced.

3. Logical AND.

A controller of the first type will function as a Boolean AND. When beams (1) and (2) are modulated wiύi binary information, energy will appear at location (4) and move on into ouφut (5) only when botii beams (1) and (2) are on. When eidier of beams (1) or (2) is off, output (5) is off.

In terms of amplitudes, when ύie smaller of tiie two input beams is zero, ύie output that equals twice die smaller one is zero. 3. Amplification.

If one of tiie input beams such as beam (1) is kept on at a substantially constam level and die second beam (2) is analog-modulated, die energy diverted into the output will be greater than and proportional to beam (2). Because energy from botii beams appears in ύie ouφut (5), ouφut (5) represents an amplified version of die modulated input at beam (2). As a result, the maximum amount of energy, when both beams are equal is tiie combination of both beams; namely, twice tiie level of beam (2). Energy for accomplishing amplification comes from the constam power beam (1), and is determined by the modulated control beam (2). If ύie control beam (2) is binary, ύie ouφut (5) will also be binary, at up to double die level of beam (2).

More is occurring here than the simple changing of amplitude such as occurs when a beam of a certain cross-section is focused onto a small spot. Beam (2) has a modulated envelope widi a certain maximum amoum of energy. This is able to be represented by die actual amplitude of the beam.

The power beam (1) also contains energy, and has an amplitude greater man or equal to that of the control input beam (2), but carries no information. Imerference causes energy from both beams (1) and (2) to be diverted into die output (5) only when beam (2) is on, and in proportion to beam (2). Since all of die modulation changes in beams (1) and (2) appear in die output (5), the ouφut (5) will contain information only from beam (2) simply because only the control inpm at beam (2) has any information.

The modulated envelope in die ouφut (5) has up to twice ύie amoum of energy - not just amplitude - as ύie control input (beam (2)). As a result, ύie action of the presem invention truly is amplification and does not have a residual output level when die control beam is off, as do die energy

beam amplifiers of die prior art.

Eidier type of special interference is able to be used to amplify. If die second type is used, beam (1) must be the power beam, as its energy does not contribute to ouφut (5) when die control input (beam (2)) is off. S 4. Gating.

If ύie first type of special interference is used, then either input is able to provide power because they both work die same. Additionally, turning off the power beam (1) prevents any feed- through of energy from beam (2). This permits the amplifier to be gated off by simply turning off beam (l). 0

5. Limiting.

The level of the constam power at beam (1) establishes a certain saturation level. Generally diat level is equal to the level of beam (1) (depending on die optics used).

When die control beam (beam (2)) is below die saturation level, the modulated input of beam (2) is amplified in die outpm (5). Above diat level, all of die energy available for diversion into ouφut (S) has been diverted. Widi no more energy from beam (1) available, increasing die level of die control inpm (beam (2)) is unable to cause more energy from beam (1) to appear in die ouφut (5). As a result, die presem invention functions as a limiting amplifier.

As with its electronic counterpart, noise or odier irregularities are able to be clipped off of die modulated envelope. How cleanly it does this depends on die optics used. Heavily over-driving a device is able to cause leakage in some cases. Certainly, if the second type of special interference is used, there will be leakage by design. Increases in inpm level above the saturation level diat do leak through will not be amplified, but will be attenuated by the mask (6), which prevents some of the excess energy from reaching outpm (S). Imperfect optics may direct some of the input energy from beam (1) to some location whereby the interference is unable to divert diat portion of die energy into die ouφut (5). Such energy has simply been made unavailable for diversion by die optical arrangemem.

6. Inverter and NOT.

When destructive interference takes place at location (3) of an amplifier, the energy from both beams is diverted to die areas of constructive interference such as location (4), which removes energy from location (3). Figs. 3 and 3A show a modified arrangemem whereby mask (6) has been replaced with mask (7) (shown in cross section.) diat has a hole in it at location (3).

Energy diverted into ouφut (S) is directly proportional to the control input beam (2). As a result, output (5) is called an "uninverted" ouφut. Energy diat passes through location (3) into ouφut (8) is inversely proportional to the control inpm (2). As a result, ouφut (8) is called an "inverted" ouφut.

The ny-Tim-im uninverted ouφut (when die control inpm is smaller than the power inpm) is equal to die power beam that simply passes through into output (8). As a result, ύie uninverted output is not amplified.

If ύie anφlifier is driven into saturation, the excess energy that is not diverted into ύie uninverted ouφut (5), exits via output (8). The output phase changes depending on ύie input levels and phases. This aspect will be covered below in greater detail in ύie discussion on direshold detection. If ύie control input (2) of this inverter is modulated wiύi binary information, then this arrangemem performs ύie logical NOT function. 7. Phase demodulator and comparer.

Next is an examination of how the presem invention operates given changes in phase and frequency, reference to eidier Figs. 2A - 2E or 3 - 3A.

The presem invention is very phase-sensitive. The operational phase is established by the constam power beam of an anφlifier simply because it is unmodulated and serves as a reference beam to compare other beams to.

To produce a maximum ouφut at ouφut (5), die input beams (1) and (2)) must be exactly 180 degrees out of phase at location (3). Any other phase difference, greater than zero, will reduce ύie energy level at location (3), and increase it at location (4) in proportion to the phase difference between beams (1) and (2) in a broad band arrangemem. As ύie optics are made more precise, and especially as construction accuracy nears wavelength dimensions, devices are able to be built so that energy that arrives between locations (3) and (4) will be blocked by ύie image componem separator, resulting in narrower bandwiddi and sha-pemng of die phase selectivity of the device.

If ύie control input beam (2) is in-phase wiύi power beam (1) at location (3), constructive rather than destructive interference will occur at location (3). In diat case, no energy will be diverted to location (4), and ouφut (5) will be off. Amplitude changes in die control input beam (2) will not feed through to ouφut (5).

Phase modulating the control (beam (2)) produces an amplitude modulated ouφut at ouφut (5). If the optics are very broadband, allowing energy from ύie space between locations (3) and (4) to enter ouφut (5), a slight phase-πx-dulated componem will slip through into ouφut (5). However, narrow band optics will allow energy to exit into ouφut (5) only from locations such as location (4) where complete constructive interference occurs, and the energy from both input beams is in-phase.

If ύie control beam (2) has both phase and amplitude components, output (5) will contain an aπφUtude-modulated combination of boύi ύie aπφhtude-modulated part and the phase-modulated part. As a result, the highest-level ouφut resulting from den-odulating die phase changes is also affected by die amplitude of ύie input.

As a result, ύie presem invention is able to be used as a phase demodulator by phase modulating die control beam (2) of an aπφlifier.

The phase demodulator actually compares die phase of beams (1) and (2). If one is a constam standard, and die otiier is phase-modulated, die ouφut is demodulated into an AM signal. If both inputs are phase-modulated, die ouφut will be of a high level when the two signals are out of phase, at location (3), and at a lower level when they are in phase. Since constructive interference occurs at location (4), the phase of the ouφut at ouφut (5) will be ύie common phase between them.

Ou ut amplitude of broad band designs will vary proportionately over the full range of phase

differences. Narrow band designs will vary proportionately, only over die bandpass range. 8. Multi-frequency operation and active filtering.

Each control signal inputted to beam (2) that is to be amplified must have a matching power beam (1). In order for interference to take place and amplification to occur, tiie power beam (1) must 5 contain energy at the same frequency as the information to be amplified in the control (beam (2)).

Ci is produced at slightly differem locations for each frequency (like the colors of a rainbow). If the ouφut openings are large enough at location (4) to accommodate a large number of precise frequency locations, then the device will function in a broad-band manner. If it is designed to only allow through energy at a small number of frequency locations, then the device will be narrow-band. 10 If the control input beam (2) is provided with energy having multiple wavelengths, interference will occur only with those frequencies that appear simultaneously in both inputs. (Not considered are harmonics, heterodyning and die use of special materials, which are also able to be used widi the presem invention.)

As a result, ύie presem invention operates as a very sharp filter. If a frequency-multiplexed 15 beam is provided, it is able to be used to pick out information that matches each power beam frequency and amplify it, ignoring all odier frequencies if die first type of imerference is used. If die second type of interference is used, die matching frequencies will be amplified and die others will be attenuated.

In die case of light, a single color is able to be extracted, so diat any information carried by it is able to be examined. Given die phase and frequency sensitivity of die presem invention, microwave 20 and even visible portions of tiie spectrum are able to be divided imo channels as narrow as die audio channels in die radio bands. Naturally, the degree of separation possible also depends on die ability of die optics to produce standard power beams that are extremely pure.

If die power beam (1) is supplied with energy having two or more frequencies, the ouφut (S) will contain amplified signals extracted from a wide-band inpm diat match die combination of 25 frequencies in the power beam (1), while rejecting all others. If die power source, such as a laser, happens to have lines at several nearby wavelenguis, they will often be modulated in unison. If diat is die case, the presem invention will also amplify them in unison, if they are within the design bandwidth of die active filter used.

Fig. 4 shows three example spectrums of a broad band energy beam amplifier. The power 30 inpm labeled widi its beam number (1) has five frequencies. (PI) through (P4) have the same level; (P5) is smaller. The control input labeled widi its beam number (2) has 10 input frequencies at various levels.

Using formulae 3, above, the ouφut amplitude during amplification is: if (1) > (2) then

->-> <^αnbπ-cd output) ™ ^ uauol Input)

And, during saturation die ouφut amplitude is: if (2) > (1) then

^oorobine ampul) ™ & om iipw)

If ύie power input - 0 at the frequency in question, the combined ouφut (5) - 0

The combined ouφut shown in Fig. 4 has only three signals. This is how they are derived using the formulae above.

1. Cl, C3, C4, C6, C7, C9, and CIO have no matching power beam and, consequently, no matching output. These frequencies are filtered out from ouφut (5); however, they will appear in die inverted output (8) if it is available.

2. C2 matches PI, producing ouφut Ol twice as large as C2.

3. C5 matches P3, producing output 02 twice as large as CS.

4. C8 matches PS, producing ouφut 03 twice as large as P2. Because C8 is larger than P2, the amplifier is in saturation at this frequency. If the second type of special interfeience is used, limiting action occurs, as wiύi formulae 5, and will not clipped at the P2 level.

5. P4 finds no match in die control input, so that no outpm is produced at this frequency. However, because it is a part of the constam power beam, it is ready and waiting for a matching signal to show up in die control input.

Note that the amplifier is able to be in saturation at one frequency and at cutoff at another frequency simultaneously.

Each of tiie frequencies produces interference, and die resulting interaction within the presem invention, in all its forms, completely independently of die otiier frequencies. The exception is when two frequencies are so close together that imerference does occur between ύie two. In that case, a cascade series of amplifiers, each having a slightly differem frequency in succession, will allow information phase- modulated into ύie first of ύie cascaded amplifiers to successively transfer ----formation to higher and higher (or lower and lower) frequencies until the information has been successfully transferred from one frequency band to another.

Another exception is ύie interference of one frequency wiύi its harmonics. This effect is also able to be used in die presem invention for transferring information from one base frequency to another.

As a result of independent operation at each frequency, every one of ύ-e means, methods and apparatus of die presem invention are able to be operated in a multi-frequency mode, whether they are binary flip flops, cascaded amplifiers, limiters, or odier devices.

Certain physical materials are able to react to various frequencies so that tiieir characteristics change in such a way that die change affects oύier frequencies. Such materials are usable to divide and multiply carrier wave frequencies. By including such materials as the energy transmission medium within the presem inventions, the characteristics of the medium affect interference. For example, if P4 is twice the frequency of Cl, a frequency doubling medium will cause interference to take place so as to produce a modulated outpm at P4 and possibly twice P4 and Cl, depending on tiie material. Negative or inverted filtering is also able to be accomplished widi die presem invention. The inverter of Figs. 3 and 3A directs energy away from output (8) when a frequency match occurs. As a result, ouφut (8) will be similar to the control input beam (2), but lacking ύie frequencies that find an equal match with die power input beam (1). If they are unequal, die difference amplitude will output. The following section relates to the interfacing of energy beam amplifiers and die various

power levels and phase relationships needed to accomplish large-scale amplification. 9. Cascaded amplifiers.

Fig. S shows three amplifiers connected in a cascade series. Beam (9) is die control beam of the first amplifier, and beam (10) is its power beam. Location (11) is the equivalent of location (3) in Figs. 2A - 2E, where destructive interference takes place when both beams are on. Energy is diverted by interference to beam (12). An image componem separator (11 A) blocks energy at location (11) and allows it to pass imo beam (12) and on to location (13). The maximum amoum of amplification from this first amplifier is double die level of power beam (10) when all of ύie energy from beams (9) and (10) are diverted to beam (12). The second amplifier is at locations (13) and (14). Power is supplied by beam (15). Again, when beam (12) is on, di occurs at location (13), and ci occurs at location (14) by the diversion of energy from beams (IS) and (12). The image componem separator (13A) at location (13) prevents energy at location (13) from reaching location (16), while permitting energy from location (14) to pass. The maximum amplification at this stage is twice the maximum of beam (12) when die level of beam (15) is sufficiently high. As a result, when beam (9) is at maximum, location (14) contains energy from beams (9), (10), and (15) for a maximum level of 4 times beam (9). When beam (9) is off, no interference occurs at location (11), so beam (12) is off, causing beam (14) to be off. As a result, beam

(14) is controlled by beam (9).

The tiiird amplifier is at locations (16) and (17). The maximum amoum of amplification here is twice the maximum of beam (14), so beam (18) is set at 4 times beam (9). When beam (9) is at maximum, di at location (16) causes the energy from beams (18) and (14) to be diverted to die ci location (17), which passes directiy to the outpm (beam (19)) of the cascade series. An image componem separator (16 A) prevents energy from location (16) from passing into output (beam (19)), while allowing energy diverted imo location (17) to pass imo the ouφut (beam (19)). As a result, outpm beam (19) contains energy from beams (9), (10), (15) and (18), for a total energy 8 times that of beam (9). When beam (9) is off, beam (14) is also off, which causes beam (19) to be off.

The ouφm at beam (19) is completely controlled by beam (9). Since beams (15) and (18) are larger than beam (9) and are eidier diverted or not diverted into ouφut (19) by interference being controlled by beam (9), beam (9) is controlling die large beams. Naturally, one is able to add as many amplifiers as needed so diat a small beam is able to control very large beams. The process is analog in nature, but is also able to operate with binary signals by using either high or low input signals inpm to beam (9).

Using die first type of special interference, shutting off any of tiie power beams (beams (10),

(15) or (18)) shuts off ouφm beam (19). As a result, this embodimem of die presem invention is a gated cascade amplifier, also called a gated avalanche amplifier since a small inpm causes energy to avalanche through the arrangemem and into die ouφm.

Using die second type of interference, shutting off any of die power beams reduces the total ouφm at beam (19), but does not terminate die ouφut because of feed-dirough of the control beams at each stage. Only beam (9) has complete control of die series.

All of the features of ύie ----dividual amplifiers apply also to the cascaded amplifiers. All of die amplifiers within oύier arrangements disclosed in mis application are able to be replaced with a cascade series.

Using the first type of interference, if any of ύ-e amplifiers is driven into saturation, the series is effectively in saturation. If multiple frequencies are used, all of them will be amplified, but if one is missing in eitiier ύie control beam or the power beam, the missing one will not be amplified. Any of die anφlifiers, other than ύie first one, are able to be fitted wiύi an inverted output, or ouφm beam (19) is able to be directed into an inverter. In this way, an amplified inverter is able to be produced.

Many interesting and useful arrangements are able to be made when die power beams have differing characteristics. For example, if each power inpm has multiple frequencies widi differem amplitude distributions for die differem power inputs, then tiie subset of those frequencies that have common characteristics will be selected out, as each stage acts as an active filter, allowing to pass only energy which it has been designed to pass as shown in Fig. 4 and described in section 8 above.

Similarly, when one of the arrangements is narrow-band and die others are broad-band, the total bandwiddi will be that of the ---arrow-band arrangemem.

10. Multi inpm AND.

In die cascade amplifier above, tiie power beams, as well as ύie controlling input beam (9), are able to be provided with equal energy levels. Using die first type of special interference, shutting off any one of tiie inputs shuts off ouφut beam (19), resulting in a multi-input AND. The means and method of connecting controllers to oύier energy beam devices, including oύier controllers, is also illustrated in Fig. 5. In order to operate, each ouφut, such as beam (12), must have a frequency, a phase, an amplitude, a cross-section, an energy level, beam direction(s) and consistency and, if modulated, a timed modulation envelope. Each succeeding device, such as the next amplifier, location (13) and beam (14), has die same requirements. In order to interconnect multiple devices, the interconnection optics must take into account all of tiie requirements. If necessary, additional optical elements are able to be added between die various locations and/or beams, such as between beam (12) and location (13), to correct any problems and to match ύie output of one anφlifier to ύie input of another. The same is true when very complex images are used that essentially integrate a large number of these basic devices operating in parallel and in series wiύi oύier parallel devices.

11. Feedback, Bistable arrangements.

Fig. 6. shows how to include a feedback circuit in die presem invention. Beam (1) supplies energy for die amplifier. Image component separator (6) prevents energy that has not been diverted by interference from entering die output (5). When ύie control beam (beam (2)) is on along wiύi ύie power beam (beam (1)), di occurs at location (3), and ci occurs at location (4), resulting in outpm (5).

As outpm (5) is able to be picked up at multiple locations, a portion of output (5) is redirected by optical elements (20) and (21) such as mirrors. This beam reenters the amplifier as another control beam, the "feedback beam" (22). It is oriented so that die feedback energy is in phase wiύi ύie comrol beam (beam (2)) and out of phase with ύie power beam (beam (1)).

Initially, ouφut (5) is off. An inpm pulse of die control beam (beam (2)) is amplified. A portion of the outpm feeds back into the inpm in phase with this control beam (beam (2)). Regenerative feedback causes die ouφut (5) and the feedback beam (22) to increase suddenly. How much it increases depends on how strong the feedback beam (22) is. If die feedback beam (22) is a small portion of ouφut (5), amplification will be enhanced, just as with its electronic counterpart. If a large portion of ouφut (S) is used for feedback, die ouφut (5) will increase suddenly to die point of saturation.

At this point output (S) is on; what happens when that input pulse turns off? If feedback beam (22) is only enough to enhance amplification, the ouφut (5) will be reduced to its lowest ouφut level. If feedback beam (22) has driven output (5) to die poim of saturation, then ouφut (5) will remain on, making the arrangemem bistable. Control beam (beam (2)) is, as a result, a "set" input. Pulsing either the feedback beam (22) or die power beam (beam (1)) off momentarily will reset the arrangemem.

This arrangemem is very sensitive. It is able to be built to respond to a single photon. As a result, it is able to be used as a very sensitive frequency- and phase-sensitive energy sensor. For most digital applications this energy beam circuit may be too sensitive. In order to be able to control its sensitivity, the process of level detection needs to be disclosed. 12. Threshold detector.

Fig. 7 is a direshold detector using an inpm summing stage and a phase detector using an amplifier. The direshold detector takes advantage of two important processes: energy beam summing and die phase filtering and sensitivity of energy beam amplifiers.

The cross-hatch in this drawing helps to identify the beams, but whether it is on or off depends on die relationships described below.

Energy beam summing is a very important concept of interfacing energy beam signals diat have various switching times and amplitudes. Summing occurs at location (23). The multiple input sum image is separated, just as otiier images are separated. In tiiis case mask (24), shown in cross-section, permits only energy from the summing location to be transmitted into beam (2) of the amplifier.

For most applications, inputs (25) and (26) are 180 degrees out of phase widi each odier. When more inputs are added, they will have one of these two phases, and not some phase in the middle. The reason for this is shown in Fig. 8 Fig. 8 shows four vector graphs, illustrating four possibilities of the sums of two energy vectors. The general formula uses tiie standard vector triangle (8A), Combinations of any two vectors (sides of die triangle) produce die third. The length of the arrow is its amplitude, and its angle from 12 o'clock vertical is die phase angle, widi straight up representing 0 phase, and straight down representing a phase of 180 degrees. By maintaining the phase relationships of die input beams (25) and (26) of Fig. 7, so their phase relationship is eitiier 0 or 180, the sum at summing location (23) will be represented by graphs (8B), (8C) and (8D) of Fig. 8. Because the input phase angles are eidier in or completely out of phase, die vector triangle degenerates imo a straight line. The sum will have the phase of tiie larger of the two and an amplitude equal to die difference between diem.

If one beam is kept at a constam amplitude and ύie oύier rises slowly from zero, ύie constam beam will initially determine the phase of the sum (8B). When the beams are equal, the ouφut amplitude drops to zero (8Q. When it becomes larger man ύie constam beam, ύie ou ut will suddenly shift phase (8D) so as to match ύie greater of the two. This sudden phase change that occurs even wiύi a slowly changing input is the basis for level or direshold detection in ύie presem invention.

In Fig 7, beam (26) is called ύie "threshold-controlling beam. " It is held at a substantially constam level. That level determines die direshold which will be detected. Beam (25) is called die "trigger beam" because it will trigger the ouφm (5) when its level rises above the threshold. When beam (25) is off, ύie sum at summing location (23) has its phase determined by threshold-røntrolling beam (26). The separated energy is directed imo beam (2) of an amplifier having a power beam (1) held at a substantially constam level, as described above in ύie discussion of amplification.

By setting ύie path length properly, ύie phase of beam (2) at this time will drive die phase detecting amplifier into cutoff by producing constructive interference at location (3), resulting in no energy at location (4). Ouφut (5) is off.

It doesn't matter what die sum at summing location (23) is. As long as it produces a sum having a phase equal to the threshold-co---trolling beam 26, ouφut (5) will remain off. Compare energy level (8B) of Figs. 8 and 8A. This means that, as input (25) rises, it is able to fluctuate in any manner below ύie direshold without changing ouφut (5). The graph at Fig. 8A shows how ύie trigger input is able to fluctuate below the THRESHOLD without causing ouφut (5) to change. When ύie trigger beam (25) equals ύie threshold-controlling beam (26) as shown at (8C) of (OUTPUT (5)) remains off because the amplitude of beam (2) is zero.

When trigger beam (25) rises above the threshold (8C), beam (2) is of the proper phase to produce an amplified (OUTPUT (5)) at (8D). The result is a very sensitive threshold detector wiύi an adjustable direshold. The direshold is able to be adjusted statically or dynamically by changing die level of beam (26) or by providing additional inputs to summing location (23) that are eitiier in phase widi the threshold-controlling beam (26) or in phase with ύie trigger beam (25). 13. Schmitt trigger and Set reset Bistable arrangemem. The threshold detector is able to be made bistable or into a schmitt trigger, by die addition of feedback. Fig. 9 is a threshold detector like Fig. 7, widi die addition of two more inputs to die slimming location (23).

The cross-hatch in this drawing helps to identify ύie beams, but whether a particular beam is on or off depends on die relationships described below. A portion of outpm (5) is directed using any conveniem optical system such as mirrors (29) to provide die feedback inpm beam (27). Beam (27) arrives at summing location (23) out of phase with the direshold-comroJling beam (26).

The progress of its operation is illustrated in die graphs of Fig. 10.

Initially, ouφut (5) is off, and power beam (1) and die tlιreshold-controlling beam (26) are on.

Trigger beam (25) comes on at a level which is less than the level of beam (26), and adds algebraically with beam (26). The direshold has not been reached (as described above for die direshold detector,) so die phase detector at locations (3) and (4) remains cut off.

The rising level of trigger beam (25) reaches the direshold, and output (5) begins to turn on. 5 A portion of output (5) is directed to beam (27). Because it is in phase widi the trigger beam (25), it produces regenerative feedback, which forces die phase detector into, or closer to saturation. As a result, outpm (5) is fully on.

Whether the arrangemem is bistable or schmitt trigger depends on die level of die beam (27). If it is smaller than die direshold-controlling beam (26), ύie arrangemem will be a schmitt trigger. 10 When ύie trigger input goes below ύie direshold, die direshold will be reached, die phase at summing location (23) will reverse, and the phase detector will go again into cutoff.

If die feedback beam (27) is larger than tiie ύireshold-controlling beam (26) die arrangemem is bistable. When ύie trigger beam (25) goes off, beam (27) will still have enough energy to maintain the energy at summing location (23) above the threshold. Trigger beam (25) is tiien able to be called a 15 "set" inpm.

To reset tiie arrangemem beam (28) is added as a reset inpm. An energy pulse in beam (28) arrives at summing location (23) in phase widi die threshold-controlling beam (26), raising die direshold above die level of die feedback beam (27), reversing die phase at summing location (23) and forcing die direshold detector off. 20 14. Multi-input AND and OR, and fuzzy logic element.

Fig. 11 shows how, by using eidier Fig. 7 or Fig. 9, a multi-input AND is able to be made by setting die direshold level as needed. The threshold-controlling beam (26) is set so diat die direshold matches the type of operation desired. In Fig. 11, die AND THRESHOLD shows how the threshold is set just below die maximum sum diat occurs when all of die trigger inputs are on, in tiiis case beams 25 (25) and (28), beam (28) being in phase widi beam (25). Since all of the trigger inputs have to be on to reach die direshold, and as many trigger inputs as one wants are able to be added to die arrangemem, it functions as a multi-input AND.

If the urreshold-controlling beam (26) is set low, so that tiie threshold is reached when any one of die trigger beams comes on, die OR THRESHOLD has been reached and ouφut (S) turns on. This is 30 a tnulti-input OR.

If the arrangemem of Fig. 7 is used, die levels of all of die input beams are able to be set so that saturation occurs with the addition of any beam that puts summing location (23) above the threshold. If die arrangement of Fig. 9 is used, tiien its inputs are set so diat it functions as a schmitt trigger. 5 Since die threshold is able to be set anywhere between die AND THRESHOLD and die OR

THRESHOLD, die arrangemem is able to be used in fuzzy logic and neuro circuitry. The direshold- controlling beam (26) is able to function as an adjustable "weight" input. 15. NAND.

Having die basic controller operation shown in previous figures, logic interconnections are able

to be depicted using conventional logic symbols having labels to match ύie beam positions used. Where duplicate numbers are used in all of ύie logic diagrams of tiiis specification, they refer to the beam types as described in Figs. 2A - 2D, 3, and 3A as they apply to the individual logic element. For example, Fig. 12 shows an AND symbol widi a (5) outpm directed into an inverter symbol at beam (2). The AND symbol refers to die arrangement of Figs. 2A - 2D, and die inverter symbol refers to die arrangemem of Figs. 3 and 3A. Of course, cascade arrangements are able to be used as described in section 9 above.

Fig. 12 shows die logic diagram of a NAND. (30) is an AND as in Figs. 2A - 2D. (31) is an inverter as in Figs. 3 and 3A, however, die inverter shown in U. S. Patem 5,092,802 will also work as tiiis inverter. Input (1) of inverter (31) is provided widi a power beam (P). (30) provides die AND function on the inpm signals (NAND1) and (NAND2), and (31) performs the NOT function to produce ύie NAND function output at (NAND3).

15. NOR.

Fig. 13 shows ύie logic diagram of a NOR. Two inverters (32) and (33) are provided widi power beams from (P). Inputs (NOR1) and (NOR2) are inverted before entering AND (34). This produces a NOR output, (NOR3).

16. XOR.

Fig. 14 shows the logic diagram for an XOR that does not change phase in its ouφut as does die prior art Exclusive OR. (XORl) and (XOR2) are inputs to a summing location (35). This summing location operates just like die summing location depicted in (8B), (8C), and (8D) of Figs. 8 and 8A and location (23) of Fig. 7 and Fig. 9.

The summing location (35) produces a biphase ouφut whenever XORl or XOR2 are on by themselves. The ouφut of suinming location (35) is directed to two ANDs (36) and (37), which function as phase detectors. Although tiie AND symbol is used, die second type of special interference is able to be used in mis instance.

A constam power beam is supplied to AND (36) from P. Power is also supplied to AND (37); however, it is first phase-shifted by 180 degrees by phase shifter (38). This phase shift is able to be accomplished by an optical element, or by simply positioning die components properly. As a result, when XORl is on by itself, AND (36), acting as a phase detector provides an ouφut, and AND (37) does not. When XOR2 is on by itself it is 180 degrees om of phase wiύi XORl, and so is ύie ouφm of summing location (35). Now AND (36) is cut off, and AND (37) provides the ouφm.

Having separated ύie two phases, they are now able to be brought back together. The output of AND(36) is phase-shifted by 180 degrees by phase shifter (39), or positional process. The ouφuts of AND (37) and die phase-shifted AND (36) are brought together at another summing location (40). This provides an XOR output at XOR3 mat produces ύie same ouφm phase regardless of whether XORl or XOR2 is ύie XOR that is on by itself. XOR3 is off if XORl and XOR2 are on or off together. 17. Binary Half Adder.

The binary half adder is shown in Fig. 15. The Exclusive OR of Fig. 14 is shown using

conventional XOR symbol (41). The two inputs (42) and (43) are apportioned into both XOR (41) and AND (44). The Exclusive OR provides die SUM ouφut, while AND (44) provides die CARRY outpm.

18. Clocked Bistable. 5 Fig. 16 shows the logic diagram for clocking an energy wave bistable device (FF). (FF) is a bistable arrangemem such as the one shown in Fig. 9. Beam (25) provides a set pulse, and beam (28) provides die reset pulse.

In order to produce set and reset pulses of the proper timing, the CLK clock input pulses have to be directed first to die set inpm beam (25) and then to the reset input beam (28) of (FF). 10 Power is provided to flip flop (FF) as shown in Fig. 9. Initially, (FF) is reset, and output (5) is off. Inverter (47) is powered by a beam from (P). Its outpm is on, enabling AND (46).

The pulses into CLK are directed to ANDs (45) and (46). There is as yet no energy coming om of delay line (48), so AND (45) is disabled. The first pulse from (CLK) passes through the enabled AND (46) and sets die flip flop (FF) through beam (25), turning on ouφut (5). IS A portion of ouφut (5) goes through delay path (48). Before energy exits delay path (48), this first clock pulse from (CLK) terminates to insure that the set function is successful.

The delayed energy from delay path (48) enables AND (45) and is directed into inverter (47). This causes inverter (47) to turn off, disabling AND (46).

The second pulse from CLK arrives when AND (46) is disabled and AND (45) is enabled. 20 This causes energy to pulse through to beam (28), resetting (FF).

Energy continues to ouφut from delay path (48) to insure that the reset operation is successful. The second pulse from CLK terminates before die end of die delay time of (48). Then energy from delay path (48) terminates, enabling AND (46) and disabling AND (45); and die clocking cycle is complete. 25 19. Differentiation, and die binary digit counter.

Fig. 17 (17A) is a graph of a typical bistable signal, found at die ouφut of a clocked bistable arrangemem such as Fig. 16, which is to be used as an inpm to a binary digit counter. A binary digit counter is essentially a bistable device diat is able to be clocked by a binary pulsed beam having variable-length pulses. Fig. 18 shows the logic diagram of a binary digit counter. 30 (CFF) in Fig. 18 is die clocked flip flop of Fig 16. It requires clock pulses at (CLK) to be shorter dian die time of delay path (48) of Fig. 16. Variable-length pulses are shown in graph (17A) of Fig. 17 and as the inpm to the binary digit counter, Fig. 18, (BI).

The first binary pulse starts at time (51), Fig. 17. To produce standard-length pulses, a portion of (BI) is directed into a summing location (49), and another portion into a delay path (50) in Fig. 18. 35 Summing location (49), as yet, has no otiier inpm. Its output is on, which starts the first clock pulse for CLK.

The delay path (50) ouφuts no energy until time (52) shown in Fig. 17, at which time energy exits delay path (50) which causes di at summing location (49), terminating the first clock pulse imo CLK. The delay time of delay path (50) is shorter than the enabling delay time of delay path(48) in

Fig. 16.

CLK remains off as long as botii inputs to slimming location (49) remain on, until time (53), when ύie BI input shuts off, and the di at summing location (49) stops. There is still energy in delay path (SO), so summing location (49) begins ouφutting another pulse. However, because di is used at summing location (49), this pulse is 180 degrees om of phase with the first pulse shown by the cross- hatch in Fig. 17. This shows the process of differentiation of an energy pulse.

The phase detecting controllers in CFF do not respond to this phase of energy, so these pulses are ignored. When all of tiie energy in delay path (50) is exhausted, die reverse phase pulse at CLK terminates at time (54). At time (55) the process is repeated, and CFF clocked for a second time. This produces die CFF output BC having wave form (17C) at one-half the clocking frequency of the (BI). As a result, single binary digits are counted in binary.

In order for all but a small amount of energy to pass through die differentiation process, through (49), the pulse rise and/or fall rates must be faster than the delay period through (50). If rise or fall times are too slow, men di will be established or removed before die ouφut from (49) is able to reach its maximum value. As a result, rapidly rising or falling pulses will pass through ύie differentiation process, while slowly ^{ -ha gin pulses will be attenuated.

This differentiation process is able to be applied to a great variety of energy beam circuits, just as its electronic counterpart is a versatile tool in electronics.

20. Binary counter. A cascade series of binary digit counters, BDC, through BDC _N , are shown in Fig. 19. Each

BDC, binary digit counter, is as shown in Fig. 18. Each BC ouφut of each BDC is directed imo die BI input of ύie next BDC.

A series of pulses to be counted enters at ύie first input (56) and graph (56). Operation is the same as wiύi electronics, except that when using ύie presem invention wiύi light, the process counts at light speed.

Graphs (57) dirough (59) show a ύie typical binary count outpm of matching beams (57) through (59). As many binary digit counters as are needed are able to be added, as shown by ... BCD _N widi its ouφut (60).

21. Gated Oscillator. Fig. 20 shows a Gated Oscillator. An anφlifier (61) provides amplification. Its power beam (1) is supplied by beam (G). Its control input beam (2) is supplied by the ouφut of an inverter (63). The inverter's power comes from beam (P). Initially, inverter (63) is on, which directs energy to beam (2) of amplifier (61), which in turn is amplified to produce an ouφm at output (5).

A portion of ouφut (5) is directed into delay path (62) to provide a delayed feedback signal. The cumulative delay through die whole circuit is ύie frequency determimng device, but ύie delay time dirough (62) is die primary determining factor. The outpm of delay path (62) turns off inverter (63). Outpm (8) turns off anφlifier (61) and ouφm (5), completing ύie cycle.

The oscillator continues to turn the energy beams on and off as long as power is supplied to anφlifier (61). When constam level input (G) is gated off, oscillation stops. When (G) is gated on.

oscillation starts again. 22. Square Wave Oscillator.

Fig. 21 shows a square wave oscillator. FF is the set/reset bistable arrangemem of Fig. 9. Power is supplied from (G) to beam (1). A portion of ouφut (S) is directed into delay path (64), which 5 provides die primary time delay that functions as tiie primary frequency determining device. The ouφm of delay path (64) is directed into die reset input (28) of die bistable device (FF).

Initially, ouφm (5) is off. Input (25) is gated on and held on during die time oscillation is to take place. The threshold is reached, and outpm (5) turns on. After die delay dirough delay path (64), FF is reset because tiie reset inpm (28) is out of phase with the set signal at inpm (25). It brings die

10 combined inputs down below die direshold, and ouφm (5) nuns off. (28) continues to hold die combined input below die direshold until die pulse termination at (5) is able to pass dirough delay (64), completing the cycle.

Gating inpm (25) off causes die combined input to be below the threshold, so that oscillation stops. It is self starting when input (25) comes on. Oscillation is also able to be gated off by turning off

15 die gated power source G.

The gated oscillator shown above in Fig. 20, will operate using analog as well as digital signals, depending on how hard die delayed feedback signal drives die amplifier. As a result, it is able to produce a sine wave waveform envelope. The Square Wave Oscillator of Fig. 21, on the other hand, is binary in nature because of the flip flop FF. As a result, it produces a square waveform ouφut SQR.

20 23. One Shot.

Only one modification is needed to change die square wave oscillator of Fig. 21, to perform a one shot function. If the set pulse is gated on for a shorter period of time dian die time delay through the delay padi (64), then the arrangemem will cycle through one cycle only, and shut off. 24. D-type Bistable Function.

25 Fig. 22 shows a D-type bistable arrangemem using a set/reset flip flop as in Fig. 9. Beam (D) is modulated with input binary data. A portion is directed into AND (65) and Inverter (66). The output (8) of inverter (66) is directed into AND (67). AND (65) sets flip flop FF through inpm (25) when it ouφuts, and AND (67) resets FF, dirough input (28) when it is on. Enabling beam (E) is directed to die odier two inputs of AND (65) and AND (67).

30 Fig. 22 also has a graph to show how dύs D-type bistable function works. Initially at time

(68), (D) is off. Any changes in the state of (D) have no affect on die state of (FF) as long as ANDs (65) and (67) are off, as shown at times (68) and (69). At time (70), enabling pulse (E) comes on, enabling ANDs (65) and (67), and causing the state of (D) to be written into FF. At time (70), (D) is off, which causes AND (65) to be off and inverter (66) to be on. This in turn, turns on AND (67),

35 because it had also been enabled by (E). AND (67) ouφuts a reset pulse to inpm (28) diat lasts as long as (E) is on. The state of (D) now appears at ouφm (5) of FF. After enabling pulse (E) terminates, changes in (D) again have no affect on (FF).

At time (71), another enabling pulse enables ANDs (65) and (67). This time (D) is on, which turns inverter (66) off, which keeps AND (67) off. A portion of beam (D) is now able to turn on AND

(65), which pulses input (25), setting flip flop FF. The new state of (D) now appears at ouφut (5) of FF. Again ύie enabling pulse (E) terminates before (D) changes state.

If (D) were to change state before (E) terminates, the last state of (D) will be written into FF as long as (D) retains this final state long enough for the complete set, or reset cycle to take place. 25. Frequency Demultiplexer.

Fig. 23A shows die logic diagram for a f equency demultiplexer. A group of controllers (as in Figs. 2A - 2E) are used as filters, F, through F _N . A frequency multiplexed inpm beam (72) is apportioned om to ύie control input of each of ύie filters. Each power beam (73) through (76) is provided with a differem frequency of energy. The ouφuts (77) dirough (80) contain only one frequency each that is ai-φUtude-modulated widi what-ever infoπnation that frequency was modulated wiύi during the multiplexing process.

Fig. 23B shows a frequency demultiplexer where all of the controllers share common components (3), (72), and (81). However, each frequency will have a separate ouφut location, (77) dirough (80), just as the differem frequencies in sunlight separate to make a rainbow. The separate frequency power beams are able to be combined into a single input power beam (81) in order to make the geometry of separation easier to work widi; however, diat is not required.

25. Time Division Demultiplexer.

Demultiplexers have the common quality of providing multiple ouφuts from a common inpm. Fig 23A is able to be used as a time division demultiplexer by using a differem power beam arrangement.

In tiiis case, input (72) has time division multiplexed information, that is, serial pulses. Inputs (73) dirough (76) have a sequenced series of pulses, The controllers function as amplifying ANDs. As each controller is pulsed sequentially in its turn, whatever information is on beam (72) will ouφut imo the separate outputs (77) through (80). From that point on, further processing, such as setting a bistable device, is able to be accomplished.

Fig. 23C shows how ANDs F, through F _N are able to share a common di location. Beam (72) enters a complex controller. The orientations of each of the inputs (73) through (76) are differem from each other in order to provide different positions for the ou uts (77) through (80). As the power inputs are sequenced, each bit frame will be picked off from beam (72) and diverted off into a differem direction to provide a separation of the ouφuts.

26. Time Division Multiplexer.

Time division multiplexers are essentially parallel to serial converters. Fig. 24 shows tiie logic diagram of a time division multiplexer. Input beams (81) through (84) are provided widi parallel information. See graph in Fig. 24A. The power inputs (85) through (88) sequentially pulse the controllers CR, through CR** configured as ANDs. The ouφuts of die controllers are directed into a common ouφut beam (90) at location (89).

Ouφut beam (90) tiien contains a series of pulses taken sequentially from the data on beams (85) through (88). Summing of die signals does not ordinarily occur at location (89) because no two inputs are on at the same time.

27. Phase I x-lring.

Phase locking is essentially time division phase demultiplexing followed by in-phase multiplexing. Fig. 25 is die logic diagram for a phase locking arrangemem. A fluctuating phase beam (91) is apportioned om to ύie control inputs of an array of controllers configured as phase demodulators. The power inputs (92) dirough (95) are each provided wiύi a differem phase of constant- level energy. The phase-fluctuating beam exists at one phase, then switches to another, and tiien another. It may be random or in some pattern. At whatever phase it happens to be at any given moment, at least one of die phase demodulators PD, through PD _N will produce an ouφut because enough phase demodulators are provided so as to cover the entire range of phase fluctuations. Once demultiplexed, each of die controller outputs undergoes a phase shift (96) dirough (99) so that, as each one turns on, its energy will arrive at location (100) and on into ouφm (101), having the same phase as die others.

The power inputs (92) through (95) each have a constam phase, although phase-adjusted from one another. The phase at each controller output is also constam, when on, having been produced by ci within the controller. As a result, die introduction of specified phase shifts allows the remultiplexed output (101) to have any phase pattern we build it to produce. To produce phase locking, those phase shifts are simply made to outpm a substantially constam phase.

If the phase-fluctuating beam is, for example, an outside laser diat is to be matched with another laser, energy from botii lasers will exit (101) having die same phase. If tiie phase fluctuating beam (91) is amplitude modulated, output (101) will likewise be amplitude modulated. If, for example, phase-fluctuating beam (91) comes from an optical fiber, its information needs to be impressed onto a beam from a local laser so diat it is able to be processed photonically. This arrangemem will do diat.

28. Conclusion The presem invention has made it possible for these many common logic steps to be implemented using energy beams. Each of diem have two interesting qualities that add to tiieir uniqueness. First, tiiey are able to be made quite small. As described in die Disclosure above, differences occur when these processes are carried out, and devices built, at wavelengdi dimensions. The more precise die optical arrangements, die more precise each controller is able to accomplish its design task, whether narrow band or broadband.

The second important feature is that many of these procedures are able to be accomplished by controllers diat occupy common locations. A complex inpm beam is able to be operated on by a group of otiier input beams having slightly differem geometries, pulse timing, frequencies, phases, and levels. Together, these operations are able to be incorporated as pixel operations which are tiny parts of dynamic images diat are very much larger. By properly timing, orienting, modulating, and selecting the individual rays, or sets of rays diat make up complex images, whole computers that accomplish light speed computing are able to be constructed in a very small space.

The principle of frequency multiplexing broadband devices, interspersed widi specific narrow band devices, allows complex operations that have total control over the energy streams being used.

Dynamic images produced this way are able to shuttle data, addresses, timing, storage, and every other kind of information from pixel to pixel, at wavelength dimensions.

Both multiplexing and demultiplexing, whether frequency division or time division, are basically separating, sorting, and coinbining operations. By arranging an array of many controllers, in appropriate sequences, and appropriately gating each embodimem, many differem processes are able to be performed. These include time and frequency division multiplexing and demultiplexing, address decoding, sorting and channel switching.

Optical switching, such as is used in fiberoptic networks, optical computing, the control and manipulation of holographic and dynamic images are also able to be performed using the presem invention. When acoustical waves are used, various imaging processes such as sonar and sonagrams are able to be inφlemented.

The pixel-by-pixel control of otiier images and die organization of application characteristics allow many tilings to be done that were difficult or impossible in the past. These include the manipulation of moving particle streams both large and small for industrial machining, electron microscopes, laser cutting, and odier processes which require energy, or moving particles, to be controlled with precision.

Most importantly, these operations will become more widespread and valuable as optical computers, including those constructed using die presem invention, are constructed for die purpose of mampulating and comrolling tiie many input and ouφut peripherals that are also able to be built using ύie presem invention

This disclosure provides more dian just the basic invention using special interference. It also lays oat die principles Of componem inteπ-onnection, and energy beam manipulation for implementing and organizing controllers wiύi other energy beam processes. It provides die basic means and methods for acconφl-ishing tasks that where previously confined to die realm of slow electronics. As a direct result of tiiis extensive disclosure, practically any logical organization or architecture that is common in the electronic world is able to be implemented widi this new energy beam technology, and especially in light. This is because specific examples of a variety of logic arrangen-e s are fully explained, showing precisely how the components are to be interconnected to accomplish complex logical tasks.

Man has been studying light, sound, particle waves and odier forms of energy for centuries. The principle of superposition has been well understood for over 150 years, yet die conventional methods, even die optical ones, have run into a wall of difficulty placed diere by the basic physics of the fundamental processes used to create them.

The world of photonic computing was at an impasse. The present invention sweeps away that wall, by using tiie he-re-to-fore overlooked special interference. The present invention establishes an entirely new technology. As a direct result, ύie presem invention is expected to be the foundation of 21st-century computer science.