Technical Field

This invention is generally related to communication devices and more particularly to digital communication devices.

Many digital communication devices use zero-crossings of the demodulated signal to determine signal polarity. In zero IF (Intermediate Frequency) applications, signal phase information can be extracted from the zero crossings of the in-phase and quadrature signals. The phase relationship of the in-phase and Quadrature components is then used to estimate and reconstruct the originally transmitted information signal. The number of zero-crossings is directly related to the deviation of the carrier signal. The higher the deviation the more the number of zero crossings. However, the increase in deviation results in an increase in the bandwidth. Narrow band FM modulation (such as minimum shift keying (MSK)) puts a significant limitation on the bandwidth thereby leaving designers with the sole option of increasing the number of zero-crossings. Patent number 4,322,851 issued to Ian A. W. Vance on March 30,

1982 teaches one method of increasing the number of zero-crossings. This patent uses four mixers to produce four signal components, i.e. two in- phase and two Quadrature components. In other words, in addition to the two mixers that are needed to produce the in-phase and quadrature components two more mixers are used to produce additional zero- crossings. The additional mixers provide additional in-phase and quadrature components at 45° and 135°. These two components will be used with phase components at 0° (in-phase) and 90° (quadrature phase) to decode the transmitted signal. A significant problem with this approach is the need for two additional mixers. As is known in the art, mixers generally draw much supply current or may require significant power from the local oscillate signal. The addition of two more mixers is

therefore particularly troubling in portable communication devices which use battery power. It is one goal of this invention to conserve energy particularly in portable radio applications. In addition to the added mixers, two more phase shifters and filters are needed at the local oscillator and the output of the mixer, respectively. The phase shifters produce appropriate local oscillator signals to produce mixed signals at 45° and 135° away. The added filters remove undesired high frequency signals from the output of the two additional mixers. The problem of added current drain is exacerbated by the filters and phase shifters. The prior art is therefore inefficient in recovering digitally modulated signals. A scheme is desired to recover digitally modulated communication signals without a significant demand on current drain and/or unnecessarily additional circuit components..

Brief Description of the Drawings

FIG. 1 is a block diagram of a communication device with an efficient zero-crossing generator/detector in accordance with the present invention. FIG. 2 shows the vector relationship of the four vectors as generated by the block diagram of FIG. 1.

FIG. 3 shows a block diagram of a summer and subtracter in accordance with the present invention.

FIG. 4 shows the vector relationship of the vectors of the block diagram of FIG. 5.

FIG. 5 shows a block diagram for producing additional zero crossings in accordance with the present invention.

Detailed Description of the Preferred Embodiment

Some digital demodulation schemes (such as Frequency Shift Keying (FSK)) estimate transmitted data by estimating a phase rotation direction at zero-crossing points. The performance of such demodulation schemes varies with the number of zero crossing points. An increase in the number of measurements of a phase rotation direction at zero-crossing points improves the performance of the demodulator, and increases the maximum transmit bit rate with acceptable performance.

In general, a received signal is converted to zero IF either directly or via multiple conversion stages. In fact, a zero IF signal may be acquired either by directly converting the received signal to zero IF or first going through an intermediate conversion stage. Traditionally, the conversion is accomplished via two mixers which produce the in-phase (i(t)) and the quadrature phase (q(t)). These two signals are then filtered via two low pass filters (LPF) where the undesired high frequency components are removed. The i and q signals are used in the recovery of the transmitted information signal using methods known in the art. One technique determines the polarity of the bit by estimating the phase rotation direction. A phase rotation direction may be determined by sampling q waveforms at i zero crossings and i waveforms at q zero-crossing points. However, the number of zero crossings formed by the in-phase and quadrature components are insufficient to recover an information signal transmitted via narrowband FM and/or under noisy conditions. Indeed, in narrowband FM some bits may be missed (wrongly estimated) due to the lack of a sufficient number of zero crossings. Noise could similarly affect the detection of the signal. The present invention proposes the use of summers and subtractors at the outputs of only two mixers to increase the number of i and q components, hence increasing the number of zero crossings.

To better understand the principles of the present invention reference is made to the drawings and in particular to FIG. 1. This figure shows relevant portions of a communication device 100 having a zero IF demodulator including an efficient zero-crossing generator. A radio frequency signal, such as FSK, received at the antenna 102 is converted to zero IF via two mixers 104 and 108 An oscillator 112 provides the local oscillator (LO) signal for 104. The local oscillator signal for the mixer 108 is supplied through a 90° phase shifter 106. Alternatively, the radio frequency signal may be converted down to a non-zero IF before a final conversion to zero IF. The output signals of the mixers 104 and 108 are filtered at LPFs 110 and 114 to produce the i(t) and q(t) signals. These signals are added in the summer 116 and subtracted in the subtractor 118 to produce additional in-phase and quadrature signals, iι(t)) and qι(t), respectively. As will be shown mathematically, these additional channels are 45° and 135° away from i(t).

In general, a received RF signal coupled from the antenna 102 is mixed with cos (w _c t) and -sin (w _c t) to generate i and q signals at the outputs of mixers 104 and 108, respectively. When the received RF signal is represented as cos (wct + q (t)), the mixing operation may be mathematically described as:

i _m (t) = cos (w _c t + q (t)) cos (w _c t) (1)

= - [cos (2 w _c t + q (t)) + cos (q (t))]

This signal is filtered via filters 110 and 114. The filtered signals are represented as:

i (t) = cos (q (t)) (2)

q (t) = cos (q ((t) - 90°)) (3)

= sin (q (t)) (4)

The filtered signals are applied to limiters 120 and 122 before being coupled to a zero-crossing detector 128. These limiters provide zero crossing information on the i and q channels. The filtered signals are added and subtracted at 116 and 118, respectively to produce:

i _l (t) = cos (q (t) - 45°) (5)

_qι (t) = cos (q ((t) - 135°)) (6)

= sin (q (t) - 45°) (7)

The generation of these two signals ii (t) and qi (t), are shown as additional new ii and qi axes in the phase diagram 200 of FIG. 2. Note that the creation of these two additional components ii and qi is accomplished without any additional mixers, phase splitters, or low pass filters.

The accuracy of equations 5 and 7 is pursued using the following trigonometric equation.

cos a + cos β = 2 cos ( ( (a -β)) cos ( (α +/ϊ)) (8)

Using this relationship, ii (t) may be expressed as:

i, (0 = (cos ((θ(t))) + cos (θ(t) - 90°)) (9)

Equation 9, may be expressed in terms of i (t) and q (t) using

Equations 2 and 3:

i,(t) = (i(t) + q (t)) (10)

Similarly, q. (t) may be generated using the following equation:

q _λ (t) = (-i(t) + q(t)) (12)

Equations 10 and 11 indicate that L (t) and q. (t) may be generated by summing and subtracting i (t) and q (t).

The second in-phase ii (t) and quadrature qι(t) components result in additional zero crossings. The outputs of the summer 116 and the subtracter 118 are coupled to the zero-crossing detector 128 via limiters 124 and 126, respectively. These limiters work in conjunction with limiters 120 and 122 to provide the detector 128 with additional zero crossings which are detected therein. The detection of zero crossings may be accomplished via D flip flops with edge triggered clock inputs. In general, the known phenomenon that when the i waveform (time domain) goes from positive to negative, or from negative to positive, the phase trajectory in a phase diagram crosses the q axis. Also, q values at i zero crossings indicate the direction of phase rotation. The zero crossings of i and q waveforms may be viewed as the phase crossings of i and q axes in the phase diagram 200. A positive phase axis-crossing means that the phase trajectory crosses i or q axis in a positive direction (counter-clockwise). Similarly, a negative phase axis-crossing means that the phase trajectory crosses i or q axis in a negative direction (clockwise). The zero-crossing detector 128 sets its output high if the phase trajectory crosses i or q axis in a positive direction. A low output is produced when the phase trajectory crosses i or q axes in the negative direction. The detection of zero crossings is translated to information signal at a controller 130. The controller 130 may be any micro-controller or microcomputer available from various semiconductor

manufacturers. The audio portions of the decoded information signal is coupled to a speaker 134. The data portion is displayed on a display 132 automatically or upon request or otherwise used or processed.

A significant feature of this invention is the fact that demodulation can be done digitally without the use of conventional A D converters. In other words, only simple limiters or compactors are needed to convert the baseband signals into a 0 and 1 representation. The need for A/D converters is eliminated by performing the addition and subtraction functions in the analog domain. Consequently, the additional i and q signals are generated in the summer 116 and subtractor 118 before application to the limiters.

In summary, an efficient zero-crossing generator utilizes a summer and a subtractor to generate additional zero crossings. In general, zero crossings provide vital information on the phase of the signal. Indeed, by generating zero crossings one could readily recover the transmitted information signal. Additional zero crossings are highly desired in systems particularly those having low modulation indices. Instead of mixing a received signal with a local oscillator utilizing additional mixers and phase shifters, the present invention generates the additional zero crossings by adding and subtracting the initial in-phase and quadrature components. This manipulation of the mixer signal is highly efficient and may be accomplished with minimum additional components. It is known that the summation or the subtraction of unit quadrature vectors results in a coefficient (amplitude) other than one (i.e. Vϊ). Due to this change in the amplitude, it is required to scale the components to a point where the resultant vector has an amplitude of one.

FIG. 3 shows an analog circuit technique using operational amplifiers which can be used to accomplish the needed inversions, summations and magnitude scaling. The i and q components are inverted via inverters 302 and 304. The inverted signals are then added and scaled at the summer 306. The output of this summer is the second quadrature signal qι(t). A second summer 308 adds and scales the inverted i and q signals to produce the first in-phase component ii (t) (scaled). It is noted that although devices 306 and 308 are both adders they provide summing and subtracting functions, respectively by inversion actions that take place on the signals before they are applied thereto. As mentioned, since the manipulation of i and q signals is implemented in vector forms some

scaling is necessary, particularly if additional in-phase and quadrature signals are desired beyond ii (t) and qι(t).

In alternative embodiments of the present invention additional in- phase and quadrature signals may be generated to improve the performance of the demodulator by the generation of more zero crossings. Additional signals are particularly helpful in systems having a small modulation index. These additional signals are generated by continuing the progression of the summer/subtractor circuit. FIG. 4 shows a specific case where four additional vectors, i2(t), q2(t), i3(t), q ₃ (t) are generated. These vectors are used in the phase domain to represent the additional in- phase and quadrature signals.

FIG. 5 shows the circuit implementation for the additional vectors. The differing values of resistors used in the circuits are for the purpose of scaling the input phase components. As mentioned, this scaling is necessary to produce vectors with uniform amplitudes. Obviously, the progression could be continued indefinitely to generate ever more vectors. Generally, it is desirable to progress the number of i and q vectors by powers of 2 so that the vectors are all equally spaced. Thus we go from the original 2 vectors (i(t), q(t)) to 4 vectors (i(t), q(t), iι(t), qι(t)) to 8 vectors, (i(t), q(t), iι(t), qι(t)), ijjtt), qsstt)), i ₃ (t), q ₃ (t)) etc.

Other circuit techniques, using fully differential operational amplifiers or differential operational transconductance amplifiers may be used to efficiently generate additional zero crossings. Current mode summing can be used as well. These various other circuit topologies have their unique advantages and disadvantages and can be employed to meet certain other requirements.

A significant benefit of the present invention is the elimination of additional mixers, phase shifters and filters as suggested by the prior art. The elimination of these additional components results in significant current savings which is highly desired in portable communication devices. The summer and the subtractor which provide the additional in- phase and quadrature components are traditionally low current consuming devices as compared to mixers.

An additional benefit of the present invention is the ease with which the number of in-phase and quadrature components may be increased beyond those suggested by the prior art. By adding and subtracting the original in-phase and quadrature components from the output signal of the

summer and subtractor one could generate additional i and q signals with significant ease. These additional components produce even more zero- crossing points which would in turn provide for very narrowband communication.

What is claimed is: