Access Control Apparatus

The invention relates to an access control system for access to a secure area and specifically to the use of a mobile communication device with a colour display to transmit a specific code that is used for access control, to a reader for accepting or rejecting the code transmitted. The invention utilises Visible Light Communication (VLC) and envisages the use of, but is not limited to, Computer/Laptop or Smartphone or PDU or iPad™ or Tablet or Pad or Palmtop or Phablet as a VLC transmitter.

Access control at its simplest is a user provided with a single key that permits access to a secure area, by the action of, for example, opening a door. Multiple users can be provided with the same key, reducing the level of security, or each user can be provided with their own unique key thus maintaining high security levels. The access door or controller recognises each unique key that is approved or has been given access rights and the system can grow with the number of users. If needed, the complexity of the unique keys can be increased to increase the robustness of the security. Transmission of the key can by physical means, electronic means for example with a push button pad for key entry by a user, or by other means such as audio. Unlike many transmission systems, one for access rights should respond with an entry allowance or refusal within a few seconds to avoid user frustration and delay. This can present technical challenges and the present invention seeks to improve the existing access systems.

According to an aspect of the present invention there is provided an access controller code transmission device comprising a mobile communication apparatus with a colour display, the transmission device arranged to use visible light to communicate an access code to an access code reader and controller controlling access to a secure area, the transmission device comprising; a calibration portion, a synchronisation portion arranged so as to provide synchronisation with an access code reader, a generator generating a data telegram and a transmission portion arranged to transmit an access code in the form of at least one data telegram.

This invention intends to use a Computer/Laptop or Smartphone or PDU or iPad™ or Tablet or Pad or Palmtop or Phablet as a VLC transmitter, for transmission of a specific code used for access control. This type of device has been referred to throughout the specification as a mobile communication apparatus with a colour display. Although digital as well as analogue VLC is known and there are existing examples of two way VLC communication, the transmission of an access code with VLC presents particular issues to be addressed. Some of the issues are highlighted here at points 1 to 7 below;

1 . The said transmitters may need to employ a coding and a method to identify the device and/or the user of the device so that access rights can be confirmed or rejected. This coding may require a level of security. The transmitting device may also be required to receive data from the access control system or from an administrator so that encryption/ciphering can be made more robust.

2. All Visual Display Screens will have a refresh cycle, which is typically 60Hz.

Therefore the transmission rate of a colour is fixed and static for the duration of one refresh time, and can only transmit at a speed dictated by this refresh rate. If 3-bit colour is used, then a data rate maximum of 180-bits per second can be transmitted at 60Hz. Addressing the synchronisation issue could be useful so that the bits or colour codes presented are not ignored, dropped or unread, which is difficult to achieve with a single point detector and/or a detector with a given operational response time. So the transmission speed may have to be reduced by a given fraction in order to maintain a stable or adequate bit time. In many other cases, we have found that the assumption has been made that the transmitting light source is an LED or RGB LED with exacting control over its on- off state for each bit time, thus no indication is given for the rate of transmission.

3. The most significant problem is that even though colour standardisation is in

place for most devices, the actual colours from each device differ, which is not a significant problem for RGB detection, but it becomes more problematic for chrominance discerning detection. So there will be a trade-off between the amount of data that can be sent, and the accuracy of the detection, whenever differing light sources are to be used.

4. Synchronisation may also present a further problem, where all prior art has not taken into account that bit times can vary due to framing errors, which of course may be specific to the computer devices used, rather than dedicated optical transmitter drivers. This becomes more problematic when several exposures of light colours have an identical digital code for example, a succession of green codes that follow each other due to potential timing errors.

5. Should a specific point RGB photodiode or phototransistor be used for the

receiver, then from a given screen, the overall colour, which is the average of each pixel, must be detected, where the apparatus can only work if the detector is looking at one RGB pixel, or whole numbers of pixels thereon after, until there is a more substantial average where a percentage of one pixel error or more, is less significant.

6. The use of a camera for colour detection is also subject to synchronisation

problems, where this is an impractical solution if the camera scans at a similar rate to the refresh rate of the transmitter, or more problematic if it scans at a standard rate, which is slower than the screen refresh rate of most devices. To use a camera will require high speed technology with advanced optics, which, for some applications, could be acceptable, however, for mainstream door access control readers, the cost of such camera technology and software would be a disadvantage compared to near field communication (NFC) solutions that already exist. Equally, the higher power demand, to support a high speed camera with the complex software may to too restrictive.

For chrominance detection, the camera will need to be able to discern minute changes in colour or colour contrast accurately and repeatedly - which is not possible with lower cost cameras, particularly in lit areas, where the ambient light will have an overriding impact.

7. The transmitting device, if held by hand may tilt or skew or may not be located at the sensor or detector, and proximity may leave room for ambient light to affect the reading.

In the apparatus envisaged there is a VLC transmitter with both calibration, to address the issue with photodiodes and other photo-detector having differing responses and 'reading' and emitting different colours, and synchronisation so that both the

transmission, and later, the receiving of the VLC communication is able to be in step with the VLC reader. A telegram is generated, and here a telegram is a term used to describe a data packet comprising a preamble, which may include security, the data of the access code and a synchronisation code. Transmission of the data telegram is also arranged.

Further features of the invention are set out in the claims, specific details of which are in the dependent claims. In an embodiment the colour display of an access code transmission device is arranged to provide colour for transmission chosen from the range of; at least two RGB colours, chrominance, a chrominance pattern, adaptable digital chrominance pattern. In an embodiment the calibration portion of the access code transmission device comprises electronic noise filtering apparatus and can also comprise apparatus arranged to effect amplification, digitisation and comparison steps for subsequent identification of a colour transmitted. The synchronisation portion of an access code transmission device in an embodiment is comprises apparatus arranged to effect bit transition synchronisation. In an embodiment there is an access code transmission device, wherein the synchronisation portion is arranged to utilise one of the colours in the colour display for clock synchronisation. One of the colours may be selected from the range; red, blue and green. In an embodiment the colour display comprises two or more colour zones for the code transmission device, and may include an orientation portion for alignment of the two or more colour zones.

According to a further aspect of the present invention, there is provided an access controller code reader arranged to receive an access code from a transmission device comprising a mobile communication apparatus with a colour display using visible light to communicate an access code, the access code being readable by a controller controlling access to a secure area, the code reader comprising; a detection portion arranged to detect calibrated colour data, a synchronisation and matching portion arranged so as to interpret calibrated colour data in the form of a data telegram, a transmission portion arranged to transmit a successful code match of a data telegram to an access controller so as to allow access to a controlled area.

In an embodiment of an access controller code reader the detection portion comprises a photodiode or a photo transistor. In an embodiment an access controller code reader further comprises an audio receiver device arranged to receive an audio signal from a transmission device. In an embodiment of an access controller code reader there may be an audio transmission device for transmitting an audio signal to the access code transmission device. In an embodiment, an access controller code reader may further comprise a key transmission and receiver component, wherein the key may be an audio signal. The access controller code reader of an embodiment may include presence detection apparatus arranged to detect one or more of the characteristics; presence, location and orientation of access code transmission device. In one example

embodiment an access controller code reader having presence detection apparatus comprises a pseudo finger device arranged with capacitive properties to mimic human touch and arranged so as to detect presence in response to the contact of the transmission device with the code reader and arranged so as to cause the transmission device to transmit and the code reader to respond with an acknowledgement signal.

According to a further aspect of the present invention there is provided an access controller arranged to control access to a secure area in response to an acceptable access code, the controller comprising an access code transmission device, an access controller code reader, a comparator apparatus arranged to conduct a compare operation with an access code submitted and output a result, a check apparatus arranged to release a lock and allow access to the secure area in response to the comparison, wherein the access code transmission device comprises a mobile communications apparatus with a colour display and the transmission device is arranged to use visible light to communicate an access code to an access controller code reader. In an embodiment an access controller may comprise near field

communication between the access code transmission device and the access controller code reader.

According to further aspects of the present invention there is provided a method of operation of the access code transmitter and the controller code reader.

Embodiments of the invention will now be described in detail, by way of example only, with reference to the accompanying drawings, in which:

Figure 1 is a schematic illustration of the code transmitting and code reading/ detection system according to an embodiment of the invention; Figure 2 shows graphical plots illustrating the response for the same colour sequence, of colour Hexadecimal numbers, but using two different screen types - Type A and Type B;

Figure 3 is a block diagram of a digital chrominance method of operation;

Figure 4 is a block diagram illustrating summation, averaging and a reference means;

Figure 5 illustrates an example of the principle of synchronous data transmission;

Figure 6, like Figure 4, shows colours for a transmission device, where a summed and average reference means for detection is used;

Figure 7 illustrates another method for clock timing in synchronous transmission;

Figure 8 illustrates a stepped sweep across a linear chrominance bar, for a given screen type;

Figure 9 shows a graph of response of the colours blue, green and red;

Figure 10a, 10b and 10c illustrate the interaction of RGB photodiodes where: 10a shows the relationship between Green and Red where Blue is not introduced, 10b shows the relationship between Green and Blue where Red is not introduced, and 10c shows the relationship between Red and Blue where Green is not introduced;

Figure 1 1 , shows a response of a voltage (mV) in response to Blue, with a linear setting between 0, and point 5 on the X axis;

Figure 12 illustrates a practical measurement of the RGB photodiode;

Figure 13 illustrates the same measurement test as in Figure 12 but increasing only the colour red;

Figure 14 shows the stages of operation of the invention as a process flow chart; Figure 15 shows an illustration of modulation when a fixed colour is present on the display;

Figure 16 illustrates one variation of colour detection using two photodiodes;

Figure 17 illustrates a data shape or data sweep over time with a limited number of colours;

Figure 18 illustrates a data shape or data sweep over time with a number of colours and a more complex 3D key;

Figure 19 shows elements of the invention for transmitting colour;

Figure 20a shows a colour display for transmission with multiple colour zones, specifically 4 zones;

Figure 20b shows the device and display of Figure 20a with realignment;

Figure 21 illustrates one embodiment of the invention, including a sonic or ultrasound output;

Figure 22 shows a block schematic for the code reader of the invention; and

Figure 23 illustrates an enhancement of the embodiment illustrated in Figure 21 , where RF (radio frequency) or NF (near field) communication is included.

In this invention, a RGB photo-detector, such as a RGB photodiode or phototransistor or a bi-colour version, or discrete detectors, which have peak sensitivity to Red (R), Green (G) and Blue (B) light, is used. Some background to the invention and the use of RGB detectors is provided.

RGB photodiodes can vary in spectral peak sensitivity and amplitude, and depending on the colours revealed to the detectors, wavelength sensitivity overlaps are significant enough to prevent accurate and repeatable primary and secondary colour detection, based on a variable transmitter technologies, spectra and settings. To give an example this in HEX code, presenting a Blue light (#0000FF), to a RGB sensor, will output a given voltage or current from the Blue diode. Increasing the Green light (#00++FF) will not only increase the output from the Green diode, but also the Blue diode, and the Red diode. This is called 'bleeding'. This relationship is the same, but at differing levels of 'bleeding', over the entire combination of the three colours. So in reality, although it is possible to add or change the colours to enable a 3-bit binary code to be derived from a single displayed colour, made up of Black or White (which must be used for true 3-bit transmission), Red, Green, Blue, Cyan, Magenta or Yellow, a degree of matching or calibration may be required.

Using a 3-bit example, where 3 bits are used to give each colour a unique code, is as follows:

Black = 000

Red = 001

Green = 010

Blue = 01 1

Cyan = 100

Magenta = 101

Yellow = 1 10

White = 1 1 1

This set of data is subject to amplitude and transmission accuracy.To deal with the amplitude, one method according to the invention is to use the following normalisation solution:

White can be used as a digit if the calibration is performed at a White screen to normalise the diode outputs, or for normalisation. In this case, the RGB colours are much closer to each other, and are all higher, or high in overall voltage. Equally, black can be used as a digit where the colour's voltage will be at or near Zero (allowing for 'dark current' and/or ambient light leakage). Of course, the logic after normalisation or equalisation (calibration if required) can be derived from the analogue model, where a reference voltage is established by the averaging sum:

Vref = (∑V _R V _G VB )/3

Each RGB diode Voltage/s above this reference (V _re f) is/are the active primary colour or colours, and voltage/s below are the inactive primary colours, and therefore RGBCMY + Black and White can be discerned, if White and/or black are discernible, by other means such as similarity or differences with respect to V _re f and/or a given set point/s.

Figure 2 illustrates the above argument for two differing screen types - Type A and Type B. In this case, the reference is normalised to Zero Percent (X axis) - thereon, a positive percentage is the active primary colour, and a negative percentage is an inactive primary colour of R, G or B. As can be seen, the variance between screen types, if viewed in isolation would not appear different, however, the difference can be seen, and the ability to discern the primary and secondary colours is achievable, but could be only just discernible. Further variations may not be so well defined or as easily discernible, particularly if ambient light is introduced, which may, for LED lighting, tend towards the Blue spectrum. However, with certainty, the differences between the two screens is indicative that further colours will not easily be discernible using this technique, which is limited to true primary and secondary colours:

The true primary and secondary colours are the exact RGB balances or ΌΟ', 'FF' Hex numbers. However, for better assessment of the colours, with a tighter defined difference between the RGB on-off states, the true primary colours may have a slight adjustment to include a small amount of another colour for example. The voltage logic (on-off comparison) is better defined by the following, for one given average display type e.g. below are colours on an Apple iPhone™, which may also be applicable to an extent, for other smartphone models:

Colour HEX

Yellow D5FF00

Magenta D700FF

Cyan 00FFEC

Green 36FF00

Red FF0065

Blue 6200FF The normalisation, in effect, cancels out the effect of colour bleeding between the RGB diode sensitivity curves, which differs slightly from each other, and it can be a useful function to introduce for better or more defined detection.

Photodiodes are also very noisy because their outputs are low in voltage or current, particularly for detecting light from a said device screen. In this case, it presents an issue for any system, because a screen refresh rate is at or near to the same frequency of a 1 10/230VAC supply, which will cause 'hum pick-up' (EMI) if the circuit is not suitably screened - however, the optical parts cannot be shielded. Therefore, filtering out noise at this frequency will also attenuate the data. So design care and a certain amount of hum cancelation may be required, as well as using higher frequency noise attenuation to eliminate RFI. Good supply isolation may also be required to reduce low frequency pick up.

The RGB balance of a RGB photodiode also varies according to temperature and/or manufacturing variations, with Blue being always the lower in amplitude due to the lower 'screen' energy in this spectrum, and lower semiconductor sensitivity. Coupled with a difference in the said screen spectrum and amplitude, with ambient light affects, the outputs from the RGB didoes cannot be used to accurately discern colours outside the primary and secondary colour range, and in some instances, even these colours may not be accurately and repeatedly discerned, particularly with some screen setting adjustments by the user, coupled with difference in the screen technology. To example this some screens can be set to a warm or cool colour as a user preference, where the colours can be significantly different.

Therefore, in some cases, the numbers of detectable colours could reduce the bit- number magnitude (i.e. a 3-bit sequence may only include a count of 4 or 6, and not 8). If on the other hand, a more repeatable system can be used to eliminate the expected differences, then with RGB detection, it is possible for a count of thousands to be derived, which is more than 3-bits, and up to a possible maximum of 24-bits

(#RRGGBB).

Therefore the second part of this invention is to first detect what the colour balance is set to, for at least one of the colours, up to all colours (White), and to include no colours (Black), so that it can discern the primary and secondary colours more accurately and repeatedly across all transmitter models. This, at a primitive level, could be a calibration or amplification adjustment, be it fixed or variable, at the manufacturing stage to equalise the colours to give a common voltage level when a screen, of a typical device is displaying White (#FFFFFF), whereon the colour discerning algorithm is conducted as described above.

However, this is not exact, because the calibration is only performed with one type of display and one type of setting, where even the same models in a series from a manufacturer can vary.

A third embodiment of this invention will comprise a calibration method, performed in real time, with the actual presented transmitting device used for the calibration source. In this case, the calibration can be conducted by first displaying one or more known colour/s, to include black and/or white, that is known to both the transmitting device and the receiving detecting device for example, the smartphone displays a White colour, and the detecting device knows that this is a White colour, so it will normalise the voltages from each diode so that the colours can be discerned more accurately as described above.

This is different to common meaning of equalisation because it is normalisation for a reference transmission or calibration transmission, which is known, which may comprise a preamble for serial data conversion, and once it is detected and set, it remains fixed at that setting for the duration of the telegram, which is sent in less than one second, typically, and which is acceptable for access control use, which is not a continuous transmitter. This invention's type of normalisation is adaptive only for non-data.

However, it can be demonstrated that a different screen or device, even when calibrated to White, will not show the same ratios for other colours and/or other colour

combinations. Equally, the ratios of other inactive colours can also change, where the logic of 'equalisation' does not work, so it will still be limited to a few colours more, beyond the primary and secondary colours. This can also be down to amplification variances of the sensor, and the difference or interaction between each diode, and the temperature it is at, as well as he offset errors of any amplification. So the technology solution requirements will be limited to a given RGB sensor type, where a lower cost unmatched RGB sensor will require a different approach with specific problems to be solved.

Specifically, if a given reading means, described below, is employed, the importance of the normalisation means may be less, to a point of not requiring normalisation, where it can operate without normalisation or in some cases, without equalisation. One approach to this is to perform a calibration for a single average screen type, or to perform differing amplification levels for each RGB diodes using fixed resistances or variable resistances and/or software adjusting the levels at the possessor stage. To reduce or eliminate equalisation, then either the colour variances may decrease

(meaning a smaller number) and/or the system must be calibrated using a calibration phase during the use of a given transmitting device:

A fourth part of the invention can take the real time calibration method a stage further. This embodiment is required where longer security codes need to be sent, and the telegram should last no longer than one second. In this case, it can be solved in two ways, using one or the other or both:

The first way is to increase the number of detectable colours above the primary and secondary colours, and this can be achieve in a combination of certain ways or different ways:

1 A calibration phase as described above can present some of the colours, up to all of the expected colours, in a predefined way for example, first a White colour is presented, which may be a preamble of Black and White, and a snap-shot of the RGB diode output voltages are taken for either normalisation (or this may be a pre-calibrated level), or as an un-calibrated snapshot. This snapshot is the ratio levels of RGB voltages and it is the pattern representing White where if this ratio occurs during a second presentation of White during the data telegram transmission phase, it will be repeated, to within acceptable tolerances, and therefore identifiable as White (#FFFFFF), by comparing the calibration snapshot to the presented data readings. The calibration run can be repeated for more colours, and up to the amount of colours required to be adequately discerned, where more repeatable colour combination will not necessarily require a calibration snapshot, it can be estimated with a fair degree of accuracy if the performance of the RGB diode is known and typical. This will require a database of RGB voltages and responding colours to be stored during calibration so that it can be compared during the data transmission. The voltages may be further coded as a percentage number based on an average so that the numbers are compressed (bias removed). The other method is to measure a few readings of colours and/or colour combinations, then understanding the performance of the RGB diode detectors, and the relationship between them, an adaptive quasi-analogue chrominance graph can be established or calculated based on the actual transmitting device's colours or colour interpretation, to a great or lesser accuracy, depending on the number of calibration samples given, to a point where the calibration time is unacceptably long, based on the transmission speed. Furthermore, luminance measurement and subsequent equalisation or normalisation is not required, because this has been 'calibrated out' during the calibration phase. Again, this is achievable because of the short time required to transmit the telegram, and the low probability that that changes can occur during this time. Also, the RGB diode's linearity, repeatability, short term stability and interaction will be similar for its calibration envelope and between devices.

In practice, a true chrominance graph with or without luminance measurement will be distorted to account for the weaker responses at certain colour frequencies and amplitudes, as well as an increase in noise at lower amplitudes and inter-colour 'bleeding'. This is an important factor to consider, so variation in the colour transmission, as well as detection algorithms may be more variable with certain colours, at certain amplitudes.

The embodiment of this invention does not use a 'shape' for detection, which is an analogue way of performing the chrominance comparison. Instead, it uses a comparison method to discern any deviation of a displayed colour from a snap-shot of a previously displayed colour, used during the calibration phase, or a mathematically calculated/estimated/projected snap-shot that is calculated through empirical means, where the snap-shots are specific to a given display type, and are adapted to be able to detect the colours of a specific screen that is in use at that time. Therefore, the chrominance graph is adjusted each time, and therefore, adaptive.

In reality, once the colours have been detected, the chrominance detection method is inconsequential; the individual diode output voltages do not have to be labelled as a colour, where the colour becomes irrelevant. In other words, the voltages now become an adequately repeatable, predictable, variable, ratio of the analogue inputs. Furthermore, because of the colour bleeding of each of the RGB diodes, the relationship of the sensor outputs differs from the actual chrominance of the colour presented. As a consideration, the Red, Green and Blue diodes can be swapped around but still communicate in the same way, and so a 'colour shape' does not have to be used.

As described above, figure 2 is indicative that screen variations in colour wavelength and/or colour intensity will occur, compounded by variations in the detection circuit, where this will make a static chrominance method of detection far less reliable to a point of not workable.

Another way may encompass any one or both of the above approaches, but in this case, the system uses not one, but more than one colour source, channel or zone for the same transmitted data, and more than one RGB detector for the same received data, which in essence splits the transmitted data into more than three primary colours described in all prior art. In this case, alignment of the colour sources or independently controllable colour zones within the same device display, will be aligned with each relevant RGB detector.

Different shades of Grey may also be considered for the coding, which will be the RGB levels set to first White, and then R+G+B is equally decreased in intensity.

When the RGB diode outputs a voltage for the respective display colour, and it has stabilised, then this voltage is read, or the average of the stabilised voltage is read, then digitised.

In this approach, and possibly in other approaches, it must be understood that for each colour display or frame, the timing of the display can vary, not only on a frame by frame basis, bust also during the actual screen refresh, and this is further compounded by differing response times of the circuits for each colour due to diode response times, filtering, amplifier response times and processing time or timing. This could be further worsened with more than one colour zones as described above. Therefore, when the signals are digitised, there will be a degree of bit detection delay, which may vary across all bits. This may or may not affect the ability to communicate in an

asynchronous way as opposed to a synchronous way, which is important to understand: A further approach of this invention is to accurately transpose the analogue detection of the detected data into a digital format, convert this into a serial format and send this to an access controller, be it synchronous, or asynchronous. In this case, the serial format will be suitable for access control integration. The access control system may also send data back to the reader, but it is not intended to use the same approach to the

transmitter i.e. smartphone, unless there is a requirement, meaning that the detector will also be a transmitter to perform duplex or half duplex VLC.

Bit synchronisation will now be described. For different zones, all data will not be in exact timing synchronisation, and the analogue digits (levels) may comprise a slightly variable mark-space or mark-mark or space-space ratio which is down to the device's software and/or interruptions and/or the screen refresh rate and timing if not

synchronised to V-SYNC and/or the relevant response times of the colours and the filter/detection circuit and so on. This is sometimes known as 'wow and flutter'.

[Although in data terms, this is often called wander and/or jitter]. So to achieve synchronization and to allow for wow and flutter, a discrete clock can also be

transmitted. This clock can be synchronised or embedded with the actual colour sequence, requiring one bit, or it can comprise a separate screen zone, of one colour, for example, Black-White on-off or any other RGB colour sequence, thus allowing one bit for timing, and one or two additional data bits. The clock sensor may comprise any one of the above said sensor technologies or may comprise a visible light sensitive phototransistor:

Of course, synchronization of the individual bits can be achieved because the sensor voltage transitions, due to the colour changes, and is therefore detectable, but this will assume that the colours change at each displayed colour, which is more probable, but for where a series of 1 's, for example, is transmitted at one or more colour zones, one after the other, the transition will no longer be there, and therefore no longer detectable unless there is a reference that does change. This is generally not an issue for asynchronous transmission, where as long as the transmission clock or time is adequately stable, then telegram detection can be successful. This can be avoided by always sending a different colour, or at least one colour zone is changed, at each bit time - but is then does reduce the amount of data that can be transmitted, and it is more difficult to programme. There may be indication of a colour transition between identical Bytes (colours) which are seen at each screen refresh. This may or may not be detectable, however if it is detectable, then any clock signal or identifier bit may not be required. However, this will relay on synchronising the bit rate to the screen refresh time.

Again, if the transmission time, accounting for screen refresh times, is stable or predictable, the asynchronous detection can be used. If not, then a clock means of some sort will need to be included in the transmission code if there is likely to be a succession of identical colour codes or digital Bytes.

Security aspects will now be described.

Specifically, the colour-data sequence or code transmitted to the receiver will be different for all transmitters and/or for all transmitter operators and/or for every transmission. This is so that a device and/or a device user can be identified as the device and/or user with access rights, or without access rights. This means that each time a telegram is sent, it is unique to that transmitting device and/or user. This may require methods for securing the codes from being copied. The security may also require means for the reader or the access control system to respond or send data to the transmitting device or at least synchronising other aspects with the device such as Time/Date and so on. This communication direction may also be used to send updated encryption or cipher codes. This may be achieved by sending data back to the transmitting device in a similar optical (VLC) manner, where the device must be fitted with a camera. However, the said system may be able to use other wireless means to communicate data to the transmitting device for example, by the use of Wi-Fi,

Bluetooth, 3G and so on, which is the typical communication link with a smartphone and so on, and it is a standard feature that can be used with no additional hardware requirements for the transmitting device or the receiving device.

It is possible to set each reading device with a given random, known or programmable offset between each of the RGB voltage levels or amplification of each manufactured reader. The purpose of this means is that if a criminal were to obtain a reader, and was able to crack the codes, and test this with the reader, then presenting this to another reader, will mean that the code is different in terms of light level ratios. Therefore, when the criminal's phone is presented to the actual reader, the levels for that given identified code will be different. If it is different enough, the control system could discover this, and either block access rights, or it may issue/ask for an additional alternative

challenge, such as another PIN, which must be responded to. For example, there may be a challenge LED on the reader that illuminates, and in this case, only by keying into the phone, or reader keypad a new PIN, in perhaps real-time, the challenge can be answered.

This challenge feature could also be used in its own right, in that for additional security, where a slightly longer time at the door is less critical, a secondary PIN would be an advantage. This means that first the entry into the smartphone 'app' is gained through a password or PIN if required, then at the reader, a further password or PIN, that is, or may be different, is required. This can be performed on the smartphone and/or if required at an alternative/additional reader which has a PIN entry keyboard/pad. This may be further advanced by requiring a smartcard reader, which would provide a very secure system comprising of more than one technology or method.

A further embodiment of this invention comprises means for differing methods of compensation and/or correction, where a thermal sensor may be used to correct for analogue drift over a wide ambient temperature range, or to assist calibration or reading at a given temperature, where this can also be used as part of the arithmetic function for the above said algorithm or summation/amplification. In addition, a further White or wider wavelength light sensitive sensor may be positioned so that it is at the same zone as the RGB sensor, or at a separate zone, or in one of more of the said more than one zone, where this is used for normalising the colour amplitudes or as a normalising reference, or a reference and so on. It can also be used for compensation and correction and so on.

Transmitter to detector Alignment will now be described.

Guides/markers on the reader and the phone screen may also be provided so that alignment of one or more zones on the screen will be aligned to the respective sensors. Misalignment could be announced using a sound or indicator at the reader if alignment is out; where this may require say three zones, where the colours are different at each zone when presented, and enough not to require a calibration at that point.

This invention deviates from known devices as follows: The transmitter does not make a 'shape' from the inputted data. The transmitter merely adjusts the Hexadecimal values of the RGB light pixels, the same for all pixels in a given zone, according to the inputted data.

The receiver then uses the voltages from the RGB sensor, that is in relatively close contact with the transmitter, which are related to the said Hexadecimal codes and compares them to a database and/or arithmetic function and/or reference voltage/s of a given type/s, where these voltages will be interactional with each other depending on the colours presented, and often different, depending on the transmitter technology and settings.

Transmitter detection will now be described.

It may be necessary for the reader to detect the presence of the transmitting device and/or to set the transmitting device into communication. This can be achieved in various ways such as hall-effect or capacitance sensors or light-beam breaking optical sensors or even micro-switches and so on. For the reader to communicate with the device, this will require further communication routes such as audio triggers or light triggers to be picked up by a smartphone's camera. However, this will be difficult because the cameras can be in differing locations. However, this can be avoided if:

1 . The screen 'key' is continually cycled for a defined number of times or until cancel.

2. When the transmitter is in position with the reader, an on-screen button is

depressed to initiation the sequence for one or more given cycles.

Pixel to detector aperture alignment will now be described.

Importantly, where colour detection resolution is high, and a RGB photodiode sensor is to be used to detect the colours on a device display, comprising many thousands of pixels closely placed together, and the distance from the transmitter and detector is down to near 0mm, then the chrominance method will be less accurate or possibly unworkable if an exact number of RGB pixels is not in alignment with the detector. This is because the detection of Red, for example, is dependent on seeing all Red pixels when all pixels are in view of the sensor's red zone. If ten pixels are partly exposed to the Red detector where in total, only nine Red pixels are directly exposed, then the Red measurement error will be -10%. Therefore, in some cases, the resolution will suit this digitised approach, that is less analogue, and comprises a tolerance based measurement.

To reduce the number of missed or aliased pixels, the sensor may be placed at a greater distance than 0mm form the screen or the transmitter and/or a lens may be used to focus a larger area of the screen to the smaller sensor area. This could be achieved by lens and/or reflector and/or Fresnel lens and/or by partitioning with a white or respective partition to that the light is scattered and diffused.

Device matching will now be described.

A pre-calibration run may not be required if at first, the chrominance difference of the transmitting device is corrected. This may be performed by modifying a reader, with the same performance as all other readers, to be able to communicate with the user or the phone so that at each calibration point, the ratios of colours are set. This requires an app that can adjust the colour ratios so that it matches a given reference or standard. Once this is performed, the device will no longer require calibration, or only calibration checks.

Figures and embodiments will now be described.

Figure 1 illustrates an example of the transmitting and detection system comprising a transmitting device 1 , which may be a smartphone, in visual contact with a RGB sensor 2 or sensors 2n, where each colour's signal is amplified 3, 4, 5 by a similar amount, or a predetermined differing amount, or an adjustable amount, and at any point, may comprise electronic noise filtering (not shown). The amplified signals are then digitised and calculated by a processing unit and/or comparator 6 so that each colour presented can be identified. The part of the sensor data from the first sensor 2, or a subsequent sensor 2n may be used for bit transition synchronisation. The data is then converted into a serial data stream 7, and passed to the access controller. The colour zone 8 or zones 8, 8n, will be aligned to each respective sensor 2, 2n. The more than one colour zones 8, 8n will output independent data and/or related data, such as a clock, but will be refreshed at the same time, within the refresh speed/direction time. With more than one colour zone, singular data telegrams from the transmitter software program, is split into two or more independent colour zones and then sent to the detection system in a parallel format. The parallel format is then collated by the processor 6 and then sent as the singular data telegram to the access control system. More than one zones, will increase the data rate/size that is transmitted and/or increase the data detection accuracy.

Figure 2 illustrates the response for the same colour sequence, of exacting colour Hexadecimal numbers, but using two different screen types - Type A and Type B. In this case, the reference is normalised to Zero Percent (X axis) - thereon, a positive percentage is the active primary colour, and a negative percentage is an inactive primary colour or RGB. As can be seen, the variance between screen types, if viewed in isolation would not appear different, however, the absolute difference can be seen, and the ability to discern the primary and secondary colours is achievable, but could be only just discernible. Further variations may not be so well defined or as easily discernible, particularly if ambient light is introduced, which may, for LED lighting, tend towards the Blue spectrum. However, with certainty, the differences between the two screens is indicative that further colours will not be discernible using this technique without a matching calibration, where this is limited to true primary and secondary colours. Figure 3 illustrates the preferred embodiment, and describes the example of an adaptable digital chrominance method. In this example, the three colours from the RGB sensor 1 1 is fed into a processing device 10. On a calibration phase or sequence, the sequence of calibration colours is a known sequence, and at each display of a given calibration colour, a snap-shot of the voltages corresponding to that given calibration colour is stored in a database 9. Empirical data on how the RGB diode voltages perform and/or relate to each other, by way of an algorithm and/or empirical data base, may also be stored in this database or be related to that given database, with the purpose of predicting or inferring what a snap-should would be, or is predicted to be, for a given presented data colour that is not taken as a snap-shot during the calibration phase - this could be performed using a one or more measured calibration colour/s and/or a mixture of colours, including white and/or black, then using an arithmetic function or functions, preform a measurement and comparison between the data colour presented, and the mathematical prediction of that colour, or the mathematical colours can run until the display colour is matched, in terms of numeric similarities.

For example, this is how an ADC works, using only a DAC. So this could be a data base look up table, or an analogue method of scanning through the possible

combinations, within a given tolerance or resolution, unto the actual colour and the calculated colour, match (with a given tolerance). This may be achievable to preform without an initial calibration, or a partial calibration, wherever, for better resolution, a calibration run or more than one colour will be required.

Figure 4 illustrates one aspect, which describes the summation, averaging and reference means. In this example, the RGB sensor 1 1 , has three outputs (shown as a single line), which are summed and averaged 12, whereon the summed and averaged output is fed into each comparator for R, G and Blue, as the reference voltage. Each R, G or B diode Voltage/s above this reference (V _re f) is/are the active primary colour or colours, and voltage/s below are the inactive primary colours, and therefore RGBCMY + Black and White can be discerned, if White and/or black are discernible, by other means such as similarity or differences with respect to V _re f. The output from the 3 comparators will provide a parallel 3-bit Byte.

This method will work only if the outputs from the sensor are equalised at say, White. This may be performed and calibrated at the manufacturing stage, or it may be calibrated and equalised in real time, when White is presented on the screen, which may be a preamble. Other secondary or primary colours may also be considered for calibration.

Figure 5 illustrates an example of the principle of synchronous data transmission and detection using two colour zones, in which the first zone is the RGB data 13, and the second zone is the clock data 14. In this case, the clock data is a single on-off colour or colour transition. The clock and data may vary in timing and/or duration and/or duty cycle, however, the variation will be true for both the clock and the data, so reading accuracy will be assured with significant wow-and-flutter (Jitter) present.

Figure 6 (also relates to figure 4), where a summed and average reference means for detection is used. In this case, it can be seen that all primary colours are singularly above the reference line Ref, and all dual primary coloured secondary colours are above the reference line Ref, apart from point 15 used to example that in this case, the Green primary sensor voltage and Blue sensor voltage are above the reference voltage Ref, stating that the colour is Cyan, which in error. To correct this error, the Blue diode voltage should be shifted downwards to a point where it is below the reference voltage at the Green display point. This may require a negative bias or a decrease in the Blue amplifier's gain. Figure 7 illustrates another method for clock timing in synchronous transmission, where a clock cycle 18 is normally seen and used from a separate colour zone to synchronise data bits 16 for applications where wow and flutter is encountered, and a succession of the same data bits or Bytes 16 is possible. However, this clock cycle uses up a potential useable bit and/or colour zone. If a single zone is to be used for RGB communication and wow and flutter demands that a synchronous code is transmitted, then one of the tree colours may be used to embed a clock in the data code 17.

However, because of the bleeding from one colour to another for RGB detector diodes, there will be a degree of change seen in the remaining diode voltages, which now comprises 2 bits, even when presenting the same digital information.

This can be carefully compensated for by knowing what the performance and reaction will be for each of the data code combinations, when the clock is either at a '1 ' or a Ό', and possibly, the level of the clock may be adjusted to be different in amplitude, meaning that it can be differentiated from the total measurements and/or it could be a method for separating the clock from the data, meaning that all three colours can still be used for data to maintain the same number size, or a reduced number size. In reality, it may be better to employ more than one colour zone, where a second colour zone can have just clock data, or use one of the two or three other colours of a RGB diode. Of course, the clock could be embedded in the transmitting device, where all colours are used for display, but where the colours are shown for each frame, every other one of the colours could increase or decrease by a known amount to signify a clock cycle change, providing this change can be discerned by the detector as a clocked change of intensity, rather than a colour change. This can be further enhanced by allowing one or more of the colours to change in a detectable way, and so on. Another method may be to insert a White or black byte after each colour data code, or to insert one in between each bit only where there are successive colour bits to be sent, where the detector will understand a clock time by changes in the data colour and/or a transmission of white or black, as a clock cycle.

For a given RGB zone, with an integrated clock, it is still possible to use the colour transmission codes up to a point where the data number is close to the number it would be with using the RCB system for just data. However, the chrominance resolution may be lower when a clock is embedded, but still achieving a high count, up to a point where the clock cannot be discerned from a legitimate data colour. The clock means may be emended in one zone or two zones and/or it may alternate between zones or be complimentary or contrasting between zones in a known or unknown way.

There are many ways to solve synchronous timing, and those skilled in the art will be able to choose the most appropriate solution to include how many zones are used, which colours represent what, and if a clock zone is to include data or not.

Figure 8 illustrates a stepped sweep across a linear chrominance bar, for a given screen type, in steps of #00, #40, #7F, #BF and #FF (0, 25, 50, 75 and 100%). In this case, the sweep, from left to right, is from Red to Green to Blue and back to Red, with White at the extremities. From this, it is clear that the RGB voltage positions at each step are adequately detectable, discernible and repeatable, which is shown by the Red colour at each end. This represents a count of about twenty one, as opposed to six, when using only the primary and secondary colours. This can be further extended with a given design, to increase the count, where the more significant step changes in voltage can accommodate a better resolution.

Further details of colour detection will now be described.

In other systems the chrominance chart or CIE 1931 Chromaticity Diagram is the means for detection of the specific colour transmitted. However, figure 9 illustrates the overlap of colours when using a standard RGB photodiode, which is for this invention's intended use. In this example it can be clearly seen that any output voltage level from the Green photodiode, for example, will introduce a certain level of output voltage from the Blue photodiode and Red photodiode to a lesser extent, thus tending towards Cyan and White, even though there is no Blue or Red component in the light.

Figure 10a, 10b and 10c illustrate the interaction of the RGB photodiodes where:

10a is the relationship between Green and Red where Blue is not introduced (set to #00) as shown on the ratio graph labelled 19.

10b is the relationship between Green and Blue where Red is not introduced (set to #00) as shown on the ratio graph labelled 20.

10c is the relationship between Red and Blue where Green is not introduced (set to #00) as shown on the ratio graph labelled 21.

The RGB Photodiodes may have differing peak spectral responses, differing 'Q' values, and differing amplitudes between each diode and diode batches, further worsened by differing amplifier performance and specification tolerances. Therefore it is clear that a different detection method will be required for this invention, because relating each of the Red, Green and Blue output voltages to a chrominance graph will not be indicative of the actual output colour provided by the light source, nor will it be repeatable for differing light sources: To compound this further, the light source will vary because this application is intended to be used with many differing light sources, which include variations in display technologies such as; LCD, LED, OLED, OMOLED or LCD with Backlit LED. And to compound this still further, the screen glass may not be optically pure, with a tinted colour (in some cases), the display setting may be set to a preferred mode, for example, Warm, Neutral or Cool, with emphasis on different parts of the spectrum, and finally, ambient light, which can be more intense than the screen intensity, can make its way between the screen and the sensor directly, or indirectly if it cannot be fully isolated, bearing in mind that the screen glass, if it is exposed to an external source, can internally reflect and disperse the light across the entire viewing surface of the screen, even if the sensor is shielded and in direct contact with the glass. Finally, each colour, of each pixel, may not be an exact RGB balance, because the screen driver may not modulate the pixels precisely, and the colour generating pixels can actually vary slightly in colour across the screen, and vary between the same model types, depending on the manufacturing batch.

The intended light source in the case of portable devices is limited in brightness and is not matched to the intended RGB detectors. Figure 1 1 shows the voltage (mV) response to Blue, with a linear setting between 0, at point 1 on the X axis, and 255 at point 5 on the X axis. It can be seen that the response to the light is non-linear, with better linearity occurring at a 50% setting and above. This performance will deviate from the luminance intensity part of the chrominance graph, where it will need to be taken into account if a higher resolution is required to include colour mixtures of less than 50% or so. Again, this can very between light sources.

The chrominance graph is not intended to display shades of Grey, spanning from Black to White (although White is included) which may offer additional data or higher detection resolution.

For detection and digitisation of the colours, the methods have been described earlier. In this case, a calibration is made first, or periodically, or continually on the same or on a differing zone, which may be an additional synchronisation clock so that the

transmitting device is mapped and calibrated with the detector. The requirement is to measure the actual absolute voltages of each of the RGB photodiodes and to compare this to a database that is derived from the calibration data and/or from an optional empirical database and/or derived formula/algorithm so that the intended specific transmitter data, which is converted to a specific colour, is received, assessed and reconverted back to the same transmitted data with no error, where this can be scrambled and descrambled at any point if required. This is achieved by measurement and/or performing relative and/or average calculations of the two or three photodiode voltages, with additional thermal compensation calculations for thermal drift correction, because the intended use of this invention may be outdoors or in areas of high temperature differentials. It may also be calibrated or calculated against a reference colour and reference photo diode, which may be White sensitive or a specific colour sensitive, or more than one colour sensitive or RGB sensitive, which may reside on the same or a different colour zone, where the transmitter is transmitting a known colour, known to both the transmitter and the receiver.

The differentiator, of figure 12, is explained here.

Figure 12 illustrates a practical measurement of the RGB photodiode, where in this case, only GREEN (the uppermost curve) is increased in value, from #00 to #FF in linear incremental steps. It is clear that the chrominance theory suggests that from looking at this graph, that RED and BLUE are also present in the colour, which will resembles a light green. It can be argued that this is merely a change in luminance, but Figure 13 counters this argument using the same test but increasing only RED. In this case, the suggested colour is nearer to ORANGE.

It is also clear that the relationship of the colours, when the RGB photodiode is used in photovoltaic mode, is not linear. With a linear step change on the transmitting device, the output response is polynomial. Figure 12, upper curve (GREEN) has a polynomial approximation of (taken from the plot):

y = 0.1 17x ⁶ - 3.4259x ⁵ + 37.439x ⁴ - 195.6x ³ + 575.29x ² - 795.2x + 533.78 So there is a significant difference between using a chrominance graph, and the actual measured value to represent the same sourced colour. Therefore, one embodiment of this invention is the need to measure the RGB diode output voltages individually, and to compare the values to a look up table of the same values taken from a snap-shot of the same colour during calibration. The differences may be compared between a given error band so that any slight deviations due to noise and so on, can be ignored. This will mean that, depending on the tolerance, the measurement resolution may be more course than the transmitted resolution, which is typically 1 in 255 at best.

The snap-shot may comprise absolute readings and/or average for all three readings and/or differences between the absolute reading and the average reading and so on, and it may also be a reading taken over a given time period to cancel out any in-band or near in-band noise, such a utility supply pick-up (50 or 60 or 100 or 120 Hz and their harmonics). Pick-up and noise levels are higher at lower light levels, where this may be taken into consideration when deciding the measurement range points, which are described below.

For high data telegrams, taking a snap-shot of all possible colour combinations could take an unacceptably long time, particularly if displaying through one zone. So another embodiment of this invention is to use an empirically derived formula to simulate the performance of the RGB didoes and the relationship between them. This way, if an output from the diodes is measured, it can be translated to the actual colour presented at the transmitting device:

From Figure 12, for a given detector circuit and a given smartphone, the relationship between the RGB number and the output voltage in mV is given by the following empirically derived polynomial equations (from the curves):

Green: y = 0.1 17x ⁶ - 3.4259x ⁵ + 37.439x ⁴ - 195.6x ³ + 575.29x ² - 795.2x + 533.78

Blue: y = 0.0501 x ⁴ - 0.2835x ³ + 20.579x ² - 59.921 x + 150.56

Red: y = -0.0594x ⁴ + 1 .7141 x ³ + 9.0659x ² - 33.708x + 90.833

Where xjs the R, G or B number between 000 (no primary colour) to 255 (full primary colour), and yjs the output amplifier voltage in mV. This output (y) is true only for an increase in Green and referenced or with respect to Green. It will be different for Red and Blue and combinations thereof.

It can be seen that this is not a chrominance assessment but simply a reconstruction of the intended transmitted colour from a Hexadecimal or RGB number, back to its true Hexadecimal or RGB number after it has been distorted by the transmitting device and/or the characteristic of the RGB detection means, and it is performed

mathematically against one polynomial or one or more interacting polynomials, and certainly not using 'shapes', which is the embodiment of any prior art, where in this case, the 'shape' transmitted will not be same as the 'shape' received from the detectors unless a significant amount of processing is applied with additional correction detectors and so on. Which will demand high processing power and high cost. Also, the Hexadecimal or RGB number will be further converted to a digital Byte/s.

However, for the above formula to work for transmitting devices and/or detectors that can deviate in colour and/or intensity and/or in linearity and repeatability, then

Calibration and/or equalisation must be performed before the actual data transmission can take place. This may require a lower number than the total number of possible colour combinations by a significant fraction, and may require a selected sequence of colours, which may also be embedded in the preamble of the telegram and so on. For example, the calibration sequence may comprise a White, then Red, then Green, then Blue sequence, with additional Yellow, Magenta and Cyan colours. Black may also be indisposed to eliminate dark-voltage and/or parasitic external light. This may include further colours if require and/or differing intensities. From this calibration run, the amplitude/gain levels or characteristic for each colour sensitive diode can be normalised or compressed or equalised for a reference template (i.e. the empirical data) using one or more of the following:

1 Microcontroller/processor software digital equalisation, offset, gain or

normalisation

2 Trans-impedance photo diode amplifier gain/offset adjustment using digital potentiometers

3 Software Polynomial parameter adjustment.

However, the requirement for a calibration run may be eliminated, or reduced still further when using one or more of the following inferred/real-time calibration/compensation techniques, with the choice of using one or more of the above calibration and/or empirical formula calculation techniques:

1 A second White light sensitive photodiode detector is used at the same colour point as the RGB detector, where a comparison is made between each R, G and B detector against the White light detector which will correct for transmitter amplitude differences.

2 This may also include the summation of the RGB levels, where this will be compared to the White Level, where this should be same and if not, the above amplitude adjustment will be implemented. 3 The RGB levels will be calculated for the measurements, then summated and divided by 3 to obtain an average voltage, which will be used as a reference voltage (as described earlier) and hardware or software compared to the actual measured voltage, where deviation of each of the R, G and B voltages above the reference and/or below the reference as an absolute deviation or a relative or a percentage deviation will be indicative of the colour transmitted, where intensity compensation may or may not be calculated or it may be eliminated.

In all cases, each of the techniques may be used in various combinations. However, a mismatch in the primary colours of the transmitting device may occur with the primary response of the photodiode detector (not of the same colour wavelengths). In this case, it could be corrected for automatically using the above techniques, or the resolution may be decreased by using fewer colours, or a calibration run may be required, but just using a calibration sequence of White, and measuring each R, G and B outputs and/or a calibration sequence of R, G and B and so on.

Variable sensitivity and resolution will now be described.

From Figure 12 and 13, it is also quite clear that at certain intensities of the colours, and at certain colour combinations (not shown), the same change in transmitted magnitude will result in a smaller change in output, to a point of not being discernible from noise, drift, measurement resolution and so on, at certain intensity points, where some combinations of colours or single colours may be better defined at the same points (i.e. δ Transmitted colour intensity >, < or≠ δ Photodiode output voltage over 00 to FF). Therefore, where the change in colour magnitude results in a smaller change in output voltage/s, the Hexadecimal steps are increased in size, and where the change in colour magnitude results in a larger change in output voltage, the Hexadecimal steps are decreased in size. This approach is a compromise to achieve and maintain good detection accuracy and high data numbers.

Again, this significantly deviates from the chrominance measurement method, because in this case, the transmitter colour mixes (as describe earlier) and/or intensities and/or intensity steps are distorted for each colour and/or colour combination and/or intensity combinations in order to match the detector characteristics and not a chrominance chart, where the detection is further referenced against calibration data and/or against empirical data and/or against empirically derived formula and/or spice simulation formula and so on.

It is also not required for colour monitoring, calibration or correction. Synchronisation will now be described.

It is intended to transmit a stream or a telegram of colours from the device screen to the reader so that a required number of security type bytes can be sent, without requiring a high number of RGB detectors. Using four static colour zones or more will not be able to convey the amount of data required and also perform calibration. A higher number of zones will require greater position accuracy to align the zones with the corresponding detectors.

There is a trade-off between RGB photodiode response time (depending if used in photovoltaic or photocurrent mode), filtering response time and the screen refresh rate. So any telegram is limited to less than the screen refresh rate if the refresh rate is high or above 70Hz or more. All prior art is intentionally asynchronous, and works in a similar way to Quadrature Amplitude Modulation (broadband [ADSL] over copper). However, the transmitting device can be any type, and can have any screen refresh rate from 50Hz to 500Hz. So an application programme for generating the colour telegram must be accommodating to be more universal, where to do this, the programme will have its own time sequence, which may be locked to the screen refresh time if this is a standard code, but at less than 50 'shows' per second. Furthermore, many of the smartphone systems will have a central processor that may be conducting other tasks while running the access control application software. This will result in bit time differences and therefore telegram 'jitter'. Telegram jitter can be large enough to convert one but time to a fraction or a multiple of the design bit time to a point of not being able to locate or register the bits when using an asynchronous method of transmission and reception. To combat this possible problem, a clock must be introduced for synchronous transmission and reception. This clock will also be an embedded colour/s that must be discernible from the transmitted data. Therefor this differs again from any prior art because the 'clock colour' is not to be included in the data analysis. The rising and falling edges of the clock may be used as the trigger points for reading the data, however, allowance must be taken to allow for the slew rate of the

photodiode's response times as well as the screen uptake of the colour (settling time). In addition, allowance must be given to the clock timing if this is at a different zone because of the refresh sweep. Readings of the data must also be performed at the shortest anticipated bit time.

If the data is continually changing, and there is little chance of a continual stream of the same (identical) byte count, then an asynchronous approach can be workable in terms of using preamble to set the clock synchronisation and or to use the changing data to establish the clock cycles. In this case, even one or two successive identical data bytes could be tolerated if the jitter is within a given range.

Furthermore, successive data bytes can be reduced if pre-scrambled and then unscrambled, or if the transmitter knows that there is a successive sequence of identical bytes, then it may inject a non-data 'frame' for example, inserting Black in between each byte or at certain points to account for the worse case jitter.

Should a clock be vital to use, where screen refresh cannot be synchronised and so on, then one of the colours may be used for clock synchronisation. It is found that the Red primary colour has the least effect on the other two remaining primary colours.

Therefore a full on-off clock pattern using just the red colour can be used for timing. This will affect the other two colours as described above, but if the clock in known, then adjustments can be made to the other colours, which reaction to Red is known, when the Red colour is in off or on mode so that a correction can be applied to the remaining colours accordingly so the that resulting data is not affected. To example this, the levels of the G and B diodes will change to a different value when Red is introduced. If this change (positive) is known or can be estimated by empirical means, then the

anticipated positive levels can be subtracted to maintain the remaining colour

parameters. This can occur in reverse, and can change depending on the different values.

However, as an additional embodiment for this invention, it has been established, that only a small change in the Red colour, can be detected and have a lesser impact on the remaining colours, therefore it is possible to retain the Red colour for data, to the same or lesser extent, and to use a small detectable change embedded on the Red data channel as a clock. I must be appreciated that it is therefore possible to use other colours for the clock or to change from one colour to the other colours for the clock, where it could also be embedded on more than one colour channel and/or all three colours. The clock change in amplitude may also vary, depending on the transmitted amplitudes. Again, this is a method for discerning the clock data from the actual data, which is achievable using any of the above methods described in this invention.

During a preamble or calibration run, if the detector can establish where the bit time jitter occurs, due to refresh cycles, then a degree of anticipation can be applied to compensate for subsequent jitter errors. This will be pre-emptive method to account for predictable jitter points.

Of course, the clock could be achieved by using a separate colour zone i.e. multiple colour zones which is set out below.

The described invention anticipates the use of one or more RGB photodiodes positioned over one or more independent colours zones on the same screen.

Preferably, one colour zone will be the target.

Multiple zones in reality, the cost to perform the same operation can be proportionally higher, and the power demand will be proportionally higher, not discounting size, which will be proportionally higher than existing systems. So the means to detect more than one colour zone will require a more challenging solution using only RGB photodiodes and/or other types of photodiodes or phototransistors and the like.

Where one colour zone is used, and a clock is intended, then one of the RGB colours will be used for the clock as described earlier.

Multiple zones on different size screens can mean differing poisons of the zones on the screen between the devices because the graphics are defined by Pixels and not linear length. This can be corrected if the device screen dimensions or aspect ratios or number of pixels are known.

However, multiple zones will offer a better compromise for better detection accuracy using fewer colour changes for each zone and the speed of each frame can be reduced so that detection stability is easier, to take into account the detection response time and possible data jitter:

Multiple detectors and RGB detection will now be described.

The downside of multiple zones is that alignment between the transmitter and the detectors is more critical. However, using more than one RGB detector for more than one zone, could enable a degree of misalignment compensation where, for example, RGB or RB detection is illustrated in Figure 16 with one variation of colour detection using only two photodiodes 32 (for each zone) where each diode will have a different spectral sensitivity, for example, Red and Blue. This is a differential circuit that measured the difference between the two colours, and not the absolute values.

Therefore, colour is detected at one chrominance level.

Although most of this specification deals with colour sensitive detectors, White light (ambient or visible light) sensors under a colour filter may also be used, with the advantage of lower cost, and the ability to adjust the colour response.

Multiplexing is an option as the amplifier stages are relatively expensive, and where multiple zones may require one or more detectors, then the number of amplifiers can be reduced by multiplexing the detectors, into a fewer numbers of amplifiers. Amplifiers may also be reduced or eliminated, where if the photodiodes are operated in say the photovoltaic mode, the diode outputs can be directly injected into the microcontroller. However, this will rely on low voltage ADC accuracy and/or high impedance ADC inputs.

The effects of ambient light are known, and this has been described above. However, if the reader is exposed to ambient light when the transmitting device is not in position, then the light will have an effect on the reading. This is because in this case, parasitic ambient light can be more intense than the transmitting device screen and/or it can contain more dominant or overriding colours. Not only that, it can comprise a 50Hz cycle, that can appear to be a clocked data telegram. Therefore, a method must be used to discern real data from background influence or to minimise background light effects, when considering a RGB photodiode detection method.

A further embodiment for this invention is use a discernible preamble to define that the data transmission is to begin, where this preamble could also comprise the calibration run as well as synchronisation. Using a Red-Blue mark space pattern can be enough to overcome a colour dominance of parasitic ambient light colours so that ambient light can be ignored or always treated as non-data, whereon internal processing will not take place. Other issues include that some transmitting devices will not have a steady intensity that can be measured at slightly different points in time. The intensity is often modulated at between 50Hz to 500Hz, and the amplitude differences (p-p) can be significantly high, and due to non-linearity, intensity and colour dependant detection systems will not be able to discern the intended colour unless a correction is made. This will involve, for some chrominance detection methods, presenting multiple shapes and then to average the results to obtain the intended shape, where this is not included in the description. Taking one shape of a modulated transmission will not work unless the point of measurement is taken at one precise point that is correct.

Figure 15 illustrates the modulation when a fixed colour is presented on the screen. Traces 31 are the RGB diode outputs from one type of 'tablet'. Trace 30 is a given colour, but it shows in this case that the modulation can be different, where this is typical or one particular brand or technology that differs. In all cases, the modulation is synchronised for each RGB colour, and the frequency is stable. Also, in all cases, the colour differentials, at the base line or zero point of the AC modulated signal, between each RGB colour remain the same, although the absolute intensity is modulated. There may be a difference where some levels of colour have differing amplitude. This means that providing the variables are measured at adequately the same point in time, or at the AC zero point, the colour ratios can be discerned. This also means that unless an average of the variable/s is/are established, which may not be possible for modulation frequencies that are at or near the same data transmission frequency, then intensity embedded or absolute detection will not be possible, and therefore reliance on the ratio between each colour may be the only factor for accurate detection.

Intensity may be averaged providing there is more than one cycle of the modulation between each colour frame, where measurement but be conducted of complete cycles.

Some alternatives and deviations include that instead of, or in addition to, the changing colours, a static bar graph could be used instead, whereby the device is physically moved across the reader, or the reader is moved across the device, or the static barcode picture is moved on the screen (scrolled). In this case, the same detection method is used. And again in this case, the bar graph comprises colour codes in place of a black or white line. Each colour bar, to include black or white, may also include a clock bit or byte colour, to include back or white inserted between each said colour bar. The clock if required may be positioned on a separate bar, meaning that multiple zones may be utilised. Alternatively, the blank codes in between each data code may be simply a black space to the end of one byte and the beginning of the next can be easily defined.

A further deviation would be to utilise more than one zone of course, however, to decrease the alignment accuracy requirement, the zones may be in alignment with more than one detector, where a voting system can be used to asses which sensors are aligned with which zones, and cross zone alignment can be ignored. This becomes more of an array, and the detectors can be arranged in specific alignment for example, clusters, circles, crosses or inline.

Radar graphic or moving rings may be used where multiple zones are used, a moving graphic could emanate from a central point, then to form circular bands that move to the outer edge of the screen and so on.

Different sensor types may also be available for use with this invention: These sensors or detectors may comprise colour, including white, sensitive sections, and integrated amplifiers and processing, where the input-output function of the sensor may be light to Voltage, light to Frequency, light to TTL and so on. In this case, there will still be an amplifier used to increase the diode output voltage or current, and the amplifier output will be low impedance voltage, converted to another format that can interface into a microcontroller or similar if required.

The additional white diode can be used for compensation or reference against brightness variations and/or colour variations.

For each zone, two RBG or RGBW (where W is White) sensors may be place in parallel, or an array of sensors may be clustered together (more than two), where correction for modulation and low frequency noise can be made, and a comparison can also be made for validation.

In the case of a colour bar-graph or barcode, where the phone is swept across the sensor, the sensor, dual sensor or sensor array may be arranged along the sweep axis with optional parallel repeat so that anticipation of change can be calculated and/or noise or modulation can be reduced or eliminated, and where smoother transitions between each bar in the bar-graph is used (a gradual change in colour between each bar) then it is possible to derive a mathematical equation of the curves produced by polynomial calculation and/or differentiation and/or integration, where the mathematical function of each of the colours, and interaction of the colours can be plotted over the entire sweep, which may result in a polynomial expression or similar formula or derivation, which may be a complex of Sinewaves, exponents, squares, polynomials and so on. This formula may be relational to the polynomial response curves described above, or it may be independent.

This formula for RGB or RGBW over the sweep can then be a data shape over time, which is similar to all colours. This shape is not a shape derived by chrominance detection of a colour, but a shape derived over the entire telegram transmission of all colours i.e. it is a telegram shape. Figure 17 and 18a shows a sweep across the screen, showing a limited number of colours. Each RGB sensor will respond the colour mix accordingly as is traverses along the screen, perpendicular or

approximately/adequately perpendicular to the colour gradient/bars. It can be seen that each response of the RGB sensors resembles a shape or 'key', where in this case there are three keys, which can be called a 3D-Key when you take all three responses, which can be interactive as described above. Figure 18b shows a more complex 3D 'key', comprising more colours.

Each of the keys can be given a mathematical equation over W to tn, as an example. This equation can be assisted by using distinct and uniform bars as opposed to gradual colour transitions so that each colour can be identified, and the rate of change can be calculated against the steady state points and eliminated. The purpose of this is that different sweep speed is not influential enough to cause identification errors. However, methods can be employed so that graduation can be used, and the sweep speed effect can be reduced or eliminated.

Where multiple colour bands or blends are used on a single screen, the speed of the swipe can mean that the colour changes may be faster than the response time of the sensor. Therefore, a high speed sensor design will be required for given fast scan speeds and/or more colour transients, which of course may be limited by the actual diode surface area or travel axial length and any azimuth deviation (skew).

Additionally, the introduction of multiple zones and/or multiple sensors and/or multiple frames/screen refresh cycles can increase the amount of data transmitted. All other points regarding bit jitter, described above, also apply if applicable, not discounting a time code barcode also described above. Modulation elimination/reduction can be achieved by the use of more than one sensor, placed along the travel axis where the modulation will be the same, but the colour may be different, and knowing the speed of travel, perhaps by using a preamble, the modulation can be calculated and eliminated or reduced. This preamble can also be used to compensate for the scan speed where blended colours are used. Modulation can be reduced by calculation or measurement criterion for example, peak to peak, average, peak or trough measurement. If it is uniform and constant, then it can also be anticipated and subtracted.

Security

Where the screen is static, and the transmitting device is moved or scanned, then it is possible to take a photograph of the screen or scan it over an identical reader purchased and owned by the criminal. However, it is made more difficult because the screen colours, if photographing, will have to be known precisely if there are many colours to choose from, and the device will have to be replicated if the reader can be made to recognise the device characteristics on the first pass. However, one way to deal with this is to introduce a scrambling method into the reader, which is

programmable by the user using a manual or an automatic process for duplex/half duplex readers or it can be pre-programmed during manufacture. It can of course be 'learned' during commissioning tests. It may also comprise a manual setting of DIP switches and so on. In this case, the criminal will not be able to easily match their reader to the actual reader used at that given site. So the key from the device may be legitimate, but the reader will not be able to scramble the code correctly before it is sent to the access control system. Furthermore, the code can be further scrambled at the phone too, providing the scrambling is known to the reader and the phone.

Application areas for the invention will be described below.

It must be noted that this invention is specifically intended for access control, time and attendance and general security. Therefor the data transmission technique and requirements will be different from those inventions intended for continual data transmission or transmission lasting longer than a few seconds or less. This invention is intended to be used with any type of colour display device, at any setting (to within reason) and to asses if the display device it is in communication with, is unique to that device and/or to that user. This invention intends to use photodiode and/or phototransistor for VLC detection, and it must use only a design power that is compatible with the intended specification, which may also include small 3V lithium pill or button type batteries, or in some cases, solar-cells. The reader circuit may also have size and cost constraints, where this will of course define which solution will be workable in real-life.

Figure 14 illustrates the embodiment of the invention as a flowchart.

The invention may include aspects that relate to an access control or time and attendance or payment/money transfer method and/or building management system requiring access rights, comprising a dual-media transceiver utilising audible sound communication (ASC), approximately in the 20Hz to 20kHz band, and visible light communication (VLC), approximately in the 430THz to 790THz band, with the option of a tri-media solution, to include transceiver soft-button switching which is instigated between transceivers.

A multi-media approach has specific advantage of power and cost minimisation, and also for security, where to eavesdrop, both VLC and ASC need to be recoded and analysed, where the VLC solution will be the most difficult to detect. Using a reader activated soft-button also introduces a further level or security and operational advantages:

Of course, two-way communication can be achieved by VLC only, and this is one option, but it will rely on the camera being at the right position to detect a VLC transceiver at the reader, where some cameras are only offered at the rear of the screen, and using the camera will incur additional device power usage. This can be used as one embodiment of the invention providing it meets the criterion below, but by using VLC instead of ASC.

Duplex or half duplex communication can be envisaged. For alternative security, specific operational control of the VLC method (describe above) and acknowledgment or error alarm, communication from the reader back to the device must be performed. This can be achieved in various ways, not discounting WiFi, Bluetooth or the standard phone network.

In a typical access control reader, there is mostly included a loudspeaker or sounding device that is used for an alarm to alert the user that the door is unlocked. In this case, the reader can therefore communicate to the device using audible acoustic/sound communication through this said loudspeaker, where the transmitter or device will have a microphone, as standard, that can pick up the communicating sound from this said loudspeaker.

Therefore, it is possible, without the requirement of significant/additional components, to communicate information from the reader to the device.

Reader and device communication - The reader can communicate data or information to the device by way of one of more of the following, and for any duration:

1 . A tone or specific frequency.

2. A differing tone for differing instructions/acknowledgments.

3. A complex waveform containing more than one tone, which may be static or dynamic, with a static or dynamic amplitude for each tone and/or static or dynamic frequency representing data and/or differing

instructions/acknowledgments.

4. FSK, ASK, PSK, QAM or any other serial method for digital communication.

5. Spread Spectrum communication using one or more of the above.

Using any of the above methods may require that the device includes means for accurately detecting the data or information. This may include Fast Fourier

Transformation or similar using the devices internal DSP capability. It must be clear the reader will not have any DSP or audio filtering functionality apart from automatic volume control, if this is necessary, where this can be performed in one of many ways as described below.

Automatic volume control (AVC)

The sound from the reader to the device may be sent amongst influential background noise which can mask the intended data from being detected. Spread spectrum communication can greatly reduce or eliminate the influence of background noise, but the transmission will be set to a specific volume, which may be too high for quiet areas or too low for the background noise level. Therefore, Automatic Volume Control (AVC) can be implemented by either placing a microphone at the reader to measure the background noise and to adjust the level to be detectable for that given level of background noise, or it can be achieved by first sending a tone or data to the device, whereon if there is no acknowledgment from the device, by way of a VLC response, the reader can increase the sound level until an acknowledgment is received. This acknowledgment may be that the device will begin a calibration run. Variations on this concept may have to be employed so that faults can be detected or diagnostics can be issued or so that the volume of the reader is not too high for a given area.

One other method will be to use the device to detect background noise level and/or to analyse the background noise frequency content. Once established, the device could set a specific colour code on the screen or begin to send rudimentary data i.e. black and white pulse code modulation, which can be discerned by the reader prior to calibration, to give instruction to the reader, at which amplitude to set the ASC volume level and/or which frequency bands to transmit at. In fact, this VLC data could also be used as part of the calibration procedure.

Significantly, if a microphone is to be included in the reader for AVC or other reasons for communication (i.e.'voip'), then the device could transmit a clock frequency for data synchronisation for the VLC. Again, all of the above noise cancelation means apply.

Presence/proximity detection will be described.

In certain cases, when operating the device, it may be beneficial, preferred or easier to point or direct the device screen at the reader. In this orientation, it may not be possible (or desirable) to operate the touchscreen soft-buttons, where a constant transmission loop or repeat may be required. This will reduce the time for transmission by eliminating the need to resend a key if the first key has not been successfully received or if it was detected or presented part of the way through a telegram transmission. If the device is not intended to repeat the key or it must wait until the reader is ready, or at a defined maximum distance, which can include contact, then the reader must include presence detection so that the phone is in the correct position before commencing VLC

communication, with or without a given delay between detection and initiation.

One problem of setting a VLC telegram into operation before the device is in static contact with the reader would be where the device is moving towards the reader from a distance. During this time, a calibration run may be in action where the effect of background light will be changing as the device approaches the reader, and the intensity of the screen's light will also be changing. This may lead to an incorrect calculation of the VLC colour data and lead to communication failure, requiring a retransmission of the telegram. This approach or solution may be performed in many differing ways; for example: Using accelerometers within the device (if fitted), to asses when the device is in position by way of tapping the device on the reader, or the reader may have accelerometers or optical proximity detectors which can be the VLC RGB detector, laser/capacitance distance detectors, and pressure sensors. Once the device has been detected, then the reader may transmit an ASC signal to the device to instruct the device to commence VLC or for other means/methods which will are listed below. Pseudo-finger

One preferred method for detecting when the device is in correct visual contact/distance with the reader is to utilise the device's touch screen to initiate the VLC sequence automatically or by way of response from the reader for example using ASC. In this embodiment, a pad or embossed surface, or other suitable means, which can be referred to as a pseudo-finger, but not limited to the same dimensions, which has the equivalent characteristics of a human finger or fingers, is detected by the device when in direct contact (or in operable proximity). This is the same as a key-press which can instruct the device to commence VLC. Equally, taking the device away from the reader can be detected as a key-release, when on the instruction may be to abort transmission at or after the data telegram transmission if required. Of course, if the code repeats, then a means to turn off or abort the sequence/loop may be required by the user, so double clicks in this case may be required, or certain timing rules can be put in place for example. This will be a simple task to implement for those skilled in the art.

Significantly, taking this a stage further, if this pseudo-finger can be electrical, capacitive, charge, conductive, alterable/changed or controllable so that to the screen, it appears to touch, then release, without actually moving the pseudo-finger, then it can step though a sequence or though programs/program menus or issue an

acknowledgement using a soft-button driven menu and/or be used for one or more of the methods listed below.

Taking this another stage forwards, it may be possible to use this technique to communicate data, which will of course be limited to the speed of response of the touchscreen technology and/or the firmware used and the controllable pseudo-finger. Using this technique may or may not eliminate the need to use ASC from the reader to initiate VLC, or it may be used in conjunction with ASC. Of course, this type of two way communication can take place in a duplex or half duplex manner.

A further embellishment of this communication concept or soft key-press technique, may comprise more than one pseudo-finger, which may be apparent when looking at the reader or it may be concealed, which may be static (a single event key-press) or dynamic in operation (meaning it can be controlled electrically or physically), and/or more than one pseudo-finger shape and/or size, which may also be pre-configured or configurable. The purpose of this is to either increase the capability for faster communication and/or to present a unique pattern for that reader. This would therefore have the potential for faster communication, or it may provide a unique ID so that the device will recognise that reader as being a legitimate reader.

It is possible to use a movable or dynamic pseudo-finger/s by way of long reach transducer using piston/diaphragm, piezoelectric means or electrostatic plates and so on.

For a static or dynamic pseudo-finger, inductors and/or capacitors and/or relays and/or high impedance semiconductors could be used to compensate for circuit trace lengths and/or for finger simulation components.

The pseudo-finger and/or ASC, in combination with the VLC means for communication, is effectively a method for interaction with the device, which is equivalent to a HMI (human-machine-interface), where in this case, it is a MMI (machine-machine-interface) using the same techniques a human would use for interaction with the device.

The audio or visual feedback from the device or from the reader can be used for one or more of the following actions:

1 To initiate the calibration run for matching the device to the reader.

2 To initiate the data telegram.

3 To cancel or abort the calibration run and/or data telegram.

4 To acknowledge a successful data transmission.

5 To initiate a further calibration run and/or data telegram.

6 To be used for real time clock synchronisation for synchronous data

transmission.

7 To send an update to the encryption or cipher or to advance the key

and/or unique identity and so on. 8 To validate that it is a legitimate reader before access is granted and/or the device initiates and sends a key.

9 To alarm/warn the user of a fault or a situation.

10 To send a message or a code or an instruction to the device or to the

user.

1 1 To send a new password or PIN.

12 To instigate a further challenge, for example require a Q&A or a PIN or second/alternative PIN or another finger print.

13 To initiate one or more program/s in the device for example to set up video identification.

14 To instruct the device to increase or decrease the screen brightness

and/or change the chrominance balance, either manually or it can do this automatically.

15 To return the screen to the original settings.

16 To communicate and therefore adjust the volume level of the reader

and/or microphone gain of the device so that it can be detected by the device (this will be automatic volume control as described above.

17 To acknowledge a panic button operation and optionally, to display a

message on the device.

18 If a certain 'panic button' is operated or a button is operated in a certain way, it may be used send an instruction to the device, to immobilise the device or destroy/shred the key program and the cyphers and so on. This way, accidental operation of the given panic button outside of the reader range will not destroy the key.

One of more of the above methods may take place before, during or during, at specific points, or after the VLC telegram is in operation and/or it may interrupt the VCL telegram at one or more points during VLC communication

Devices may have voice recognition, fingerprint recognition, PIN entry to validate the user, where challenges can also be instigated part way through the process or after the process.

Panic button

This code will be issued to the access control system via VLC. In a panic situation, for example, if a criminal is holding the legitimate user of the key, hostage, then the user can stealthily send a distress code to the security personnel or the police and/or it can now block access rights for the legitimate key.

To open the door the button will be pressed one way. If for example the button was slid to the right, then the door will be opened, but an alert will be issued. Slide the button to the left and the door will remain locked and an alert will be issued. Slide it up, and the door will be locked and an alarm bell can be sounded. This can also be achieved at, for example the finger print recognition stage, or during the PIN entry.

Secret button-press 'shape'

The user may also define how an initiation of the 'key' program or app is performed. For example, it may be a relatively unique sequence to draw a shape, first to swipe a vertical line, then a line to the left, then a line down, where this can be extended, or shortened. This may be sufficient enough not to require a PIN entry at each time of use, or so that a longer PIN can be used initially, then a shorter swipe pattern may be used at each time. This may offer some protection against a phone being stolen/taken after the PIN has been entered. This may be a faster alternative to using a fingerprint recognitions system.

In summary the solution here may comprise directional or bidirectional (half duplex or full duplex) VLC, by way of including communication from the reader LED's or display, to the device's camera, which may be in direct line of sight with the LED's or influenced by them indirectly and/or this solution may comprise directional or bidirectional (half duplex or full duplex) ASC to include a microphone in the reader, which may also be used as part of an audible intercom system integrated in the reader and/or this solution may comprise directional device soft-button activation by the reader.

Figure 19 is an illustration of the summary where the device 1 is used to transmit a colour, whereon this is colour, representing data, is viewed by a detector 2, for the colour displayed on the device, for a first calibration and second data telegram sequence of data colours. The colour is amplified for Red 3, Green 4, and Blue 5, with optional White 5W and digitised for analysis, computation and data retrieval 6 using the VLC transfer from the dive 1 to the reader. A reader comprising a sound transducer 61 (loudspeaker), used first for alarm purposes, may also be used for performing ASC from the reader to the device 1 . The reader may also comprise one or more soft-button activators 60 with may be static or dynamically operated by the reader. The reader may also comprise a microphone transduce 63, which may be used for intercom purposes, to be also used for ASC from the device 1 to the reader, which may comprise

synchronisation clocks and/or data. The Reader may also comprise LED for

annunciation of the reader state for example door unlocked or door locked. These LED's may also be used for VLC communication directly, or indirectly, to the device's 1 camera (not shown).

Alternatives, here this example specifically, relates to the ability to transmit data using one or more black or white or colour zones, in a way described above, with a colour discerning resolution of between 1 -bits to n-bits, to be sent at a rate of n-times per second, where n is less than the screen refresh cycles per second, and where the touchscreen is operated manually and/or automatically by the said static or electrically alterable pseudo-finger. This may or may not include sound communication from the reader to the device and/or from the device to the reader. In this variation, the pseudo- finger may comprise two or more pseudo fingers located at different places on the touch screen with specific functional reasons, which will include;

1 . With two or more pseudo-finger points, static or dynamic, it will be possible to send VLC data using more than one zone, and for the reader to use the pseudo-finger points as locators, very much like a QR code locator means, where the touchscreen zones can be positioned on the screen, according to where the locators are, in a position and orientation that will match the respective colour detectors.

2. With two or more pseudo-finger points, static or dynamic, it will be possible to size the more than one colour zone's envelope without have to know the actual screen size and/or aspect ratio and/or screen pixel resolution. Zone sizes can only be defined by pixels unless the screen data is known by the software. Where two or more points on the touchscreen are known, then the zone sizes, orientation and position can be calculated, where the zones may be located inside the point's envelope and/or outside the points envelope.

3. The points may be indicative of an imaginary square, with equal length sides, or it may be indicative of the centre of concentric circles, with an outer or predetermined point which will define the width of the circle bands. This can be arranged in any suitable way, using n-number of points, so that where more than one colour zones are used, the zones can be alighted to the respective detectors in the reader.

4. To ensure the screen is aligned with one or more sensors and/o used for redundancy.

5. Using an electrically alterable pseudo-finger (dynamic), then further data and/or instructions and/or validation can be sent from the reader to the device.

Proximity and alternative proximity methods will now be discussed. A proximity method is described, which may include one specific method, in isolation, or in conjunction with any of the above methods, whereby the device will be transmitting a repeated contrasting colour code, enough to be discernible without calibration, to include Black and/or White, or it may be displaying a static colour, for example, White at full intensity and/or a moving bar or graphic and so on. This bar or graphic may be preferred due to the stroboscopic effect of a contrasting colour, which will/may be initially seen, where this may trigger photosensitive epilepsy.

As the device approaches the reader, the ambient light and the screen light will be present at the detector/s, and in changing proportions in most cases, where the detectors will see a change in intensity or characteristic or ratio of the RB or RGB or RGBW outputs. As the device nears the reader, the device screen will become more dominant and the changes in characteristics will begin to rapidly decrease, until the changes stabilise when the device and reader are in the final position. This is then detected by the reader whereon it will send an instruction to the device to commence either calibration, communication, or data transfer.

This can be achieved by any proximity type detector of any type, where the stabilisation point is the result of successful 'docking' of the reader with the device. A small delay may be required during stabilisation to verify that no 'bounce' has taken place.

Of course, at the point of device operation, the screen may not strobe for a given time so that a photosensitive user can be protected from directly viewing the screen whilst it is being placed on the reader. This can adapted in many ways to protect photosensitive users. A clock timing alternative is discussed -

The reader can issue a bust tone or tone code, to the device, after each colour, or each frame, has been identified and acknowledged, to instruct the device to advance to the next frame or next few frames/specific colour display. This will synchronise the transmission to the reader's clock, and also insure that each frame or exposure of a given colour is detected, detectable and/or or correct.

This may be taken a stage further, in that the tone burst or code, this may be used as a clock and/or it can be used as a challenge-response system or an instruction to change the cypher/encryption part way though the telegram transmission. The challenge can be a choice between several options that are known only to that device or user and the access control system. This could be used in a similar way that online banking is conducted where there is a given password and PIN, and the user (or device) is asked to provide a certain part of the password, so that the entire, or full, PIN and/or password is never disclosed.

One specific solution comprises a device, displaying only Black, Red and Blue colours, and mixtures thereof. This is matched by a Red and Blue photo detector/s in the reader. The purpose of this limit has many advantages in that;

The Red and Blue detector influence on each other is minimal and it is easier to calculate, anticipate and/or correct.

The initial calibration may only require full Black, full Red and full Blue or full Magenta and full Black.

Non-linearity of the detectors, can be compensated by transmitting a non-linear step in colour intensity for example, using RGB (or RB) colour steps of 0, 153, 205 and 225 for each colour, where for a RB system, a 4-bit byte can be realised with one colour shown. This combines both the RGB 3-bit, and the Chrominance methods, but with a higher level of detectability and/or a higher bit number.

Again, using the above system can be combined with communication from the reader, to the device, using audible sound communication, which may comprise 1 -bit to n-bits transmission.

Therefore, communication from the device to the reader can be one or more of;

identification, encryption, updates, message, alarm, validation, calibration, proximity, password/key or parts thereof, and communication from the reader to the device can be identification, clock/synchronisation, encryption, updates, message, alarm, validation, calibration, proximity, password/key or parts thereof.

Taking this a stage further, additional audible sound communication could take place from the device to the reader with any of the above actions.

Specific deviation using proprietary devices.

All of the above may also apply to this specific deviation, such that smart-devices can also be used, although this may also operate as a stand-alone system, with parts of, or all of the above being relevant if or when required.

In this embodiment, a specific proprietary device is used, which is an access control 'key', comprising one or more higher speed light emitting diodes (as opposed to a LCD screen), of one or more colours, including ultraviolet or infrared, and if required for duplex or half duplex communication, one or more photodiodes or phototransistors of one or more sensitivities to one or more colours, including ultraviolet or infrared may be used. This is essentially the same in principle to a smart device screen, but it has higher intensity and higher speed light emitting elements (Pixels), and fewer of them, where screen refresh rates will not be a limiting factor in this case.

The access control reader or control system is essentially the same, comprising one or more photodiodes or phototransistors of one or more sensitivities to one or more colours, including ultraviolet or infrared, and if required for duplex or half duplex communication, one or more higher speed light emitting diodes, of one or more colours including ultraviolet or infrared may be used.

The 'key' may be constructed in a specific way, for example, a wearable item similar to a watch or bracelet, a key-fob, a card, a badge and so on.

The power for the 'key' may be one or more of the following:

1 Internal battery/super capacitor

2 Photovoltaic Cell

3 Mutual Induction coil

4 Mutual Capacitance plate

5 Thermopile

6 Vibration/Accelerometer

7 Piezoelectric generator 8 Power connector taking power from the reader or other charging means such as USB. This could be an edge connector or exposed pads for contact connection.

The key may be pre-programmed and/or programmed via the VLC means and/or via one or more of the above means.

The Reader may also comprise one or more of the above power coupling methods for example; it may comprise a light source able to power the key using an integral Photovoltaic Cell.

The intended data rate will be governed by the response of the LED, the detector, the detector amplifier, and the processor. This may be in the region of 2kHz to 10kHz, equating to 2-kbps to >10-kbps using one colour, and up to 6-kbps to 30-kbps using RGB, although a degree of alignment and/or mixing will be required, because RGB LED colour will be stratified.

Empirical testing with Blue LED's, the frequency for transmission was about 12kHz, and detectable at a current of 0.5mA peak, driving through a blue LED.

Further embodiments of this concept can include biometric pairing, where a wrist band is able to monitor heartbeat characteristics and/or pressures and/or body temperatures and/or vein/artery patterns and/or chemical composition of the person's perspiration and so on, such that the wearer can be identified and therefore the key will only operate if the wearer matches the profile stored on the key. This can also be used to detect if the wearer is under duress and therefore notify the security team via that access control system.

In another alternative the 'pseudo-finger' principle is used, in that if the device is in contact with the reader, as described earlier, them the touchscreen of the device can be activated to perform specific tasks.

Here, two or more pseudo-finger points, which may be static or controllable, are used to define the position or location of multiple colour zones on the device touchscreen so that they become aligned with the relevant zone sensors. With two touch points, it is possible to use them to assign the X-Y of the zones, as well as possibly assigning the rotation to a degree. However, using three touch points will allow rotation

compensation, as for example, a QR code does. An artificial horizon will now be described. One specific embodiment of this part of the invention now uses one, or it may use more for better accuracy or scaling, touch points to be able to define the more than one colour zone's X-Y coordinates as well as the rotation (the angle at which the device is presented to the reader), by using the device's artificial horizon capability (accelerometers and/or gyroscope). Most smartphone devices or phablets/tablets will have the capability of understanding it's orientation in space, relative to the gravitational pull. Therefore, more than one colour zone can be orientated about that one touch point so that the X-axis is adequately level to 'sea level', or at least the reference for the X-Y axis is at a known orientation. Taking the artificial horizon principle, considering only aircraft 'roll' and not 'pitch' or 'yaw', it can be seen that on a typical artificial horizon display, the Blue zone or sky could be used for one zone, and the Brown zone or earth could be used for a second zone, where the reader is level and fixed. This can then be further sectioned to include more than two colour zones for data and/or clock timing.

Figure 20a illustrates the multi-colour zone 8a to 8n, comprising four zones. The zones are aligned with the respective detectors 2a to 2n when the device is positioned onto the reader, and it is centred at the point where the pseudo-finger 60 makes contact with the touchscreen of the device, and in this case, the zones are shown to match the reader detector positions for each zone. However, if the device is rotated and/or offset, as shown in figure 20b, the device will re-align the centre point of the zones, and rotate the zones to match the same zone orientation as shown in figure 20a, which is using the artificial horizon principle, so that the correct zones are aligned with the correct detectors, thus allowing 360-Degree rotation of the device, without losing the correct alignment of the zones to the reader's detectors.

A certain amount of offset is also show in figure 20b, however, there will be a limit of how much can be tolerated, before one or more of the sensors are no longer in direct view of the screen. In this case, is cab ne calculated by the device, and an alarm can be given.

In cases where a touch screen is not available, then the zones may be combined into one zone or if alignment is possible, then multiple zones can be used. Where one or two zones are used, instead of say four, then it can still be made to operate, but over a longer period of time. Of course, the reader may also comprise more detectors than there are zones. In this way, rotation can be derived by using a calibration run of alternate colours, where given detectors will respond according to the zone position and orientation, where this may also be used in conjunction with the various methods described above, with the purpose of being able to accommodate device placement flexibility, whilst transmitting the maximum amount of data. Each detector/s will be part of a two dimensional array, where a given number of detectors will see the same area of the zone, where each zone is an alternate colour, for example, Blue and Red and/or Black and/or White. Detectors detecting the boundary of the zones can be ignored, and the more detectors you have, the better the zone location will be.

An active pseudo-finger is a described. The pseudo-finger can not only be an operator for the device touchscreen, it may also be combined with, or compliment, or comprise, an operational power-switch or activation-switch, which may comprise proximity means in the reader. This switch is used to switch on the optical detectors or other circuits and/or to wake up the controller. This will be particularly useful for battery operated readers. Of course, the reader can be listening or looking for a device to be present, then the controller can initiate power on detection, however, a switch may be more precise and could decrease the amount of time that the system is in operation or eliminate false operations due to nearby presence of devices, not in position or in physical contact.

The switch may be designed to simply operate the reader for the duration of the communication, when if not complete, or complete, the controller will override the switch to close the system down. This is to prevent the reader switch continually holding on the power if held manually.

When the switch is activated, an alarm may also be repeated on the other side of the door so that an occupant may be notified of another user or an attempted intrusion. Optical/chrominance linearization

Although it is possible to correct for non-linear detector response using colour RGB number, it is possible to achieve using a calibration run of 0, 127 and 255 intensities of a given colour or colours. From this, an approximate non-linear curve can be

established. However, using three points for each colour, linearization may not be required, meaning that for a two colour detector, a byte of 3-bits can be established, because a full Off and full On' are far enough away from a mid-point, which can span a longer range for detection. It is to be noted that in the case of battery power, audible communication from the reader to the device may be limited, but still utilised if required. Audible communication from the device to the reader may also be utilised in conjunction with visible light communication. Audible communication is a MODEM.

It is known to provide communication from a device (for example, a smartphone) to a reader using audio based Near Field Communication where the audio frequency range is in the ultrasonic band, although the description also covers audible frequencies in the sonic band.

This of course is similar to a modem or acoustic coupler, known modems will

communicate through FSK, ASK or PSK, with the latest introduction of QAM -

If a device emits acoustic waves, when transmitting and receiving, this can easily be intercepted. This therefore requires a far higher degree of protection and encryption to prevent a hacker stealing the code or information.

However, one embodiment of this invention is to utilize both Acoustic Near Field

Communication (ANFC) and Visible Light Near Field Communication (VLC), which are dependent upon each other, and therefore offer a far higher degree of protection from interception, as well as increasing the speed and/or density of the communication. In this case, it is a dual media type of transmission where both means will need to be intercepted and/or related, which is significantly more difficult to achieve, or closer to impossible without the highest degree of effort, which will be a superior deterrent based on a similar level of security implementation.

The concept of using acoustic and relational or dependent light near field coupling from one device to another device, through a transparent media (air or water) is not known nor is it implemented, researched or used for any other solution, variation or application. The third step is to adapt the system, with secure measures, for use in access control and other security/fiscal transaction related applications.

Taking this a stage further, the light means for communication will be of the

chrominance or partial chrominance (less resolution) technique as described previously, but not specifically. Specifically, any smart device's screen or LED can be used if either or both are available and controllable, and most smart devices will have at least an audio output and an optional microphone.

In this case, the relationship between the two said media types is specific in that both will be required for successful acceptance of authentication and that one or the other means cannot operate independently.

The relationship between the optical and the audio may be in real time and/or it may be in series and/or in parallel. For example, encoding may be performed where a given color displayed on the screen will instruct that a given set of frequencies or combination of frequencies are used for the audio transmission, or the color display sequence may be part of the authentication code or PIN and so on. It may also be used in a 'question and answer' way or deterministic way and so on.

In a similar way, or to realize the above, the reader may also transmit an audio code back to the device, where the completion of the task will only be allowed if the device transmits back an optical response and/or an audio response. In all cases, the authorization will not be executed or accepted without both valid audio and light communication.

This in part, specifies that both Light and sound data is transmitted from the device to the reader where for authentication, both must be correct, and optionally, the reader will transmit sound and/or light to the device, where this may also be used for authentication or for operation.

In each case, the data or code of sound and/or light or parts of the data or code of sound and/or light may be used for authentication and/or acknowledgment and/or instruction and/or validation and/or timing and/or encryption and/or token passing and/or cipher/encryption change instructions and/or FSK/ASK/PSK/QAM changes and so on. Figure 21 illustrates one embodiment of this invention, comprising a device 70, with at least a display 72 or light source and a sonic and/or ultrasonic output port 78, a reader 73 comprising at least a light detector/s 75 and a microphone port 77. Further embodiments may also comprise one or more of the following; a light source 74 in the reader in direct or indirect range of the device camera 71 , a sonic and/or ultrasonic output port 76 for communication to the device microphone port 79, a serial link or wireless output 80 so that communication to a controller and/or computer can be achieved, with the option to communicate to the device via wireless network means. Figure 22 illustrates one example of the implementation of the reader, where those skilled in the art will be able to realize such a circuit from the block schematic.

Variations on the concept can be implemented without deviating from the key functions illustrated in figure 21 .

The Acoustic communication received in the reader may treat the signal in differing ways for example it may be compressed or expanded at first before it is processed. It may be analogue processed and/or digitally processed for example, a processor or controller may be used to perform Fourier transforms or similar systems. For

Demodulation of ASK, FSK, PSK, QAM, and any other format, the system may use analogue filtering techniques and/or it may use digital filter techniques where the analogue and/or digital techniques may comprise low pass and/or band pass and/or high pass filtering, which may also comprise demodulating filtering, which may be band pass. It may be a ultra wideband, comprising many filters and filter techniques. The digital technique for reading the acoustic codes may also be by Fast Fourier

Transformation or similar technique thus eliminating the need for all, or some, of the analogue filtering, or it may be used to further enhance the analogue filtering

techniques.

Near field communication (NFC) may also be envisaged, or with an RFID

transmitter/receiver. Devices with NFC can easily be 'hacked' and unauthorised access is relatively simple, particularly for University campuses, where the concentration of technology students, able to perform these tasks, can be relatively high. So to integrate this with VLC would add an additional layer of security, which will make it more secure, and therefore, more attractive for higher risk areas.

However, NFC is not available on all devices and tablets, so to use NFC or RFID, in conjunction with VLC would not be as universally accepted as using acoustic

communication in conjunction with VLC as described in addendum 5.

If, however, NFC or RFID does migrate to be more dominant, then it is possible, as a further embodiment, to use NFC and/or Acoustic NFC (ANFC) in conjunction with VLC. This has the advantage of being universal in terms of device use, where DESFire cards and the like can be used. In reality, a mixture of ANFC and NFC/RFID could be implemented for better security over NFC alone, and in principle, this could be one embodiment of this invention, however, both these can be 'listened to' from a remote location, whereas VLC and one or either NFC and/or ANFC would comprise the best level of security due to the complexity of VLC 'hacking'.

Figure 23 illustrates an enhancement of Figure 21 , where in this case, RFID or NFC communication 82 can be transmitted from the device to the reader and/or RFID or NFC communication 81 can be transmitted from the reader to the device.

Therefore, the option of dual or triple media technologies can be used or considered as part of the embodiment of this invention.

In all cases of multimedia transmitting or transceiving, it is important to stress the inter- dependability of each media type in that they may be part of the access code and/or identity and/or communication between the device and the reader, one way or both ways, simplex, duplex or half-duplex and/or used for timing or synchronization and/or calibration and/or falsification or masking and/or for user information or

alarming/annunciation. The media types may also interleave and/or alternate in functionality. Depending what the media is used for, a system must never be able to function in its task using one media type alone.

Various modifications may be made to the described embodiments without departing from the scope of the present invention. The invention described may include other methods of operation and function or include other apparatus components, for example, there may be other types of sensor used. The communications device may have any type of display that provides optically clear media, different sensor types. The portion of the transmitter device can be for any task, for synchronisation, for calibration, it could be a portion of a telegram, portion of screen, portion of colour.