CN101159821A - Signal Output Method of Optical Receiver Area Array Sensor - Google Patents

Abstract

For the optical receiver composed of imaging optical system and area array photoelectric sensor, the invention introduces an output selection scheme. The pixel positions that respond to useful light signals are first identified, and then these pixel signals are selectively output, excluding the output of those pixels that only respond to ambient light. If the area array photoelectric sensor adopts sampling output, the integrity of the output information cannot be guaranteed for the pulse light signal. This invention introduces synchronous sampling technology and provides a method for obtaining synchronization.

Description

Signal Output Method of Optical Receiver Area Array Sensor

technical field

The invention belongs to the field of optical wireless communication, in particular to a signal output method of an optical receiver with an imaging optical system and an area array photoelectric sensor.

Background technique

Optical communication systems can be divided into wired communication and wireless communication from the channel point of view. The current main technology of optical cable communication is optical fiber communication, which has become the backbone technology in the transmission network. In terms of the characteristics of the emitted light used in the optical wireless communication system, there are currently two main types: laser communication and infrared communication. Laser communication is generally used in relatively long-distance point-to-point communication. One of the main technologies of infrared communication is the IrDA (International Infrared Data Association) standard for short-range data exchange of portable devices.

A general optical communication system receiver is mainly composed of an optical system, a photodetector and a signal processing circuit. The optical system collects and filters the optical signal in the space to focus on the photodetector; the photodetector converts the optical signal into an electrical signal and outputs it to the signal processing circuit; the signal processing circuit performs corresponding amplification, filtering, detection, demodulation, Decoding and other final output data. In such an optical communication system, one emitting light source usually corresponds to one photodetector. Even in an optical fiber communication system in which multiple signals are multiplexed and finally demultiplexed at the receiving end, one detector corresponds to one optical signal in space. The photoelectric receiver used for communication generally adopts the direct detection mode, that is, the phase information of the optical signal is not considered, but only the energy information of the optical signal is considered. The output of the detector is a continuous analog electrical signal, and the form of the signal depends on the modulation of the emitted light. form. In the case of pulsed light, the output is also a pulsed electrical signal.

Visible light communication is an emerging communication technology after laser and infrared. With the rapid development of the lighting LED industry, the use of lighting LEDs to realize visible light communication will become a supplement or replacement of the current communication system, especially in indoor short-distance high-speed wireless communication applications. The realization principle and basic composition of the visible light communication system are similar to those of the laser and infrared communication systems. The wireless communication devices used indoors are usually portable devices, such as notebook computers, mobile phones, audio and video players, digital cameras, game consoles, etc. An easy way to connect these devices to the communication network through optical communication is to use their own photoelectric imaging system to receive visible light communication signals, such as cameras on laptops and mobile phones. One of the most possible application scenarios is to use the LED array to send multiple optical signals, and form images of these LEDs on the image sensor through the imaging optical system of each portable device, and different detection pixels correspond to different LEDs. If these detection pixels can independently detect and output corresponding LED light signals, then multiple-channel or multiple-access optical communication can be realized.

The photodetector of the unit realizes continuous photoelectric conversion, and the corresponding output is a continuous analog signal, and the subsequent processing circuit is usually amplification, filtering, etc. If the detector of the area array is not large in scale, such as only 32×32 pixels, then the output of each pixel can be treated independently, and each detection pixel and its subsequent circuits constitute an independent output channel. There should be 1024 independent parallel continuous output channels for 32×32 area array detectors. This is not very difficult to realize for modern integrated circuit technology.

However, current image sensors (including future area sensors), whether CCD or CMOS, such as SONY's ICX418AKL (8mm CCD image sensor for color cameras) and KODAK's KAI-0330 (648×484 CCD image sensor), KAC -9628 (648×488 CMOS image sensor) is an integration of area array photodetectors and output sampling/processing circuits, and its output is the sampling value of the detector at a certain moment, rather than continuous output.

The output method adopted by the general CCD image sensor is to transfer the response signals of all detection pixels to the vertical CCD linear array arranged at intervals with the column pixels at the same time for storage, and then transfer the signals on each vertical CCD to the On a horizontal CCD line array, each time is equivalent to one line of signals, and the horizontal CCD is driven by the clock to serially output one by one. The output mode of the CMOS image sensor is similar to that of the CCD image sensor, except that the output of the CMOS column signal is not the value at the same time, but the output after the row-by-row scan or the unit-by-element scan.

The sampling output of CCD and CMOS image sensor is a discrete analog signal rather than a continuous analog signal. The discrete analog signal becomes a digital signal after the subsequent Correlated Double Sampling (CDS) analog-to-digital conversion. The reason for adopting the sampling output method is that, on the one hand, the number of pixels in the area array is too large, and it is difficult to plan millions or even tens of millions of parallel processing and output channels in integrated circuits; on the other hand, as an image or video Application, do not need too high output frequency, stored and then serial output or scan output can already meet the requirements of use.

However, if a similar image sensor is used as the detection device of a multi-channel optical receiver, in addition to the parameters such as detection sensitivity, response time, and response frequency band involved in conventional photoelectric detection, there will be two problems: one is the selection of output pixels problems; one is the synchronization of output sampling and received light.

When an image sensor is used to detect optical communication signals, the responding pixels are not all the pixels, but may be only some of the pixels, which is related to the relative positions of the sending light source and the optical receiver. So how to shield the response signals of other irrelevant pixels to ambient light and only output effective communication signals has a great impact on improving the data transmission rate and reducing redundant signals. Especially when the sampling output method is still used, if the output pixel can be selected, then under the same output clock drive, the output rate for some pixels will increase sharply.

The optical signal sent by the optical wireless communication system is a modulated signal, usually using pulse amplitude modulation (PAM), pulse frequency modulation (PFM), pulse width modulation (PWM) or pulse position modulation (PPM). Regardless of the modulation method, its performance is an intermittent optical signal. If the modulated signal can be output continuously, the integrity of the information contained in the signal can be maintained. But if the detector adopts sampling output, theoretically speaking, only by increasing the sampling frequency of the output to more than twice the bandwidth of the optical pulse signal according to the Nyquist sampling theorem can it be possible to maintain the integrity of the information. If the expected communication data rate is high, the corresponding optical pulse frequency is high, and this sampling rate is difficult to meet.

In the present invention, the output of the area array photoelectric sensor is characterized by discreteness in space and time. In order to improve the communication rate, a kind of output pixel selection and output sampling is introduced in the optical receiver using the area array photoelectric sensor. A method for synchronizing irradiation with light pulses.

Contents of the invention

The main content of the present invention is a receiver in an optical wireless communication system. Its main feature is that the optical receiver adopts an area array photoelectric sensor. In the area array sensor, a specific pixel signal can be selected for output, and the output sampling time and illumination pulse Synchronized in time.

The invention is based on a light transmitting unit and a light receiving unit. The basic composition includes a light sending point source or light sending array, an imaging optical system at the receiving end, an area array photodetector and its output and processing circuit. Obviously, such a light receiving unit can also be used with an image/video acquisition system.

The optical sending unit sends modulated pulsed optical signals according to the agreement of the communication system, and these signals represent the data to be sent. The optical signal is imaged on the area array photodetector after passing through the space channel and the optical system. The area array detector outputs the photoelectric conversion signal under the control of the output circuit, and then sends the signal to the processing circuit for demodulation, decoding and other processing, and then outputs the final receiving signal. The data.

In practical communication applications, in addition to responding to the useful signal of the light source sent by the transmitting end, the area array detector also responds to ambient light and may occupy more pixels. If the response signals of all the pixels are output, many unnecessary signals will be generated in the subsequent circuit, making it difficult to extract useful signals, and the effective transmission rate cannot be improved. To solve this problem, the present invention introduces a useful pixel output selection program in the receiving process: adopt two methods to confirm the useful pixel, and then select the output to reduce or eliminate the response output of the non-communication optical signal.

One is an image processing method based on the physical characteristics of the sending light source. First, in the initial imaging, the light source, such as a light emitting diode (LED), is used as the target to be detected in the entire image, and its position in the area array sensor is determined by pattern recognition. Whether the LED is bright or dim, current image processing capabilities are readily available. This method is equivalent to processing in space.

The second is to use auxiliary identification methods. The LED at the sending end cooperates with the identification of the receiving unit to send a signal with a specific pulse frequency and brightness, and the receiving end collects multiple images. Assuming that the ambient light does not change in a short period of time, or the way of change is different from the light pulse frequency of the LED, then by comparing multiple images collected and performing correlation calculations, you can find the response point that matches the specific LED light source. This locates its position on the area sensor. This method is equivalent to joint processing of space and time.

The above two methods can be used alone or together; they can be used after initiating a communication request, and can also be dynamically reused for further correction during the communication process.

When the position of the useful pixel is determined at a certain time, change the timing and frequency of the output driving signal, or the row and column address selection signal, so that it only selects the output of the useful pixel without considering the response of other pixels. Improve the efficiency of the transmission channel.

In a common optical communication system, the transmitted optical signal is a pulse, and the location of the data information, whether it is the pulse interval, pulse amplitude or pulse width, requires the output of the photoelectric sensor not to lose this information. However, general area array sensors do not output continuously but sample output. The sampling frequency and time are for low-rate image/video acquisition, so there is no matching relationship with high-speed communication optical pulses, and a lot of pulse information will be lost if applied directly. In order to solve this problem, the present invention introduces a measure of synchronous sampling in the optical receiver, that is, to ensure that the output at the sampling moment can completely record the information of the irradiated light pulse. Regardless of the pulse modulation method, the sampling circuit ensures that the photoelectric conversion unit is activated at the beginning of the pulse period to enter the charge accumulation process, converts the energy of the optical signal into charge, and outputs the charge at the end of the pulse period. In such a cycle, the information of each light pulse can be completely collected, and will not be confused with the information of adjacent light pulses.

When there is no photoelectric sensor with continuous output, the capture of the synchronous moment can adopt the search method. One of the methods is that the light source first sends a leading pulse train with a fixed pulse period and pulse width, and the output sampling circuit samples and outputs to the sensor with the same period. The synchronization moment is related to the phase. First, the sampling trigger pulse is sent with any phase, and then the phase of the subsequent trigger pulse is adjusted to change in the direction of output increase according to the size change between two consecutive output values. After several cycles, you can The maximum output value is captured and tracked continuously.

After solving the problem of output selection and synchronization, the image sensor can be used in a high-speed multi-channel optical receiver, and the high-frequency optical pulse signal can be output in real time without being interfered by other pixel signals.

Description of drawings

Fig. 1 is a composition diagram of the present invention. An LED light source (or an LED array) at the sending end forms an image on the image sensor through the photographic optical system, and the covered pixel responds to the LED light signal, converts it into an electrical signal and selects it for output.

Figure 2 shows the composition of a conventional camera system using a CMOS image sensor.

Figure 3 shows that when the sampling period is shorter than the period of the pulse signal, two different states of the signal can always be collected.

Fig. 4 is the dual output mode shift register of the present invention, the output is all discrete analog signals, can output the video signal as conventional serial, also can select the signal output of any pixel.

FIG. 5 is a process for identifying the imaging position of the communication light source used in the present invention on the image sensor.

Fig. 6 shows the phase relationship between the modulated light pulse sequence at the transmitting end, the sampling trigger time of the image sensor at the receiving end, and the output signal when synchronizing.

Figure 7 is the composition of the synchronous capture and tracking circuit output by the image sensor.

FIG. 8 is an input-output characteristic of an amplitude comparator of an image sensor output synchronization circuit.

FIG. 9 is a schematic diagram of a synchronous capture and tracking process of a trigger signal sampled and output by the image sensor at the receiving end.

Detailed ways

The optical receiver based on the area array photoelectric sensor can theoretically perform space division multiplex communication through the imaging system. However, to simplify the description of the present invention, a single sending light source is taken as an example. Figure 1 is a simplified optical communication system. At the optical sending end, the sent data stream is modulated by a modulator (1-1) into a pulse signal and sent out through a light-emitting diode (1-2). At the light receiving end, through the imaging optical system (1-3), the light source at the sending end is imaged on the image sensor (1-4) to form a light spot (1-5). The image sensor (1-4) is composed of many pixels, and each small square in the figure represents a pixel, but the 16×16 small squares shown in Figure 1-4 do not represent the actual size of the image sensor. Often it can be much larger. Usually, this kind of optical wireless communication system is mostly used for indoor short-distance communication, so the concentrated illumination area (1-5) of the LED (1-2) on the image sensor (1-4) often covers multiple pixels, Even hundreds of pixels, while the situation in Figure 1 is only represented by dozens of pixels. Other pixels outside this area also receive light, but instead of useful communication light, it is ambient light.

The row address decoder (1-6) in Fig. 1 outputs the row coding address under the control of the DSP controller (1-9), and selects the pixel of a certain row to output to the shift register (1-7). In the camera mode, a line of signals saved by the shift register (1-7) is serially output to the DSP signal processor (1-9) for processing to obtain the final image or video signal. Here DSP comprehensively expresses the relevant processing functions of discretely output video signals, including video amplification, analog-to-digital conversion, line/frame synchronization, image processing, etc. In the communication mode, the row address decoder (1-6) under the control of the DSP (1-9) transfers a specific bit or bits of a row signal transferred to a bit register (1-7) through a multiplexer Switches (1-8) are transferred directly into the DSP. In this way, any one or several picture elements in the image sensor can be selected for output by utilizing the cooperation of the address decoder and the MUX.

The output of conventional image sensors is for taking pictures or videography, so it outputs whole frames one by one. If it is a standard video, the output rate is at most dozens of frames per second. Although the total data rate is relatively high, up to several megahertz, for each pixel, there is still only a sampling output rate of dozens of times per second. This is too low for communication.

If the situation in Figure 1 is in the communication state, and the response signals of all pixels are serially output according to the conventional video output method, and then a useful communication signal is selected from the output signal, then two problems will be faced: One is that the response rate of a single pixel cannot be improved at all due to the limitation of the bandwidth of the output channel, and the other is that a large number of output signals are background light signals, which are useless. One of these two problems limits the data rate of optical transmission, and the other makes it difficult to extract useful signals. If each pixel is not output in parallel, even if measures are taken to increase the overall data output rate, it is still not feasible in practice.

The output mode of a conventional image sensor is shown in FIG. 2 , which is a typical output mode of a CMOS image sensor. The CMOS image sensor (2-1) is driven by the row address decoder (2-2), progressively or interlaced, and transfers the response signal of the photoelectric sensor to the horizontal shift register (2-3), which is similar to the analog TV line scan Different from a line of continuous analog signals output during the period, the signals here are discrete analog signals. The shift register is then driven by the master clock to output the signal bit by bit to the video amplifier (2-5) for proper amplification to match the analog-to-digital converter, and the analog-to-digital converter (2-6) converts the discrete analog signal into a pure The digital signal is sent to the digital signal processing unit (2-4) for corresponding image processing and storage. The output rate of the horizontal shift register determines the output rate of all pixels. In this driving mode, the output rate of any pixel is equal to the frame rate, depending on the image sensor and video standard, it is about tens to hundreds of times per second. For example, the frame rate of KODAK's KAC-9628 is 30fps, and its number of pixels is 648(H)×488(V).

The scheme used in the present invention no longer adopts the conventional output mode, but selectively outputs the response signal of the pixel. For the situation in Figure 1, only select the pixel signal output under the coverage of the light spot (1-5), so that the total output signal can be basically guaranteed not to be blocked and disturbed by the ambient light signal, and the total data output bandwidth can be determined by the useful Signal exclusivity thus greatly increases the data rate.

If you want to select the output of the useful pixel, you must first determine the position of the useful pixel in the image sensor, which can be expressed in the form of X and Y coordinates before it can be sent to the row and column address decoder for decoding and selection.

For image sensors, image output is a basic function, and it is a relatively mature technology to use image processing methods to locate specific light sources. Assuming that visible light LED is used as the light source at the sending end, a relatively simple positioning method is to cooperate with the transceiver to make the LED at the sending end fully bright. At this time, the image output by the image sensor at the receiving end will have a very obvious light spot in the background. This light spot The shape of the LED is the shape when it emits light. It can be recognized by the general algorithm of image recognition. In order to identify the imaging point of the light source more effectively and quickly, the cooperation of the light source will be very effective. One of the matching methods is to send periodic pulses as shown in 3-1 in Figure 3, the pulse duty cycle is 50%, and the period is greater than the frame period that the image sensor can output, as shown in 3-2 in Figure 3. Such a setting can enable the output image to capture different situations when the light source is bright and dark, and provide more relevant information for identification.

The identification process is shown in Figure 5. At the receiver side, as much feature information as possible about the light source is obtained first, and then multiple images I _n (u, v), n=1, 2, 3, . . . are collected continuously. If the wavelength of the light source is not considered, then binary processing of the image into a black-and-white image will greatly reduce the calculation amount of image processing. The selection of the threshold T during the binary processing can consider the contrast between the brightness of the ambient light and the brightness of the communication light source, which will further highlight the communication light source LED on the image and limit other background objects. For example, when the brightness of the LED is relatively high, you can choose a larger T. In this way, in the thresholded image, more ambient light is filtered out. Several consecutive images U _n (u, v) after thresholding processing, n=1, 2, 3..., respectively perform absolute value subtraction operations with their respective adjacent images to generate a new image F _n (u, v ), n=1, 2, 3.... In the absence of relatively fast movement between the transceivers and the flickering of the background light, the result of the subtracted image is that if the LEDs are all bright or dark when adjacent images are collected, there will be only There are few bright spots, and if the adjacent images are collected when the LED is bright and dark, the output result after subtraction will show a prominent LED spot. In F _n (u, v), select a picture of V _n (u, v) with the most adjacent bright spots, and perform pattern recognition on V _n (u, v), it is very easy to distinguish it and get its image sensor The specific location coordinates (u _i , v _i ) on where i=1, 2, 3, . . . , M. M is the number of pixels covered by the LED. These picture elements all respond to the same LED light signal, so their output signals are basically the same. According to the strength of the signal or the position of the pixel in the light spot, you can choose the one with the largest output or the one in the middle, or you can choose to output more than one. Adding multiple output signals can obtain diversity gain.

No matter what type of image sensor, it is not difficult to add output selection function. One of the simple methods of output selection is to replace the shift register (2-3) in Figure 2 with the circuit in Figure 4. Figure 4 combines the shift register with the multiplexer to realize the bit selection function. The method of realization is to make each bit output of the shift register (4-1) be connected to a multiplex switch 4-2 simultaneously except being connected to the next bit. A multi-way selector switch can select one or several outputs among numerous input channels. 4-2 is a high-speed analog switch. For the sake of convenience, only one output is shown in the figure. In fact, it can be constructed to select several outputs at the same time. Each bit input of 4-1 and 4-2 corresponds to a column pixel of the image sensor. After determining the position coordinates (u _i , v _i ) of the LED like this, it is supposed to select a pixel i=K output in M pixels, so the row address decoder (2-2) is arranged at u _K , setting the multi-way selector switch (4-2) at v _K , the signal of the pixel (u _K , v _K ) can be continuously output.

Whether it is an area array sensor capable of continuous output or an image sensor with sampling output, concentrating the output channels from all pixels to one or only a few pixels solves the problem of output selection, but to achieve high-speed communication, the sampling output image The sensor must also address the issue of synchronizing the discrete sampling of the output with the light pulse.

In order to explain the solution to this problem, assuming that the modulation method used in optical communication is a representative pulse amplitude modulation (PAM), then the optical emission source for communication will send pulses with different amplitudes at a fixed period and pulse width. A size-based rating represents one or more digits of digital information. If it can be divided into 4 levels, each pulse can represent 2bit information, and when it only needs to represent 1bit information, it degenerates into OOK modulation, that is, the presence or absence of pulses represents 0 or 1. The requirement for synchronous sampling of pulses in this modulation mode is that at the beginning of each pulse period, the sampling circuit starts the charge accumulation process of the photoelectric conversion unit, and samples the accumulated results at the end of the pulse or period.

Figure 6 shows a situation where an ideal synchronization state is reached. Corresponding to the data stream 6-1 to be sent, the optical pulse after OOK modulation is shown in 6-2. However, if PAM modulation is used, corresponding to data stream 6-3, the modulated optical pulse waveform should be as shown in 6-4. 6-5 is a synchronous sampling trigger pulse, the leading edge of which starts the photoelectric conversion process of the photodetector of the pixel, and its trailing edge triggers the output process, transferring the charge accumulated by the detector, or the voltage (if it has been converted) to the output circuit superior.

Taking the PAM modulation signal 6-3 as an example, at time t1, the light pulse reaches the image sensor, at this time the sampling signal 6-5 starts, and the photoelectric converter on the image sensor starts to convert the light energy into electric charge and accumulate it. At the elapsed time t2, the sampling signal resets the photoelectric converter and at the same time transfers the signal to output, and obtains the pulse signal between t2 and t3 as shown in 6-7. As the cycle continues, the optical pulse sequence 6-4 is converted into an electrical pulse signal sequence 6-7 for output. As for the light pulse sequence 6-2, it will be converted into an electrical pulse signal 6-6 for output.

The phases of the sampling signal 6-5 and the optical pulse 6-4 are strictly synchronized. In addition to ensuring the correct acquisition of the signal, it can also reduce the influence of environmental noise on the output signal caused by the sampling trigger being too late for the falling edge of the pulse. This is because after the light pulse disappears, the ambient light does not disappear, and the photoelectric conversion process still exists.

To obtain the above-mentioned synchronization state, the receiving end undoubtedly needs to know when the optical pulse has arrived, but it cannot know until the optical pulse is not obtained. Therefore, a search, capture and then track approach is required.

Fig. 7 shows the composition of the synchronous pulse phase capture circuit constructed in the present invention, which is a closed-loop adjustment system. 7-1 represents the image sensor and its peripheral output circuit, etc. After the image sensor receives the light pulse signal, it outputs a signal U under the trigger of the control pulse signal Pt. The size of U is not only related to the intensity of the irradiating light, but also related to the phase of the trigger signal Pt. The Pt signal comes from a pulse signal generator (7-6).

The signal U output by the image sensor is shaped by a peak hold circuit (7-2) and then converted into a digital signal by an ADC analog-to-digital converter (7-3). This signal is divided into two outputs, one is directly sent to the amplitude comparator (7-4), and the signal P2 generated after being delayed by the delayer (7-5) for T time is also sent to the amplitude comparator, and T is the optical pulse period time. The amplitude comparator compares these two signals to generate a signal S related to the amplitude of P1 and P2. The input-output characteristics of the amplitude comparator are shown in Figure 8, where △=P1-P2. The pulse generator (7-5) changes the phase shift of the generated pulse according to S: when S>0, the output signal is increasing, indicating that the phase adjustment direction is correct, and the phase of the pulse Pt continues to change along the original adjustment direction; when S< When 0, the output signal is decreasing, indicating that the phase adjustment direction is wrong, and the phase of the pulse Pt should be changed against the original adjustment direction; when S=0, it indicates that the signal is at the synchronization point and does not need to be changed.

Figure 9 is now used to illustrate the acquisition process of the sync pulse phase. Assume that the communication light source LED first sends a leading pulse to establish a communication relationship, with a period of T and a pulse width of τ, as shown in 9-1. The pulse generator 7-5 first generates an initial pulse of any phase, such as the first pulse in the sequence 9-2, to trigger the image sensor to enter the charge accumulation process of photoelectric conversion, as shown in 9-3, and then the Pt drops Output the signal along the edge, as shown in the first pulse of 9-4, and reset the photoelectric converter. The output pulse amplitude of 9-4 is proportional to the overlap time width between the Pt pulse 9-2 and the input light pulse 9-1. At this time, because P2 has no output yet, it is considered to be 0, so △=P1-P2>0, S>0. But since there is no reference point for phase adjustment in the initial state, it can be stipulated that the phase of the next pulse of Pt is delayed by a δ value relative to the previous pulse, thus generating the second pulse in 9-2. The signal U output after the end of the pulse is the second pulse of 9-4. At this time, the output of the amplitude comparator outputs a signal S less than zero according to the difference between the amplitudes of the two pulses before and after. Therefore, the next Pt pulse output by the pulse generator will adjust the phase in the opposite direction to generate the third pulse. Since the third pulse is closer to the exact phase than the second, its output becomes larger, such as the third pulse of 9-4. At this time, the signal S is positive, so the signal generator continues to adjust the phase along the original adjustment direction, and generates the fourth pulse of 9-2. This recursively generates the fifth output pulse. This is the maximum output position, but the circuit doesn't know it, and produces excessive phase adjustment on the next trigger pulse, causing the output pulse amplitude to decrease. At this time, since S<0, the phase of the next trigger pulse is adjusted in the opposite direction. Since the possible maximum point 9-5 has been found, the amplitude comparator will adjust the magnitude to be smaller, so that it gradually returns to the maximum point and stabilizes to enter the phase tracking process. After the preamble pulse ends, the transmitting end can send the data sequence (9-6), and the data signal (9-7) will be generated thereafter.

Since the pulse phase is tracked, and there is interference at the zero point of S, the adjustment step size of the Pt pulse phase affects the synchronization acquisition time and accuracy, but in practice, stable tracking can be achieved only by modifying the parameters.

So far, an embodiment of the present invention has been basically clearly described in terms of system configuration, characteristic functions, circuit composition, and algorithm.

The above-mentioned embodiments provided by the present invention are only one of methods for realizing the functions proposed by the present invention. It is well known that the circuits and algorithms to achieve the same function are diversified. In different communication network applications, the types of area array photoelectric sensors and the forms of received optical signals may also be different. Therefore, in order to achieve high-speed optical communication using area array photoelectric sensors However, the introduced selective output and synchronous sampling do not deviate from the basic concept and protection scope of the present invention.

Claims (6)

1. A light receiving unit is composed of an imaging optical system, an area array photoelectric sensor and corresponding output selection, control and signal processing circuits. 2. The photosensitive wavelength band of the area array photoelectric sensor described in claim 1 is visible light or invisible light. 3. The output of each pixel of the area array photoelectric sensor described in claim 1 is selected by the output selection circuit, and the output selection circuit can select all the pixel signals to output, and can also select a specific one or a part of the pixel signals to output. 4. The output selection circuit according to claim 3 selects the output pixel according to the imaging position of the light source sending the useful light signal on the area array sensor. The positions of these pixels on the area array sensor are identified and determined according to the geometric characteristics of the light source reflected on the output image, or the correlation of the specific frequency optical pulse response is identified and determined based on multiple images continuously output. 5. The output control circuit of the area array photoelectric sensor as claimed in claim 1 synchronizes the output sampling of the picture element with the communication light pulse frequency. 6. The synchronization described in claim 5 refers to the time relationship determined between the output sampling moment of the pixel and the illumination pulse, the output sampling moment of each pixel can be the same or different, and its purpose is to ensure that each selected The output of the pixel is the largest and the noise is the smallest.

Images