CN201191225Y - Weak pulse light signal detection device - Google Patents

Abstract

A weak pulse light signal detection device, the device includes a high-sensitivity photomultiplier tube, a preamplifier, a high-speed digital signal processor and a computer, the high-speed digital signal processor is divided into a high-speed A/D unit, an automatic threshold setting unit, a single photon discrimination output unit, an adaptive pulse width matching filter unit, a pulse extraction and positioning output unit and a demodulation unit, and the high-speed digital signal processor realizes single photon discrimination and communication signal processing through software. The utility model can solve the difficult problems of pulse light signal detection and extraction under different application occasions, background noise and pulse distortion. It is based on a high-speed DSP chip, has low cost, high integration, certain intelligence, high sensitivity, strong anti-noise ability, low bit error rate, simple structure and light weight, and can be used for the detection of weak pulse light signals modulated by pulse position in free space laser communication.

Description

Weak pulse light signal detection device

technical field

The utility model relates to a weak pulse light signal detection device under the background noise, the device can accurately detect and extract the weak pulse position modulation pulse light signal under the background noise, and then accurately extract the pulse The optical signal is correctly demodulated, and the demodulation result is displayed and stored in real time.

Background technique

With the continuous development of pulsed laser technology, laser is more and more widely used in the field of free space laser communication, the transmission distance is also increasing, and the application occasions are becoming more and more diversified. At the same time, with the increase of laser transmission distance, transmission The pulse distortion and attenuation caused by the channel are becoming more and more serious, the received energy has reached the level of a few photons per microsecond or even lower, and the width of the received pulse has also ranged from hundreds of nanoseconds to tens of microseconds or even longer. Usually free-space laser communication mostly adopts the pulse position modulation method, so the distortion of the pulse seriously affects the correct demodulation of the signal, thus putting forward higher and higher requirements for the photoelectric detection technology. The traditional analog photoelectric detection technology has been difficult to meet the communication requirements. Compared with traditional photoelectric detection technology, single photon detection technology has higher sensitivity and can be used to detect weak light signals. Based on the uncertainty and unpredictability of the signal during communication, various requirements are put forward for the reception of weak optical signals:

①Adopt single-photon detection technology to meet ultra-high sensitivity requirements;

②It can extract signals from the background noise environment, and can automatically set the single-photon discrimination threshold according to different application scenarios and different background noise situations, and can also resist certain burst noise;

③Able to automatically adjust the parameters of the matched filter according to the pulse width of the received signal, accurately extract the pulse position, and correctly demodulate the extracted pulse signal to meet the performance requirements of the communication bit error rate;

④ It can work continuously in real time, with a high degree of device integration, stable and reliable work, and low cost.

The existing single-photon detection technology mostly uses imported single-photon counting cards, which are expensive, have poor development flexibility, and cannot meet specific needs. The scheme of single-photon energy screening is mainly used to solve the problem of high sensitivity. However, for actual laser As far as the communication application system is concerned, it is usually in the background light noise, and the characteristics of the background noise will also change with the different application occasions; in addition, the traditional single photon detection technology usually uses the method of counting the signal pulse in a fixed pulse width period. Identification and extraction are difficult to adapt to occasions where the pulse width changes, and it is easy to cause bit errors.

Contents of the invention

The problem to be solved by the utility model is to overcome the above-mentioned difficulties encountered in the application of the traditional photoelectric detection technology and the general single photon detection technology in the free space laser communication, and provide a weak pulse light signal detection device, which can solve the problem in different applications. The problem of detection and extraction of pulsed optical signals under occasions, background noise and pulse distortion, based on high-speed DSP chip, low cost, high integration, certain intelligence, high sensitivity, strong anti-noise ability, low bit error rate, simple structure And light in weight, it can be used for the detection of pulse position modulation weak pulse light signals in free space laser communication.

The technical solution of the utility model is as follows:

A weak pulse light signal detection device is characterized in that it includes a high-sensitivity photomultiplier tube, a preamplifier, a high-speed digital signal processor and a computer, and the high-speed digital signal processor is divided into a high-speed A/D unit, an automatic threshold setting A unit, a single photon discrimination output unit, an adaptive pulse width matched filter unit, a pulse extraction and positioning output unit and a demodulation unit.

The detection method of the weak laser pulse laser signal by using the above-mentioned device comprises the following steps:

①Preparation: The high-speed digital signal processor is powered on and reset, and the crystal oscillator of the high-speed digital signal processor provides the main clock signals f ₀ and T ₀ , and outputs f ₁ and T ₁ through the internal frequency divider as The high-speed A/D unit data collects the A/D clock signal, and then sends the clock signal to the adjustable counter, initially sets the count value, and provides the adaptive pulse width matching filter unit with a clock signal f with an initial matched filter pulse width T _{2 2} , _T2 ;

② Photoelectric conversion: the high-sensitivity photomultiplier tube converts the weak light signal into an electrical signal and amplifies it through a small-signal low-noise preamplifier circuit and sends it to the high-speed A/D unit of the high-speed digital signal processor to convert the analog signal is a digital signal;

依据误码率性能的要求和背景噪声设置甄别阈值为： $D=Q*\sqrt{N_0},$ 其中Q为正整数，即甄别阈值为背景噪声的倍数，随之转入甄别模式，单光子甄别输出单元依次对高速A/D单元输出的数字信号进行甄别，确保在噪声情况下无输出，即输出为“0”，在有信号到达时，输出为“1”；③ Adaptive threshold screening: After the high-speed digital signal processor is in a stable working state, the high-speed digital signal processor clears all registers of the adaptive threshold setting unit, and first works in the threshold mode to receive the high-speed A/D unit The output digital signal is statistically averaged within 1 second to obtain

According to the requirements of bit error rate performance and background noise, the screening threshold is set as: $\mathrm{D.}=Q*\sqrt{{\mathrm{<figure-callout\; id="0"\; label="N"\; filenames="Y20082005806100113.png"\; state="\{\{state\}\}">N</figure-callout>}}_0},$ Where Q is a positive integer, that is, the discrimination threshold is a multiple of the background noise, and then enters the discrimination mode, and the single-photon discrimination output unit sequentially discriminates the digital signal output by the high-speed A/D unit to ensure that there is no output in the case of noise, that is The output is "0", when a signal arrives, the output is "1";

④ Adaptive pulse width matched filtering and maximum likelihood detection: According to the maximum likelihood judgment criterion and the principle of matched filtering, the initial likelihood judgment threshold N' is determined, and the adaptive pulse width matching filter unit is within the pulse width T ₂ , for single The output results of the photon discrimination output unit are accumulated, and the accumulated result N is compared with the initial likelihood judgment threshold N', if it is greater than that, it is regarded as a signal pulse, and then the number of the longest continuous "1" at this time is counted, and the automatic adjustment The accumulated pulse width _T2 and the likelihood judgment threshold N' ensure accurate matching of the pulse signal, and according to the maximum likelihood criterion, sequentially perform adaptive pulse width matching filtering and maximum likelihood detection signal processing on subsequent input signals;

⑤Pulse extraction and positioning: The pulse extraction and positioning output unit completes the detection and positioning of weak pulse signals;

⑥Demodulation and result display: The demodulation unit performs real-time demodulation of the pulse position information accurately extracted by the pulse extraction and positioning output unit, and sends the demodulation result to the subsequent computer for storage and display to complete communication.

The utility model has the advantages of:

① The utility model uses a high-speed digital signal processor (DSP) combined with a high-speed A/D to realize the self-adaptive setting of the discrimination threshold and pulse width. Therefore, the device is small in size, light in weight, high in integration, and high in cost performance;

② The utility model adopts the single photon counting detection technology, which has high detection sensitivity and can reach the single photon level;

③The utility model adopts the technology of automatically setting the screening threshold, which can automatically set the judgment threshold in different application occasions according to the background noise intensity and bit error performance requirements, which reduces the requirements for the signal-to-noise ratio of the received signal, broadens the application occasions, and reduces the Manual setting and adjustment links;

④ The utility model adopts the technology of self-adaptive pulse width, which can be adjusted to the optimal pulse width according to the actual pulse width based on a preset suboptimal pulse width when the received pulse signal is distorted, thereby obtaining The best matched filtering effect ensures the accurate extraction of the pulse position and the subsequent correct demodulation, which improves the bit error performance of the system. At the same time, because it is based on the maximum likelihood judgment, it also has a strong ability to resist burst noise;

⑤ The utility model solves the problem of accurately extracting weak pulsed light signals in different applications, background noise and pulse distortion, and can accurately demodulate them based on the modulation method adopted by the sending end to meet the communication performance requirements. It can be widely used in free space laser communication, has broad application prospects, can promote the continuous practical use of related technologies, and has good market prospects.

Description of drawings

Figure 1 is a schematic diagram of the overall structure of the weak pulse light signal detection device of the present invention;

Fig. 2 is the schematic diagram of division of internal logical function units of DSP3 in the weak pulse light signal detection device of the present invention;

Fig. 3 is the timing diagram of the weak pulse optical signal detection device of the utility model based on the high-speed DSP, automatically setting the threshold value and the single-photon discrimination and counting technology of the self-adaptive pulse width;

Fig. 4 is a flow chart of the weak pulse optical signal detection device of the present invention based on high-speed DSP, automatic threshold setting and self-adaptive pulse width single photon discrimination and counting technology.

In the figure: 1-high-sensitivity photomultiplier tube, 2-pre-low noise amplifier circuit, 3-high-speed digital signal processor (DSP), 4-computer, 31-adjustable frequency divider, 32-high-speed A/D unit , 33-single-photon discrimination automatic threshold setting unit, 34-single-photon discrimination output unit, 35-adaptive pulse width matching filter unit, 36-pulse extraction and positioning output unit, 37-settable counter, 38-demodulation unit.

Detailed ways

Below in conjunction with embodiment and accompanying drawing, the utility model will be further described, but should not limit the protection scope of the utility model with this.

The utility model device takes the free space laser communication in a certain occasion as an example, adopts the pulse position modulation mode, and the pulse width of the signal received is expected to be 250ns, the signal strength is less than 1nw, and the bit error rate requirement is better than Pe<=10- 5.

Please refer to Fig. 1 and Fig. 2 at first, Fig. 1 is the overall structural diagram of the weak pulsed light signal detection device of the present invention, Fig. 2 is the internal logic function unit division diagram of DSP3 in the weak pulsed light signal detection device of the present invention, can be seen from the figure , the utility model weak pulse light signal detection device, including high-sensitivity photomultiplier tube 1, preamplifier 2, high-speed digital signal processor 3 and computer 4, described high-speed digital signal processor 3 is divided into high-speed A/D unit 32. Automatic threshold setting unit 33, single photon discrimination output unit 34, adaptive pulse width matching filter unit 35, pulse extraction and positioning output unit 36 and demodulation unit 38, single photon discrimination and communication signal processing realized by software.

Please refer to FIG. 3 . FIG. 3 is a timing diagram of the device and method for detecting weak pulsed light signals of the present invention. It can be seen from Fig. 3 that the weak pulsed light signal detection device of the present invention is provided with a main clock by a 160MHz high-frequency crystal oscillator with a high frequency stability in the DSP3 - the frequency is f ₀ and the period is T ₀ , and then the main clock is divided by the internal adjustable Frequency divider 31 is set to divide by four to obtain the clock frequency f ₁ and cycle T ₁ for performing 40MHz high-speed A/D conversion on the signal to be detected, and the detection signal is sampled in cycle T ₁ and judged at the falling edge of the pulse The threshold is judged, and the output "1" or "0" signal is sent to the digital buffer for storage; in addition, the clock frequency f ₂ and cycle T ₂ for matching filtering are obtained after the clock f ₁ is set by another adjustable counter 37, Accumulate the digital signals in the buffer period T ₂ , detect and extract the pulse signal after matched filtering, if the best matching filtering effect is not achieved, the adjustable The counter is adjusted to achieve the best matching effect, so as to accurately extract the position of the pulse signal.

Please refer to Fig. 4 at last, Fig. 4 is the DSP program flowchart of the detection device of the present invention. It can be seen from Figure 4 that after the DSP is powered on and reset, it first obtains the clocks f ₁ and f ₂ by setting bits 31 and 37, and obtains the initial likelihood judgment threshold by setting bit 35, and then starts the high-speed A/D32 to process the detected analog signal. Analog-to-digital conversion, and the automatic threshold setting unit 33 works in the threshold mode, automatically sets the single-photon discrimination threshold according to the background noise and performance requirements, and then the single-photon discrimination unit 34 judges whether it is a single photon according to the set threshold, and through the available The adjustment counter 37 counts it, and then the adaptive pulse width matching filter unit 35 performs matching filtering and maximum likelihood judgment on the counting result, and the pulse extraction and positioning output unit 36 extracts the pulse position according to the previous judgment result and sends it to demodulation The unit 38 performs demodulation, and at the same time corrects the parameters of the adjustable counter 37 by counting the number of the longest continuous "1" to achieve the best matched filtering effect, which is completed by the revised matched filtering and likelihood detection For the detection of the subsequent input signal, the final demodulation unit 38 will correctly demodulate the received signal according to the received pulse position information, and send the correct demodulated result to the computer 4 in real time through the serial port, and finally the computer 4 will pass the serial port. The received signal is stored and displayed in real time.

The concrete device that present embodiment adopts is: described high-sensitivity photodetector 1 is Hamamatsu (HAMAMATSU) photomultiplier tube; Described pre-low noise amplifier circuit 2 is its special-purpose small signal low noise amplifier module; Computer PC4 is P4 Computer, follow-up adjustable frequency divider 31, high-speed A/D unit 32, single-photon discrimination automatic threshold setting unit 33, single-photon discrimination output unit 34, adaptive pulse width matching filter unit 35, pulse extraction and positioning output unit 36, Both the adjustable counter 37 and the demodulation unit 38 are realized by a single-photon discrimination and signal processing module 3 implemented by TMS320VC5416-160 fixed-point high-speed DSP as the main chip and software. Provides many multi-function instructions, adopts pipeline work mode, and has strong parallel computing ability. Wherein the 40MHz high-speed A/D unit 32 converts the input data and sends it to the DSP data port, and controls the realization of the above-mentioned functions through software programming; the main clock f ₀ is provided by the 160MHz high-stability crystal oscillator in the chip; the main clock can be set for frequency division The device 31 divides the frequency by four to obtain f ₁ =40MHz, and the period is 25ns, that is, the A/D conversion rate is 40Mbps. The value obtained after the A/D conversion is compared with the single photon decision threshold, and the decision output is "0" or "1". ; After the initial frequency division of the clock by the adjustable counter 9, f ₂ =4 MHz and a cycle T ₂ of 250 ns are obtained, which are used as the initial matched filter pulse width.

Combining with Fig. 1, Fig. 2, Fig. 3 and Fig. 4, the specific process of weak pulse light signal detection of the single photon discrimination and counting technology of automatically setting the threshold and self-adaptive pulse width in the embodiment of the utility model is as follows:

并作为单光子甄别的阈值取值的参照；① First, turn on the device before the signal arrives to make each component enter the working state, and set the adjustable frequency divider 31 and the adjustable counter 37 to obtain clocks f ₁ and f ₂ respectively. At this time, no signal arrives, and DSP3 works at Threshold mode, that is, what PMT1 detects is background noise n(t), after the signal is converted by 40M high-speed A/D unit 32 after pre-amplification 2, single photon discrimination automatic threshold value setting unit 33 performs it within 1S Statistically averaged

And as a reference for the threshold value of single photon discrimination;

②Secondly, according to the best judgment criterion, the judgment threshold based on the required bit error rate performance is set to be no less than twice the background noise, because the minimum error rate is The code rate can be expressed as:

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; e</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}}{<span\; class="notranslate">\; \&Integral;</span>}_{\mathrm{<span\; class="notranslate">\; Q</span>}}^{<span\; class="notranslate">\; \&infin;</span>}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span>{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; dx</span>}$

In the formula $Q=\mathrm{D.}/\sqrt{{\mathrm{<figure-callout\; id="0"\; label="N"\; filenames="Y20082005806100113.png"\; state="\{\{state\}\}">N</figure-callout>}}_0},$ Indicates that when judging the "0" code, the judgment threshold D exceeds the multiple of the average noise, which contains the concept of signal-to-noise ratio. The embodiment of the utility model considers the performance requirement of bit error performance P _e <= 10 ^-5 , referring to the above formula The discrimination threshold D is taken as 4 times of the background noise, namely $\mathrm{D.}=4*\sqrt{{\mathrm{<figure-callout\; id="0"\; label="N"\; filenames="Y20082005806100113.png"\; state="\{\{state\}\}">N</figure-callout>}}_0},$ It is equivalent to the signal-to-noise ratio (SNR) requirement of the received signal as signal-to-noise ratio (SNR)>=4. Compared with traditional photoelectric detection, the signal-to-noise ratio (SNR)>= under the same bit error rate performance requirements 16 requirements, reducing the requirements for receiving signals, that is, improving the receiving sensitivity and receiving performance. After the single-photon discrimination automatic threshold setting unit 33 sets the discrimination threshold according to the background noise and bit error rate performance requirements, the DSP3 turns into the discrimination mode;

③ At this time, the weak optical signal is converted into an electrical signal by the high-sensitivity photodetector 1 and amplified by the small-signal low-noise amplifier circuit 2, then s(t) is sent to the subsequent circuit for processing, and at the time set by the adjustable frequency divider 3 The period T ₁ =25ns is converted to 40M high-speed A/D by the high-speed A/D unit 32, and is screened by the single-photon discrimination output unit 34 according to the discrimination threshold set above to ensure that there is no output in the case of noise, that is, output s (j) is "0", and there is an output when a signal arrives, that is, the output s(j) is "1";

④ Subsequently, according to the maximum likelihood decision criterion and the principle of matched filtering, the adaptive pulse width matched filter unit 35 accumulates the automatically set threshold judgment output results within a preset certain pulse width T ₂ =250ns, and according to the accumulated result S ₀ (j)=“N” According to the maximum likelihood criterion, compare the accumulated result N with the initial likelihood decision threshold N′ to judge whether it is a real optical signal pulse rather than burst noise, if it is greater than that, it is considered as a signal pulse, Then count the longest continuous "1" number at this time, cooperate with the pulse extraction and positioning output unit 36 to control the adjustable counter 37 to automatically adjust the accumulated pulse width _T2 and the likelihood judgment threshold N ', to ensure the accuracy of the pulse signal Accurate matching, and according to the maximum likelihood criterion, sequentially perform adaptive pulse width matching filtering and maximum likelihood detection signal processing on subsequent input signals. The pulse extraction and positioning output unit 36 cooperates with the aforementioned adaptive pulse width matched filter unit based on the maximum likelihood decision criterion, adaptively sets the likelihood decision threshold N′ according to the output of the matched filter, and sets the pulse width T ₂ through the feedback of the adjustable counter. At the same time, based on the results of matched filtering and maximum likelihood judgment, it can accurately judge whether it is a real optical signal pulse rather than burst noise, and eliminate burst noise, so that it is accurate when the maximum likelihood judgment result is "1". The position of the pulse signal is stored at this moment and sent to the subsequent demodulation and decoding process in real time, that is, the useful weak pulse light signal is extracted from the received light signal and the pulse position is precisely positioned, and the weak pulse light signal is completed. Detection and positioning.

The reason for using matched filtering is that among all linear filters in a white noise environment, the signal-to-noise ratio output by the matched filter is the largest. The response function h(j) of the optimal matched filter is the mirror translation function of the input signal s(j). Ideally, the pulses after discrimination and judgment are rectangular series of length P, so h(j) takes a length in practice is the unit rectangle series of P, then the output is:

${\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span>$

And in terms of noise the output noise is:

${\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span>$

Since n(j), n(j-1), n(j-2), ... n(j-p+1) are not correlated with each other, N ₀ (j) will not be enhanced by correlation, so the matched filter output The maximum peak signal-to-noise ratio of the input will be higher than the original input signal-to-noise ratio, which is more conducive to accurate extraction of pulses and finding the pulse position. For the present utility model, the initially set pulse length is the unit rectangle of P=10 (that is, the actual width of the detected pulse is considered to be close to T ₂ =10×T ₁ =250ns) for matching filtering, and then through the detected output situation The time length of the matched filtering is automatically adjusted to P=14 (that is, the actually detected pulse width T ₂ =14×T ₁ =350s), so as to achieve an optimal matched filtering.

⑤Based on the previous steps, the pulse extraction and positioning output unit 36 can eliminate the burst noise and accurately judge whether the received signal is a useful pulse light signal under the background noise situation, and can accurately extract the position of the pulse at the same time, and the feedback control can The modulation counter adjusts the pulse width, and finally the demodulation unit 38 demodulates the received signal according to the pulse position information accurately detected and extracted, and finally the demodulation result is sent to the P4 computer 4 in real time through the serial port, and the P4 computer 4 Receive, store and display in real time through the serial port to complete the communication.

Claims (1)

1, a kind of faint pulsed optical signals pick-up unit, it is characterized in that comprising high sensitivity photomultiplier (1), prime amplifier (2), high speed digital signal processor (3) and computing machine (4), described high speed digital signal processor (3) is divided into high-speed a/d unit (32), automatic threshold is provided with unit (33), single photon examination output unit (34), self-adaptation pulsewidth matched filtering unit (35), DISCHARGE PULSES EXTRACTION and location output unit (36) and demodulating unit (38).

Images