CN116405205A - A method for generating quantum random numbers after real-time entropy evaluation - Google Patents

Abstract

The invention provides a method for generating a real-time entropy evaluation post-processing quantum random number. The quality of the random number determines the privacy and the safety of a communication system, side information is inevitably introduced in the quantum random number generation process due to quantum detection, the quantum noise entropy content of the system must be strictly estimated, and the random number is post-processed by using an extractor which is proved by a safety information theory. The invention provides a method for reconstructing the space distribution of quantum state phases in real time in a continuous variable quantum random number generation scheme and measuring the deviation between the space distribution and an ideal value, and updating entropy evaluation in real time when the deviation reaches a re-judgment threshold value, so that the scale of a post-processing matrix is adjusted in real time; the seeds for constructing the random extractor are updated in real time, and the information theory safety generalized hash post-processing of the multipath quantum random numbers is realized in parallel on the basis. The invention provides a random number real-time entropy evaluation post-processing scheme with intensive cost and strong expansibility, which can effectively improve the actual safety and practicability of a quantum random number generator.

Description

A method for generating quantum random numbers after real-time entropy evaluation

technical field

The invention belongs to the technical field of encryption, and in particular relates to a method for generating quantum random numbers after real-time entropy evaluation.

Background technique

Random codes are the key to encryption technology and fundamentally determine the security of information systems. Quantum random number generators are based on the inherent uncertainty of quantum physics and have provable randomness from information theory, providing the safest random number generation scheme. In the past decade or so, research in this field has focused on breaking through higher yield thresholds and passing more comprehensive statistical test suites. Recently, the field of information security has increasingly realized that the research focus of random numbers should shift to the rigorous evaluation of their entropy sources. Strictly speaking, any flaw in the physical implementation of a quantum random number generator may leak information related to generating random numbers, so-called side information. Correlations such as classical or quantum could be exploited by an eavesdropper to guess the measurement. In the presence of side information, the maximum amount of randomness that can be extracted is determined by the quantum conditional minimum entropy. The entropy evaluation in the continuous variable quantum random number generation scheme reported in the past is done once and for all. Instead of continuously monitoring the dynamic characteristics of the system and proving its security, the statistical test suite that is also applicable to pseudo-random numbers is applied to the output results of privacy enhancement. For testing, only the minimum entropy real-time, reliable prior evaluation can guarantee its security and randomness.

Among the quantum random number generation schemes, the random number extraction scheme based on the fluctuation uncertainty measurement of the quantum orthogonal components of the vacuum state is particularly promising for practical application. The vacuum state is not affected by external physical quantities and cannot be controlled or correlated by attackers. In practical applications, with the continuous increase of communication distance and communication time, higher requirements are bound to be placed on the generation rate of quantum random numbers and real-time security, and a problem that cannot be ignored is that the physical entropy source is During working hours, its entropy may change under the influence of variable factors such as temperature and noise, which affects the high stability and safety of the quantum random number generator under long-term working conditions.

Contents of the invention

Aiming at the problems of the prior art, the present invention provides a method for generating a quantum random number after real-time entropy evaluation, which specifically includes the following steps:

Perform entropy evaluation on the original true random sequence; reconstruct the distribution of the vacuum state Wigner function in the phase space, use the KLD relative entropy to measure the gap between the experimental vacuum state and the ideal vacuum state, and grasp the change of the quantum state purity in real time, according to the laboratory safety environment The obtained minimum entropy lower bound fluctuation range obtains the exact threshold value. If the actual runtime fluctuation exceeds this threshold value, it is considered to be attacked or interfered by a third party, and then the minimum entropy evaluation is re-evaluated to obtain the real-time Toeplitz post-processing matrix scale;

Based on the hardware conditions of the field programmable logic gate array and the scale of the Toeplitz matrix, design a three-stage parallel pipeline algorithm whose matrix size can be dynamically changed with the real-time evaluated quantum entropy, and update the random seed for constructing the Toeplitz hash matrix in real time. Real-time generalized hash extraction for information-theoretic security;

Based on the advantages of parallel computing of Field Programmable Logic Gate Array, real-time extraction of more than three different quantum sideband modes is realized, and multi-channel real-time parallel high-speed generation of safe quantum random numbers.

Further, the generation method of the original true random sequence is:

Build a balanced zero-beat detection system for the vacuum state of the light field, extract the vacuum quantum fluctuation as the entropy source for generating true random numbers, make the background light and the vacuum field interfere with a 50/50 optical beam splitter, and output light through the beam splitter The two beams of light signals that are strong and basically equal are subjected to balanced zero-beat detection to obtain a photocurrent signal, and the photocurrent signal obtained by the balanced zero-beat detection is mixed with the radio frequency signal and filtered by a filter, and then converted to an analog-to-digital by an analog-to-digital converter Get the original true random sequence.

Further, the optimal sampling range needs to be determined in advance when performing entropy evaluation on the original true random sequence.

Further, the method to determine the optimal sampling range is:

Use a voltage amplifier to continuously adjust the AC voltage gain, use an oscilloscope to collect the amplified outputs of each channel, and perform frequency statistics on the collected data. When the frequency beyond the sampling range of the oscilloscope is approximately equal to the frequency of the highest frame in the middle, in the collected true random sequence The proportion of "0" and "1" is similar, which is the optimal sampling range of random numbers.

Further, the three-stage parallel pipeline algorithm with variable matrix scale includes:

The first level is the construction of the sub-matrix, and the sub-seeds are sequentially selected from the BRAM through the address to construct a random sub-matrix;

The second stage is the operation of the sub-matrix, and the constructed sub-matrix is multiplied with the original random sequence to obtain the sub-matrix multiplication result;

The third stage is the XOR of the sub-matrix results, and performs XOR on the sub-matrix multiplication results obtained in each cycle.

Further, the construction of the sub-matrix includes two modules: storage of random seeds and selection of random seeds.

Further, the dynamic matrix scale transformation after the real-time entropy evaluation includes the following three steps in total;

The first step is to establish a minimum entropy physical model that removes side information; calculate the transfer function of the balanced zero-beat detection system, and calibrate the power spectrum output by the device; under the optimal sampling range of the ADC, the signal variance, Conditional signal variance and conditional excess noise variance; establish the distance between system parameters and safety parameters, that is, confidence intervals, and strictly calibrate the minimum entropy lower bound.

The second step is real-time entropy evaluation feedback; detect the change of the minimum entropy lower bound, and prevent quantum attacks, reconstruct the distribution of the vacuum state Wigner function in the phase space, and use the KLD relative entropy to measure the experimental vacuum state and the ideal vacuum state The gap between quantum states can be grasped in real time, and the real-time post-processing matrix scale can be obtained by combining the minimum entropy lower bound fluctuation range; after obtaining the real-time matrix scale, take four parallel post-processing channels as an example: the host computer transmits eight Bit instructions to programmable gate arrays to represent sixty-four matrix sizes for four parallel post-processing channels;

The third step is to dynamically change the size of the post-processing matrix inside the FPGA; first, construct the selection circuit, assuming that the largest matrix size is j×k, and the size of the sub-matrix is j×q, it can be seen that there are k/q different The sub-matrix multiplication result of XOR is the total number of cycles, which is also the total number of sub-matrices contained in different matrix sizes, here we use y to represent. After receiving the instruction, decode and analyze to obtain the matrix size of the corresponding channel. After completing a large matrix multiplication, select the corresponding sub-matrix multiplication result according to the corresponding matrix size, XOR the total number of cycles to realize the dynamic matrix size transformation after real-time entropy evaluation .

Further, when collecting optoelectronic signals, the processing clock of the field programmable logic gate array is reduced to half of the sampling clock of the analog-to-digital converter, the buffer is used to drive the global clock and the global reset high fan-out signal is inserted into the critical path Stage pipeline registers reduce logic delay; use register replication method for high fan-out registers to reduce register fan-out.

Further, the background light is a continuous wave laser output by a LD-TC40 semiconductor laser with a center wavelength of 1550nm;

Described radio frequency signal is to adopt the HP8648A type signal generator of 100kHz～4000MHz to produce the signal of different frequency bands;

The balanced zero beat detection adopts 1.6GHz PDB480C-AC type balanced detector to detect _;

Described filter is the ZFM-11+ type mixer of frequency range 1MHz-2GHz and the frequency is the BLP-100+ type low-pass filter of 100MHz;

The field programmable logic gate array model is Xilinx Kintex-7 XC7K325T;

The analog-to-digital converter model is ADS42LB69.

The present invention has one of the following technical effects:

(1) The present invention realizes random dynamic selection of a true random seed for constructing a hash post-processing matrix. Store n sub-seeds generated by shifting m true random seeds and a number of true random numbers with a length of a bits in different on-chip block memories of the FPGA, and the true random numbers of a bits randomly represent 0~2 ^a -1 ; Divide 2^a into m equal parts, each time a matrix multiplication is completed, select the corresponding true random seed through the size of the random number to construct a new Toeplitz hash matrix; A new a-bit true random number is written in the memory address to eliminate the periodicity of use. Randomly dynamically selecting a true random seed to construct a hash post-processing matrix can improve the security of random numbers and eliminate the non-random factors introduced by using a constant true random seed to construct a Toeplitz matrix.

(2) The present invention realizes real-time entropy evaluation and real-time dynamic change of Toeplitz matrix in the true random number generation process. While true random numbers are generated at high speed, raw random data is collected in real time for entropy evaluation, and the real-time extraction ratio is obtained and the exact size of the Toeplitz hash matrix is calculated; y types of matrices are preset in the field programmable logic gate array Scale, according to the real-time entropy evaluation results, the host computer transmits corresponding instructions to the FPGA through the UART serial port; the FPGA decodes and analyzes the instructions to obtain the matrix size of each channel, after completing a large matrix multiplication The selection circuit selects the corresponding sub-matrix multiplication result to XOR the total number of cycles to realize the dynamic transformation of the Toeplitz hash matrix. It ensures the real-time high-speed generation of random numbers and effectively guarantees its security.

(3) The excellent timing sequence designed by the present invention ensures the high stability and safety of the quantum random number generator under long-term working conditions. During the 12-hour long-term test, the generated random numbers all passed NIST, Diehard and TestU01 test.

(4) The high-speed random sequence produced by the present invention makes full use of the parallel computing advantages of logic operation hardware units, realizes the generalized hash post-processing of multi-channel quantum random numbers in parallel, generates quantum random numbers in real time and at high speed, and finally can reach 11Gbts/s Generate rate.

Description of drawings

The drawings illustrate various embodiments, generally by way of example and not limitation, and together with the description and claims serve to describe embodiments of the invention. Where appropriate, the same reference numbers will be used throughout the drawings to refer to the same or like parts. Such embodiments are illustrative, and not intended to be exhaustive or exclusive embodiments of the apparatus or method.

Fig. 1 is a schematic diagram of a balanced zero-beat detection system of the present invention;

Fig. 2 is the spectrogram of vacuum noise among the present invention;

Fig. 3 is a schematic diagram of the signal-to-noise ratio of the present invention;

Fig. 4 is the post-processing algorithm flowchart of the present invention;

In the figure 1, 1550nmLD-TC40 semiconductor laser; 2, optical attenuator; 3, 50/50 beam splitter; 4, low-pass filter; 5, photoelectric detector; 6, balanced zero-beat detector; 7, hybrid 8. Signal generator; 9. 100M low-pass filter; 10. Oscilloscope; 11. Analog-to-digital converter; 12. Field programmable gate array; 13. UART serial port; 14. Host computer.

Detailed ways

It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and embodiments.

An embodiment of the present invention provides a method for generating a quantum random number after real-time entropy evaluation, including the following steps:

Step 1: Build a balanced zero-beat detection system for the vacuum state of the light field. The balanced zero-beat detection system of the light field vacuum state is shown in Fig. 1 . The LD-TC40 semiconductor laser 1 with a central wavelength of 1550nm is used to continuously output continuous wave laser light as the background light of the balanced zero-beat detection system. After passing through the attenuator 2, it interferes with the vacuum light through the 50/50 optical beam splitter 3 to produce basically equal light intensity. The reflected light and the transmitted light; the reflected light and the transmitted light are respectively detected by a pair of photodetectors 5 with high quantum efficiency and symmetrical performance. The two detectors convert the detected optical signal intensity fluctuations into broadband photocurrent signal fluctuations, and the difference signal of the photocurrent signals is proportional to the orthogonal component fluctuation of the vacuum field, and amplifies it to the macroscopic level. As shown in Figure 3, the signal-to-noise ratio of quantum noise and low-frequency part noise is above 10dB, which can be used as the quantum entropy source of this type of quantum random number generator. In this step, a 1.6GHz PDB480C-AC balanced detector is used for balanced zero-beat detection.

Step 2: As shown in Figure 1, the photocurrent output of the balanced zero-beat detector 6 is mixed with the RF signal 8 respectively through the ZFM-11+ type mixer 7 with a frequency range of 1MHz-2GHz, and then passed through a frequency range of 100MHz BLP-100+ type low-pass filter 9 performs filtering to reduce the bandwidth to 100M, so as to satisfy the Nyquist double sampling law; then it is collected by an analog-to-digital converter 11 with a model ADS42LB69 and a maximum sampling rate of 250MS/S .

Step 3: Collect the beat frequency signal to establish the minimum entropy physical model to remove side information. The ADC collects the beat frequency signal output by the balanced zero-beat detection system at each frequency within the detector bandwidth, uses the average periodogram method to reconstruct the power spectrum, and calculates the power The spectrum is calibrated. Under the optimal sampling range of the ADC, the signal variance, conditional signal variance and conditional excess noise variance are obtained through the calibrated power spectrum. Establish the distance between the system parameters and the security parameters, that is, the confidence interval, strictly calibrate the minimum entropy lower bound and obtain the fluctuation range of the minimum entropy.

Step 4: Real-time entropy evaluation and feedback. Collect the original random sequence, detect the change of the lower bound of the minimum entropy, and take precautions against possible quantum attacks. For example, when encountering a thermal attack, using the minimum entropy calculation of the quantum condition will increase its entropy value. At this time, the vacuum state Wigner function The distribution in the phase space will become wider and shorter, and the KLD relative entropy is used to measure the gap between the experimental vacuum state and the ideal vacuum state, and the quantum attack can be detected in time by grasping the change of the quantum state purity in real time. Combined with the minimum entropy fluctuation range for threshold judgment, decide whether to re-evaluate the minimum entropy of the quantum condition, and obtain the matrix size of the post-processing.

Step 4: Based on the concurrency characteristics of field programmable logic gate arrays, design a three-stage parallel pipeline algorithm with variable matrix size, and update random seeds in real time to construct Toeplitz hash matrix for generalized hash extraction of quantum random numbers. Based on the concurrency characteristics of the field programmable gate array, a large matrix operation is split into multiple small matrix operations of appropriate size. Dividing a large matrix operation of one cycle into sub-matrix operations of multiple cycles relieves the resource pressure and timing pressure of the field programmable logic gate array 12 . The variable matrix size module includes an instruction decoding module and a matrix size selection module. The instruction decoding module analyzes the instructions transmitted by the host computer according to the designed instruction set, and obtains the corresponding channel information and matrix size information. At the same time, the module pre-stores y optional matrix sizes and the corresponding clock cycles required for y different complete large matrix operations. By changing the channel information and matrix scale information obtained through analysis, the maximum number of operating clock cycles and the number of sub-seed selection cycles are changed to realize the dynamic transformation of the matrix scale. The first stage of the three-stage parallel pipeline is the construction of the submatrix. As shown in Figure 3, the digital signal output by the analog-to-digital converter is transmitted to the field programmable gate array through the LVDS differential bus. Select m true random seeds used to construct the Toeplitz matrix, write a C program to shift them to generate a total of n sub-seeds and write them into a text file, and store the text file in the on-chip block random access memory of the field editable gate array, and at the same time The on-chip block random access memory then stores multiple a-bit true random numbers, randomly representing 0 to 2 ^a -1, divides 2^a into m equal parts, and selects an a-bit true random number every time a matrix multiplication is completed, and passes The size of the random number randomly selects a random seed to construct a new Toeplitz hash matrix, and at the same time writes a new true random number to cover the original true random number. The second stage of the three-stage parallel pipeline is the multiplication of sub-matrixes. Use an AND gate instead of a multiplier to realize multiplication between single bits; use an exclusive OR gate to realize addition. Through the combination of AND gate and XOR gate, a matrix "multiply-add" operation can be completed. The third stage of the three-stage parallel pipeline is a sub-matrix operation result addition module. In the first cycle of the sub-matrix operation, the register is used to cache the result of the sub-matrix operation, and then the current value of the register is XORed with the result of the sub-matrix operation in each cycle. The output result is obtained after multiple cycles of XOR operation. In this step, in order to alleviate the resource consumption of the field programmable gate array and optimize the timing, the following methods are used: 1) the digital signal output from the analog-to-digital converter to the field programmable gate array is buffered by using a dual-port asynchronous FIFO, The processing clock of the field programmable gate array is reduced to half of the sampling clock of the ADC, so that the original random sequence involved in the calculation of each clock is twice the original, and the timing pressure is relieved. 2) Insert pipeline registers for overly long combinatorial logic to reduce the number of stages of combinatorial logic; use register replication for high fan-out registers to reduce fan-out and improve drive capability; insert multi-stage pipeline registers in critical paths to reduce logic delays. 3) Use buffers to drive high fan-out signals such as global clocks and global resets to reduce fan-out, improve drive capability and optimize timing.

Step 5: Based on the three-stage parallel pipeline algorithm with variable matrix size designed in step 3, the hardware realizes real-time entropy evaluation and dynamic change of Toeplitz hash matrix size. There are three steps involved. The first step is real-time entropy evaluation. Repeat step 3 regularly, collect the original random sequence, detect the minimum entropy lower bound, reconstruct the distribution of the vacuum state Wigner function in the phase space, use the KLD relative entropy to measure the gap between the experimental vacuum state and the ideal vacuum state, and grasp the quantum state purity in real time. Variation, combined with the minimum entropy fluctuation range to obtain the postprocessing matrix size. The second module is the serial port instruction sending module. The present invention designs a simple instruction set. The size of each command is equal to the bit width of the serial port, which is x-bit. Encode the high z bit of the x-bit command as a one-hot code, indicating which channel of the z channels the command is valid for; binary code the remaining bits of the x-bit command, indicating y different matrix sizes. After obtaining the real-time matrix scale, the upper computer 14 sends instructions to the field programmable gate array 12 through the UART serial port 13; the third step is the matrix dynamic change module, and a three-stage parallel pipeline algorithm with variable matrix scale is designed in step 4 , the field editable gate array parses the instruction sent by the host computer after receiving it, obtains channel information and matrix scale information, and changes the maximum operation clock cycle number and sub-seed selection cycle after completing the current large matrix operation Number, realize the dynamic change of post-processing matrix scale.

The present invention calculates the entropy value introduced by the classical noise, deducts the classical noise entropy from the total entropy value of the original data, and calculates the size of the pure quantum entropy contained in the original data, which is used as the parameter for extracting the true random number in the post-processing of the data. Therefore, when the random number is extracted subsequently, the extraction ratio of the random number is also the highest. Multiple random seeds and continuously updated true random numbers are stored in the Field Editable Gate Array, which realizes the random selection of true random seeds for constructing Toeplitz, and eliminates the non-random factors introduced by invariant seeds. Through the overall planning of field programmable gate array logic resources, a three-stage parallel pipeline post-processing algorithm with variable matrix scale is established. After real-time entropy evaluation, the real-time Toeplitz hash matrix scale is determined, and the quantum random number is realized. Real-time safe high-speed generation.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto, any person familiar with the technical field within the technical scope disclosed in the present invention, according to the technical solution of the present invention and its Any equivalent replacement or change of the inventive concept shall fall within the protection scope of the present invention.

Claims (9)

1. The method for generating the real-time entropy evaluation post-processing quantum random number is characterized by comprising the following steps of:

entropy evaluation is carried out on the original true random sequence; reconstructing the distribution of the vacuum state Wigner function in the phase space, using the difference between the KLD relative entropy measurement experiment vacuum state and the ideal vacuum state, grasping the change of the quantum state purity, giving a threshold value according to the minimum entropy fluctuation range in the laboratory safety environment, considering that the fluctuation exceeds the threshold value in the actual operation, and then carrying out the operation of minimum entropy reevaluation and post-processing matrix adjustment; obtaining a real-time Toeplitz post-processing matrix scale;

based on hardware conditions of a field programmable gate array and Toeplitz matrix scale, designing a three-level parallel pipeline algorithm with the size of the matrix dynamically changing along with quantum entropy estimated in real time, updating and constructing random seeds of a Toeplitz hash matrix in real time, and carrying out information-theory safe real-time generalized hash extraction on quantum random numbers;

based on the parallel computing advantage of the field programmable gate array, real-time extraction under more than three different quantum sideband modes is realized, and the safe quantum random numbers are generated in real time and at high speed by multiple channels.

2. The method according to claim 1, wherein the generating method of the original true random sequence is:

a balanced homodyne detection system of a light field vacuum state is built, vacuum quantum fluctuation is extracted to serve as an entropy source for generating a true random number, background light and a vacuum field are made to interfere through a 50/50 optical beam splitter, two light signals with basically equal light intensity are output through the beam splitter, balanced homodyne detection is carried out to obtain a photocurrent signal, the photocurrent signal obtained through balanced homodyne detection is mixed with a radio frequency signal and filtered through a filter, and then analog-digital conversion is carried out through an analog-digital converter to obtain an original true random sequence.

3. The method of claim 1, wherein the entropy evaluation of the original true random sequence requires a prior determination of an optimal sampling range.

4. A method according to claim 3, characterized in that the method of determining the optimal sampling range is:

the gain of the alternating voltage is continuously regulated by using a voltage amplifier, the output of each amplified path is collected by using an oscilloscope, the frequency count is carried out on the collected data, and when the frequency exceeding the sampling range of the oscilloscope is approximately equal to the frequency of the highest frame in the middle, the proportion of 0 and 1 in the collected true random sequence is similar, so that the random number optimal sampling range is obtained.

5. The method of claim 1, wherein the matrix-scale variable three-stage parallel pipeline algorithm comprises:

the first stage is the construction of a submatrix, wherein a random submatrix is constructed by sequentially selecting submatrices from BRAM through addresses, namely an on-chip block memory of the FPGA;

the second stage is the operation of the submatrix, and the constructed submatrix and the original random sequence are subjected to multiplication operation to obtain a submatrix multiplication result;

the third stage is the exclusive or of the submatrix result, and the submatrix multiplication result obtained in each period is exclusive or.

6. The method of claim 6, wherein the construction of the submatrices comprises two modules of random seed storage and random seed selection.

7. The method of claim 1, wherein the real-time entropy-estimated dynamic matrix scale transformation includes the following three steps;

the first step is to build a minimum entropy physical model for removing side information; calculating a transfer function of the balanced homodyne detection system, and calibrating a power spectrum output by equipment; under the optimal sampling range of the analog-to-digital converter, the signal variance, the conditional signal variance and the conditional excessive noise variance are obtained through calibrating the power spectrum; establishing a distance between a system parameter and a safety parameter, namely a confidence interval, and strictly calibrating a minimum entropy lower bound;

the second step is real-time entropy assessment feedback; detecting the change condition of the minimum entropy lower bound, preparing quantum attack, reconstructing the distribution of a vacuum state Wigner function in a phase space, measuring the difference between the experimental vacuum state and the ideal vacuum state by using KLD relative entropy, grasping the change of the quantum state purity in real time, and obtaining the real-time post-processing matrix scale by combining the fluctuation range of the minimum entropy lower bound; after obtaining the real-time matrix size, take four parallel post-processing channels as an example: the upper computer transmits eight-bit instructions to the editable gate array through the UART interface to represent sixty-four matrix scales of four parallel post-processing channels;

the third step is the post-processing matrix scale dynamic transformation inside the field programmable gate array; firstly, a selection circuit is constructed, after an instruction is received, the matrix scale of a corresponding channel is obtained through decoding and analysis, and after one large matrix multiplication is completed, the corresponding submatrix multiplication result or the total period number is selected according to the corresponding matrix scale, so that the dynamic matrix scale transformation after the real-time entropy evaluation is realized.

8. The method of claim 1, wherein the processing clock of the field programmable gate array is reduced to half of the sampling clock of the analog-to-digital converter when the optoelectronic signal is collected, the buffer is used to drive the global clock and the global reset high fan-out signal, and a multistage pipeline register is inserted in the critical path to reduce logic delay; register replication is used on high fan-out registers to reduce the fan-out of the registers.

9. The method of claim 2, wherein,

the background light is continuous wave laser output by adopting a semiconductor laser with a central wavelength of 1550nm LD-TC 40;

the radio frequency signals are signals with different frequency bands generated by adopting an HP8648A type signal generator with the frequency of 100 kHz-4000 MHz;

the balanced homodyne detection adopts a PDB480C-AC type balanced detector of 1.6GHz to detect _；

The filter is a ZFM-11+ type mixer with the frequency range of 1MHz-2GHz and a BLP-100+ type low-pass filter with the frequency of 100 MHz;

the field programmable gate array model is Xilinx Kintex-7 XC7K325T;

the model number of the analog-digital converter is ADS42LB69.

Images