Quantized signal fusion detection method for distributed radar under communication constraints

Abstract

The present invention proposes a distributed radar quantization signal fusion detection method under communication constraints, including constructing an unquantized local test statistic based on the echo signal; quickly solving the quantization threshold and quantization value of each radar station based on the minimum mean square error quantization criterion; optimizing the design of the quantization threshold and quantization value according to the number of quantization bits, thereby constructing a quantized local test statistic; constructing the quantization statistic under communication constraints according to the communication factor setting; fusing the quantization statistics under communication constraints to determine whether the target exists. The present invention mainly solves the problem that the existing fusion detection algorithm is difficult to apply in engineering due to the long time of solving the quantization threshold and ignoring factors such as communication delay, packet loss rate, bit error rate, etc. The quantization threshold solution can be completed quickly, and the system can maintain a good detection probability under the premise of greatly reducing the system communication burden, which is conducive to engineering implementation.

Description

Technical Field

[0001] The present invention belongs to the field of radar technology, and in particular relates to a distributed radar quantization signal fusion detection method under communication constraints. Background Art

[0002] Single radar station target detection can effectively detect targets when the target signal-to-noise ratio is strong. However, with the rapid development of technology, modern radar targets tend to be stealthy and miniaturized, and thus have a smaller signal-to-noise ratio, which brings great challenges to traditional single radar station target detection. Compared with single radar detection, distributed radar detection can obtain target echoes at different angles through multiple local radar stations and accumulate target energy, thereby achieving higher detection performance. However, directly transmitting the statistics of local radar stations to the fusion center will cause a huge communication burden on the system. Therefore, it is necessary to study effective local statistics data compression methods to reduce the amount of communication data. In addition, in practice, there is channel noise in the communication channel, so the local signal transmission process will be affected by communication factors such as communication delay, packet loss rate, bit error rate, etc., resulting in transmission information loss, thereby reducing the detection probability. Therefore, it is necessary to study the impact of the above communication factors on the detection performance of distributed radar detection.

[0003] In the literature [H. Zang, H. Wang, Y. Chen, C. Gao, H. Liu, and Y. Cao. Quantizer design for distributed detection based on bhattacharyya distance. In 2016 CIE International Conference on Radar (RADAR), pages 1-4, Oct 2016.], the local radar statistics are quantized and compressed based on the maximization of the Bhattacharyya distance criterion, and the sequential quadratic programming algorithm is used to obtain the optimal design of the quantization threshold. Finally, the fusion center completes the fusion decision of the quantized statistics based on the Neyman-Pearson criterion.

[0004] In the literature [J.Lu, S.Zhou, Z.Shao, and H.Liu, “Distributed detection with local least squares quantization,” in 2019IEEE International Conference on Signal and Information and Data Processing, 2019.], local statistics are quantized and compressed based on the minimum mean square error quantization criterion, and the quantization threshold of each local radar station is solved according to the alternating optimization algorithm. Finally, the Neyman-Pearson criterion is used in the fusion center to make the fusion decision of the quantized statistics of the local radar stations.

[0005] The above detection methods all assume that the background noise power is known, ignoring the impact of factors such as communication delay, packet loss rate, bit error rate, etc. on detection performance. However, in practice, the noise power is often unknown, and the above communication factors often lead to a decrease in noise detection performance. In addition, the solution of the quantization threshold in the above method uses a numerical optimization method with high computational complexity, which takes a long time to calculate and is not conducive to engineering applications. Summary of the invention

[0006] In order to solve the above problems existing in the prior art, the present invention provides a distributed radar quantization signal fusion detection method under communication constraints. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0007] The present invention provides a distributed radar quantization signal fusion detection method under communication constraints, comprising:

[0008] S100, acquiring an echo signal of each radar station in the distributed radar system, and constructing an unquantized local test statistic according to the echo signal of each radar station;

[0009] S200, quickly solving the quantization threshold and quantization value of each radar station based on the minimum mean square error quantization criterion;

[0010] S300, constructing a quantized local test statistic according to the unquantized local test statistic, the quantization threshold and the quantization value;

[0011] S400, constructing quantitative statistics under radar station communication constraints according to the communication factor of the distributed radar system and the quantified local test statistics;

[0012] S500, fusing the quantized statistics under the limited communication of the radar station to obtain a global test statistic, and using the global test statistic to determine whether the target exists.

[0013] Beneficial effects:

[0014] First, under the condition that the background noise power is unknown, the present invention provides a fast quantization threshold solution method based on the minimum mean square error quantization criterion, which can greatly reduce the calculation complexity of solving the quantization threshold. Compared with the existing implementation scheme, the quantization threshold solution can be completed more quickly, which is convenient for engineering application;

[0015] Second, under the premise that the background noise power is unknown, the present invention incorporates factors such as communication delay, packet loss rate, bit error rate, etc. into the design of the fusion detection algorithm, and proposes a quantitative signal fusion detector based on the Neyman-Pearson criterion, which can greatly reduce the system communication burden while ensuring good detection performance. It can also be used to evaluate the impact of communication factors such as communication delay, packet loss rate, bit error rate, etc. on the detection performance of distributed radar quantitative signal fusion, which is beneficial to practical engineering applications.

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flow chart of a distributed radar quantized signal fusion detection method under communication constraints of the present invention;

[0018] Figure 2 It is a sub-flow chart of the present invention for quickly solving the quantization threshold and quantization value based on the minimum mean square error quantization criterion;

[0019] Figure 3 It is a curve diagram showing the average calculation time and quantization mean square error of the quantization threshold optimization solution algorithm proposed in the present invention, the existing alternating optimization algorithm, and the sequential quadratic programming algorithm as a function of the number of quantization bits;

[0020] Figure 4 It is a curve diagram showing the detection probability changing with the signal-to-noise ratio under different quantization bit number settings of the present invention;

[0021] Figure 5 It is a curve diagram showing the variation of detection probability with communication delay under different signal-to-noise ratio settings of the present invention;

[0022] Figure 6 is a curve diagram showing the variation of detection probability with packet loss rate under different signal-to-noise ratio settings of the present invention;

[0023] Figure 7 It is a curve diagram of detection probability changing with bit error rate under different signal-to-noise ratio settings of the present invention. DETAILED DESCRIPTION

[0024] The present invention is further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.

[0025] The purpose of the present invention is to address the shortcomings of the existing technology and propose a distributed radar quantization signal fusion detection method under communication constraints. The method assumes that the noise power is unknown and incorporates factors such as communication delay, packet loss rate, bit error rate, etc. into the design of the fusion detection algorithm. The quantization threshold solution can be completed more quickly, which can greatly reduce the system communication burden while ensuring good detection performance. It can also be used to evaluate the impact of communication factors such as communication delay, packet loss rate, bit error rate, etc. on the detection performance of distributed radar quantization signal fusion, which is convenient for engineering applications.

[0026] The technical solution of the present invention is summarized as follows: under the condition of unknown background noise, the local radar constructs the original local test statistic according to the echo signal; given the number of quantization bits, the quantization threshold and the quantization value are solved according to the fast minimum mean square error quantization threshold optimization solution algorithm proposed in the present invention; each local radar station constructs a quantization statistic according to the quantization threshold and the quantization value; the quantization statistic under communication constraints is constructed according to the communication delay, packet loss rate and bit error rate set by the system; the fusion center fuses the quantization statistics under communication constraints according to the Neyman-Pearson detector to determine whether the target exists.

[0027] refer to Figure 1 The present invention provides a distributed radar quantization signal fusion detection method under limited communication, including:

[0028] S100, acquiring an echo signal of each radar station in the distributed radar system, and constructing an unquantized local test statistic according to the echo signal of each radar station;

[0029] The distributed radar system of the present invention includes a signal fusion center and N local radar stations, and the clocks of the radar stations are synchronized and the observations are independent of each other.

[0030] As an optional implementation manner of the present invention, S100 includes:

[0031] S110, obtaining the echo signal r of the i-th radar station in the distributed radar system i,0 , expressed as:

[0032]

[0033] Among them, z i,0 Represents a noise echo, which has a mean of 0 and a variance of Complex Gaussian distribution of i Represents the target echo, which has a mean of 0 and a variance of The complex Gaussian distribution of Indicates that there is no assumption about the target, Indicates that there are assumptions about the target;

[0034] S120, constructing an unquantized local test statistic y according to the echo signal of the i-th radar station i , expressed as:

[0035]

[0036] Among them, r i,k It represents the echo signal of the kth reference unit corresponding to the i-th radar station, which has a mean of 0 and a variance of The complex Gaussian distribution, K i is the number of reference units of the i-th radar station.

[0037] S200, quickly solving the quantization threshold and quantization value of each radar station based on the minimum mean square error quantization criterion;

[0038] As an optional embodiment of the present invention, refer to Figure 2 , S200 includes:

[0039] S210, define the normalized quantization threshold of the i-th radar station The first two elements of Initialize m=1 and set the precision factor Δ p ;

[0040] S220, iteratively solving the second to Mth parameters of the i-th radar station in sequence according to the binary search method i Normalized quantization thresholds, where M i represents the number of quantization intervals of the i-th radar station;

[0041] S220 of the present invention includes:

[0042] S221, define the lower bound of the m+1th normalized quantization threshold of the ith radar station as The upper bound is

[0043] S222, define the m+1th normalized quantization threshold of the ith radar station as And according to the m+1th normalized quantization threshold Calculate the error function It is expressed as:

[0044]

[0045] in,

[0046] S223, if Then execute S224, otherwise execute S225;

[0047] S224, if Then And execute S222; if Then And execute S222 to obtain the m+1th normalized quantization threshold of the i-th radar station:

[0048] S225, if m <M i -1, then set m=m+1 and execute S221;

[0049] S226, if m≥M i -1, then the optimized normalized quantization threshold of the i-th radar station is

[0050]

[0051] S230, according to the normalized quantization threshold Solve the optimized quantization threshold τ of the i-th radar station i ;

[0052] S230 of the present invention includes:

[0053] S231, initialize m=1, according to the normalized quantization threshold The mth optimized quantization threshold of the ith radar station is calculated as:

[0054]

[0055] in, The preset signal-to-noise ratio for the ith radar station;

[0056] S232, if m <M i , then set m=m+1 and execute S231;

[0057] S233, if m ≥ M i , then the optimized quantization threshold value of the i-th radar station is Among them, the first element τ i,0 is the preset quantization threshold, defined as τ i,0 =0.

[0058] S240, according to the quantization threshold value τ of the i-th radar station i , solve the quantized value v of the i-th radar station i .

[0059] S240 of the present invention includes:

[0060] S241, initialize m=1, according to τ i,m-1 , τ i,m Calculate the mth optimized quantized value v of the i-th radar station i,m for:

[0061]

[0062] S242, if m <M i , then set m=m+1 and execute S241;

[0063] S243, if m≥M i , then the optimized quantized value of the i-th radar station is

[0064]

[0065] S300, constructing a quantized local test statistic according to the unquantized local test statistic, the quantization threshold and the quantization value;

[0066] S400, constructing quantitative statistics under radar station communication constraints according to the communication factor of the distributed radar system and the quantified local test statistics;

[0067] This step is based on the average communication delay μ of the i-th radar station set by the distributed radar system i,t , Communication delay variance Packet loss rate d i , bit error rate i , the fusion detection window length ΔT and the quantized local quantization statistic q i , construct the quantitative statistics of the i-th radar station under communication constraints for:

[0068]

[0069] in, represents a uniformly distributed random number between 0 and 1, c i =log2M i represents the number of quantization bits of the i-th radar station, and the quantization statistics of all local radar stations under communication constraints are recorded as

[0070]

[0071] S500, fusing the quantized statistics under the limited communication of the radar station to obtain a global test statistic, and using the global test statistic to determine whether the target exists.

[0072] As an optional implementation manner of the present invention, S500 includes:

[0073] S510, constructing a global test statistic based on the Neyman-Pearson criterion It is expressed as:

[0074]

[0075] Among them, λ i is the signal-to-noise ratio of the ith radar station, is the Dirac function, expressed as:

[0076]

[0077] S520, according to the given global false alarm probability P fa , solve the detection threshold γ and randomization probability factor κ offline;

[0078] S520 includes:

[0079] S521, using Matlab software to generate global test statistics under the assumption that the target does not exist A total of random numbers , among which Indicates 100 / P fa Round up;

[0080] S522, sort the generated random numbers from small to large, and record the sorted random numbers as vector t mc , then t mc for vector, t mc The elements with different values ​​are denoted as Initialize l=1;

[0081] S523, will t mc Medium or larger The number of elements is denoted by N l , and t mc Equal to The element is denoted by N e ,in Indicates taking The lth element of

[0082] S524, if Then let l=l+1 and execute S523, otherwise, the detection threshold is The randomization probability factor is

[0083] S530, based on the global test statistic The detection threshold γ and randomized probability factor κ determine whether the target exists.

[0084] S530 of this step includes:

[0085] S531, determine the global test statistic Is it greater than the detection threshold γ? If it is greater than the detection threshold γ, the target is judged to exist; If it is less than the detection threshold γ, then it is determined that the target does not exist; if If it is equal to the detection threshold γ, then execute S532;

[0086] S532, generate a random number u that obeys a uniform distribution from 0 to 1, and determine whether u is greater than a randomization probability factor κ. If u≤κ, determine that the target exists, otherwise determine that the target does not exist.

[0087] The present invention proposes a distributed radar quantization signal fusion detection method under communication constraints, including constructing an unquantized local test statistic based on the echo signal; quickly solving the quantization threshold and quantization value of each radar station based on the minimum mean square error quantization criterion; optimizing the design of the quantization threshold and quantization value according to the number of quantization bits, thereby constructing a quantized local test statistic; constructing the quantization statistic under communication constraints according to the communication factor setting; fusing the quantization statistics under communication constraints to determine whether the target exists. The present invention mainly solves the problem that the existing fusion detection algorithm is difficult to apply in engineering due to the long time of solving the quantization threshold and ignoring factors such as communication delay, packet loss rate, bit error rate, etc. The quantization threshold solution can be completed quickly, and the system can maintain a good detection probability under the premise of greatly reducing the system communication burden, which is conducive to engineering implementation.

[0088] The numbers of the above steps are only for more clearly illustrating the implementation scheme of the present invention, and the sequence of the numbers is not limited.

[0089] The technical effects of the present invention are further illustrated in conjunction with the following simulation experiments.

[0090] 1. Simulation conditions:

[0091] Assume that the number of radar stations N = 3, the working parameters and target signal-to-noise ratio of each radar station are the same, the preset radar signal-to-noise ratio is 10 dB, the fusion center accumulation window length is 0.5 s, and the system false alarm probability is set to 10 -4 .

[0092] Hardware environment: CPU is Intel Xeon E5-2667 v4, main frequency is 3.20GHz, memory is 128GB, and 64-bit operating system.

[0093] Software environment: Microsoft Windows 10 Professional Edition, MATLAB R2022b simulation software.

[0094] 2. Simulation content and results

[0095] Simulation 1: Under the above conditions, the quantization threshold and quantization value solution method proposed in the present invention and the existing alternating optimization algorithm and sequential quadratic programming algorithm are simulated to obtain the average time curve and the quantization mean square error curve with the quantization bit number. The results are as follows: Figure 3, where the alternating optimization method comes from the literature [J.Lu, S.Zhou, Z.Shao, and H.Liu, “Distributed detection with local least squares quantization,” in 2019 IEEE International Conference on Signal and Information and Data Processing, 2019.], and the sequential quadratic programming method comes from the literature [H.Zang, H.Wang, Y.Chen, C.Gao, H.Liu, and Y.Cao.Quantizer design for distributed detection based on bhattacharyyadistance. In 2016 CIE International Conference on Radar (RADAR), pages 1-4, Oct 2016.].

[0096] from Figure 3 It can be seen that the algorithm proposed in the present invention and the existing alternating optimization algorithm and sequential quadratic programming algorithm can all be optimized to converge to the same quantization mean square error, but the calculation time required by the present invention is much less than that of the existing alternating optimization algorithm and sequential quadratic programming algorithm. Taking the 3-bit quantization bit number as a reference, the average calculation time required by the present invention is only about 0.1% and 0.04% of that of the alternating optimization algorithm and the sequential quadratic programming algorithm, indicating that the present invention can solve the quantization threshold and quantization value in a shorter time and use it in engineering applications.

[0097] Simulation 2: Under the above conditions, the curve of detection probability varying with signal-to-noise ratio is simulated when the communication delay is 0.2s, the packet loss rate is 1%, the bit error rate is 0.01%, and the quantization bit number is 1 bit, 2 bits, 3 bits, and infinite bits respectively. The results are as follows: Figure 4 shown.

[0098] from Figure 4 It can be seen that as the number of quantization bits increases, the detection probability also increases; taking 0.6 detection probability as a reference, when 3-bit quantization is used, the signal-to-noise ratio loss is less than 0.6 decibels, indicating that the present invention can greatly reduce the system communication burden while enabling the system to still have good detection performance.

[0099] Simulation 3: Under the above conditions, the curve of detection probability varying with communication delay is simulated when the quantization bit number is 2 bits, the packet loss rate is 0, the bit error rate is 0, and the target signal-to-noise ratio is 10 dB, 12 dB, 14 dB, and 16 dB respectively. The results are as follows: Figure 5 shown.

[0100] from Figure 5 It can be seen that as the communication delay increases, the target detection probability decreases accordingly; for a target with a 12dB signal-to-noise ratio, when the communication delay increases from 0s to 0.5s, the detection probability decreases significantly, by about 32%, indicating that the present invention can be used to evaluate the system detection probability under different communication delays, which is beneficial to engineering applications.

[0101] Simulation 4: Under the above conditions, the curve of detection probability varying with packet loss rate is simulated when the number of quantization bits is 2 bits, the communication delay is 0, the bit error rate is 0, and the target signal-to-noise ratio is 10 dB, 12 dB, 14 dB, and 16 dB respectively. The results are as follows: Figure 6 shown.

[0102] from Figure 6 It can be seen that as the packet loss rate increases, the target detection probability decreases accordingly; when the packet loss rate increases from 0.0001 to 0.1, the detection probability decreases slightly, about 5%; and when the packet loss rate increases from 0.1 to 0.6, for the target with a 12dB signal-to-noise ratio, the detection probability decreases significantly, about 34%, indicating that the present invention can be used to evaluate the system detection probability under different packet loss rates, which is beneficial to engineering applications.

[0103] Simulation 5: Under the above conditions, the curve of detection probability versus bit error rate is simulated when the quantization bit number is 2 bits, the communication delay is 0, the packet loss rate is 0, and the target signal-to-noise ratio is 10 dB, 12 dB, 14 dB, and 16 dB respectively. The results are as follows: Figure 7 shown.

[0104] from Figure 7 It can be seen that as the bit error rate increases, the target detection probability decreases accordingly; when the bit error rate increases from 0.00001 to 0.01, the system detection probability decreases slowly, only about 1%. When the system bit error rate increases from 0.01 to 0.5, the system detection probability decreases significantly. For a target with a 12dB signal-to-noise ratio, the system detection probability decreases by about 50%, indicating that the present invention can be used to evaluate the system detection probability under different bit error rates, which is beneficial to engineering applications.

[0105] In summary, the present invention can quickly solve the quantization threshold and quantization value under the condition of unknown background noise power, and can maintain a good detection probability of the system under the premise of greatly reducing the communication burden of the system. It can be used for the detection performance evaluation of distributed radar quantization signal fusion under communication constraints, which is beneficial to the engineering realization and application of distributed radar systems.

[0106] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A distributed radar quantization signal fusion detection method under communication constraints, characterized in that: include: S100, acquiring an echo signal of each radar station in the distributed radar system, and constructing an unquantized local test statistic according to the echo signal of each radar station; S200, quickly solving the quantization threshold and quantization value of each radar station based on the minimum mean square error quantization criterion; S300, constructing a quantized local test statistic according to the unquantized local test statistic, the quantization threshold and the quantization value; S400, constructing quantitative statistics under radar station communication constraints according to the communication factor of the distributed radar system and the quantified local test statistics; S500, fusing the quantized statistics under the limited communication of the radar station to obtain a global test statistic, and using the global test statistic to determine whether the target exists; S200 includes: S210, define the normalized quantization threshold of the i-th radar station The first two elements of Initialize m=1 and set the precision factor Δ p ; S220, iteratively solving the second to Mth parameters of the i-th radar station in sequence according to the binary search method i Normalized quantization thresholds, where M i represents the number of quantization intervals of the i-th radar station; S230, according to the normalized quantization threshold Solve the optimized quantization threshold τ of the i-th radar station i ; S240, according to the quantization threshold value τ of the i-th radar station i , solve the quantized value v of the i-th radar station i， S400 includes: According to the distributed radar system, the average communication delay μ of the i-th radar station is set i,t , Communication delay variance Packet loss rate d i , bit error rate i , the fusion detection window length ΔT and the quantized local quantization statistic q i .

2. The distributed radar quantization signal fusion detection method under communication limitation according to claim 1 is characterized in that: S100 includes: S110, obtaining the echo signal r of the i-th radar station in the distributed radar system i,0 , expressed as: Among them, z i,0 Represents a noise echo, which has a mean of 0 and a variance of Complex Gaussian distribution of i Represents the target echo, which has a mean of 0 and a variance of The complex Gaussian distribution of Indicates that there is no assumption about the target, Indicates that there are assumptions about the target; S120, constructing an unquantized local test statistic y according to the echo signal of the i-th radar station i , expressed as: Among them, r i,k It represents the echo signal of the kth reference unit corresponding to the i-th radar station, which has a mean of 0 and a variance of The complex Gaussian distribution, K i is the number of reference units of the i-th radar station.

3. The distributed radar quantization signal fusion detection method under communication limitation according to claim 1 is characterized in that: S220 includes: S221, define the lower bound of the m+1th normalized quantization threshold of the ith radar station as The upper bound is S222, define the m+1th normalized quantization threshold of the ith radar station as And according to the m+1th normalized quantization threshold Calculate the error function It is expressed as: in, S223, if Then execute S224, otherwise execute S225; S224, if Then And execute S222; if Then And execute S222 to obtain the m+1th normalized quantization threshold of the i-th radar station: S225, if m <M i -1, then set m=m+1 and execute S221; S226, if m ≥ M i -1, then the optimized normalized quantization threshold of the i-th radar station is 4. The distributed radar quantization signal fusion detection method under communication restriction according to claim 3 is characterized in that: S230 includes: S231, initialize m=1, according to the normalized quantization threshold The mth optimized quantization threshold of the ith radar station is calculated as: in, The preset signal-to-noise ratio for the i-th radar station; S232, if m <M i , then set m=m+1 and execute S231; S233, if m ≥ M i , then the optimized quantization threshold value of the i-th radar station is Among them, the first element τ i,0 is the preset quantization threshold, defined as τ i,0 =0.

5. The distributed radar quantization signal fusion detection method under communication limitation according to claim 4 is characterized in that: S240 includes: S241, initialize m=1, according to τ i,m-1 , τ i,m Calculate the mth optimized quantized value v of the i-th radar station im for: S242, if m <M i , then set m=m+1 and execute S241; S243, if m ≥ M i , then the optimized quantized value of the i-th radar station is 6. The distributed radar quantization signal fusion detection method under communication limitation according to claim 5 is characterized in that: Constructing quantitative statistics of the i-th radar station under communication constraints for: in, represents a uniformly distributed random number between 0 and 1, c i =log2M i represents the number of quantization bits of the i-th radar station, and the quantization statistics of all local radar stations under communication constraints are recorded as 7. The distributed radar quantization signal fusion detection method under communication limitation according to claim 6 is characterized in that: S500 includes: S510, constructing a global test statistic based on the Neyman-Pearson criterion It is expressed as: Among them, λ i is the signal-to-noise ratio of the ith radar station, is the Dirac function, expressed as: S520, according to the given global false alarm probability P fa , solve the detection threshold γ and randomization probability factor κ offline: S530, based on the global test statistic The detection threshold γ and randomized probability factor κ determine whether the target exists.

8. The distributed radar quantization signal fusion detection method under communication limitation according to claim 7 is characterized in that: S520 includes: S521, using Matlab software to generate global test statistics under the assumption that the target does not exist A total of random numbers , among which Indicates 100 / P fa Round up; S522, sort the generated random numbers from small to large, and record the sorted random numbers as vector t mc , then t mc for vector, t mc The elements with different values ​​are denoted as Initialize l=1; S523, will t mc Medium to large The number of elements is denoted by N l , and t mc Equal to The element is denoted by N e ,in Indicates taking The lth element of S524, if Then let l=l+1 and execute S523, otherwise, the detection threshold is The randomization probability factor is 9. The distributed radar quantization signal fusion detection method under communication limitation according to claim 8, characterized in that: S530 includes: S531, determine the global test statistic Is it greater than the detection threshold γ? If it is greater than the detection threshold γ, the target is judged to exist; If it is less than the detection threshold γ, then it is determined that the target does not exist; if If it is equal to the detection threshold γ, then execute S532; S532, generate a random number u that obeys a uniform distribution from 0 to 1, and determine whether u is greater than a randomization probability factor κ. If u≤κ, determine that the target exists, otherwise determine that the target does not exist.

Images