Distributed radar low-flux signal level fusion detection method and device

Abstract

The invention provides a distributed radar low-flux signal level fusion detection method and device. The method comprises the steps that firstly, parallel real signal construction processing is carried out through radars, detection statistics are obtained, original complex signals are converted into real numbers from complex numbers, the data size is reduced to a certain degree, and the pressure of processing all the original complex signals through a fusion center is avoided; and secondly, signal screening processing is further carried out on the detection statistical magnitude through double-threshold screening and a least square quantization (LSQ) method, and finally, an actual quantized value obtained through screening processing, a fusion statistical magnitude and a judgment threshold are sent to a fusion center for target detection. As the original complex signals are screened for multiple times before being sent to the fusion center, and the screening processing is executed in parallel on each radar, the data processing efficiency is improved, the data processing amount of the fusion center is reduced, and the communication bandwidth required by data transmission is reduced while the target detection precision is ensured.

Description

Technical Field

[0001] The present invention relates to the field of radar data processing technology, and in particular to a distributed radar low-throughput signal level fusion detection method and device. Background Art

[0002] A distributed radar system generally includes multiple radar stations and a fusion center. For a distributed radar system, in order to obtain better detection performance, multiple radar stations often need to send all the original signals in the observation area to the fusion center through wireless communication links for signal fusion. Under this architecture, a large communication bandwidth is often consumed.

[0003] Existing distributed radar signal-level fusion detection methods include, for example: Method 1 "A distributed radar low-communication quasi-signal fusion target detection and processing method", which combines distance unit position coding and echo signal data quantization coding that exceeds the local radar threshold, further performs data compression coding, and encrypts the radar echo signal data containing the target signal and transmits it to the fusion center for judgment. Method 2 "A distributed radar target detection method based on Doppler channel maximum quantization", which sends the Doppler channel maximum value to the fusion center for distributed signal fusion detection, and adopts the least squares quantization LSQ algorithm to quantize the Doppler maximum value, thereby reducing the communication bandwidth and reducing the computational cost. "

[0004] However, since the amount of data transmitted by the above method is still large, there is still the problem of high required communication bandwidth. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a distributed radar low-throughput signal level fusion detection method and device.

[0006] The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0007] In a first aspect, the present invention provides a distributed radar low-throughput signal level fusion detection method, which is applied to a distributed radar system. The distributed radar system includes: a fusion center and multiple radars, including:

[0008] All radars receive the echo signal respectively and obtain the original complex signal;

[0009] All radars perform real signal construction processing on the original complex signal to obtain detection statistics, and construct fusion statistics based on the detection statistics;

[0010] All radars perform double threshold screening on the detection statistics to obtain screening statistics, and use the screening statistics to calculate the decision threshold;

[0011] All radars quantize the screening statistics based on the least squares quantization (LSQ) method to obtain the actual quantized value, and send the actual quantized value, fusion statistics, and decision threshold to the fusion center;

[0012] The fusion center obtains the fusion target detection result based on the actual quantization value, fusion statistics and decision threshold.

[0013] Optionally, all radars perform real signal construction processing on the original complex signal to obtain detection statistics, and construct fusion statistics based on the detection statistics, including:

[0014] All radars construct real signals based on energy processing algorithms to process the original complex signals and obtain the detection statistic r k ;

[0015]

[0016] Among them, L k represents the number of pulse accumulation of the kth radar during the coherent processing time, x k (l) represents the original complex signal of the kth radar under the lth pulse, where l = 1, 2, ..., L d , L d Indicates the total number of pulses;

[0017] Based on the detection statistic r k , the fusion statistic r is constructed using the following formula;

[0018]

[0019] Among them, N d represents the total number of radars, Represents the local radar signal-to-noise ratio ρ k R k Any value r in k Taylor expansion of the process;

[0020]

[0021] represents f k (r k ,ρ k ), f k (r k ,ρ k ) represents the local radar signal-to-noise ratio ρ k R k The result of log-likelihood ratio transformation is represents f k (r k ,ρ k )exist The function value expanded at , O(r k ) represents the test statistic r k Higher-order infinitesimals.

[0022] Optionally, all radars perform double threshold screening on the detection statistics to obtain screening statistics, and use the screening statistics to calculate the decision threshold, including:

[0023] Each radar reduces the communication bandwidth by a factor α k Calculate the corresponding screening threshold g k ;

[0024] The test statistic r k Greater than the screening threshold g k The value of , which constitutes the screening statistic t k ;

[0025] Based on the screening statistic t k Calculate the false alarm probability p f ;

[0026] Using the screening statistic t k and false alarm probability p f , and obtain the corresponding decision threshold g through binary search.

[0027] Optionally, the screening threshold g k It is expressed as:

[0028] g k =-ln(α k ).

[0029] Optionally, based on the screening statistic t k Calculate the false alarm probability p f ,include:

[0030] t k Encode and obtain the encoding set S;

[0031]

[0032] Among them, N d represents the total number of radars, k represents the kth radar, D k Represents the first parameter, g k Indicates the screening threshold,

[0033] Based on the coding set S, the false alarm probability p is calculated using the following formula: f ;

[0034]

[0035] Pi (S|H 0 ) represents the second parameter, p i (S|H 0 ) represents the third parameter,

[0036]

[0037] Where i represents the i-th possibility of the coding set S, and N represents t k Failed to pass g k The set of Y represents t k Through g k The collection of Z i represents the false alarm probability p f The threshold under the i-th possibility, Represents the screening statistic t k In the case of no target H 0 The probability density function of .

[0038] Optionally, the decision threshold g is expressed as:

[0039]

[0040] p i (S|H 1 ) represents the fourth parameter, Represents the screening statistic t k In the case of target H 1 The probability density function of .

[0041] Optionally, all radars quantize the screening statistics based on the least squares quantization LSQ method to obtain actual quantized values, and send the actual quantized values, fusion statistics and decision thresholds to the fusion center, including:

[0042] S201, all radars obtain the initialization quantization threshold vector T k , initialize the total square error E k And the initialization error square residual E;

[0043] T k =[T k,0 ,T k,1 ,…,T k,j ,…,T k,J ] T ;

[0044] Among them, T k,j represents the quantization threshold of radar k in the jth quantization interval, j = 0, 1, ..., J, J represents the total number of quantization intervals, (·) T Represents the transpose operation;

[0045] S202, initialize the total square error E k As the first total square error

[0046] S203, based on the initialization quantization threshold vector T k and the screening statistic t k , the intermediate quantization value v of the kth radar in the jth quantization interval is calculated using the following formula: k,j ;

[0047]

[0048] S204, based on the intermediate quantization value v k,j The optimal quantization threshold T of the kth radar in the jth quantization interval is calculated using the following formula: k,j ′;

[0049]

[0050] v k,j+1 represents the intermediate quantization value v of the kth radar in the j+1th quantization interval k,j+1 ;

[0051] S205, using the optimal quantization threshold T k,j ′ initializes the quantization threshold vector T k Update to get the current quantization threshold vector T k ′;

[0052] S206, using the optimal quantization threshold T k,j ', the second total square error E is calculated using the following formula k '

[0053]

[0054] Among them, T k,j-1 ′ represents the optimal quantization threshold of the kth radar in the j-1th quantization interval, Represents the screening statistic t k In the case of target H 1 The probability density function under ;

[0055] S207, determine the second total square error E k ′ and the first total square error Is the difference less than or equal to the initialization error square residual E?

[0056] S208, when the condition of S207 is met, based on t k In the current quantization threshold vector T k′, and obtain the interval search result. Then, use the interval search result to convert t k Quantized to the actual quantized value v k ;

[0057] S209: Send the actual quantization value, fusion statistics and decision threshold to the fusion center.

[0058] Optionally, the fusion center obtains a fusion target detection result based on the actual quantization value, the fusion statistic, and the decision threshold, including:

[0059] Based on the actual quantized value v k And the fusion statistic r obtains the global optimal statistic

[0060] Determining the global optimal statistics Is it greater than or equal to the decision threshold g?

[0061] If so, it is determined that there is a target in the spatial resolution unit;

[0062] If not, it is determined that there is no target in the spatial resolution unit.

[0063] Alternatively, the global optimal statistic is expressed as:

[0064]

[0065] in, represents the global optimal statistic, N d represents the total number of radars, k represents the kth radar, P(v k |H 0 ) represents the actual quantized value v k In the absence of the assumption H 0 The probability mass function under k |H 1 ) represents the actual quantized value v k In the presence of H 1 The probability mass function under , r represents the fusion statistic.

[0066] In a second aspect, the present invention provides a distributed radar low-flux signal level fusion detection device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, when the distributed radar low-flux signal level fusion detection device is running, the processor and the storage medium communicate through the bus, and the processor executes the machine-readable instructions to perform the steps of the distributed radar low-flux signal level fusion detection method as described in the first aspect above.

[0067] The present invention provides a distributed radar low-throughput signal level fusion detection method and device. The distributed radar low-throughput signal level fusion detection method is applied to a distributed radar system, which includes: a fusion center and multiple radars, and the method includes: all radars receive echo signals respectively to obtain original complex signals; all radars perform real signal construction processing on the original complex signals to obtain detection statistics, and construct fusion statistics based on the detection statistics; all radars perform double threshold screening on the detection statistics to obtain screening statistics, and use the screening statistics to calculate the decision threshold; all radars perform quantization processing on the screening statistics based on the least squares quantization LSQ method to obtain actual quantization values, and send the actual quantization values, fusion statistics and decision thresholds to the fusion center; the fusion center obtains the fusion target detection result based on the actual quantization values, fusion statistics and decision thresholds. In the present invention, firstly, the parallel real signal construction processing performed by each radar is used to obtain the detection statistic, so that the original complex signal is converted from a complex number to a real number, which reduces the data volume to a certain extent and avoids the pressure of processing all the original complex signals by the fusion center; secondly, the detection statistic is further subjected to signal screening processing through the double threshold screening and least squares quantization LSQ method, and finally the actual quantization value, fusion statistic and decision threshold obtained by the screening processing are sent to the fusion center for target detection. Since the original complex signal has been screened multiple times before being sent to the fusion center, and the screening processing is performed in parallel on each radar, the data processing efficiency is improved while the data processing volume of the fusion center is reduced, and the communication bandwidth required for data transmission is reduced while ensuring the target detection accuracy.

[0068] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 A schematic diagram of a flow chart of a distributed radar low-throughput signal level fusion detection method provided by an embodiment of the present invention;

[0070] Figure 2 Performance comparison of dual threshold fusion detection based on different pass rate conditions provided by the embodiments of the present invention;

[0071] Figure 3 Comparison of the detection performance using quantization processing provided by the embodiment of the present invention with that without quantization;

[0072] Figure 4 A schematic diagram of the structure of a distributed radar low-throughput signal level fusion detection device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0073] 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.

[0074] In order to reduce the communication bandwidth required for data transmission while ensuring target detection accuracy, an embodiment of the present invention provides a distributed radar low-throughput signal level fusion detection method. Figure 1 The present invention provides a schematic flow chart of a distributed radar low-throughput signal level fusion detection method. Figure 1 As shown, including:

[0075] S101. All radars receive echo signals respectively to obtain original complex signals.

[0076] S102. All radars perform real signal construction processing on the original complex signal to obtain detection statistics, and construct fusion statistics based on the detection statistics.

[0077] Optionally, S102 may specifically include:

[0078] All radars construct real signals based on energy processing algorithms to process the original complex signals and obtain the detection statistic r k ;

[0079]

[0080] Among them, L k represents the number of pulse accumulation of the kth radar during the coherent processing time, x k (l) represents the original complex signal of the kth radar under the lth pulse, where l = 1, 2, ..., L d , L d Indicates the total number of pulses;

[0081] Based on the detection statistic r k , the fusion statistic r is constructed using the following formula;

[0082]

[0083] Among them, N d represents the total number of radars, Represents the local radar signal-to-noise ratio ρ k R k Any value in Taylor expansion of the process;

[0084]

[0085] represents f k (r k ,ρ k ), f k (r k ,ρ k) represents the local radar signal-to-noise ratio ρ k R k The result of log-likelihood ratio transformation is represents f k (r k ,ρ k )exist The function value expanded at , O(r k ) represents the test statistic r k Higher-order infinitesimals.

[0086] S103, all radars perform double threshold screening on the detection statistics to obtain screening statistics, and use the screening statistics to calculate the decision threshold.

[0087] Optionally, S103 may specifically include:

[0088] Each radar reduces the communication bandwidth by a factor α k Calculate the corresponding screening threshold g k ;

[0089] The test statistic r k Greater than the screening threshold g k The value of , which constitutes the screening statistic t k ;

[0090] Based on the screening statistic t k Calculate the false alarm probability p f ;

[0091] Using the screening statistic t k and false alarm probability p f , and obtain the corresponding decision threshold g through binary search.

[0092] Optionally, the screening threshold g k It is expressed as:

[0093] g k =-ln(α k );

[0094] Optionally, based on the screening statistic t k Calculate the false alarm probability p f ,include:

[0095] t k Encode and obtain the encoding set S;

[0096]

[0097] Among them, N d represents the total number of radars, k represents the kth radar, D k Represents the first parameter, gk Indicates the screening threshold,

[0098] Based on the coding set S, the false alarm probability p is calculated using the following formula: f ;

[0099]

[0100] P i (S|H 0 ) represents the second parameter, p i (S|H 0 ) represents the third parameter,

[0101]

[0102] Where i represents the i-th possibility of the coding set S, and N represents t k Failed to pass g k The set of Y represents t k Through g k The collection of Z i represents the false alarm probability p f The threshold under the i-th possibility, Represents the screening statistic t k In the case of no target H 0 The probability density function of .

[0103] Optionally, the decision threshold g is expressed as:

[0104]

[0105]

[0106] p i (S|H 1 ) represents the fourth parameter, Represents the screening statistic t k In the case of target H 1 The probability density function of .

[0107] S104. All radars quantize the screening statistics based on the least squares quantization (LSQ) method to obtain actual quantized values, and send the actual quantized values, fusion statistics, and decision thresholds to the fusion center.

[0108] Optionally, S104 may specifically include:

[0109] S201, all radars obtain the initialization quantization threshold vector T k , initialize the total square error E k And the initialization error square residual E;

[0110] Tk =[T k,0 ,T k,1 ,…,T k,j ,…,T k,J ] T ;

[0111] Among them, T k,j represents the quantization threshold of radar k in the jth quantization interval, j = 0, 1, ..., J, J represents the total number of quantization intervals, (·) T Represents the transpose operation;

[0112] S202, initialize the total square error E k As the first total square error

[0113] S203, based on the initialization quantization threshold vector T k and the screening statistic t k , the intermediate quantization value v of the kth radar in the jth quantization interval is calculated using the following formula: k,j ;

[0114]

[0115] S204, based on the intermediate quantization value v k,j The optimal quantization threshold T of the kth radar in the jth quantization interval is calculated using the following formula: k,j ′;

[0116]

[0117] v k,j+1 represents the intermediate quantization value v of the kth radar in the j+1th quantization interval k,j+1 ;

[0118] S205, using the optimal quantization threshold T k,j ′ initializes the quantization threshold vector T k Update to get the current quantization threshold vector T k ′;

[0119] S206, using the optimal quantization threshold T k,j ', the second total square error E is calculated using the following formula k '

[0120]

[0121] Among them, T k,j-1 ′ represents the optimal quantization threshold of the kth radar in the j-1th quantization interval, Represents the screening statistic t k In the case of target H1 The probability density function under ;

[0122] S207, determine the second total square error E k ′ and the first total square error Is the difference less than or equal to the initialization error square residual E?

[0123] S208, when the condition of S207 is met, based on t k In the current quantization threshold vector T k ′, and obtain the interval search result. Then, use the interval search result to convert t k Quantized to the actual quantized value v k ;

[0124] S209: Send the actual quantization value, fusion statistics and decision threshold to the fusion center.

[0125] Further, in the embodiment of the present invention, when the condition of S207 is not satisfied, S201-S208 are repeated until the condition of S207 is satisfied, and the processing step of S208 is executed on the basis that the condition of S207 is satisfied.

[0126] In addition, in the embodiment of the present invention, the quantization process can be expressed as:

[0127] v k =v k,j (T k,j-1 ≤t k ≤T k,j );

[0128] S105. The fusion center obtains the fusion target detection result based on the actual quantization value, the fusion statistic and the decision threshold.

[0129] The embodiment of the present invention provides a distributed radar low-throughput signal level fusion detection method, which is applied to a distributed radar system. The distributed radar system includes: a fusion center and multiple radars. First, a detection statistic is obtained by constructing and processing parallel real signals performed by each radar, so that the original complex signal is converted from a complex number to a real number, which reduces the data volume to a certain extent and avoids the pressure of processing all the original complex signals by the fusion center. Secondly, the detection statistic is further subjected to signal screening processing by a double threshold screening and least squares quantization LSQ method, and finally the actual quantization value, fusion statistic and decision threshold obtained by the screening processing are sent to the fusion center for target detection. Since the original complex signal has been screened multiple times before being sent to the fusion center, and the screening processing is performed in parallel on each radar, the data processing efficiency is improved while the data processing volume of the fusion center is reduced, and the communication bandwidth required for data transmission is reduced while the target detection accuracy is guaranteed.

[0130] Optionally, S105 may specifically include:

[0131] Based on the actual quantized value v k And the fusion statistic r obtains the global optimal statistic

[0132] Determining the global optimal statistics Is it greater than or equal to the decision threshold g?

[0133] If so, it is determined that there is a target in the spatial resolution unit;

[0134] If not, it is determined that there is no target in the spatial resolution unit.

[0135] Alternatively, the global optimal statistic is expressed as:

[0136]

[0137] in, represents the global optimal statistic, N d represents the total number of radars, k represents the kth radar, P(v k |H 0 ) represents the actual quantized value v k In the absence of the assumption H 0 The probability mass function under k |H 1 ) represents the actual quantized value v k In the presence of H 1 The probability mass function under , r represents the fusion statistic.

[0138] In order to verify the effectiveness of a distributed radar low-throughput signal-level fusion detection method provided by an embodiment of the present invention, a simulation experiment was also conducted in the embodiment of the present invention.

[0139] Figure 2 The performance comparison of dual threshold fusion detection under different pass rate conditions provided by the embodiment of the present invention is shown in FIG. Figure 2 As shown in the figure, four pass rate comparison channels (1%, 3%, 0.1%, 0.01%) are set. The target echo signals of different radar channels are statistically independent of each other, and the signal-to-noise ratio is the same. Each radar uses a classic energy detector as the test statistic, and the false alarm probability controlled by the fusion center is 10 -6 .from Figure 2 It can be seen from the results that if the communication bandwidth is reduced to 1 / 1000, the signal-to-noise ratio loss does not exceed 0.5 dB, which proves that the dual-threshold fusion detection of the embodiment of the present invention provides a mechanism for using less detection performance loss in exchange for a larger communication bandwidth reduction.

[0140] Figure 3The comparison of the detection performance using quantization processing provided by the embodiment of the present invention and the detection performance without quantization is shown in FIG. Figure 3 As shown in the figure, a total of 4 radar sites are set up. The signal processing algorithm and parameters of each radar site are the same. The cumulative number of radar pulses is 16. The local radar uses the unit average constant false alarm detection algorithm to obtain local statistics in each Doppler channel. The number of reference units is 32, the number of quantization intervals is 6, the initial pass rate is 0.4%, and the overall false alarm rate of the system is 10 -6 , LSQ quantification is performed on the statistic. The LSQ algorithm requires a preset signal-to-noise ratio. When the preset signal-to-noise ratio is different, the detection probability curve obtained is as follows Figure 3 As shown, the horizontal axis represents the signal-to-noise ratio of a single radar. Figure 3 It can be seen that when the preset signal-to-noise ratio is 4dB, the detection performance is optimal. Figure 3 As can be seen from the small figure, the signal-to-noise ratio loss is less than 0.3dB compared with unquantized fusion.

[0141] The distributed radar low-throughput signal-level fusion detection method of the embodiment of the present invention greatly reduces the communication bandwidth required for signal-level fusion detection through a series of data compression means such as parallel calculation, double threshold screening, and secondary quantization (quantization processing), so that the information transmitted from the radar station to the fusion center is concise and effective. Then, an optimal fusion algorithm based on the NP criterion is designed in the fusion center to make a global decision to determine the presence or absence of the target. At the same time, in order to ensure that the signal-to-noise ratio loss is small under the premise of the required communication bandwidth, an LSQ quantization strategy based on different preset signal-to-noise ratios is set, and the preset signal-to-noise ratio value corresponding to the best detection performance is given. Finally, the ability to approximate signal-level fusion detection with a point-trace communication bandwidth under low communication volume conditions is achieved. Numerical results show that when the preset signal-to-noise ratio of the present invention is 4dB, the detection performance is optimal; at the same time, based on an initial pass rate of 0.4% and 6 quantization intervals, the present invention can reduce the communication bandwidth to 1 / 1000, and the signal-to-noise ratio loss caused does not exceed 0.5dB. Therefore, the present invention provides a mechanism for using less detection performance loss in exchange for a larger communication bandwidth reduction. This research will greatly support the practical application of distributed radar detection and reduce computing and communication costs.

[0142] The method provided in the embodiment of the present invention can be applied to an electronic device. Specifically, the electronic device can be: a desktop computer, a portable computer, an intelligent mobile terminal, a server, etc., which is not limited in the embodiment of the present invention.

[0143] Based on the same inventive concept, an embodiment of the present invention also provides a distributed radar low-throughput signal level fusion detection device. Figure 4The structural diagram of the distributed radar low-flux signal level fusion detection device provided by the embodiment of the present invention includes: a processor 710, a storage medium 720 and a bus 730, the storage medium 720 stores machine-readable instructions executable by the processor 710, when the distributed radar low-flux signal level fusion detection device is running, the processor 710 communicates with the storage medium 720 through the bus 730, and the processor 710 executes the machine-readable instructions to execute the steps of the above method embodiment. The specific implementation method and technical effect are similar and will not be repeated here.

[0144] The storage medium may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk storage. Optionally, the storage medium may also be at least one storage device located away from the aforementioned processor.

[0145] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0146] It should be noted that the terms "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention.

[0147] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification.

[0148] Although the present invention is described herein in conjunction with various embodiments, in the process of implementing the claimed invention, those skilled in the art can understand and implement other changes of the above disclosed embodiments by viewing the drawings and the disclosed content. In the description of the present invention, the term "comprising" does not exclude other components or steps, "one" or "an" does not exclude multiple situations, and the meaning of "multiple" is two or more, unless otherwise clearly and specifically limited. In addition, certain measures are recorded in different embodiments, but this does not mean that these measures cannot be combined to produce good results.

[0149] 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 scope of protection of the present invention.

Claims

1. A distributed radar low-throughput signal level fusion detection method, applied to a distributed radar system, the distributed radar system comprising: Fusion center and multiple radars, characterized by, including: All radars receive the echo signal respectively and obtain the original complex signal; All the radars perform real signal construction processing on the original complex signal to obtain detection statistics, and construct fusion statistics based on the detection statistics; All the radars perform double threshold screening on the detection statistics to obtain screening statistics, and use the screening statistics to calculate a decision threshold; All the radars quantize the screening statistics based on the least squares quantization (LSQ) method to obtain actual quantized values, and send the actual quantized values, the fusion statistics and the decision threshold to the fusion center; The fusion center obtains a fusion target detection result based on the actual quantization value, the fusion statistic and the decision threshold.

2. The distributed radar low-throughput signal level fusion detection method according to claim 1, characterized in that: All radars perform real signal construction processing on the original complex signal to obtain detection statistics, and construct fusion statistics based on the detection statistics, including: All radars perform real signal construction processing on the original complex signal based on the energy processing algorithm to obtain the detection statistic r k ; Among them, L k represents the number of pulse accumulation of the kth radar during the coherent processing time, x k (l) represents the original complex signal of the kth radar under the lth pulse, where l = 1, 2, ..., L d , L d Indicates the total number of pulses; Based on the detection statistic r k , the fusion statistic r is constructed using the following formula; Among them, N d represents the total number of radars, Represents the local radar signal-to-noise ratio ρ k R k Any value r in k Taylor expansion of the process; represents f k (r k ,ρ k ), f k (r k ,ρ k ) represents the local radar signal-to-noise ratio ρ k R k The result of log-likelihood ratio transformation is represents f k (r k ,ρ k )exist The function value expanded at, O(r k ) represents the test statistic r k Higher-order infinitesimals.

3. The distributed radar low-throughput signal level fusion detection method according to claim 1, characterized in that: All radars perform double threshold screening on the detection statistics to obtain screening statistics, and use the screening statistics to calculate a decision threshold, including: Each radar is based on a communication bandwidth reduction factor α k Calculate the corresponding screening threshold g k ; The detection statistic r k is greater than the screening threshold g k The value of t constitutes the screening statistic k ; Based on the screening statistic t k Calculate the false alarm probability p f ; Using the screening statistic t k and the false alarm probability p f , and obtain the corresponding decision threshold g through binary search.

4. The distributed radar low-throughput signal level fusion detection method according to claim 3 is characterized in that: The screening threshold g k It is expressed as: g k =-ln(α k )。 5. The distributed radar low-throughput signal level fusion detection method according to claim 3, characterized in that: Based on the screening statistic t k Calculate the false alarm probability p f ,include: t k Encode and obtain the encoding set S; Among them, N d represents the total number of radars, k represents the kth radar, D k Represents the first parameter, g k Indicates the screening threshold, Based on the coding set S, the false alarm probability p is calculated using the following formula: f ; P i (S|H0) represents the second parameter, p i (S|H0) represents the third parameter, Where i represents the i-th possibility of the coding set S, and N represents t k Failed to pass g k The set of Y represents t k Through g k The collection of Z i represents the false alarm probability p f The threshold under the i-th possibility, Represents the screening statistic t k Probability density function in the no-target case H0.

6. The distributed radar low-throughput signal level fusion detection method according to claim 5, characterized in that: The decision threshold g is expressed as: p i (S|H1) represents the fourth parameter, Represents the screening statistic t k Probability density function in the case of target H1.

7. The distributed radar low-throughput signal level fusion detection method according to claim 6, characterized in that: All radars quantize the screening statistics based on the least squares quantization LSQ method to obtain actual quantized values, and send the actual quantized values, the fusion statistics and the decision threshold to the fusion center, including: S201, all radars obtain the initialization quantization threshold vector T k , initialize the total square error E k And the initialization error square residual E; T k =[T k,0 ,T k,1 ,…,T k,j ,…,T k,J ] T ; Among them, T k,j represents the quantization threshold of radar k in the jth quantization interval, j = 0, 1, ..., J, J represents the total number of quantization intervals, (·) T Represents the transpose operation; S202, the initialized total square error E k As the first total square error S203, based on the initialization quantization threshold vector T k And the screening statistic t k , the intermediate quantization value v of the kth radar in the jth quantization interval is calculated using the following formula: k,j ; S204, based on the intermediate quantization value v k,j The optimal quantization threshold T of the kth radar in the jth quantization interval is calculated using the following formula: k,j ′; v k,j+1 represents the intermediate quantization value v of the kth radar in the j+1th quantization interval k,j+1 ; S205, using the optimal quantization threshold T k,j ' Initialize the quantization threshold vector T k Update to get the current quantization threshold vector T k ′; S206, using the optimal quantization threshold T k,j ', the second total square error E is calculated using the following formula k ' Among them, T k,j-1 ′ represents the optimal quantization threshold of the kth radar in the j-1th quantization interval, ft k|H1 (tk) represents the probability density function of the screening statistic tk in the target case H1; S207: Determine the second total square error E k ′ and the first total square error Is the difference less than or equal to the initialization error square residual E? S208, when the condition of S207 is met, based on t k In the current quantization threshold vector T k ′ to search for the corresponding interval, obtain the interval search result, and use the interval search result to convert t k Quantized to the actual quantized value v k ; S209: Send the actual quantized value, the fusion statistic, and the decision threshold to the fusion center.

8. The distributed radar low-throughput signal level fusion detection method according to claim 1, characterized in that: The fusion center obtains a fusion target detection result based on the actual quantization value, the fusion statistic and the decision threshold, including: Based on the actual quantized value v k And the fusion statistic r obtains the global optimal statistic Determine the global optimal statistic Is it greater than or equal to the decision threshold g? If so, it is determined that there is a target in the spatial resolution unit; If not, it is determined that there is no target in the spatial resolution unit.

9. The distributed radar low-throughput signal level fusion detection method according to claim 1, characterized in that: The global optimal statistic is expressed as: in, represents the global optimal statistic, N d represents the total number of radars, k represents the kth radar, P(v k |H0) represents the actual quantized value v k The probability mass function, P(v k |H1) represents the actual quantized value v k The probability mass function under the assumption H1 exists, and r represents the fusion statistic.

10. A distributed radar low-throughput signal level fusion detection device, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the distributed radar low-flux signal level fusion detection device is running, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of the distributed radar low-flux signal level fusion detection method as described in any one of claims 1 to 9.

Images