Method and system for obtaining impulse responses of multiple human targets in complex environments

Abstract

The application discloses a method for obtaining multi-human target impulse responses in a complex environment, which comprises the following steps: obtaining radar echo data of a target area; performing MTI processing; performing target rough detection based on a CA-CFAR algorithm; performing decision accumulation and obtaining a final accumulated decision curve; performing peak value judgment; and performing processing on the basis of a Clean-optimized impulse response reconstruction algorithm to obtain a final multi-human target impulse response echo in a complex environment. The application further discloses a system for implementing the method for obtaining multi-human target impulse responses in a complex environment. The application can not only realize multi-human target judgment and impulse response acquisition in a complex environment, but also has higher reliability, better accuracy and higher efficiency.

Description

Technical Field

[0001] This invention belongs to the field of digital signal processing, specifically relating to a method and system for acquiring the impulse responses of multiple human targets in a complex environment. Background Technology

[0002] As people's living standards improve, health monitoring radar is gradually entering people's daily lives, bringing endless convenience to people's lives.

[0003] Currently, when acquiring the impulse response of multiple human targets based on health monitoring radar, it is necessary to first detect the presence of targets within the monitoring range, then obtain the distance gate of the detected target through signal processing, and finally extract the impulse response of each human target.

[0004] Existing solutions for target detection generally employ three methods: The first method utilizes characteristic data such as the energy distribution, standard deviation, variance, and information entropy of radar echo signals, employing a local peak detection algorithm; the second method combines MTI with a CA-CFAR decision-making method to achieve target detection; and the third method calculates the correlation between the radar echo signal and the transmitted signal template, detects targets by finding the peak of the correlation, sets a threshold for the correlation to control the algorithm's operation, and outputs the number of targets, target distance, and impulse response of each target after the computation loop ends.

[0005] The first approach involves performing feature calculations on radar echo signals, including energy distribution, standard deviation, and information entropy. By differentiating the energy and information complexity of different target echoes, local peak detection of the feature curves is performed. Ultimately, this approach enables the detection of human targets within the detection environment and the acquisition of target detection results.

[0006] The second approach first performs MIT processing on the radar echo to eliminate interference from stationary background echoes, and then combines CA-CFAR to perform target determination on the echo signal after interference elimination, thereby realizing the detection of human targets in the detection environment.

[0007] The third approach is called the Clean approach. This approach first calculates the cross-correlation of radar echoes, then detects the peak value of the cross-correlation curve, and finally obtains the clean echo of the human target based on the detection results.

[0008] However, all three existing schemes have their own drawbacks: The first scheme, in complex environments, suffers from multipath interference between multiple human targets and between the human target and its surroundings, leading to inaccurate target detection methods such as energy distribution, thus failing to accurately determine target information. The second scheme, while reducing background interference to some extent, faces excessive interference between multiple targets in complex environments, making it impossible to completely filter out background interference and thus also hindering accurate target determination. The third scheme not only suffers from the first scheme's flaw—its tendency to inaccurate target detection—but also introduces a threshold parameter whose impact on target detection results is unquantifiable, requiring extensive experimental data fitting, a complex and time-consuming process. Summary of the Invention

[0009] One of the objectives of this invention is to provide a method for obtaining the impulse response of multiple human targets in complex environments that is highly reliable, accurate, and efficient.

[0010] The second objective of this invention is to provide a system for acquiring the impulse response of multiple human targets in the complex environment described above.

[0011] The method for obtaining the impulse response of multiple human targets in a complex environment provided by this invention includes the following steps:

[0012] S1. Acquire radar echo data of the target area;

[0013] S2. Perform MTI processing on the radar echo data obtained in step S1;

[0014] S3. Based on the data obtained in step S2, perform coarse target detection using the CA-CFAR algorithm;

[0015] S4. Based on the target coarse detection results obtained in step S3, perform decision accumulation and obtain the final accumulation decision curve;

[0016] S5. Based on the accumulated decision curve obtained in step S4, determine the peak value;

[0017] S6. Based on the peak value judgment result obtained in step S5, the impulse response reconstruction algorithm based on Clean optimization is used for processing to obtain the final impulse response echoes of multiple human targets in complex environments.

[0018] Step S3, which involves performing coarse target detection based on the data obtained in step S2 using the CA-CFAR algorithm, specifically includes the following steps:

[0019] The i-th echo R(i,j) of the radar echo matrix is ​​processed to obtain the coarse target detection result.

[0020] Initial values ​​for the parameters: echo window size N for a single batch processing, and the total false alarm rate P detected within the echo window. fa,N The number of times the target is detected within the echo window, K; the data window size M for the CA-CFAR algorithm to process the echo; and the protected data window size G when the CA-CFAR algorithm processes echo data.

[0021] The probability distribution of signal amplitude is described by the following formula:

[0022]

[0023] In the formula, G(R(i,j); α,λ) is the echo fitting gamma probability distribution function; α is the shape parameter and E() is the mean function, Var() is the variance function; λ is the scale parameter and Γ(α) is the gamma function;

[0024] False alarm rate P of single echo monitoring fa,i The following formula applies to the threshold T:

[0025]

[0026] In the formula, z is the data variable when the CA-CFAR data window is known; γ() is the incomplete Gamma function; T′ is the actual CA-CFAR operation threshold and T′=T·Z, Z is the window mean obtained when the CA-CFAR algorithm processes echo data;

[0027] When obtaining the coarse target detection result within a single echo window, the total false alarm rate P accumulated by the K / N window decision is obtained. fa,N The false alarm rate P of the i-th echo fa,i The relationship between them satisfies:

[0028]

[0029] In the formula, y is a random variable; The binomial coefficients are and Let y be the probability of false alarms under the set K / N window conditions;

[0030] Based on the total false alarm rate P fa,N Solve for the threshold T;

[0031] Using the obtained threshold T as data input, the actual computational threshold T′ is calculated during echo processing within the window, and finally the coarse detection result of the target is obtained.

[0032] The above is based on the total false alarm rate P fa,N The threshold T is solved by the following steps:

[0033] Through the Gamma cumulative distribution function G cdf Obtain the threshold T;

[0034] The threshold T is calculated using the following formula:

[0035]

[0036] In the formula Gamma cumulative distribution function G cdf The inverse function of .

[0037] Step S4, which involves accumulating decision data based on the coarse target detection results obtained in step S3 and acquiring the final accumulation decision curve, specifically includes the following steps:

[0038] In the two target accumulation processes, the first step is K / N window decision accumulation; the decision δ of the target echo within the window is accumulated to obtain the K / N decision packet; the final accumulation decision curve is obtained by accumulating K / N decisions within the window; and the judgment result D after the l-th K / N judgment is... KN (l,j), the cumulative decision curve expression is obtained as:

[0039]

[0040] In the formula, D(j) is the accumulated decision curve; n′ is the number of decisions required for a total of n echoes to accumulate in the K / N window decision given N.

[0041] Step S5, which involves determining the peak value based on the accumulated decision curve obtained in step S4, specifically includes the following steps:

[0042] After obtaining the accumulation decision curve, the peak value of the accumulation decision curve is determined, and several peak values ​​are obtained.

[0043] Using the formula R(i,tao) k ) = peak k To obtain the final location information tao k And the target quantity num.

[0044] Step S6, based on the peak value judgment result obtained in step S5, uses a Clean-optimized impulse response reconstruction algorithm to process the data and obtain the final impulse response echoes of multiple human targets in complex environments. Specifically, this includes the following steps:

[0045] Based on the peak value determination results, a Clean-optimized impulse response reconstruction algorithm is used for processing; the human target impulse response echo is calculated using the following formula:

[0046]

[0047] In the formula R clean (i) represents the final human target impulse response echo; s(t) represents the original transmitted signal template.

[0048] This invention also provides a system for acquiring the impulse response of multiple human targets in a complex environment, comprising a data acquisition module, a data processing module, a coarse detection module, a curve acquisition module, a peak value judgment module, and a response acquisition module; the data acquisition module, data processing module, coarse detection module, curve acquisition module, peak value judgment module, and response acquisition module are connected in series; the data acquisition module acquires radar echo data of the target area and uploads the data to the data processing module; the data processing module performs MTI processing on the received radar echo data and uploads the data to the coarse detection module; the coarse detection module performs coarse target detection based on the CA-CFAR algorithm based on the received data and uploads the data to the curve acquisition module; the curve acquisition module performs decision accumulation based on the received data and acquires the final accumulation decision curve, and uploads the data to the peak value judgment module; the peak value judgment module performs peak value judgment based on the received data and uploads the data to the response acquisition module; the response acquisition module processes the received data using a Clean-optimized impulse response reconstruction algorithm to acquire the final impulse response echoes of multiple human targets in the complex environment.

[0049] The method and system for acquiring the impulse response of multiple human targets in complex environments provided by this invention, through the innovative design of the overall algorithm, can not only realize the identification and impulse response acquisition of multiple human targets in complex environments, but also has higher reliability, better accuracy and higher efficiency. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the method flow of the present invention.

[0051] Figure 2 This is a schematic diagram comparing the detection effects of the method of the present invention with those of existing technologies.

[0052] Figure 3 This is a schematic diagram illustrating the acquisition effect of the pure echo of the human target by the method of the present invention and the prior art.

[0053] Figure 4 This is a schematic diagram comparing the detection effects of the method of the present invention with the second scheme in the background art.

[0054] Figure 5 This is a schematic diagram of the functional modules of the system of the present invention. Detailed Implementation

[0055] like Figure 1 The diagram shown is a flowchart of the method of the present invention: The method for obtaining the impulse response of multiple human targets in a complex environment disclosed in this invention includes the following steps:

[0056] S1. Acquire radar echo data of the target area;

[0057] S2. Perform MTI processing on the radar echo data obtained in step S1;

[0058] S3. Based on the data obtained in step S2, perform coarse target detection using the CA-CFAR algorithm (this algorithm is a CA-CFAR algorithm with improved detection threshold); specifically including the following steps:

[0059] The i-th echo R(i,j) of the radar echo matrix is ​​processed to obtain the coarse target detection result.

[0060] Initial values ​​for the parameters: echo window size N for a single batch processing, and the total false alarm rate P detected within the echo window. fa,N The number of times the target is detected within the echo window, K; the data window size M for the CA-CFAR algorithm to process the echo; and the protected data window size G when the CA-CFAR algorithm processes echo data.

[0061] The probability distribution of signal amplitude is described by the following formula:

[0062]

[0063] In the formula, G(R(i,j); α,λ) is the echo fitting gamma probability distribution function; α is the shape parameter and E() is the mean function, Var() is the variance function; λ is the scale parameter and Γ(α) is the gamma function;

[0064] False alarm rate P of single echo monitoring fa,i The following formula applies to the threshold T:

[0065]

[0066] In the formula, z is the data variable when the CA-CFAR data window is known; γ() is the incomplete Gamma function; T′ is the actual CA-CFAR operation threshold and T′=T·Z, Z is the window mean obtained when the CA-CFAR algorithm processes echo data;

[0067] When obtaining the coarse target detection result within a single echo window, since the steps of this invention only have two outputs δ0 / δ1, meaning the decision result satisfies a binomial distribution, the total false alarm rate P accumulated by the K / N window decision can be obtained. fa,N The false alarm rate P of the i-th echo fa,i The relationship between them satisfies:

[0068]

[0069] In the formula, y is a random variable; The binomial coefficients are and Let y be the probability of false alarms under the set K / N window conditions;

[0070] Based on the total false alarm rate P fa,N The threshold T is calculated using the following steps:

[0071] Because according to P fa,N Directly solving for T is quite complex; in practice, it can be simplified to using the Gamma cumulative distribution function G. cdf To obtain the threshold T for the judgment;

[0072] The threshold T is calculated using the following formula:

[0073]

[0074] In the formula Gamma cumulative distribution function G cdf The inverse function;

[0075] Using the obtained threshold T as data input, the actual computational threshold T′ is calculated during the echo processing within the window, and finally the coarse detection result of the target is obtained.

[0076] S4. Based on the coarse target detection results obtained in step S3, perform decision accumulation and obtain the final accumulation decision curve; specifically including the following steps:

[0077] In the two target accumulation processes, the first step is K / N window decision accumulation; the decision δ of the target echo within the window is accumulated to obtain the K / N decision packet; the final accumulation decision curve is obtained by accumulating K / N decisions within the window; and the judgment result D after the l-th K / N judgment is... KN (l,j), the cumulative decision curve expression is obtained as:

[0078]

[0079] In the formula, D(j) is the accumulation decision curve; n′ is the number of decisions required for a total of n echoes to accumulate in the K / N window decision given N.

[0080] S5. Based on the accumulated decision curve obtained in step S4, determine the peak value; specifically, this includes the following steps:

[0081] After obtaining the accumulation decision curve, the peak value of the accumulation decision curve is determined, and several peak values ​​are obtained.

[0082] Using the formula R(i,tao) k ) = peak k To obtain the final location information tao k and target quantity num;

[0083] S6. Based on the peak value judgment result obtained in step S5, the impulse response reconstruction algorithm based on Clean optimization is used for processing to obtain the final impulse response echoes of multiple human targets in complex environments; specifically, the following steps are included:

[0084] Based on the peak value determination results, a Clean-optimized impulse response reconstruction algorithm is used for processing; the human target impulse response echo is calculated using the following formula:

[0085]

[0086] In the formula R clean (i) represents the final human target impulse response echo; s(t) represents the original transmitted signal template.

[0087] The effects of the method of the present invention are described below:

[0088] like Figure 2 The diagram shows a comparison of the detection effects of the method of the present invention and the prior art; in the diagram, a) is a schematic diagram of the original experimental echo image, b) is a schematic diagram of the energy peak detection result, c) is a schematic diagram of the standard deviation peak detection result, d) is a schematic diagram of the variance peak detection result, e) is a schematic diagram of the entropy peak detection result, and f) is a schematic diagram of the detection result of the method of the present invention; through Figure 2 As can be seen, when the environment contains four human targets, the first solution in the background art (corresponding to b) to e) cannot achieve accurate detection of the human targets. However, the method of the present invention achieves accurate target detection.

[0089] like Figure 3 The diagram illustrates the acquisition results of pure echoes from a human target using the method of this invention and existing technologies. In the diagram, a) is a schematic diagram of the original experimental echo image, b) is a schematic diagram of the energy peak detection result, c) is a schematic diagram of the standard deviation peak detection result, d) is a schematic diagram of the variance peak detection result, e) is a schematic diagram of the entropy peak detection result, and f) is a schematic diagram of the acquisition result obtained by the method of this invention. Figure 3 As can be seen, the method of the present invention successfully obtained the pure echoes of four human targets, i.e., their impulse responses; proving that the method of the present invention has good effectiveness.

[0090] like Figure 4 The diagram shows a comparison of the detection effects of the method of this invention and the second scheme in the background art; where a) is a schematic diagram of the original echo image of the experiment, b) is the decision map obtained by the second scheme in the background art, and c) is the decision map obtained by the method of this invention; through Figure 4 As can be seen, the decision-making of this issuance method is significantly better than the second scheme in the background technology, and it has advantages in multi-objective determination in complex environments. Figure 4 This also proves that the method of the present invention has better coarse detection results.

[0091] like Figure 5 The diagram shows the functional modules of the system of this invention: The system disclosed in this invention for acquiring the impulse response of multiple human targets in a complex environment includes a data acquisition module, a data processing module, a coarse detection module, a curve acquisition module, a peak value judgment module, and a response acquisition module; the data acquisition module, data processing module, coarse detection module, curve acquisition module, peak value judgment module, and response acquisition module are connected in series; the data acquisition module is used to acquire radar echo data of the target area and upload the data to the data processing module; the data processing module is used to perform MTI processing on the received radar echo data and upload the data to the coarse detection module; the coarse detection module is used to perform coarse target detection based on the CA-CFAR algorithm based on the received data and upload the data to the curve acquisition module; the curve acquisition module is used to perform decision accumulation based on the received data and acquire the final accumulation decision curve, and upload the data to the peak value judgment module; the peak value judgment module is used to perform peak value judgment based on the received data and upload the data to the response acquisition module; the response acquisition module is used to process the received data using a Clean-optimized impulse response reconstruction algorithm to acquire the final impulse response echo of multiple human targets in a complex environment.

Claims

1. A method for obtaining impulse responses of multiple human targets in a complex environment, comprising the following steps: S1. Acquire radar echo data of the target area; S2. Perform MTI processing on the radar echo data obtained in step S1; S3. Based on the data obtained in step S2, perform coarse target detection using the CA-CFAR algorithm; specifically, this includes the following steps: The first radar echo matrix Dao Echo The process is performed, and the coarse detection result of the target is obtained after processing. Initial values ​​for the parameters: echo window size N for a single batch processing, and the total false alarm rate detected within the echo window. The number of times the target is detected within the echo window, K; the data window size M for the CA-CFAR algorithm to process the echo; and the protected data window size G when the CA-CFAR algorithm processes echo data. The probability distribution of signal amplitude is described by the following formula: In the formula Fit the gamma probability distribution function to the echo; For shape parameters and , It is a mean function. It is the variance function; The scale parameter and ; It is the gamma function; False alarm rate of single echo monitoring The following formula applies to the threshold T: In the formula For data variables when the CA-CFAR data window is known; It is an incomplete Gamma function; The actual calculation threshold of CA-CFAR and , The window mean obtained when processing echo data using the CA-CFAR algorithm; When obtaining coarse target detection results within a single echo window, the total false alarm rate accumulated by the K / N window decision is obtained. With the False alarm rate of echo The relationship between them satisfies: In the formula It is a random variable; The binomial coefficients are and ; To be under the set K / N window conditions The probability of a false alarm; Based on the overall false alarm rate Solve for the threshold T; Using the calculated threshold T as data input, the actual threshold is calculated during echo processing within the window. The calculations ultimately yield the coarse detection results of the target. S4. Based on the target coarse detection results obtained in step S3, perform decision accumulation and obtain the final accumulation decision curve; S5. Based on the accumulated decision curve obtained in step S4, determine the peak value; S6. Based on the peak value judgment result obtained in step S5, the impulse response reconstruction algorithm based on Clean optimization is used for processing to obtain the final impulse response echoes of multiple human targets in complex environments.

2. The method for obtaining the impulse response of multiple human targets in a complex environment according to claim 1, characterized in that... The above is based on the total false alarm rate The threshold T is solved by the following steps: Through the Gamma cumulative distribution function Obtain the threshold T; The threshold T is calculated using the following formula: In the formula Gamma cumulative distribution function The inverse function of .

3. The method for obtaining the impulse response of multiple human targets in a complex environment according to claim 2, characterized in that... Step S4, which involves accumulating decision data based on the coarse target detection results obtained in step S3 and acquiring the final accumulation decision curve, specifically includes the following steps: In the two stages of goal accumulation, the first step is Accumulated experience in decision-making at the window; Decision on target echo within the window Accumulate to acquire Decision package; for processing within the window The decision accumulation process yields the final accumulation decision curve; for the l-th iteration... Judgment result after judgment The expression for the cumulative decision curve is obtained as follows: In the formula To accumulate decision curves; For the common Dao echo in The number of decisions required for window decision accumulation given N.

4. The method for obtaining the impulse response of multiple human targets in a complex environment according to claim 3, characterized in that... Step S5, which involves determining the peak value based on the accumulated decision curve obtained in step S4, specifically includes the following steps: After obtaining the accumulation decision curve, the peak value of the accumulation decision curve is determined, and several peak values ​​are obtained. Using formula To obtain the final location information And the target quantity num.

5. The method for obtaining the impulse response of multiple human targets in a complex environment according to claim 4, characterized in that... Step S6, based on the peak value judgment result obtained in step S5, uses a Clean-optimized impulse response reconstruction algorithm to process the data and obtain the final impulse response echoes of multiple human targets in complex environments. Specifically, this includes the following steps: Based on the peak value determination results, a Clean-optimized impulse response reconstruction algorithm is used for processing; the human target impulse response echo is calculated using the following formula: In the formula This is the final target impulse response echo of the human body; This is the original transmitted signal template.

6. A system for implementing the method for acquiring the impulse response of multiple human targets in a complex environment as described in any one of claims 1 to 5, characterized in that... The system includes a data acquisition module, a data processing module, a coarse detection module, a curve acquisition module, a peak value judgment module, and a response acquisition module; these modules are connected in series. The data acquisition module acquires radar echo data of the target area and uploads the data to the data processing module. The data processing module performs MTI processing on the received radar echo data and uploads the data to the coarse detection module. The coarse detection module performs coarse target detection based on the CA-CFAR algorithm and uploads the data to the curve acquisition module. The curve acquisition module performs decision accumulation based on the received data and acquires the final accumulation decision curve, then uploads the data to the peak value judgment module. The peak value determination module is used to determine the peak value based on the received data and upload the data to the response acquisition module; the response acquisition module is used to process the received data using a Clean-optimized impulse response reconstruction algorithm to obtain the final impulse response echoes of multiple human targets in complex environments.

Images