Robust detection method of mismatched signal in main-lobe interference background

Abstract

The application discloses a mismatch signal robust detection method under a main lobe interference background and relates to the technical field of signal processing, and comprises the following steps: obtaining a to-be-detected signal, constructing a multivariate hypothesis testing problem, constructing a penalty detector, solving the multivariate hypothesis testing problem, and obtaining a detection statistic of the to-be-detected signal; in the solving process, unknown parameters are estimated, and the estimation result is used for solving the multivariate hypothesis testing problem; according to the detection statistic, a distance gate estimation result and a fuzzy number estimation result are obtained; whether the fuzzy number estimation result is equal to preset target distance fuzzy area prior information is judged, if the fuzzy number estimation result is equal to the preset target distance fuzzy area prior information, a real target exists in the to-be-detected signal, and a target distance gate position is obtained, and if the fuzzy number estimation result is not equal to the preset target distance fuzzy area prior information, a real target does not exist in the to-be-detected signal. The application can improve the detection performance under a signal mismatch condition.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of signal processing, and particularly relates to a mismatched signal robust detection method under main lobe interference background. BACKGROUND

[0002] Frequency diverse array (FDA) radar is a new radar system emerging in recent years. Unlike phased array (PA) radar and multiple-input multiple-output (MIMO) radar, FDA radar introduces a small frequency difference between each transmitting unit, so that the transmitting pattern of the FDA radar is not only related to the angle, but also related to the distance. The frequency diverse array multiple-input multiple-output (FDA-MIMO) radar combining FDA radar and MIMO radar not only enjoys the spatial diversity advantage of MIMO radar, but also has a controllable degree of freedom in the distance dimension. Early researches on FDA-MIMO radar are mostly focused on static targets, such as parameter estimation, clutter suppression, interference (especially main lobe interference) identification and suppression, etc.

[0003] Under the scenario of FDA-MIMO radar detecting targets in the air, false targets are easily disturbed, and the false targets are similar to real targets in features. Both the false targets and the real targets are located in the main lobe detection beam. Using the traditional method, the false targets increase the difficulty of target detection, and the existence of main lobe interference increases the detection false alarm rate. In addition, due to the existence of a certain width of the beam, there is a deviation between the apparent parameters of the target detection and the real target parameters, and the mismatch of the steering vector will lead to the decline of the target detection performance.

[0004] Therefore, it is urgent to provide a mismatched signal robust detection method to improve the defects existing in the prior art. SUMMARY

[0005] In order to solve the above problems existing in the prior art, the application provides a mismatched signal robust detection method under main lobe interference background. The technical problems to be solved by the application are realized by the following technical scheme:

[0006] The application provides a mismatched signal robust detection method under main lobe interference background, comprising:

[0007] Obtaining a to-be-detected signal; wherein the to-be-detected signal comprises a real target signal plus a noise signal, or a main lobe interference signal plus a noise signal, or a noise signal;

[0008] According to the to-be-detected signal, a multiple hypothesis testing problem is constructed;

[0009] A penalty detector is constructed by using a likelihood ratio detection criterion and a model order selection criterion, and the multiple hypothesis testing problem is solved to obtain a detection statistic of the to-be-detected signal; in the solving process, unknown parameters are estimated, and the estimation results are used to solve the multiple hypothesis testing problem; wherein the unknown parameters include a signal complex coordinate coefficient vector, a subspace signal distance ambiguity vector and a distance gate serial number;

[0010] According to the detection statistic, a distance gate estimation result and an ambiguity number estimation result are obtained; it is judged whether the ambiguity number estimation result is equal to preset target distance ambiguity region prior information, if yes, there is a real target in the to-be-detected signal, and the target distance gate position is obtained, if not, there is no real target in the to-be-detected signal.

[0011] The beneficial effects of the present application are as follows:

[0012] The robust detection method for mismatched signals in the main lobe interference background provided by the present application uses a subspace signal model to model the mismatched signal detection problem in a to-be-detected window as a multiple hypothesis testing problem, and uses a likelihood ratio detection criterion and a model order selection criterion to solve the problem; in the solving process, unknown parameters are estimated, and the estimation results are used to calculate a detection statistic to realize signal number estimation in the to-be-detected window; by jointly using signal number and ambiguity number estimation results, and combining preset target distance ambiguity region prior information, real target signal detection judgment is realized.

[0013] The present application will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is a flow chart of the robust detection method for mismatched signals in the main lobe interference background provided by the embodiment of the present application;

[0015] Figure 2 is a schematic diagram of the robust detection method for mismatched signals in the main lobe interference background provided by the embodiment of the present application;

[0016] Figure 3 is a schematic diagram of the received signal in the fast time dimension provided by the embodiment of the present application;

[0017] Figure 4a is a histogram of the source number estimation probability of the robust detection method for mismatched signals under different signal-to-noise ratios provided by the embodiment of the present application;

[0018] Figure 4bIt is a histogram of the source position estimation probability of the mismatch signal robust detection method provided by the embodiment of the application under different signal-to-noise ratio conditions;

[0019] Figure 5a It is a schematic diagram of the real target detection performance curve changing with the signal-to-noise ratio when the angle deviation is 0.

[0020] Figure 5b It is a schematic diagram of the real target detection performance curve changing with the signal-to-noise ratio when the angle deviation is 3dB beam width. DETAILED DESCRIPTION

[0021] The application will be described in further detail below with reference to specific embodiments, but the embodiments of the application are not limited thereto.

[0022] Please refer to Figure 1 and Figure 2 , Figure 1 It is a flowchart of the mismatch signal robust detection method in the main lobe interference background provided by the embodiment of the application, Figure 2 It is a schematic diagram of the mismatch signal robust detection method in the main lobe interference background provided by the embodiment of the application, and the mismatch signal robust detection method in the main lobe interference background provided by the application comprises:

[0023] S101, acquiring a to-be-detected signal; wherein the to-be-detected signal comprises a real target signal plus a noise signal, or a main lobe interference signal plus a noise signal, or a noise signal.

[0024] Specifically, in the embodiment, for the FDA-MIMO radar, the array comprises a uniform linear array with an array element spacing of , the number of transmitting array elements and receiving array elements is and respectively. Considering that there is a real target at a far-field position, the real target is located at a distance and an angle , after distance-dependent compensation, the real target signal received by the FDA-MIMO radar is represented as:

[0025] (1);

[0026] ;

[0027] ;

[0028] wherein, represents the received real target signal, represents a target complex-valued scattering coefficient, represents a target transmit steering vector, Indicates the target receiving guidance vector. Indicates the target angular frequency. Indicates the target distance frequency. Represents the speed of light. Indicates the wavelength of the transmitted signal. This indicates the transpose operation. Indicates the frequency offset. This represents the distance ambiguity region where the real target is located, i.e., the target distance ambiguity number.

[0029] After the FDA-MIMO radar transmits a signal, the decoy generator, upon intercepting the transmitted signal and delaying it, generates interference similar to the target, i.e., mainlobe deception interference. The interference generated by the decoy after the delay is set to be located at a distance... ,angle At the location, after main value range compensation, the main lobe interference signal received by the FDA-MIMO radar is represented as:

[0030] (2);

[0031] ;

[0032] ;

[0033] in, Represents the interference complex scattering coefficient. This represents the interference signal transmission steering vector. Indicates the interference signal reception steering vector. Indicates the frequency of the interference angle. Indicates the interference distance and frequency. Indicates the ambiguity number of the interference signal distance. , ;

[0034] ;

[0035] in, This indicates the FDA-MIMO radar's ability to identify ambiguities. , Indicates 3dB beamwidth. This indicates the apparent angle parameter of the beam of the FDA-MIMO radar in the detection state.

[0036] At the receiver of the FDA-MIMO radar, such as Figure 3 As shown, Figure 3 This is a schematic diagram of the received signal in the fast time dimension provided in an embodiment of the present invention. The received signal is in the fast time dimension, and the detection window may include real targets and pseudo-randomly distributed main lobe interference. A certain distance gate sampling signal within the detection window is denoted as... is a set of integer distance gate index, distance gate sample signal includes noise signal and possible target signal or main lobe interference signal. In order to estimate noise covariance matrix, training sample data is obtained in fast time sampling outside the protection unit is a set of integer training sample data index. Due to the existence of pseudo-random distribution main lobe interference, it is difficult to distinguish real target signal and main lobe interference signal in fast time dimension, and the detection problem of the received signal is described as: judging whether there is a real target at the apparent angle position of the detection window.

[0037] In an ideal case, for a detection window with a window length of , assuming that there are signals of interest, the received signal detection problem is described as a multiple hypothesis testing problem, which is represented as:

[0038] (3).

[0039] Wherein, represents a set of distance gate index of the signal of interest, and .

[0040] Considering the non-ideal case of apparent parameters, there is a deviation between the interference angle and the apparent angle, that is, the mismatch of the steering vector of the detection signal in formula (3), which leads to the decline and even failure of the detection performance. How to re-model the detection signal and improve the signal detection performance under the mismatching condition becomes the focus of the description of the present application.

[0041] S102, constructing a multiple hypothesis testing problem according to the detection signal.

[0042] Specifically, in this embodiment, for the steering vector mismatch problem, the robustness of the subspace model is used for modeling, which can improve the detection performance under the mismatching condition. It is assumed that the detection signal is located in a known subspace, and based on the subspace model, the multiple hypothesis testing problem is represented as:

[0043] (4).

[0044] Wherein, represents the null hypothesis, that is, the hypothesis that there is no signal of interest, represents the hypothesis that there are signals of interest, represents the distance gate sample signal, represents the training sample data, represents the​​​​​ received noise data inside the distance gate, denotes noise data in the i-th training sample data, denotes noise data in the i-th training sample data, denotes an integer set including distance gate numbers, denotes an integer set including training sample data numbers, denotes a subspace where the signal to be detected is located, denotes a dimension of the signal subspace, denotes a number of transmitting array elements, denotes a number of receiving array elements, denotes a signal complex coordinate coefficient vector, denotes a set of distance gate numbers where the signal of interest is located.

[0045] Further, unlike the conventional subspace model, the conventional subspace model sets that the signal is located in a completely known signal subspace; considering that there are unknown signal ambiguity numbers and angle deviations at the same time, an excessively high subspace dimension will result in great detection performance loss, the subspace model provided in the embodiment is only for angle mismatch amount, and the signal subspace still has unknown parameters, i.e., the ambiguity number is an unknown parameter; the subspace where the signal to be detected is located is represented as:

[0046] (5);

[0047] wherein, denotes the i-th steering vector of the subspace where the signal is located, denotes a transmitting steering vector, denotes a receiving steering vector, denotes an unknown subspace signal distance ambiguity number, denotes a known angle of the subspace where the signal is located, denotes a known angle of the i-th subspace where the signal is located, the angle can be selected as a certain angle near the apparent angle (within one beam width), denotes an angle number. S103, a likelihood ratio detection criterion and a model order selection criterion are adopted to construct a penalty detector, and a multiple hypothesis testing problem is solved to obtain a detection statistic of the signal to be detected; in the solving process, unknown parameters are estimated, and the estimation results are used to solve the multiple hypothesis testing problem; wherein the unknown parameters include a signal complex coordinate coefficient vector, a subspace signal distance ambiguity vector and a distance gate number.

[0048] S103, a likelihood ratio detection criterion and a model order selection criterion are adopted to construct a penalty detector, and a multiple hypothesis testing problem is solved to obtain a detection statistic of the signal to be detected; in the solving process, unknown parameters are estimated, and the estimation results are used to solve the multiple hypothesis testing problem; wherein the unknown parameters include a signal complex coordinate coefficient vector, a subspace signal distance ambiguity vector and a distance gate number.

[0049] ​Specifically, in the present embodiment, the following unknown parameters exist in the multiple hypothesis testing problem: a signal complex coordinate coefficient vector , a subspace signal distance ambiguity number vector , a distance gate sequence number set , a penalty detector is constructed by jointly using a likelihood ratio test (LRT) criterion and a model order selection (MOS) criterion; the model order selection criterion includes a Bayesian information criterion (BIC) and a generalized information criterion (GIC).

[0050] The penalty detector is expressed as:

[0051] (6);

[0052] (7);

[0053] (8);

[0054] (9);

[0055] (10);

[0056] (11);

[0057] (12);

[0058] ;

[0059] wherein represents a detection statistic, represents a signal matrix to be detected, represents a penalty term related to the model order selection criterion, represents a hypothesis under a signal number estimation result, represents a detection threshold, represents a subspace signal distance ambiguity number vector, represents a distance gate sequence number set in which a signal of interest is located, represents an integer, represents a probability density function of a signal to be detected, represents a length of a signal window to be detected, represents a conjugate transpose operator, represents a distance gate sampling signal, denotes the number of training sample data, denotes the i-th training sample data, denotes the log-likelihood ratio of the signal to be detected under the hypothesis, denotes the probability density function of the signal to be detected under the hypothesis, denotes the autocorrelation matrix of the data after removing the target signal from the signal to be detected under the hypothesis, denotes the autocorrelation matrix of the signal to be detected under the hypothesis, denotes the estimated background covariance matrix, denotes the number of signals of interest under the hypothesis, denotes the trace of a matrix.

[0060] Further, the model order selection criterion includes the Bayesian information criterion and the generalized information criterion, and the penalty term related to the model order selection criterion is denoted as:

[0061] (13) ;

[0062] denotes the penalty term based on the Bayesian information criterion;

[0063] denotes the penalty term based on the generalized information criterion;

[0064] wherein, denotes the number of unknown signals, including signal complex coordinate coefficients, subspace signal distance ambiguity numbers, and distance gate sequence numbers, denotes the adjustable factor of the generalized information criterion, and , denotes the dimension of the receiving subspace.

[0065] In this embodiment, it can be known from analyzing the formula (8) and the formula (13) that the key to calculating the detection statistic is the estimation of the unknown parameters; wherein,

[0066] The signal complex coordinate coefficient vector is estimated, including:

[0067] The signal complex coordinate coefficient vector in the expression is maximized to obtain the estimation result of the signal complex coordinate coefficient vector , which is denoted as:

[0068] (14) ; ​

[0069] wherein, denotes a set of distance gate indices where the signal of interest is located, denotes the data received by the distance gate corresponding to the th element in the vector, denotes the index of the distance gate.

[0070] In this embodiment, the distance ambiguity vector of the subspace signal is estimated, including:

[0071] The estimation result of the signal complex coordinate coefficient vector is brought into the expression , to obtain:

[0072] (15);

[0073] wherein, denotes a set of distance gate indices where the signal of interest is located;

[0074] For the th distance gate, the estimation result of the distance ambiguity number of the subspace signal is denoted as:

[0075] (16);

[0076] ;

[0077] wherein, denotes the distance ambiguity number of the signal of interest in the distance gate corresponding to the th element in the vector, denotes the ambiguity number recognition capability of the radar.

[0078] In this embodiment, the distance gate index is estimated, including:

[0079] The power of the to-be-detected signal in the receiving dimension is sorted from large to small, and the first distance gates are selected as the estimation result of the distance gate index. For the th distance gate, the power of the to-be-detected signal in the receiving dimension is denoted as:

[0080] (17);

[0081] ;

[0082] ;

[0083] ; ​

[0084] ;

[0085] wherein, denotes the trace of a matrix, denotes the subspace in which the received dimension signal is located, denotes the serial number of the signal vector which spans the received subspace, denotes the signal vector which spans the received subspace, denotes the angle which spans the received subspace, denotes the dimension of the received subspace, denotes the received signal of the rearranged th detection unit, , denotes the rearrangement operator which rearranges column data into matrix data, denotes the th data to be rearranged, denotes the rearranged training sample data covariance matrix, denotes the number of training samples, denotes the th training sample data matrix after rearrangement;

[0086] The estimation result of the distance gate serial number is denoted as:

[0087] (18);

[0088] wherein, denotes the number of sources Assume that the distance gate serial number is ordered from large to small based on the received signal power of the distance gate.

[0089] Further, the multivariate hypothesis testing problem is solved to obtain the detection statistic of the to-be-detected signal, including:

[0090] The estimation result of the signal complex coordinate coefficient vector, the estimation result of the subspace signal distance ambiguity vector, the estimation result of the distance gate serial number, and the penalty term related to the model order selection criterion are brought into the penalty detector to calculate the detection statistic, that is, formula (18), (16), (15) and (13) are brought into formula (6), so that the detection statistic can be calculated, and the robust detection of the mismatched signal is realized.

[0091] S104, according to the detection statistic, the distance gate estimation result and the ambiguity number estimation result are obtained; it is judged whether the ambiguity number estimation result is equal to the preset target distance ambiguity area prior information, if they are equal, there is a real target in the to-be-detected signal, and the target distance gate position is obtained, if they are not equal, there is no real target in the to-be-detected signal.

[0092] Specifically, in this embodiment, according to the estimation result of the number of to-be-detected signals in the detection statistic , the estimation result of the distance gate and the ambiguity number is calculated, which is represented as:

[0093] (19);

[0094] (20);

[0095] wherein, represents the estimation result of the distance gate, represents the estimation result of the ambiguity number, represents the estimation result of the number of signals, represents the estimation result of the number of signals when the estimation result of the number of signals is , and represents the estimation result of the ambiguity number in the first distance gate. In this embodiment, the prior information of the distance ambiguity region where the target is located obtained in the search stage is set

[0096] , and each element in is compared with . When the element is equal to , wherein, is any ambiguity number estimation result in , it is judged that there is a real target in the to-be-detected signal, and the position of the distance gate where the target is located is determined , otherwise, it is judged that there is no real target signal in the to-be-detected signal. In summary, the robust detection method for mismatched signals in the main lobe interference background provided by the present application is used for the case that the main lobe interference signal and the target signal have deviation from the apparent parameters. The subspace signal model is used to model the mismatched signal detection problem in the to-be-detected window as a multiple hypothesis testing problem, and the likelihood ratio detection criterion and the model order selection criterion are used to solve the problem. In the solving process, the unknown parameters are estimated, the estimation result is used to calculate the detection statistic, and the number of signals in the to-be-detected window is estimated. Through the joint use of the estimation results of the number of signals and the ambiguity number, and the preset prior information of the distance ambiguity region where the target is located, the reliable detection of the real target signal is realized.

[0097] In an optional embodiment of the present application, the detection effect of the robust detector in the main lobe interference background provided by the above embodiment is verified through simulation experiments, specifically:

[0098] I. Simulation conditions

[0099]

[0100] ​The simulation experiment conditions of the embodiment are as follows: the transmitting array and the receiving array in the FDA-MIMO radar are set as 6-element uniform linear arrays, the element spacing is half wavelength, the radar carrier frequency is 10 GHz, the element spacing is 0.15 m, the frequency increment is 6 kHz, the training sample sampling number is 4 , the to-be-detected window length is 30, and the radar detection apparent angle is 0 degree. It is considered that there is 1 real target and 4 main lobe interferences in the to-be-detected window, which are located in the 20th, 3rd, 4th, 5th and 6th range gates respectively, and the real target distance ambiguity number is 1 obtained in the search stage.

[0101] II. Simulation content and result analysis

[0102] Please refer to Figure 4a and 4b , Figure 4a is a histogram of the signal source number estimation probability of the mismatch signal robust detection method provided by the embodiment of the application under different signal-to-noise ratio (SNR) conditions, Figure 4b is a histogram of the signal source position estimation probability of the mismatch signal robust detection method provided by the embodiment of the application under different signal-to-noise ratio (SNR) conditions. In the simulation experiment, the angle deviation between the actual signal angle and the apparent parameter is 3 dB beam width, and the selection in the subspace matrix is the uniform sampling of 4 points within one beam width, and the adjustable factor in the GIC criterion is set as 8.

[0103] As shown in Figure 4a , the mismatch signal robust detection method provided by the application has the maximum estimation probability at the correct signal source number position, and the probability of correctly estimating the signal source number is improved with the increase of the signal-to-noise ratio. When SNR=0 dB, the probability of correctly estimating the signal source number based on the BIC criterion is greater than that based on the GIC criterion; when SNR is improved to 5 dB, the estimation probability based on the BIG and GIC criteria reaches 1, which proves the ability of the mismatch signal robust detection method provided by the application to correctly estimate the number of signals of interest in the to-be-detected window. As shown in Figure 4b , when SNR=0 dB, the probability of judging that there is a signal at the correct range gate position based on the mismatch signal robust detection method provided by the application is greater than 0.75; with the improvement of the detection signal-to-noise ratio, when SNR=0 dB, the detection probability of all signals of interest is 1, which proves the ability of the mismatch signal robust detection method provided by the application to detect and judge the correct signal position. In addition, it is noted that the mismatch signal robust detection method provided by the application does not judge that there is a signal at the position without a signal under the simulation conditions, that is, it does not cause ghosting, which proves the effectiveness of the detection and judgment of the mismatch signal robust detection method provided by the application.

[0104] In order to analyze the change of the detection performance with the signal-to-noise ratio, as shown in Figure 5a and 5b ,​Figure 5a is a schematic diagram of the real target detection performance curve with the change of signal-to-noise ratio when the angle deviation is 0 provided by an embodiment of the present application, Figure 5b is a schematic diagram of the real target detection performance curve with the change of signal-to-noise ratio when the angle deviation is 3dB beam width provided by an embodiment of the present application, in order to compare the method proposed in the present application with the existing method, the method in reference 1 is taken as a comparative experiment. From Figure 5a It can be seen that, since there is no angle deviation at this time, compared with the method in document 1 (J. Zhu, S. Zhu, J. Xu and L. Lan, "Discrimination of Target and Mainlobe Jammers With FDA-MIMO Radar" IEEE Signal Processing Letters, vol. 30 pp. 583-587, 2023.), the detection performance of the mismatch signal robust detection method proposed in the present application has performance loss, indicating that the robustness is at the cost of detection performance loss. Further, from Figure 5b It can be seen that, under the condition of angle deviation, the robust detection method proposed in the present application is superior to the method in document 1. When the detection probability is 0.9, the detection performance of the mismatch signal robust detection method proposed in the present application is improved by more than 2.4dB compared with the method in document 1, indicating that the robust detection ability of the mismatch signal robust detection method proposed in the present application is superior to the existing detection method under the condition of signal mismatch. In addition, compared with Figure 5a and Figure 5b It can be found that the detection performance of the mismatch signal robust detection method proposed in the present application is consistent under different angle deviations, because the detection performance of the mismatch signal robust detection method proposed in the present application is irrelevant to the deviation amount.

[0105] It is to be understood that the terminology "first", "second", and the like used herein is merely intended to differentiate one element from another element, and does not imply or suggest any actual relationship or sequence between the elements. Also, the terms "comprise", "comprising", or any other variant thereof are intended to cover non-exclusive inclusions, such that an item or apparatus that comprises a list of elements does not exclude other elements not expressly listed. An element defined by the phrase "comprising a... " does not exclude the presence of additional identical elements in the item or apparatus comprising the element. The terms "connected", "coupled", and the like are not limited to direct connections or physical connections, but can include indirect connections or indirect physical connections between devices or elements such as electrical connections, whether direct or indirect. The terms "upper", "lower", "left", "right", and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are used only for the purpose of facilitating the description of the present application and simplifying the description, and thus cannot be construed as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus cannot be construed as limiting the present application.

[0106] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific feature or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. The illustrative expressions of the above terms in the present specification do not necessarily refer to the same embodiment or example. Also, the specific feature or characteristic described can be combined in any suitable manner in one or more embodiments or examples. In addition, a person skilled in the art can combine and combine different embodiments or examples described in the present specification.

[0107] The above is a further detailed description of the present application in conjunction with specific preferred embodiments, and cannot be considered as limiting the specific implementation of the present application to these descriptions. For those of ordinary skill in the art, without departing from the concept of the present application, a number of simple deductions or substitutions can be made, which should be considered as falling within the scope of protection of the present application.

Claims

1. A robust detection method for mismatch signals under main lobe interference background, characterized in that, include: Acquire the signal to be detected; wherein the signal to be detected includes a real target signal plus a noise signal, or a main lobe interference signal plus a noise signal, or a noise signal; Based on the signal to be detected, a multivariate hypothesis testing problem is constructed; wherein, the subspace containing the signal to be detected... Represented as: ； in, Denotes the first subspace containing Zhang Cheng's signal. One guiding vector, Indicates the launch steering vector. Indicates the receiving guide vector. The distance ambiguity number represents the unknown subspace signal. This represents the angle of the known subspace containing the Zhang Cheng signal. Represents the known first The angle of the subspace where the Zhang Cheng signal is located. Indicates the angle sequence number; A penalty detector is constructed using the likelihood ratio detection criterion and the model order selection criterion to solve the multivariate hypothesis testing problem and obtain the detection statistics of the signal to be detected. During the solution process, unknown parameters are estimated, and the estimation results are used to solve the multivariate hypothesis testing problem. The unknown parameters include the signal complex coordinate coefficient vector, the subspace signal distance ambiguity vector, and the distance gate number. Based on the detection statistics, the distance gate estimation result and the fuzzy number estimation result are obtained; it is determined whether the fuzzy number estimation result is equal to the preset prior information of the distance fuzzy region where the target is located. If they are equal, then there is a real target in the signal to be detected, and the distance gate position of the target is obtained. If they are not equal, then there is no real target in the signal to be detected.

2. The robust detection method for mismatched signals under main lobe interference background according to claim 1, characterized in that, The multivariate hypothesis testing problem is expressed as: ； in, This represents the null hypothesis, which is the assumption that no signal of interest exists. Indicates the assumption that One signal of interest This represents the distance gate sampling signal. Represents training sample data, Indicates the first Received noise data within a distance gate, Indicates the first Noisy data in the training sample data The set of integers representing the distance to the gate index. A set of integers representing the indices of the training sample data. This represents the subspace where the signal to be detected is located. Describes the dimension of the signal subspace. Indicates the number of transmitting array elements. Indicates the number of receiving array elements. Represents the complex coordinate coefficient vector of the signal. This represents the set of gate numbers at which the signal of interest is located.

3. The robust detection method for mismatch signals under main lobe interference background according to claim 2, characterized in that, The penalty detector is represented as: ； ； ； ； ； ； ； ； in, This represents the detection statistics. Represents the matrix of signals to be detected. This represents the penalty term related to the model order selection criterion. This represents the assumptions underlying the signal number estimation results. Indicates the detection threshold. Represents the subspace signal distance ambiguity vector. Represents integers, express Assume the probability density function of the signal to be detected. Indicates the length of the signal window to be detected. This represents the conjugate transpose operator. This represents the distance gate sampling signal. Indicates the number of training sample data. Indicates the first One training sample data, express Assume the log-likelihood ratio of the signal to be detected. Let represent the probability density function of the signal to be detected under the null hypothesis. express Assume the autocorrelation matrix of the data after removing the target signal from the signal to be detected. express Assume the autocorrelation matrix of the signal to be detected, This represents the estimated background covariance matrix. This indicates the number of hypothetical signals of interest.

4. The robust detection method for mismatch signals under main lobe interference background according to claim 3, characterized in that, The model order selection criterion includes the Bayesian information criterion and the generalized information criterion, and the penalty term associated with the model order selection criterion... Represented as: ； This represents a penalty term based on the Bayesian information criterion; This represents a penalty term based on the generalized information criterion; in, Indicates the number of unknown signals, including Each signal complex coordinate coefficient, The number of subspace signal distance ambiguities and Each distance gate number, The adjustable factor representing the generalized information criterion. This indicates the dimension of the receiving subspace.

5. The robust detection method for mismatch signals under main lobe interference background according to claim 3, characterized in that, Estimating the complex coordinate coefficient vector of the signal includes: For expression The signal complex coordinate coefficient vector is maximized to obtain the estimated result of the signal complex coordinate coefficient vector. , represented as: ； in, This represents the set of gate numbers at which the signal of interest is located. express The Middle The data received by the distance gate corresponding to each element Indicates the index of the distance gate.

6. The robust detection method for mismatch signals under main lobe interference background according to claim 5, characterized in that, Estimating the distance ambiguity vector of the subspace signal includes: The estimation result of the complex coordinate coefficient vector of the signal Substitution expression In the middle, we get: ； Regarding the first The estimation results of the distance ambiguity number of the subspace signal using a distance gate. Represented as: ； in, express Inner The distance ambiguity number of the signal of interest within the distance gate corresponding to each element. This indicates the radar's ability to identify fuzzy numbers.

7. The robust detection method for mismatch signals under main lobe interference background according to claim 4, characterized in that, Estimating the distance gate number includes: Sort the power of the signals to be detected in the receiving dimension from largest to smallest, and select the top power. The estimated result of the distance gate is used as the distance gate number for the nth distance gate. A distance gate is used to calculate the power of the signal to be detected in the receiving dimension. , represented as: ； ； ； ； ； in, This indicates finding the trace of a matrix. This indicates the subspace where the received dimensional signal resides. This indicates the index of the signal vector received by Zhang Cheng in the subspace. This represents the signal vector of Zhang Cheng's receiving subspace. This represents the angle of Zhang Cheng's receiving subspace. Indicates the dimension of the receiving subspace. Indicates the rearranged first The received signal of each unit to be detected. , This represents a rearrangement operator that rearranges column data into matrix data. Indicates the first One set of data to be rearranged. This represents the covariance matrix of the rearranged training sample data. Indicates the number of training samples. Indicates the rearranged first... A training sample data matrix; The estimation result of the distance gate number Represented as: ； in, Represents the number of information sources Assume the range gate numbers are sorted from largest to smallest based on the received signal power of the range gate.

8. The robust detection method for mismatch signals under main lobe interference background according to claim 7, characterized in that, Solving the multivariate hypothesis testing problem yields the detection statistics for the signal to be detected, including: The detection statistic is calculated by substituting the estimation results of the signal complex coordinate coefficient vector, the estimation results of the subspace signal distance ambiguity vector, the estimation results of the distance gate number, and the penalty term related to the model order selection criterion into the penalty detector.

9. The robust detection method for mismatch signals under main lobe interference background according to claim 8, characterized in that, Based on the detection statistics, the distance gate estimation result and the fuzzy number estimation result are obtained, including: Based on the estimation results of the number of signals to be detected in the aforementioned detection statistics The estimated results of the distance gate and fuzzy number are expressed as follows: ； ； in, This represents the estimation result of the distance gate. This represents the estimation result of the fuzzy number. The signal number estimation result is: Time Estimation results of the number of ambiguities in the signal within the distance gate.

Images