Multichannel constant modulus signal detection method without electromagnetic interference a priori

Abstract

This invention discloses a method for detecting multi-channel constant-mode signals without prior electromagnetic interference, belonging to the field of radar signal processing technology. The method includes modeling multi-shot data received by a multi-channel antenna array to obtain high-dimensional multi-channel data; introducing weight vectors to project the high-dimensional multi-channel data onto a single channel for dimensionality reduction, resulting in single-channel weighted data; constructing a generalized likelihood ratio test detector without prior spatial electromagnetic interference information based on the probability density functions under two assumptions: the presence and absence of multi-channel constant-mode signals; and building an optimization model based on the generalized likelihood ratio test detector without prior spatial electromagnetic interference information. A hybrid optimization framework combining alternating optimization and alternating direction multiplier methods is used to solve the optimization model, obtaining the detection parameters and completing the multi-channel constant-mode detection.

Description

Technical Field

[0001] This invention belongs to the field of radar signal processing technology, specifically relating to a method for detecting multi-channel constant-mode signals without prior electromagnetic interference. Background Technology

[0002] Multichannel signal detection, a core fundamental task in array signal processing, plays an indispensable role in numerous fields such as radar detection, cognitive radio, wireless communication, autonomous navigation, medical ultrasound, and seismic location exploration due to its ability to simultaneously sense, separate, and identify multi-source signals. It has long been a focus of extensive attention and in-depth research in academia and engineering. Constant-mode signals, which effectively reduce the nonlinear distortion of transmitter power amplifiers and improve spectral efficiency, are widely used in radar pulse modulation and modern wireless communication systems. This makes high-reliability detection of constant-mode signals in the presence of spatial interference a fundamental issue in array signal processing. In noisy environments, spatial diversity techniques offer stronger interference suppression and signal resolution capabilities compared to single-channel reception, attracting widespread research attention. Radienxe et al. proposed a nonlinear product processor for detecting coprime array signals in spatially correlated noise scenarios, achieving performance comparable to traditional beamforming reconnaissance schemes and exhibiting excellent robustness to noise correlation. Olesya et al. pointed out the signal-to-noise ratio (SNR) gate constraint problem in incoherent multi-antenna signal detection and proposed a method for constructing combined decision statistics, effectively improving signal detection performance and detectability. Mati et al. proposed an invariant signal subspace matching criterion for source counting in uniform linear and rectangular arrays, maintaining robust detection performance under finite sample conditions in both white and colored electromagnetic noise environments. Le et al. designed a convolutional neural network model for array structures, overcoming multipath effects and low signal-to-noise ratios by deeply mining multichannel spectrogram features, significantly improving the accuracy and effectiveness of underwater acoustic signal detection. Xin et al., focusing on applications of uniform linear arrays, proposed an optimization method based on minimum mean square error, achieving joint solution for Signal of Interest (SOI) detection and Direction of Arrival (DOA) estimation through a data-driven iterative regularization strategy.

[0003] Most existing multi-channel constant-mode signal detection methods either require prior knowledge of the direction and power of electromagnetic interference or need to collect secondary auxiliary data to calculate the interference-noise covariance matrix. However, in complex and ever-changing real-world scenarios, electromagnetic interference exhibits randomness and dynamic characteristics, making it difficult to accurately obtain its prior information, which can easily lead to significant degradation in detection performance. At the same time, the acquisition and preprocessing of secondary data greatly increases the computational complexity and storage overhead of the detection system, making it difficult to meet the real-time requirements of fields such as radar and communication, and easily causing problems such as weak multi-channel constant-mode signals being overwhelmed by interference.

[0004] To address the shortcomings of existing multi-channel constant mode signal detection methods, which rely on prior knowledge of electromagnetic interference, require secondary data support, and have high computational complexity, there is an urgent need for a multi-channel constant mode signal detection method that does not require prior information on electromagnetic interference, does not rely on secondary data, and is low in complexity and highly reliable. Summary of the Invention

[0005] The purpose of this invention is to overcome the problems of existing multi-channel constant mode detection methods that rely on prior knowledge of electromagnetic interference, require secondary data support, and have high computational complexity, and proposes a multi-channel constant mode signal detection method that does not require prior knowledge of electromagnetic interference.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for detecting multi-channel constant-mode signals without prior electromagnetic interference, comprising the following steps: Model the multi-shot data received by the multi-channel antenna array to obtain high-dimensional multi-channel data. Introduce weight vectors to project the high-dimensional multi-channel data onto a single channel for data dimensionality reduction processing to obtain single-channel weighted data. Based on single-channel weighted data, a generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference is constructed according to the probability density function under the two assumptions of the existence and non-existence of multi-channel constant mode signals. An optimization model is constructed based on a generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference. A hybrid optimization framework combining alternating optimization and alternating direction multiplier method is adopted to solve the optimization model, obtain the detection parameters, and complete multi-channel constant mode detection.

[0007] Furthermore, the multi-shot data received by the multi-channel antenna array is modeled, including joint modeling of multi-channel constant mode signals, spatial interference, and additive noise; High-dimensional multi-channel data is shown in the following formula:

[0008] in, For multi-channel data reception matrix, This refers to the number of channels in the antenna array or the number of sensors. For the number of snapshots, For the complex amplitude of the multi-channel constant mode, For the direction of arrival of the multi-channel constant mode, For multi-channel constant mode in direction The guide vector at that location, For the waveform vector of multi-channel constant mode, For the amount of spatial interference, For the first The direction of arrival of spatial interference. For the first Spatial interference in direction The guide vector at that location, For the first One spatial interference vector, The matrix is ​​an additive white Gaussian noise matrix. The variance of additive noise; Introducing weight vectors projects high-dimensional multi-channel data onto a single channel for dimensionality reduction, resulting in single-channel weighted data. This includes applying constraints to the weight vectors. Projection dimensionality reduction is performed; the probability density function under the two assumptions of the existence and non-existence of multi-channel constant mode signal is a probability density function constructed based on single-channel weighted data; The single-channel data is shown in the following formula:

[0009] in, For weight vectors, For multi-channel data reception matrix, For the complex amplitude of the multi-channel constant mode, For the waveform vector of multi-channel constant mode, For the first The direction of arrival of spatial interference. For the first Spatial interference in direction The guide vector at that location, For the first One spatial interference vector, The matrix is ​​an additive white Gaussian noise matrix. weight vector The conjugate transpose of . for transpose, for transpose; The probability density function under the assumption of the existence of multi-channel constant-mode signals is shown in the following equation:

[0010] in, For weight vectors, weight vector The conjugate transpose of . For multi-channel data reception matrix, For the complex amplitude of the multi-channel constant mode, For the waveform vector of multi-channel constant mode, For the first The direction of arrival of spatial interference. For the first Spatial interference in direction The guide vector at that location, For the first One spatial interference vector, For the direction of arrival of the multi-channel constant mode, For multi-channel constant mode in direction The guide vector at that location, For the number of snapshots, The variance is the additive noise.

[0011] Furthermore, an optimization model is constructed based on a generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference. This includes removing prior constraints on interference direction and interference power to obtain alternative forms of the probability density function under the two assumptions of the presence and absence of multi-channel constant mode signals. An alternative form of the probability density function under the assumption of the existence of a multi-channel constant-mode signal is shown below:

[0012] An alternative form of the probability density function under the assumption that multi-channel constant-mode signals do not exist is shown below:

[0013] in, Represents the weight vector Noise variance Signal complex amplitude and signal waveform Perform a maximization solution. Represents the weight vector The solution is to maximize the noise variance. For weight vectors, weight vector The conjugate transpose of . The variance of additive noise, For the complex amplitude of the multi-channel constant mode, For the waveform vector of multi-channel constant mode, For multi-channel data reception matrix, For the number of snapshots, For the direction of arrival of the multi-channel constant mode, For multi-channel constant mode in direction The guide vector at that location.

[0014] Furthermore, the optimization model for the generalized likelihood ratio test detector, which does not require prior information on spatial electromagnetic interference, is constructed by using the generalized likelihood ratio test statistic based on constant modulus constraints, taking the logarithm of the likelihood ratio and ignoring the constant term, and constructing a non-convex optimization objective model corresponding to the generalized likelihood ratio test. The optimization model based on the generalized likelihood ratio test detector that does not require prior information about spatial electromagnetic interference is shown in the following equation:

[0015] in, For the generalized likelihood ratio test statistic of constant modulus signals, Assuming the existence of multi-channel constant-mode signals, The assumption that multi-channel constant-mode signals do not exist; The detection threshold corresponding to the constant modulus signal; For constant modulus constraints; An alternative form of the probability density function is given under the assumption of the existence of multi-channel constant-mode signals; An alternative form of the probability density function is given under the assumption that multi-channel constant-mode signals do not exist.

[0016] Furthermore, a hybrid optimization framework combining alternating optimization and alternating direction multiplier method is adopted, including... Least square solution Substituting the probability density function under the assumptions of the presence and absence of the multi-channel constant-mode signal, a simplified form is obtained. Based on this simplified form, the unknown parameters are updated iteratively through alternating optimization, as shown in the following equation:

[0017]

[0018]

[0019]

[0020] in, Indicates the number of iterations for alternating optimization; In the process of solving the problem using the alternating direction multiplier method, auxiliary variables are introduced to decouple the optimization problem from the Lagrange multipliers; The decoupled optimization problem is transformed into a simplified optimization problem that can be solved iteratively through logarithmic transformation.

[0021] Furthermore, the detection parameters are the optimal detection parameters obtained through multiple iterations and convergence.

[0022] Furthermore, completing the multi-channel constant mode detection includes detecting the multi-channel constant mode signals based on the detection parameters.

[0023] Secondly, the present invention provides a multi-channel constant-mode signal detection system that does not require prior electromagnetic interference, comprising: The data modeling and dimensionality reduction module is used to model the multi-shot data received by the multi-channel antenna array to obtain high-dimensional multi-channel data. The high-dimensional multi-channel data is then projected onto a single channel to perform dimensionality reduction processing, resulting in single-channel weighted data. The detector construction module is used to construct a generalized likelihood ratio test detector that does not require prior information about spatial electromagnetic interference, based on single-channel weighted data and the probability density function under the two assumptions of the presence and absence of multi-channel constant mode signals. The optimization model solving module is used to construct an optimization model for a generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference. It adopts a hybrid optimization framework that combines alternating optimization and alternating direction multiplier method to solve the optimization model, obtain the detection parameters, and complete multi-channel constant mode detection.

[0024] Thirdly, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a method for detecting multi-channel constant-mode signals without prior electromagnetic interference.

[0025] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor, and a method for detecting multi-channel constant-mode signals without prior electromagnetic interference.

[0026] Compared with the prior art, the present invention has the following beneficial technical effects: This invention proposes a multi-channel constant mode signal detection method that does not require prior electromagnetic interference. It fully leverages the combined design advantages of weighted vector projection dimensionality reduction and generalized likelihood ratio test (GLRT), breaking the dependence of existing multi-channel constant mode detection methods on prior electromagnetic interference information and secondary data. By establishing an optimization objective model that does not require prior electromagnetic interference, the method employs alternating optimization (AO) and alternating direction multiplier method (ADMM) to decouple non-convex constraints from the objective function. For the complex constraint solving problem involved in the optimization process, it transforms it into a simplified problem that can be efficiently iterated, obtaining the optimal detection parameters through multiple iterations. In practical applications, this method can form a deep notch in the direction of unknown electromagnetic interference, effectively suppressing multi-source electromagnetic interference and avoiding performance degradation of multi-channel constant mode signal detection due to missing interference information. Simultaneously, it significantly reduces system computation and storage overhead, maintaining excellent detection probability and signal parameter estimation accuracy even under low signal-to-noise ratio, low signal-to-interference ratio, and limited snapshot conditions. This significantly improves the reliability and real-time performance of multi-channel constant mode detection in fields such as electronic detection and electromagnetic spectrum monitoring, effectively avoiding problems such as weak multi-channel constant mode signals being overwhelmed by interference. This invention is a low-complexity and highly reliable multi-channel constant mode detection method that does not require prior information about electromagnetic interference, does not rely on secondary data, and achieves efficient suppression of multi-source electromagnetic interference and accurate detection of multi-channel constant mode signals by optimizing the target construction and solution strategy, thus meeting the real-time and reliability requirements of fields such as electronic detection and electromagnetic spectrum monitoring. Attached Figure Description

[0027] The accompanying drawings described herein are for illustrative purposes only and are not intended to limit the scope of the invention in any way. Furthermore, the shapes and proportions of the components in the drawings are merely schematic to aid in understanding the invention and do not specifically limit the shapes and proportions of the components. In the drawings: Figure 1 This is a flowchart of the multi-channel constant mode signal detection method of the present invention that does not require prior electromagnetic interference.

[0028] Figure 2 This is a structural diagram of the multi-channel constant mode signal detection system of the present invention that does not require prior electromagnetic interference.

[0029] Figure 3 This is an electronic device diagram of the multi-channel constant mode signal detection method of the present invention that does not require prior electromagnetic interference.

[0030] Figure 4 The NMSE values ​​for convergence under different SNR values ​​in this embodiment of the invention are shown.

[0031] Figure 5 This is the beam pattern when the input signal-to-noise ratio is 10dB in an embodiment of the present invention.

[0032] Figure 6These represent the detection probabilities under different SNR values ​​in this embodiment of the invention.

[0033] Figure 7 The detection probabilities are for different SIR values ​​in the embodiments of the present invention.

[0034] Figure 8 This represents the detection probability for different snapshot numbers in embodiments of the present invention. Detailed Implementation

[0035] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0036] Example 1 See Figure 1 A method for detecting multi-channel constant-mode signals without prior electromagnetic interference includes the following steps: Modeling is performed on the multi-shot data received by the multi-channel antenna array to obtain high-dimensional multi-channel data. Weight vectors are introduced to project the high-dimensional multi-channel data onto a single channel for dimensionality reduction, resulting in single-channel weighted data. Based on the single-channel weighted data, a generalized likelihood ratio test detector (GRP) without prior information on spatial electromagnetic interference is constructed according to the probability density functions under the assumptions of signal presence and absence. An optimization model is built based on the GRP test detector without prior information on spatial electromagnetic interference. A hybrid optimization framework combining alternating optimization and alternating direction multiplier method is used to solve the optimization model, obtain the detection parameters, and complete the multi-channel constant mode detection.

[0037] This invention employs a multi-channel constant mode detection method based on generalized likelihood ratio test (GLRT) and alternating optimization (AO). By constructing an optimization objective function that requires no prior electromagnetic interference, and introducing weight vectors to achieve dimensionality reduction of multi-channel data, the detector parameters are solved using alternating optimization and alternating direction multiplier method (ADMM). Ultimately, highly reliable multi-channel constant mode detection is achieved. The specific steps are as follows: Step 1: Multi-channel data modeling and dimensionality reduction have The antenna array of each channel or sensor receives A multi-channel data matrix for quick snapshots It can be represented as: (1) in It is the signal of interest, that is, the waveform vector of the multi-channel constant mode, and its corresponding direction of arrival. The guide vector at that location is The amplitude is ; and the first spatial interference vectors From direction The interference is received at the location, and these disturbances are modeled as independent, identically distributed, zero-mean, circularly symmetric white Gaussian vectors. Furthermore, It is an additive white Gaussian noise matrix whose elements follow a complex Gaussian distribution. They are independent and identically distributed.

[0038] Existing methods assume the existence of the signal. The probability density function (PDF) of the received data matrix is ​​defined as follows: or in, Add noise covariance matrix to spatial interference. Indicates the first The power of the interference. Then, by utilizing the assumption that the signal does not exist... By applying the probability density function, the corresponding generalized likelihood ratio test (GLRT) detector can be obtained. This detector requires a set of directions of arrival. With interference power set Prior knowledge or targetless auxiliary data used to estimate the interference plus noise covariance matrix are essential. However, this information or data may be unavailable in real-world scenarios, leading to inefficient or even completely ineffective detection schemes. Furthermore, the high-dimensional structure of multi-channel data implies high computational costs. Therefore, this invention proposes a novel multi-channel constant-mode signal detection method that does not require prior knowledge of electromagnetic interference and is independent of high-dimensional structures.

[0039] To address the dimensionality issue, this invention draws upon the weighting concept widely used in beamformers, namely, introducing a weight vector. , will the original A multi-channel data matrix is ​​projected onto a single channel. Furthermore, combined with... This constraint can be expressed as: (2) Under the Gaussian noise assumption Follows a complex Gaussian distribution Therefore, Suppose the probability density function of equation (2) is: (3) Step 2: Construction of a GLRT detector without prior electromagnetic interference First, after the weighting process in equation (2), assume... and The probability density functions are as follows: and For the weighted signal detection in equation (2), the generalized likelihood ratio test (GLRT) can be expressed as: (4) in To detect threshold, and They represent the assumptions respectively. and Unknown parameters.

[0040] When the number of interferences and the direction of arrival are known, additional constraints can be introduced (i.e., for...). ,make To avoid its influence on the detection of the signal of interest, under this setting, after introducing the direction-of-arrival notch constraint, the numerator in equation (3) can be rewritten as: (5) Although equation (5) represents Under the assumption of complete interference suppression scenario The ideal form, but in real-world scenarios, such prior information is difficult to obtain. Although equation (5) is in the ideal form, it is difficult to obtain such prior information. This is obtained under the constraint, but here we mainly focus on the case after removing the constraint, thus obtaining the following alternative form. ,Right now: (6) Similarly, if there is no direction from which it originates. The incident signal (i.e. The following formula can be used as an alternative form. : (7) Equations (6) and (7) do not require prior knowledge of spatial interference; they are based on ideal scenarios where the direction of arrival is known. and Alternative forms; when applying the generalized likelihood ratio test (GLRT), these two equations can efficiently suppress interference corresponding to the direction of arrival in space. The specific proof is as follows: Using formula (2) (i.e.) Equation (6) can be rewritten as: (8) because: (9) First, analyze the molecules. Properties of the expansion norm. It will obtain the dominant energy term The sum of the intersection terms. The intersection terms can be expressed as: (10) in It is the noise received by the m-th antenna. yes The m-th element. When At that time, using the law of large numbers (LLN), we have: ,in It is the expectation operator. Due to the assumption of the interference sequence... Independent and with zero mean, additive noise It is zero-mean white noise and independent of interference, therefore we can obtain: (11) (12) therefore, In other words, the contribution of the cross term vanishes in the expectation and is therefore negligible in the asymptotic case.

[0041] On the other hand, from the perspective of principal component analysis, the dominant energy comes from the terms. The cross terms are low-energy perturbations and do not significantly affect the maximization process of equation (8). Therefore, these cross terms can be safely ignored during the optimization process.

[0042] In summary, the objective function in equation (8) can be approximated as: (13) Furthermore, under the condition of complete orthogonality (interference-interference orthogonality, interference-noise orthogonality), equation (8) is completely equivalent to equation (13).

[0043] Furthermore, under the assumptions of Gaussian white noise and the law of large numbers, Substituting into equation (13), we can obtain the simplified form: (14) Applying the logarithmic operator to equation (14) and ignoring irrelevant terms, we obtain: (15) In fact, the generalized likelihood ratio test only requires the maximum value of the probability density function under the null hypothesis and the alternative hypothesis. From the derivation of equation (3) to equation (15), it can be seen that equation (6) can be used as an effective alternative to equation (5), and its interference suppression capability will be explained below.

[0044] Clearly, maximizing equation (15) is equivalent to: (16) The optimal value can be obtained by adjusting the objective function in equation (16). The partial derivatives are set to 0, and the result is: (17) Substituting equation (17) into equation (16) and ignoring the constant term, we get: (18) It can be observed that maximizing equation (15) is equivalent to minimizing the weighted spatial disturbance response. This will force In summary, maximizing equation (15) can achieve interference suppression.

[0045] Therefore, it was removed , Equation (6) under this constraint can suppress interference and is a good alternative to equation (5). Therefore, equation (6) can be directly used as... This process requires no spatial interference information (or even the amount of interference).

[0046] Similar to the analysis of equation (6), equation (7) essentially also means hour Therefore, just like equation (6), spatial interference information is not required when using equation (7).

[0047] Therefore, for CM-SOI, a generalized likelihood ratio test (GLRT) detector that does not require prior electromagnetic interference is constructed, in the following form: (19) in, and The results are given by equations (6) and (7) respectively; This represents the detection threshold corresponding to CM-SOI, and its value is determined by the expected false alarm probability (denoted as ). )Sure.

[0048] Step 3: Solve the optimization model constructed by (19) using a hybrid optimization framework. Equation (19) is a CM-GLRT detector for multi-channel constant-mode signal detection without prior electromagnetic interference. It involves a complex mixture of exponential and fractional terms with multiple variables. Therefore, a hybrid optimization framework combining alternating optimization (AO) and alternating direction multiplier method (ADMM) is used to solve it. The likelihood function in equation (5) is about the unknown parameter. , , and The function. For CM-SOI, to simplify the expression, Least squares solution: (20) Substituting into equation (6) And combining the linear constraints in equation (19), we can obtain: (twenty one) Similarly, The question can be written as: (twenty two) To determine the generalized likelihood ratio corresponding to CM-SOI, it is calculated using AO. and The steps are as follows: A. Solve

[0049] The detailed steps for alternating optimization are as follows: (twenty three) (twenty four) (25) (26) superscript This represents the number of iterations (IN) for Alternating Optimization (AO). The iterative steps are repeated until convergence or the user-defined maximum number of AO iterations is reached. .

[0050] a: Solution to equation (25) After applying the logarithmic operator and ignoring the constant term, equation (25) can be transformed into: (27) in .

[0051] The objective function in equation (27) with respect to Taking the derivative and setting it to 0, we get: (28) b: Solution to equation (26) Applying the logarithmic operator again and ignoring the constant term, equation (26) can be rewritten as: (29) in .

[0052] Equation (29) is a challenging linear fractional (LF) optimization problem that is difficult to solve directly. To decouple the numerator and denominator, an auxiliary variable (SV) is introduced. At this point, equation (29) can be transformed into: (30) The augmented Lagrangian function of equation (30) is constructed as follows: (31) in It corresponds to a quadratic constraint. Lagrange multipliers, This is the penalty parameter. It is solved using the alternating direction multiplier method through the following iterative steps. , and : (32) (33) (34) in This represents the number of ADMM iterations for solving equation (30), and its maximum number of iterations is denoted as . Repeat equations (32) to (34) until convergence.

[0053] 1) Solution to equation (32): After ignoring the constant term, equation (32) can be simplified to a single-variable unconstrained optimization problem based on auxiliary variables: (35) in .

[0054] The objective function in equation (35) with respect to Taking the derivative and setting it to 0, we get: (36) This equation can be solved using the Cardano formula, by selecting the real non-negative root that minimizes the objective function value in equation (35). .

[0055] 2) Solution to equation (33): Rewrite equation (32) as: (37) in .

[0056] Introduction Equation (37) is transformed into: (38) The augmented Lagrangian function of equation (38) is constructed as follows: (39) Therefore, equation (38) can be solved by the following iterative alternation: (40) (41) (42) in and Let these represent the ADMM iteration number and penalty parameter for solving equation (38), respectively. The maximum number of iterations is defined as... The solutions to equations (40) and (41) are shown below.

[0057] i) Solution of equation (40): The objective function in equation (40) can be simplified to: (43) in Subsequently, the simplified objective function is discussed with respect to... Setting the derivative to 0, we get: (44) ii) Solution to equation (41): Equation (41) can be rewritten as: (45) in The augmented Lagrangian function of equation (45) for: (46) About Setting the derivative to 0, we get: (47) Therefore, we can conclude that: (48) Substituting equation (48) into We can obtain: (49) Combining equations (48) and (49), we can obtain .

[0058] B. Solve

[0059] The solution is now required: (50) After applying the logarithmic operator, equation (50) can be transformed into: (51) The solution to equation (51) can be found in equations (25)-(26).

[0060] The invention will be further explained and illustrated through the following experiments: The experimental setup was as follows: a uniform linear array with half-wavelength element spacing was used, and the number of elements M=15. Due to the false alarm probability... Set as Target signal The direction is set to The directions of arrival of the two interference waves are respectively and Transmitted waveform With complex interference Randomized; a circularly symmetric complex Gaussian random process with additive noise of zero mean, the noise variance at each antenna is... The comparison method is based on a loading factor of [missing information]. Capon beamformer (LCAP), linearly constrained minimum variance beamformer (LCMV), minimum variance distortionless response beamformer (MVDR), generalized likelihood ratio iterative detector (GLRI), constant mode beamformer (CMA), feature space-based beamformer (EIG), CVCNN-LR, CVCNN-R, WEIS-S1, WEIS-2.

[0061] Experiment 1: Setting the Number of Snapshots N =300, the NMSE for convergence at different SNRs was calculated, and the results are as follows. Figure 4 As shown, the method proposed in this invention has the highest accuracy when the interference is known, and also has good performance when the interference is unknown. Figure 5 The beam pattern obtained by this invention when the input signal-to-noise ratio is 10dB is shown in the figure. Experiments have demonstrated that this beam pattern has strong interference suppression capabilities, that is, it can form a deep groove at the unknown ID. exist The value was 157.33 dB. exist The value was 182.93dB, demonstrating balanced interference suppression performance.

[0062] Experiment 2: Setting the Number of Snapshots N =300, calculate the detection probability under different SNR and different SIR. Figure 6 Set the signal-to-interference ratio (SIR) to 0 dB, and the signal-to-noise ratio (SNR) from... The change from 15dB to 5dB shows that the method proposed in this invention outperforms other methods at medium and high signal-to-noise ratios. Figure 7 Set the signal-to-noise ratio (SNR) = 5dB, SIR from The change from 15dB to 5dB shows that the performance of the method proposed in this invention is consistently better than that of other methods.

[0063] Experiment 3 sets SNR= Given a SIR of 5dB, calculate the detection probability for different numbers of snapshots. Figure 8 By varying the number of snapshots from 60 to 300, it can be seen that the performance of the method proposed in this invention is consistently superior to other methods.

[0064] Example 2 See Figure 2 A multi-channel constant-mode signal detection system that does not require prior electromagnetic interference includes: The data modeling and dimensionality reduction module is used to model the multi-shot data received by the multi-channel antenna array to obtain high-dimensional multi-channel data. The high-dimensional multi-channel data is then projected onto a single channel to perform dimensionality reduction processing, resulting in single-channel weighted data. The detector construction module is used to construct a generalized likelihood ratio test detector that does not require prior information about spatial electromagnetic interference, based on single-channel weighted data and the probability density function under the two assumptions of the presence and absence of multi-channel constant mode signals. The optimization model solving module is used to construct an optimization model for a generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference. It adopts a hybrid optimization framework that combines alternating optimization and alternating direction multiplier method to solve the optimization model, obtain the detection parameters, and complete multi-channel constant mode detection.

[0065] Example 3 See Figure 3 An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the aforementioned method for detecting multi-channel constant-mode signals without prior electromagnetic interference.

[0066] Example 4 A computer-readable storage medium storing a computer program, which, when executed by a processor, describes a method for detecting multi-channel constant-mode signals without prior electromagnetic interference.

[0067] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, read-only optical discs, optical storage, etc.) containing computer-usable program code.

[0068] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0069] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0070] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for detecting multi-channel constant-mode signals without prior knowledge of electromagnetic interference, characterized in that, Includes the following steps: Model the multi-shot data received by the multi-channel antenna array to obtain high-dimensional multi-channel data. Introduce a weight vector to project the high-dimensional multi-channel data onto a single channel for data dimensionality reduction processing to obtain single-channel weighted data. Based on the single-channel weighted data, a generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference is constructed according to the probability density function under the two assumptions of the presence and absence of multi-channel constant mode signals. An optimization model is constructed based on the generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference. A hybrid optimization framework combining alternating optimization and alternating direction multiplier method is adopted to solve the optimization model, obtain the detection parameters, and complete the multi-channel constant mode detection.

2. The method for detecting multi-channel constant-mode signals without prior electromagnetic interference as described in claim 1, characterized in that, The modeling of multi-shot data received by the multi-channel antenna array includes joint modeling of multi-channel constant mode signals, spatial interference, and additive noise; The high-dimensional multi-channel data is shown in the following formula: in, For multi-channel data reception matrix, This refers to the number of channels in the antenna array or the number of sensors. For the number of snapshots, For the complex amplitude of the multi-channel constant mode, For the direction of arrival of the multi-channel constant mode, For multi-channel constant mode in direction The guide vector at that location, For the waveform vector of multi-channel constant mode, The amount of spatial interference, For the first The direction of arrival of spatial interference. For the first Spatial interference in direction The guide vector at that location, For the first A spatial interference vector, The matrix is ​​an additive white Gaussian noise matrix. The variance of additive noise; The introduction of weight vectors projects the high-dimensional multi-channel data onto a single channel for dimensionality reduction, resulting in single-channel weighted data, including applying constraints to the weight vectors. Projection dimensionality reduction is performed; the probability density function under the two assumptions of the existence of the multi-channel constant mode signal and the non-existence of the multi-channel constant mode signal is a probability density function constructed based on single-channel weighted data; The single-channel data is shown in the following formula: in, For weight vectors, For multi-channel data reception matrix, For the complex amplitude of the multi-channel constant mode, For the waveform vector of multi-channel constant mode, For the first The direction of arrival of spatial interference. For the first Spatial interference in direction The guide vector at that location, For the first A spatial interference vector, The matrix is ​​an additive white Gaussian noise matrix. weight vector The conjugate transpose of . for transpose, for transpose; The probability density function under the assumption that the multi-channel constant-mode signal exists is shown in the following equation: in, For weight vectors, weight vector The conjugate transpose of . For multi-channel data reception matrix, For the complex amplitude of the multi-channel constant mode, For the waveform vector of multi-channel constant mode, For the first The direction of arrival of spatial interference. For the first Spatial interference in direction The guide vector at that location, For the first A spatial interference vector, For the direction of arrival of the multi-channel constant mode, For multi-channel constant mode in direction The guide vector at that location, For the number of snapshots, The variance is the additive noise.

3. The method for detecting multi-channel constant-mode signals without prior electromagnetic interference as described in claim 2, characterized in that, The optimization model for the generalized likelihood ratio test detector based on the absence of prior information on spatial electromagnetic interference includes removing prior constraints on interference direction and interference power to obtain alternative forms of the probability density function under the two assumptions of the presence and absence of multi-channel constant mode signals. An alternative form of the probability density function under the assumption of the existence of the multi-channel constant-mode signal is shown below: The alternative form of the probability density function under the assumption that the multi-channel constant-mode signal does not exist is shown in the following equation: in, Represents the weight vector Noise variance Signal complex amplitude and signal waveform Perform a maximization solution. Represents the weight vector The solution is to maximize the noise variance. For weight vectors, weight vector The conjugate transpose of . The variance of additive noise, For the complex amplitude of the multi-channel constant mode, For the waveform vector of multi-channel constant mode, For multi-channel data reception matrix, For the number of snapshots, For the direction of arrival of the multi-channel constant mode, For multi-channel constant mode in direction The guide vector at that location.

4. The method for detecting multi-channel constant-mode signals without prior electromagnetic interference as described in claim 3, characterized in that, The optimization model for the generalized likelihood ratio test detector based on the prior information that does not require spatial electromagnetic interference includes a generalized likelihood ratio test statistic based on constant modulus constraint, taking the logarithm of the likelihood ratio and ignoring the constant term, and constructing a non-convex optimization objective model corresponding to the generalized likelihood ratio test. The optimization model constructed based on the generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference is shown in the following equation: in, For the generalized likelihood ratio test statistic of constant modulus signals, Assuming the existence of multi-channel constant-mode signals, The assumption that multi-channel constant-mode signals do not exist; The detection threshold corresponding to the constant modulus signal; For constant modulus constraints; An alternative form of the probability density function is given under the assumption of the existence of multi-channel constant-mode signals; An alternative form of the probability density function is given under the assumption that multi-channel constant-mode signals do not exist.

5. The method for detecting multi-channel constant-mode signals without prior electromagnetic interference as described in claim 4, characterized in that, The hybrid optimization framework, which combines alternating optimization with alternating direction multiplier method, includes: Least square solution Substituting the probability density function under the assumptions of the presence and absence of the multi-channel constant-mode signal, a simplified form is obtained. Based on this simplified form, the unknown parameters are updated iteratively through alternating optimization, as shown in the following equation: in, Indicates the number of iterations for alternating optimization; In the process of solving the problem using the alternating direction multiplier method, auxiliary variables are introduced to decouple the optimization problem from the Lagrange multipliers; The decoupled optimization problem is transformed into a simplified optimization problem that can be solved iteratively through logarithmic transformation.

6. The method for detecting multi-channel constant-mode signals without prior electromagnetic interference as described in claim 1, characterized in that, The detection parameters are the optimal detection parameters obtained through multiple iterations and convergence.

7. The method for detecting multi-channel constant-mode signals without prior electromagnetic interference as described in claim 1, characterized in that, The completion of multi-channel constant modulus detection includes detecting multi-channel constant modulus signals based on the detection parameters.

8. A multi-channel constant-mode signal detection system that does not require prior electromagnetic interference, characterized in that, include: The data modeling and dimensionality reduction module is used to model the multi-shot data received by the multi-channel antenna array to obtain high-dimensional multi-channel data. The high-dimensional multi-channel data is then projected onto a single channel using a weight vector to perform data dimensionality reduction processing, resulting in single-channel weighted data. The detector construction module is used to construct a generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference based on the single-channel weighted data and the probability density function under the two assumptions of the presence and absence of multi-channel constant mode signals. The optimization model solving module is used to construct an optimization model based on the generalized likelihood ratio test detector that does not require prior information on spatial electromagnetic interference. It adopts a hybrid optimization framework that combines alternating optimization and alternating direction multiplier method to solve the optimization model, obtain the detection parameters, and complete the multi-channel constant mode detection.

9. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the multi-channel constant-mode signal detection method without prior electromagnetic interference as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the multi-channel constant-mode signal detection method without prior electromagnetic interference as described in any one of claims 1-7.

Images