A method and system for detecting the number of broadband array signal sources

Abstract

This invention proposes a method and system for detecting the number of broadband array signal sources, belonging to the field of signal source detection technology. The method includes: Step S1, using a broadband array model composed of M broadband signal measurement sensors, constructing K sets of broadband array signal sample data by receiving the transmitted signals from K sets of broadband array signal sources, where M and K are both positive integers; Step S2, dividing each of the K sets of broadband array signal sample data into a training set and a test set according to a predetermined ratio, determining the corresponding feature matrix for the broadband array signal sample data in the training set by calculating the covariance matrix at different frequency points, resulting in K sets of feature matrices; Step S3, training a deep convolutional neural network model based on the K sets of feature matrices, and using the trained deep convolutional neural network model to detect the number of broadband array signal sources contained in the broadband array signal sample data in the test set.

Description

Technical Field

[0001] This invention belongs to the field of signal source detection technology, and in particular relates to a method and system for detecting the number of broadband array signal sources. Background Technology

[0002] Signal source count detection is a crucial problem in array signal processing. For most high-resolution array direction-finding algorithms, in practical applications, the number of signal sources must be estimated beforehand to accurately estimate their directions; otherwise, false alarms or missed alarms will occur in the array direction-finding results. Therefore, signal source count detection is a vital step in accurate array direction-finding. Early signal source count detection methods mainly employed hypothesis testing, which requires manually defining a detection threshold and is easily influenced by subjective factors. Subsequently, information theory methods, smoothed rank methods, matrix factorization methods, Gaussian circle methods, and regular correlation methods were proposed and studied. These methods each have their advantages and disadvantages and are all designed for narrowband signal sources. However, the model for broadband signal source count detection differs from that of narrowband signals. Therefore, methods for narrowband signal source count detection cannot be directly applied to broadband signal source count detection. Furthermore, traditional detection methods have low robustness, and their accuracy varies significantly across different application scenarios. Therefore, new methods are needed to address the problem of broadband array signal source count detection. Summary of the Invention

[0003] To address the problem of detecting the number of signal sources in a broadband array, this invention proposes a scheme for detecting the number of signal sources in a broadband array. This scheme can achieve automatic and intelligent detection of the number of signal sources in a broadband array by extracting the features of the broadband array signal and constructing and training a convolutional neural network model.

[0004] The first aspect of this invention discloses a method for detecting the number of broadband array signal sources. The method includes:

[0005] Step S1: Using a broadband array model composed of M broadband signal measurement sensors, construct K sets of broadband array signal sample data by receiving the transmitted signals from K sets of broadband array signal sources, where M and K are both positive integers;

[0006] Step S2: Divide each of the K groups of broadband array signal sample data into a training set and a test set according to a predetermined ratio. For the broadband array signal sample data in the training set, determine the corresponding feature matrix by calculating the covariance matrix at different frequency points, and obtain a total of K groups of feature matrices.

[0007] Step S3: Based on the K sets of feature matrices, train a deep convolutional neural network model. The trained deep convolutional neural network model is used to detect the number of broadband array signal sources contained in the broadband array signal sample data in the test set.

[0008] According to the method of the first aspect of the present invention, the number of signal sources in the K groups of broadband array signal sources are different from each other, and the number of signal sources in one group of broadband array signal sources is zero, while the number of signal sources in the other groups of broadband array signal sources is a positive integer arbitrarily selected in the interval [1, D], where D represents the maximum value of the number of signal sources in the broadband array signal sources.

[0009] According to the method of the first aspect of the present invention, step S1 specifically includes:

[0010] S1-1. Select one broadband signal measurement sensor located at the center position from the M broadband signal measurement sensors contained in the broadband array model as the target sensor.

[0011] S1-2. For each of the K-1 groups of broadband array signal sources (where the number of signal sources is not zero), perform the following procedure:

[0012] (i) Each broadband array signal source in the group sends the transmission signal to the location of the target sensor;

[0013] (ii) The transmitted signal is captured by the M broadband signal measurement sensors, and the M captured signals are integrated to form broadband array signal sample data of the current broadband array signal source;

[0014] S1-3. For broadband array signal sources with zero signal sources, the noise signals captured by the M broadband signal measurement sensors are directly integrated as broadband array signal sample data when the number of broadband array signal sources is zero.

[0015] According to the method of the first aspect of the present invention, each broadband array signal source has an incident direction vector relative to the orientation of the target sensor, and the incident direction vectors of each broadband array signal source are different from each other; in step S2, the obtained K sets of feature matrices contain a description of the incident direction vector.

[0016] A second aspect of this invention discloses a broadband array signal source quantity detection system. The system includes:

[0017] The first processing unit is configured to: construct K sets of broadband array signal sample data by receiving the transmitted signals of K sets of broadband array signal sources using a broadband array model composed of M broadband signal measurement sensors, where M and K are both positive integers;

[0018] The second processing unit is configured to: divide each of the K groups of broadband array signal sample data into a training set and a test set according to a predetermined ratio; and determine the corresponding feature matrix by calculating the covariance matrix at different frequency points for the broadband array signal sample data in the training set, thereby obtaining a total of K groups of feature matrices.

[0019] The third processing unit is configured to: train a deep convolutional neural network model based on the K sets of feature matrices, and the trained deep convolutional neural network model is used to detect the number of broadband array signal sources contained in the broadband array signal sample data in the test set.

[0020] According to the system of the second aspect of the present invention, the number of signal sources in the K groups of broadband array signal sources are different from each other, and the number of signal sources in one group of broadband array signal sources is zero, while the number of signal sources in the other groups of broadband array signal sources is a positive integer arbitrarily selected in the interval [1, D], where D represents the maximum value of the number of signal sources in the broadband array signal sources.

[0021] According to the system of the second aspect of the present invention, the first processing unit is specifically configured as follows:

[0022] One broadband signal measurement sensor located at the center position is selected from the M broadband signal measurement sensors contained in the broadband array model as the target sensor.

[0023] For each of the K-1 groups of broadband array signal sources (each with a non-zero number of signal sources), the following procedure is performed:

[0024] (i) Each broadband array signal source in the group sends the transmission signal to the location of the target sensor;

[0025] (ii) The transmitted signal is captured by the M broadband signal measurement sensors, and the M captured signals are integrated to form broadband array signal sample data of the current broadband array signal source;

[0026] For broadband array signal sources with zero signal sources, the noise signals captured by the M broadband signal measurement sensors are directly integrated as broadband array signal sample data when the number of broadband array signal sources is zero.

[0027] According to the system of the second aspect of the present invention, each broadband array signal source has an incident direction vector relative to the orientation of the target sensor, and the incident direction vectors of each broadband array signal source are different from each other; the obtained K sets of feature matrices contain a description of the incident direction vector.

[0028] A third aspect of this invention discloses an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the broadband array signal source quantity detection method described in any one of the first aspects of this disclosure.

[0029] A fourth aspect of this invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the broadband array signal source quantity detection method described in any one of the first aspects of this disclosure.

[0030] In summary, the technical solution provided by this invention utilizes the spatial characteristics of each frequency point of the broadband array signal and the second-order statistical properties of the covariance matrix to construct a robust feature matrix and a deep convolutional neural network model for different numbers of broadband array signal sources. By training the deep convolutional neural network model, an intelligent identification model for the number of broadband array signal sources is obtained, ultimately achieving efficient, accurate, and intelligent detection of the number of broadband array signal sources. Attached Figure Description

[0031] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0032] Figure 1 This is a flowchart of a broadband array signal source quantity detection method according to an embodiment of the present invention;

[0033] Figure 2 This is a layout diagram of a broadband array model and a broadband array signal source according to an embodiment of the present invention;

[0034] Figure 3 This is a diagram illustrating the feature matrix according to an embodiment of the present invention;

[0035] Figure 4 This is an illustration of the test results according to an embodiment of the present invention;

[0036] Figure 5 This is a structural diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0038] The first aspect of this invention discloses a method for detecting the number of broadband array signal sources. Figure 1 This is a flowchart of a broadband array signal source quantity detection method according to an embodiment of the present invention; as follows: Figure 1 As shown, the method includes:

[0039] Step S1: Using a broadband array model composed of M broadband signal measurement sensors, construct K sets of broadband array signal sample data by receiving the transmitted signals from K sets of broadband array signal sources, where M and K are both positive integers;

[0040] Step S2: Divide each of the K groups of broadband array signal sample data into a training set and a test set according to a predetermined ratio. For the broadband array signal sample data in the training set, determine the corresponding feature matrix by calculating the covariance matrix at different frequency points, and obtain a total of K groups of feature matrices.

[0041] Step S3: Based on the K sets of feature matrices, train a deep convolutional neural network model. The trained deep convolutional neural network model is used to detect the number of broadband array signal sources contained in the broadband array signal sample data in the test set.

[0042] In some embodiments, the number of signal sources in the K groups of broadband array signal sources are different from each other, and the number of signal sources in one group of broadband array signal sources is zero, while the number of signal sources in the other groups of broadband array signal sources is a positive integer arbitrarily selected in the interval [1, D], where D represents the maximum number of signal sources in the broadband array signal sources.

[0043] In some embodiments, step S1 specifically includes:

[0044] S1-1. Select one broadband signal measurement sensor located at the center position from the M broadband signal measurement sensors contained in the broadband array model as the target sensor.

[0045] S1-2. For each of the K-1 groups of broadband array signal sources (where the number of signal sources is not zero), perform the following procedure:

[0046] (i) Each broadband array signal source in the group sends the transmission signal to the location of the target sensor;

[0047] (ii) The transmitted signal is captured by the M broadband signal measurement sensors, and the M captured signals are integrated to form broadband array signal sample data of the current broadband array signal source;

[0048] S1-3. For broadband array signal sources with zero signal sources, the noise signals captured by the M broadband signal measurement sensors are directly integrated as broadband array signal sample data when the number of broadband array signal sources is zero.

[0049] In some embodiments, each broadband array signal source has an incident direction vector relative to the orientation of the target sensor, and the incident direction vectors of each broadband array signal source are different from each other; in step S2, the obtained K sets of feature matrices contain a description of the incident direction vector. Specific Implementation

[0051] (1) Construct broadband array signal sample data with different numbers of signal sources and label them. The labels containing the same number of signal sources are classified as the same category. Data can be constructed through simulation.

[0052] Figure 2 This is a layout diagram of a broadband array model and a broadband array signal source according to an embodiment of the present invention; as follows: Figure 2 As shown, this embodiment uses seven broadband radio frequency signal measurement sensors to form an L-shaped broadband array, with the radio frequency signal frequency range of 30MHz to 300MHz.

[0053] Let the coordinates of the measuring sensor be (x... m ,y m (m = 1, 2, 3, ..., M) represents m sensors, where M represents the maximum number of sensors. Assume there are Q signal sources S. q (f j (q = 1, 2, 3, ..., Q) signals are incident on the array, and the incident direction of the q-th signal source is... in Let θ represent the azimuth angle and θ represent the elevation angle. Then, the broadband array signal can be expressed by the following formula.

[0054]

[0055] In the formula: X(f j ) represents the data received by the array; S q (f j () represents the j-th frequency data of the q-th signal source; This represents the direction vector of the q-th signal source at the j-th frequency point; Indicates the incident direction of each radiation source, where θ represents the azimuth angle, θ represents the elevation angle; c represents the signal propagation speed, which in this embodiment is the speed of light; [N1(f j N2(f) j ); N3(f j );…;N7(f j )] represents noise.

[0056] By setting different numbers of signal sources and different signal-to-noise ratios, and randomly setting the signal incident direction, the above formula can be used to obtain independent sample data of broadband array signals with different numbers of signal sources, and with annotations.

[0057] In this embodiment, three types of broadband array signal sample data were constructed, containing 2, 1, and 0 signal sources respectively, and labeled with "2", "1", and "0". The sample size of each type is 1000. The orientation of the signal source is randomly generated to ensure the independence of the sample data, and the signal-to-noise ratio of the signal is randomly selected between 6dB and 12dB.

[0058] (2) The sample data obtained in step S1 is randomly divided into a training set and a test set according to a certain ratio. In this embodiment, the ratio is 9:1. The covariance matrix R(f) of each frequency point is calculated for each sample data. j )(j=1,2,3,…,J),f j R(f) represents the frequency of the j-th frequency point. j Let R(f) represent the covariance matrix of the j-th frequency point, and use R(f) to represent the covariance matrix of the j-th frequency point. j Construct the feature matrix T of the sample data and generate the corresponding annotation information. Figure 3 This is a diagram of the feature matrix according to an embodiment of the present invention.

[0059] In step S2, the covariance matrix R(f) of each frequency point of the sample data j The formulas for calculating the characteristic matrix T are shown below.

[0060]

[0061] In the formula: X(f j )=[X1(f j ); X2(f j ); X3(f j );…;X M (f j )] represents the frequency domain data of the broadband array signal at the j-th frequency point, and "H" represents the conjugate transpose.

[0062] The final number of training and test set samples is shown in Table 1.

[0063] category Number of training set samples Number of test set samples "2” 900 100 "1” 900 100 "0” 900 100

[0064] Table 1. Number of samples of each category in the training and test sets after random partitioning.

[0065] (3) Using the feature matrix T of the broadband array signal sample data obtained from the training set as the learning features, a deep convolutional neural network model is designed. The feature matrix T of the training set is input into the network model for training, and the output is an intelligent detection model for the number of broadband array signal sources. The designed multi-layer deep convolutional neural network structure includes an input layer, a multi-layer convolutional network, a Dropout layer, a fully connected layer, a Softmax layer, and an output layer. Each layer in the multi-layer convolutional network contains a convolutional layer, a batch normalization layer, an activation function, and a pooling layer. The kernel size of each convolutional layer in the multi-layer convolutional network is set to 3×3, and the number of kernels is set to 12, 24, 36, 48, and 48, respectively, with a stride of 1. The dropout rate of the Dropout layer is set to 0.15, and the learning rate during training is set to 2×10⁻⁴.

[0066] (4) The intelligent detection model for the number of broadband array signal sources output in step S4 is tested using the feature matrix of the test set sample data obtained in step S2 to verify its effectiveness and obtain the detection accuracy of the model. The confusion matrix of the test results is as follows: Figure 4 As shown, the average recognition accuracy of this model is approximately 100%.

[0067] A second aspect of this invention discloses a broadband array signal source quantity detection system. The system includes:

[0068] The first processing unit is configured to: construct K sets of broadband array signal sample data by receiving the transmitted signals of K sets of broadband array signal sources using a broadband array model composed of M broadband signal measurement sensors, where M and K are both positive integers;

[0069] The second processing unit is configured to: divide each of the K groups of broadband array signal sample data into a training set and a test set according to a predetermined ratio; and determine the corresponding feature matrix by calculating the covariance matrix at different frequency points for the broadband array signal sample data in the training set, thereby obtaining a total of K groups of feature matrices.

[0070] The third processing unit is configured to: train a deep convolutional neural network model based on the K sets of feature matrices, and the trained deep convolutional neural network model is used to detect the number of broadband array signal sources contained in the broadband array signal sample data in the test set.

[0071] According to the system of the second aspect of the present invention, the number of signal sources in the K groups of broadband array signal sources are different from each other, and the number of signal sources in one group of broadband array signal sources is zero, while the number of signal sources in the other groups of broadband array signal sources is a positive integer arbitrarily selected in the interval [1, D], where D represents the maximum value of the number of signal sources in the broadband array signal sources.

[0072] According to the system of the second aspect of the present invention, the first processing unit is specifically configured as follows:

[0073] One broadband signal measurement sensor located at the center position is selected from the M broadband signal measurement sensors contained in the broadband array model as the target sensor.

[0074] For each of the K-1 groups of broadband array signal sources (each with a non-zero number of signal sources), the following procedure is performed:

[0075] (i) Each broadband array signal source in the group sends the transmission signal to the location of the target sensor;

[0076] (ii) The transmitted signal is captured by the M broadband signal measurement sensors, and the M captured signals are integrated to form broadband array signal sample data of the current broadband array signal source;

[0077] For broadband array signal sources with zero signal sources, the noise signals captured by the M broadband signal measurement sensors are directly integrated as broadband array signal sample data when the number of broadband array signal sources is zero.

[0078] According to the system of the second aspect of the present invention, each broadband array signal source has an incident direction vector relative to the orientation of the target sensor, and the incident direction vectors of each broadband array signal source are different from each other; the obtained K sets of feature matrices contain a description of the incident direction vector.

[0079] A third aspect of this invention discloses an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the broadband array signal source quantity detection method described in any one of the first aspects of this disclosure.

[0080] Figure 5 This is a structural diagram of an electronic device according to an embodiment of the present invention, such as... Figure 5As shown, the electronic device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, Near Field Communication (NFC), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0081] Those skilled in the art will understand that Figure 5 The structure shown is merely a structural diagram of the part related to the technical solution of this disclosure and does not constitute a limitation on the electronic device to which the solution of this application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0082] A fourth aspect of this invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the broadband array signal source quantity detection method described in any one of the first aspects of this disclosure.

[0083] In summary, the technical solution provided by this invention utilizes the spatial characteristics of each frequency point of the broadband array signal and the second-order statistical properties of the covariance matrix to construct a robust feature matrix and a deep convolutional neural network model for different numbers of broadband array signal sources. By training the deep convolutional neural network model, an intelligent identification model for the number of broadband array signal sources is obtained, ultimately achieving efficient, accurate, and intelligent detection of the number of broadband array signal sources.

[0084] Specifically, by utilizing the covariance matrices of each frequency point of the broadband array signal, the covariance matrices of each frequency point are further fused to construct a feature matrix of the broadband array signal with different numbers of signal sources. At the same time, a deep convolutional neural network model with a multi-layer neural network structure is designed. The network model is trained through a large amount of simulation or experimental sample data containing different numbers of signal sources to generate an intelligent detection model for the number of broadband array signal sources. By utilizing the covariance fusion features with strong stability and high discriminativeness and the strong learning ability of the deep learning model, a good accuracy in detecting the number of broadband array signal sources can be obtained. This detection method is not affected by subjective factors, which improves the robustness of broadband array signal source number detection and realizes automatic intelligent detection.

[0085] Please note that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. The above embodiments only illustrate several implementation methods of this application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for detecting the number of broadband array signal sources, characterized in that, The method includes: Step S1: Using a broadband array model composed of M broadband signal measurement sensors, construct K sets of broadband array signal sample data by receiving the transmitted signals from K sets of broadband array signal sources, where M and K are both positive integers; Step S2: Divide each of the K groups of broadband array signal sample data into a training set and a test set according to a predetermined ratio. For the broadband array signal sample data in the training set, determine the corresponding feature matrix by calculating the covariance matrix at different frequency points, and obtain a total of K groups of feature matrices. Step S3: Based on the K sets of feature matrices, train a deep convolutional neural network model. The trained deep convolutional neural network model is used to detect the number of broadband array signal sources contained in the broadband array signal sample data in the test set. Among them, the number of signal sources in the K groups of broadband array signal sources are different from each other, and the number of signal sources in one group of broadband array signal sources is zero, while the number of signal sources in the other groups of broadband array signal sources is any positive integer selected in the interval [1, D], where D represents the maximum number of signal sources in the broadband array signal sources. Specifically, step S1 includes: S1-1. Select one broadband signal measurement sensor located at the center position from the M broadband signal measurement sensors contained in the broadband array model as the target sensor. S1-2. For each of the K-1 groups of broadband array signal sources (where the number of signal sources is not zero), perform the following procedure: (i) Each broadband array signal source in the group sends the transmission signal to the location of the target sensor; (ii) The transmitted signal is captured by the M broadband signal measurement sensors, and the M captured signals are integrated to form broadband array signal sample data of the current broadband array signal source; S1-3. For broadband array signal sources with zero signal sources, the noise signals captured by the M broadband signal measurement sensors are directly integrated as broadband array signal sample data when the number of broadband array signal sources is zero. Wherein, the coordinates of the measuring sensor are (x... m ,y m ), m = 1, 2, 3, ..., M, representing m sensors, where M represents the maximum number of sensors; Q signal sources S q (f j The signal is incident on the array, q = 1, 2, 3, ..., Q, and the incident direction of the q-th signal source is... Let θ represent the azimuth angle and θ represent the elevation angle. Then, the broadband array signal can be represented as: Where: X(f) j ) represents the data received by the array; S q (f j () represents the j-th frequency data of the q-th signal source; This represents the direction vector of the q-th signal source at the j-th frequency point; Indicates the incident direction of each radiation source; c represents the signal propagation speed; [N1(f j N2(f) j ); N3(f j );…;N7(f j )] represents noise; By setting different numbers of signal sources and different signal-to-noise ratios, and randomly setting the signal incident direction, broadband array signal sample data with annotations are obtained under different numbers of signal sources.

2. The method for detecting the number of broadband array signal sources according to claim 1, characterized in that, Each broadband array signal source has an incident direction vector relative to the orientation of the target sensor, and the incident direction vectors of each broadband array signal source are different from each other; in step S2, the K sets of feature matrices obtained contain a description of the incident direction vector.

3. A broadband array signal source quantity detection system, characterized in that, The system includes: The first processing unit is configured to: construct K sets of broadband array signal sample data by receiving the transmitted signals of K sets of broadband array signal sources using a broadband array model composed of M broadband signal measurement sensors, where M and K are both positive integers; The second processing unit is configured to: divide each of the K groups of broadband array signal sample data into a training set and a test set according to a predetermined ratio; and determine the corresponding feature matrix by calculating the covariance matrix at different frequency points for the broadband array signal sample data in the training set, thereby obtaining a total of K groups of feature matrices. The third processing unit is configured to: train a deep convolutional neural network model based on the K sets of feature matrices, and the trained deep convolutional neural network model is used to detect the number of broadband array signal sources contained in the broadband array signal sample data in the test set. Among them, the number of signal sources in the K groups of broadband array signal sources are different from each other, and the number of signal sources in one group of broadband array signal sources is zero, while the number of signal sources in the other groups of broadband array signal sources is any positive integer selected in the interval [1, D], where D represents the maximum number of signal sources in the broadband array signal sources. Specifically, the first processing unit is configured as follows: One broadband signal measurement sensor located at the center position is selected from the M broadband signal measurement sensors contained in the broadband array model as the target sensor. For each of the K-1 groups of broadband array signal sources (where the number of signal sources is not zero), the following procedure is performed: (i) Each broadband array signal source in the group sends the transmission signal to the location of the target sensor; (ii) The transmitted signal is captured by the M broadband signal measurement sensors, and the M captured signals are integrated to form broadband array signal sample data of the current broadband array signal source; For a broadband array signal source with zero signal sources, the noise signals captured by the M broadband signal measurement sensors are directly integrated as broadband array signal sample data when the number of broadband array signal sources is zero. Wherein, the coordinates of the measuring sensor are (x... m ,y m ), m = 1, 2, 3, ..., M, representing m sensors, where M represents the maximum number of sensors; Q signal sources S q (f j The signal is incident on the array, q = 1, 2, 3, ..., Q, and the incident direction of the q-th signal source is... Let θ represent the azimuth angle and θ represent the elevation angle. Then, the broadband array signal can be represented as: Where: X(f) j ) represents the data received by the array; S q (f j () represents the j-th frequency data of the q-th signal source; This represents the direction vector of the q-th signal source at the j-th frequency point; Indicates the incident direction of each radiation source; c represents the signal propagation speed; [N1(f j N2(f) j ); N3(f j );…;N7(f j )] represents noise; By setting different numbers of signal sources and different signal-to-noise ratios, and randomly setting the signal incident direction, broadband array signal sample data with annotations are obtained under different numbers of signal sources.

4. The broadband array signal source quantity detection system according to claim 3, characterized in that, Each broadband array signal source has an incident direction vector relative to the orientation of the target sensor, and the incident direction vectors of each broadband array signal source are different; the obtained K sets of feature matrices contain descriptions of the incident direction vectors.

5. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps in the broadband array signal source quantity detection method according to claim 1 or 2.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the broadband array signal source quantity detection method according to claim 1 or 2.

Images