Spatial interference suppression method and device based on support vector data description, and medium

Abstract

The invention provides an airspace interference suppression method and device based on support vector data description and a medium, and belongs to the technical field of airspace interference suppression. Processing the received signal by using a self-adaptive spatial filtering algorithm to obtain an output signal representation based on the self-adaptive spatial filtering algorithm; solving real number frequency domain representation; the target function is minimized under the constraint condition, a new constraint is obtained by setting a variable partial derivative as zero, a Lagrange multiplier is obtained by optimizing the target function, and the reciprocal of the Euclidean distance between a signal sample in the low-dimensional feature space and the center of the hypersphere is solved to represent a numerical solution; and determining an output signal of the optimal adaptive spatial filtering algorithm. According to the method, the SINR of the output signal is obviously improved, and the interference suppression performance is obviously enhanced. According to the method, the problem of fuzziness of spatial domain filtering algorithm convergence in a spatial domain high-dynamic scene is solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of spatial domain interference suppression, and particularly relates to a spatial domain interference suppression method based on support vector data description, a device and a medium. BACKGROUND

[0002] In a complex electromagnetic environment of mountainous forest land, the electromagnetic environment faced by communication equipment is complex and changeable. On the one hand, the complex terrain of mountainous forest land will cause multipath effect, resulting in that the signal-to-interference-to-noise ratio (SINR) of the received signal fluctuates sharply in a short time; on the other hand, the spatial position and number of mobile interference sources change rapidly, so that the direction of arrival (DOA) changes rapidly. These two factors cause the SINR of the input signal of the communication receiver and the DOA of the interference to change greatly, forming a high dynamic spatial domain scene, which leads to the mismatch of the DOA estimation of the traditional spatial domain interference suppression method, and significantly reduces the communication quality of the communication receiver. In view of the problem of interference suppression in the high dynamic spatial domain scene of mountainous forest land, it is of great significance to study the interference suppression method in the high dynamic spatial domain scene to protect the communication link security of the communication equipment.

[0003] With the rapid development of cognitive radio technology, spatial domain interference suppression algorithms are emerging, which can be mainly divided into two categories: reference adaptive spatial filtering based on steering vector and reference-free blind adaptive spatial filtering. In the high dynamic spatial domain scene, the steering vector of the reference adaptive spatial filtering has DOA estimation deviation, wavefront disturbance distortion or local coherent scattering error, which leads to a significant decrease in interference suppression performance.

[0004] In related technologies, the method of beam broadening is used to improve the interference suppression performance in the environment of fast time-varying DOA, but cannot solve the problem of self-cancellation. In related technologies, DOA estimation and adaptive spatial filtering are combined to realize spatial domain interference suppression in the scene of fast time-varying DOA and large dynamic SINR. However, this method cannot distinguish between interference and communication signals, and has the risk of self-cancellation.

[0005] Blind adaptive spatial filtering uses the characteristics of the received signal itself to realize beamforming, such as high-order statistics, cyclostationarity, constant modulus characteristics and power domain characteristics. However, fast time-varying DOA, large dynamic SINR and complex interference patterns cause the waveform characteristics and power domain characteristics of the received signal to change dramatically, and the adaptive generated beam is deviated, and even the communication signal is suppressed. In view of the above problems, by introducing the zero-trap broadening technology, the zero-trap of the adaptive array and the interference direction exist a certain degree of mismatch, but it is not suitable for the scene where the zero-trap error falls in the communication signal direction. The related technology based on feature matching and decision tree method, adaptive selection of the optimal spatial filtering algorithm and its parameters, realizes the adaptive of spatial filtering algorithm, but the accuracy is low in complex channel. Therefore, in the high dynamic scene of space, the main problems of the reference adaptive spatial filtering and the blind adaptive spatial filtering are the mismatch of the direction of arrival estimation, which makes the convergence of the spatial filtering algorithm ambiguous, and causes the output signal SINR of the adaptive array to decrease significantly. In the complex electromagnetic environment of mountain and forest, the direction of arrival and the signal-to-interference-and-noise ratio of the received signal of the communication equipment change dramatically, which significantly reduces the performance of the related technology of spatial interference suppression technology. SUMMARY

[0006] The application provides a spatial interference suppression method based on support vector data description, which solves the ambiguity problem of spatial filtering algorithm convergence in high dynamic scene.

[0007] The method comprises the following steps: S101: a plurality of receiving antennas are used to form a receiving array to receive signals, and the discrete received signal waveform and signal representation of the first array element are analyzed; S102: the received signal is processed by using an adaptive spatial filtering algorithm to obtain the output signal representation of the adaptive array, and the output signal SINR representation based on the adaptive spatial filtering algorithm is obtained; S103: the discrete Fourier transform of the discrete waveform amplitude value of the output signal of the adaptive array is performed, and the modulus value is calculated to obtain the real frequency domain representation input to the adaptive SVDD algorithm; S104: a training data set with full target sample feature representation is constructed, a target function is determined, the target function is minimized under the constraint condition, a Lagrange multiplier is introduced, a cost function is obtained, and a new constraint is obtained by setting the variable partial derivative to zero, and the target function with respect to the Lagrange multiplier is maximized; S105: a radial basis function is used as a kernel function, the Lagrange multiplier is obtained by optimizing the target function, the signal state of the signal sample is determined according to the Lagrange multiplier, the reciprocal of the Euclidean distance between the signal sample and the center of the hypersphere in the low-dimensional feature space is calculated, and the numerical solution of the output signal SINR of the adaptive spatial filtering algorithm is represented; S106: calculate the numerical solution of the output signal SINR of the plurality of spatial filtering algorithms, solve the highest numerical solution corresponding to the spatial filtering algorithm, and determine the output signal of the optimal adaptive spatial filtering algorithm.

[0008] It should be further explained that the receiving array in step S101 is a uniform linear array. The receiving array receives the signals at the same angle When observing the far-field target, the discrete received signal waveform of the first array element of the receiving array is:

[0009] In the formula, N is the sampling point number of the received signal, is the complex amplitude of the signal, and ; In the spatial angle there is a communication signal, and in the spatial angle there are Q interference signals, and the signal received by the receiving array is represented as

[0010] In the formula, the receiving steering vector is: ; represents the transposition operation, M is the number of antennas of the array, and is the working frequency of the communication system, is the speed of light, is the distance between the receiving array elements, is the complex amplitude of the communication signal, is the complex amplitude of the q interference signal, represents a Gaussian white noise with a mean of 0 and a variance of .

[0011] It should be further explained that the output signal of the adaptive array in step S102 is represented as

[0012] In the formula, is the weight vector, represents the conjugate transposition operation.

[0013] It should be further explained that the output signal of the adaptive array in step S102 is represented as .

[0014] It needs to be further explained that the discrete waveform amplitude value in step S103 is discretely Fourier transformed in the following manner: The discrete waveform amplitude value of the adaptive array output signal is represented as

[0015] In the formula, N is the number of sampling points of the received signal, and ; Through the discrete Fourier transform, the complex frequency domain representation of the signal is obtained

[0016] By taking the modulus value, the real frequency domain representation input to the adaptive SVDD algorithm is obtained .

[0017] It needs to be further explained that step S104 constructs a training data set with full target sample feature representation as follows: , wherein, represents the frequency domain real representation of the i-th signal sample, is the total number of signal samples, and the objective function can be represented as:

[0018] wherein, is the hypersphere radius, is the penalty factor; Step S104 obtains a new constraint as follows:

[0019] The objective function with respect to the Lagrange multiplier is maximized as follows:

[0020] wherein, K is the kernel matrix defined as

[0021] It needs to be further explained that for the signal sample , the signal state is

[0022] wherein, the distance is represented as

[0023] The radius is ​

[0024] wherein, is a support vector on the hypersphere, if then the signal sample falls inside the hypersphere, otherwise, the signal sample falls outside the hypersphere and is identified as an anomaly.

[0025] Step S106 represents the numerical solution of the output signal SINR of the adaptive spatial filtering algorithm by the following formula,

[0026] wherein, is the number of output signal samples.

[0027] Solving the spatial filtering algorithm corresponding to the highest output signal SINR numerical solution

[0028] wherein, is the total number of spatial filtering algorithms.

[0029] The output signal of the optimal adaptive spatial filtering algorithm is represented as

[0030] wherein, is the weight vector of the optimal spatial filtering algorithm, represents the conjugate transpose operation.

[0031] According to another embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method for spatial interference suppression based on support vector data description when executing the program.

[0032] According to still another embodiment of the present application, a storage medium is also provided, on which a computer program is stored, wherein the computer program is executable on a processor to implement the steps of the method for spatial interference suppression based on support vector data description.

[0033] As can be seen from the above technical solutions, the present application has the following advantages: The spatial domain interference suppression method based on support vector data description provided by the application adopts a receiving array composed of multiple receiving antennas to receive signals, and can utilize the spatial characteristics of the signals to enhance the signal receiving capability. The discrete received signal waveform of the first array element and the signal representation are analyzed. The adaptive spatial filtering algorithm can automatically adjust the filter parameters to adapt to the changes of the signal and interference environment, effectively improve the SINR of the output signal, enhance the detectability and quality of the signal, and thus highlight the useful signal in a complex interference environment. The discrete Fourier transform is performed on the discrete waveform amplitude value of the output signal of the adaptive array, and the modulus value is calculated, so as to convert the signal into a real number frequency domain representation suitable for the input of the SVDD algorithm, and make the characteristics of the signal more suitable for the subsequent classification and recognition algorithm processing. The training data set with the full target sample characteristic representation is constructed, which can comprehensively cover the characteristics of various target signals and interference signals. By determining the objective function and optimizing it, the objective function is minimized under the constraint condition, and mathematical methods such as Lagrange multiplier are introduced, so that a high-performance SVDD model can be effectively trained, which can accurately classify and recognize the signal. The radial basis function is applied as the kernel function, which can map the low-dimensional feature space to the high-dimensional space, and enhance the classification ability of the SVDD algorithm. The Lagrange multiplier is obtained by optimizing the objective function, and the signal state of the signal sample is determined according to the Lagrange multiplier, and the numerical solution of the reciprocal representation SINR of the Euclidean distance is calculated, which provides a new method for evaluating the signal quality from the signal state and spatial distance. The numerical solution of the output signal SINR of the multiple spatial domain filtering algorithms is calculated, and the algorithm corresponding to the highest numerical solution is selected as the output signal of the optimal adaptive spatial filtering algorithm, which can fully compare and utilize the advantages of different algorithms, and ensure that the final obtained signal is the signal with the optimal quality in the current environment, and further improve the reliability and system performance of the signal transmission. Thus, the performance and reliability of the entire communication system or signal processing system in the interference environment are improved, the intelligibility and accuracy of the signal transmission are ensured, and the bit error rate is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the present application, the drawings required to be used in the description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0035] Figure 1 The flowchart of the spatial domain interference suppression method based on support vector data description; Figure 2 The flowchart of the spatial domain interference suppression method based on spatial filtering and support vector data description; Figure 3Fig. 6 is a schematic diagram of SINR of output signals of the SFSD algorithm and the comparative algorithm under different input signal SINRs when the DOA of interference changes greatly; Figure 4 Fig. 7 is a schematic diagram of an electronic device. DETAILED DESCRIPTION

[0036] The method of the present application is directed to the problem of spatial interference suppression in a high dynamic spatial scene, and is based on a spatial filtering and support vector data description (SVDD) based spatial interference suppression method (SFSD). Through the SVDD mathematical tool, the communication signal characteristics are matched and analyzed, the Lagrange multiplier method is used to obtain the hypersphere distinguishing the communication signal and the interference, the statistical average value is calculated to give the numerical solution of the output signal SINR of multiple spatial filtering algorithms, and then the spatial filtering algorithm with the highest numerical solution is selected as the optimal solution of the joint spatial filtering algorithm, so that the convergence ambiguity of the spatial filtering algorithm in the high dynamic spatial scene is solved. Finally, the interference suppression performance of the algorithm of the present application and the typical spatial filtering algorithm in the high dynamic spatial scene is compared through simulation, and the advantages of the method of the present application in interference suppression performance and robustness are verified.

[0037] The support vector data description based spatial interference suppression method according to the present application will be described in detail below. In order to illustrate but not to limit, specific details such as specific system structures, techniques, etc. are presented in order to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details.

[0038] It should be understood that when used in the specification of the present application, the term "comprising" indicates the presence of the described features, integers, steps, operations, elements, and / or components, but does not exclude one or more other features, integers, steps, operations, elements, components, and / or sets thereof. The terms "comprising", "including", "having" and their variants mean "including but not limited to", unless otherwise specifically emphasized.

[0039] Statements in this application that refer to "one embodiment" or "some embodiments" or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrase "in one embodiment" or "in some embodiments" or "in other embodiments" or "in still other embodiments" or "in yet other embodiments" in various places in this specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments necessarily mutually exclusive of other embodiments. The various illustrative components, as well as other features and characteristics, described herein can be combined with one another and such description or claims may

[0040] Computer program code for carrying out operations of the present disclosure can be written in any combination of one or more programming languages, including object oriented programming languages such as Java, Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may

[0041] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort belong to the scope of protection of the present application.

[0042] Please refer to Figure 1 The flow chart of the spatial interference suppression method based on support vector data description in an embodiment is shown in FIG. 1, and the method comprises the following steps. S101: Receiving signals by using a plurality of receiving antennas to form a receiving array, and analyzing the discrete receiving signal waveform and signal representation of a first array element.

[0043] In some embodiments, a plurality of receiving antennas are selected to form a receiving array in a certain way for receiving signals from a far-field target. Optionally, the receiving array is a uniform linear array for the sake of analysis, and can be extended to other array geometries. For the first element in the receiving array, the discrete received signal waveform is analyzed to determine parameters contained therein, such as the number of sampling points, signal complex amplitude, etc. Meanwhile, the mathematical representation of the signal received by the receiving array is given in full, taking into account the presence of communication signals and interference signals in space, and the physical meaning of each variable is explicitly stated, such as the receiving steering vector, communication signal and interference signal complex amplitude, Gaussian white noise, etc.

[0044] It should be noted that the discrete received signal waveform of the first element is a digital representation of the received signal in time and amplitude. The signal received by the receiving array is the superimposed result of the communication signal, interference signal and noise after propagation in space and reception by the antenna.

[0045] S102: The received signal is processed using an adaptive spatial filtering algorithm to obtain the output signal representation of the adaptive array, and the output signal SINR representation based on the adaptive spatial filtering algorithm is derived.

[0046] In some embodiments, for the received signal, a suitable adaptive spatial filtering algorithm is selected to determine the weight vector in the algorithm. The received signal is processed by the algorithm to obtain the specific expression of the output signal of the adaptive array. On this basis, according to the expressions of the input signal and the output signal, the expression of the output signal SINR based on the adaptive spatial filtering algorithm is derived, and the association of each parameter in the expression with the signal and the algorithm is explicitly stated.

[0047] As can be seen, the adaptive spatial filtering algorithm automatically adjusts the weight vector according to the characteristics of the received signal, so that the communication signal in the output signal can pass through without distortion, while the interference signal is suppressed. The output signal SINR represents the ratio of the signal power to the interference plus noise power in the output signal, reflecting the suppression effect of the algorithm on the interference and the fidelity of the signal.

[0048] S103: The real number frequency domain representation input to the adaptive SVDD algorithm is obtained by performing discrete Fourier transform on the discrete waveform amplitude values of the output signal of the adaptive array and taking the modulus.

[0049] In some embodiments, the discrete waveform amplitude value of the adaptive array output signal is determined first, and the number of sampling points and other parameters are determined. The discrete Fourier transform is performed on the discrete waveform amplitude value to convert the time domain signal to the frequency domain signal, and the complex frequency domain representation of the signal is obtained. Then, by taking the modulus operation, the phase information in the complex frequency domain representation is removed, and the real frequency domain representation containing only amplitude information is obtained, which is adapted to the input requirements of the SVDD algorithm. In this way, the output signal is converted into a form suitable for processing by the SVDD algorithm.

[0050] S104: Construct a training data set with full target sample feature representation, determine a target function, minimize the target function under the constraint condition, introduce a Lagrange multiplier, obtain a cost function, and obtain a new constraint by setting the variable partial derivative to zero, and maximize the target function with respect to the Lagrange multiplier.

[0051] In some embodiments, signal samples with full target sample feature representation are collected to form a training data set. The target function is determined according to the training data set, which is used to describe the relationship between the signal sample and the hypersphere, involving parameters such as the radius of the hypersphere. The constraint condition is set, and the minimization operation is performed on the target function under the constraint condition. The Lagrange multiplier is introduced to convert the constraint optimization problem into an unconstrained cost function form. By setting the partial derivative of each variable in the cost function to zero, a new constraint condition is obtained, and the target function with respect to the Lagrange multiplier is maximized.

[0052] It can be seen that the training data set of the present embodiment provides learning samples for the SVDD algorithm, and the target function defines the construction rule of the hypersphere, which is used to distinguish between communication signals and interference signals. By using the constraint condition and the Lagrange multiplier method, the original constraint optimization problem is converted into a solvable unconstrained problem. The optimal solution is obtained by setting the partial derivative to zero, and the target function with respect to the Lagrange multiplier is maximized to determine the best parameters of the hypersphere.

[0053] S105: Apply the radial basis function as the kernel function, obtain the Lagrange multiplier by optimizing the target function, determine the signal state of the signal sample according to the Lagrange multiplier, and calculate the reciprocal of the Euclidean distance between the signal sample and the hypersphere center in the low-dimensional feature space to represent the numerical solution of the output signal SINR of the adaptive spatial filtering algorithm.

[0054] In some embodiments, the radial basis function is selected as the kernel function applied in the SVDD algorithm, and a width parameter of the RBF is determined. A value of a Lagrange multiplier is obtained by optimizing an objective function. It is determined whether a signal sample falls inside or outside the hypersphere according to the Lagrange multiplier, so as to determine a signal state of the signal sample. A Euclidean distance between the signal sample and a center of the hypersphere in a low-dimensional feature space is calculated, and an inverse of the distance is obtained as a numerical solution of an output signal SINR of the adaptive spatial filtering algorithm. The radial basis function and the Lagrange multiplier are used to determine the signal state, and the numerical solution of the output signal SINR is obtained by calculating the inverse of the Euclidean distance, thereby providing a numerical basis for comparing performances of different spatial filtering algorithms.

[0055] S106: Numerical solutions of output signals SINR of the plurality of spatial filtering algorithms are calculated, and a spatial filtering algorithm corresponding to a highest numerical solution is solved, so as to determine an output signal of the optimal adaptive spatial filtering algorithm.

[0056] In some embodiments, numerical solutions of output signals SINR of a plurality of different spatial filtering algorithms are calculated in the manner of step S105. The numerical solutions are compared, and a highest numerical solution is found. A spatial filtering algorithm corresponding to the highest numerical solution is determined as the optimal spatial filtering algorithm. An output signal of the optimal adaptive spatial filtering algorithm is determined according to parameters such as a weight vector of the optimal spatial filtering algorithm.

[0057] In this way, because different spatial filtering algorithms have different processing effects on signals, the optimal algorithm with the best performance can be found by calculating numerical solutions of output signals SINR of the algorithms and comparing the numerical solutions. Because a higher numerical solution of SINR indicates that the algorithm performs better in terms of interference suppression and signal preservation, the optimal algorithm is determined, and the final output signal is obtained according to parameters related to the optimal algorithm. The spatial interference suppression performance is effectively improved, and the output signal with higher quality is obtained.

[0058] On the basis of the above embodiments, in order to further improve the reliability of the spatial interference suppression method based on support vector data description provided by the above embodiments, the following is a specific implementation manner, and the spatial interference suppression method based on support vector data description includes the following steps: This embodiment uses the SVDD mathematical tool to calculate numerical solutions of output signals SINR of a plurality of spatial filtering algorithms, and solves an optimal solution of a joint spatial filtering algorithm, so as to realize the spatial interference suppression based on SVDD. First, array received signals are processed by a plurality of spatial filtering algorithms , and output signals are obtained. Subsequently, a hypersphere distinguishing communication signals and interference is constructed by the SVDD mathematical tool in combination with the Lagrange multiplier method, and a first Numerical solution of the input signal SINR of a spatial filtering algorithm Finally, the highest numerical solution is obtained. Spatial domain filtering algorithm Effectively improves interference suppression performance, such as Figure 2 The flowchart shown is a spatial interference suppression method based on spatial filtering and support vector data.

[0059] This embodiment considers a receiving array, consisting of... M It consists of a receiving antenna.

[0060] This embodiment assumes that the receiving array is a uniform linear array. This embodiment can be easily extended to other array geometries. The receiving array is at the same angle. Observe the far-field target. Additionally, assume the discrete received signal waveform of the first element of the receiving array is as follows: (1) In the formula, N The number of sampling points for the received signal. Let be the complex amplitude of the signal, and Assuming in spatial angle There is a communication signal, in space angle exist Q If there are interfering signals, the signal received by the receiving array can be expressed as: (2) In the formula, represents the receiving guidance vector, as shown below. (3) This indicates the transpose operation. M The number of antennas in the array, and The operating frequency of the communication system. At the speed of light, For the spacing between receiving array elements, The complex amplitude of the communication signal. For the first q The complex amplitude of the interference signal, This indicates that the mean is 0 and the variance is . Gaussian white noise.

[0061] At the receiver end, an adaptive spatial filtering algorithm is needed to process the received signal. Processing is performed to ensure distortion-free communication signal output and effectively suppress interference signals. The output signal of the adaptive array can be expressed as... (4) In the formula, For the weight vector, This indicates the conjugate transpose operation.

[0062] From the output signal and input signal expressions (2) and (4) of the adaptive array, it can be seen that the output signal SINR of the adaptive spatial filtering algorithm can be expressed as: (5) Assuming there is There are several spatial filtering algorithms, and the SINR of the output signal of each spatial filtering algorithm is calculated. And compare their sizes. Solve for the highest value. Corresponding spatial filtering algorithm (6) However, in practical applications, the communication signal component in the received signal and interference signal components and the corresponding guide vector and The SINR of the output signal is difficult to obtain, making it difficult to accurately calculate the analytical solution. To address the problem of calculating the SINR of the output signal, this embodiment uses the idea of ​​feature matching, employs SVDD to fit the characteristics of the communication signal, approximates the SINR of the output signal for each spatial filtering algorithm, and solves for the numerical solution of the SINR of the output signal.

[0063] This embodiment addresses the problem of numerically solving the SINR output signal of an adaptive spatial filtering algorithm. This embodiment uses the SVDD algorithm to approximately calculate the SINR output signal of each spatial filtering algorithm.

[0064] The discrete waveform amplitude value of the adaptive array output signal can be expressed as: (7) In the formula, N The number of sampling points for the received signal, and The complex frequency domain representation of the signal is obtained through the discrete Fourier transform. (8) By calculating the modulus, a real-number frequency domain representation of the input to the SVDD algorithm is obtained. (9) Consider a training dataset with feature representations of all target samples. , ,in Indicates the first The frequency domain real number representation of a signal sample Let be the total number of signal samples, then the objective function can be expressed as: (10) in It is the hypersphere radius. is a penalty factor. Equation (10) can be minimized under the constraint (11) Introducing Lagrange multipliers and , the cost function can be expressed as (12) Setting the partial derivative of each variable to zero, we get a new constraint, i.e. (13) Substituting (13) into (12), the objective function with respect to can be maximized, i.e. (14) where K is the kernel matrix defined as (15) Among existing kernel functions, the most widely used kernel function is the Radial Basis Function (RBF). RBF can be expressed as (16) where denotes the width of RBF.

[0065] Lagrange multipliers can be obtained by optimizing equation (14). When , the signal sample falls outside the hypersphere. On the other hand, when , the signal sample falls inside the hypersphere.

[0066] For a signal sample , its signal status can be determined as follows (17) where the distance can be expressed as (18) with a radius of (19) where is the support vector on the hypersphere. If , the signal sample falls inside the hypersphere. Otherwise, the signal sample falls outside the hypersphere and is identified as an anomaly.

[0067] The embodiment can obtain the reciprocal of the Euclidean distance between the signal sample in the low-dimensional feature space and the center of the hypersphere, to represent the numerical solution of the output signal SINR of the adaptive spatial filtering algorithm, that is (20) In the formula, is the number of output signal samples.

[0068] Solving the spatial filtering algorithm corresponding to the highest output signal SINR numerical solution (22) In the formula, is the total number of spatial filtering algorithms. At this point, the output signal of the optimal adaptive spatial filtering algorithm can be represented as (23) In the formula, is the weight vector of the optimal spatial filtering algorithm, represents the conjugate transpose operation.

[0069] In summary, the method of the embodiment is based on SVDD, and the numerical solutions of the output signals SINR of a plurality of spatial filtering algorithms are calculated, and the spatial filtering algorithm with the highest numerical solution is solved , which effectively improves the interference suppression performance.

[0070] The data set, evaluation standard and control algorithm of the algorithm of the embodiment will be given in combination with the technical content of the embodiment.

[0071] Since the SVDD of the embodiment adopts a hard boundary cost function, pure noise or interference signals cannot be included in the communication signal training data set. Therefore, it is necessary to reconstruct the data set. In the complex electromagnetic environment of mountainous forest land, BPSK and noise frequency modulation (NFM) are commonly used communication signals and interference, respectively. The modulation mode of the communication signal in the data set of the embodiment is taken as an example of BPSK, and the symbol rate is 30 kHz. The interference signal adopts NFM, and the signal bandwidth is set to 5 kHz / 100 kHz / 200 kHz, to simulate narrowband interference and wideband interference, respectively.

[0072] In the complex electromagnetic environment of mountainous forest land, high dynamic spatial scenes are common, such as active jamming by a flying jammer. In such scenes, the number of interference signals in the received signal of the communication receiver is unknown, and both the SINR and the DOA of the received signal have large dynamic changes. The embodiment divides the common scene into the following two states: There are two channel states of communication signals and single interference and communication signals and multiple interference, which simulate single interference source and multiple interference source, respectively, and are named in the following embodiments as By random switching Two channels, analog interference signal number unknown scene. Respectively in the input signal SINR or interference DOA large dynamic changes in the scene, analysis of the algorithm SFSD and the performance of the control algorithm anti-jamming.

[0073] Similarly, in the embodiment, Two channel state parameters are listed in Table 1. Among them, the SINR or DOA of the received signal is randomly selected in the large dynamic range of-20dB~20dB and-90°~90°, respectively.

[0074] Table 1 Parameters of transmission channel

[0075] The evaluation criteria of the adaptive spatial filtering algorithm in this embodiment, that is, the average value of the SINR of the output signal of the adaptive spatial filtering algorithm The average value of the SINR of the output signal of the adaptive spatial filtering algorithm The average value of the SINR of the output signal of the adaptive spatial filtering algorithm (24) In the formula, The output signal SINR of the adaptive array when the input signal is the first Frame data, the calculation formula is shown in formula (5), The number of samples of the output signal of the adaptive spatial filtering algorithm in the large dynamic scene.

[0076] The average value of the output signal SINR In dB can be generally expressed as: (25) The average value of the output signal SINR Used to evaluate the spatial interference suppression performance of the adaptive spatial filtering algorithm, The higher, the better the algorithm interference suppression performance, the stronger the ability to extract communication signals and suppress interference signals and background noise.

[0077] This embodiment uses a classic spatial filtering algorithm to compare its anti-interference performance with the proposed SFSD algorithm under different channel environments. During the simulation, the spatial filtering algorithm set used by SFSD consists of an anti-interference algorithm based on DOA and Minimum Variance Distortionless Response (MVDR) (DOA-MVDR) and PI arrays with varying numbers of antennas. This embodiment uses a classic PI array (PI-Array) and a constant modulus algorithm (CMA-Array), as well as an anti-interference algorithm based on DOA and CBF (DOA-CBF), to compare its anti-interference performance with the SFSD algorithm under different channel environments. For the PI-Array, the number of array elements is set to 4.

[0078] For CMA-Array, the number of array elements is set to 2. For DOA-CBF and DOA-MVDR, the signal incident angle is obtained from the DOA based on MVDR. For DOA-MVDR, the baseline signal is the input signal SINR. And interfere with DOA of The array receives signals under the channel. Table 2 shows the parameters of the algorithm in this embodiment and the comparison algorithm.

[0079] Table 2. Parameters of the algorithm in this embodiment and the comparison algorithm.

[0080] This embodiment analyzes the SFSD algorithm under different input signals SINR. and interference DOA The spatial interference suppression performance of the algorithm in this embodiment is evaluated and compared with the control algorithm to verify the superiority of the interference suppression performance of the algorithm in the high dynamic spatial scenario.

[0081] Consider a single interference source and multiple interference sources Under two randomly switching channel conditions, and using a test dataset, the interference suppression performance of the algorithm in this embodiment under different input signal SINRs was simulated and analyzed for large dynamic changes in interference DOA. The performance was then compared with a control algorithm. Figure 3 As shown in the figure, when the interference DOA changes dynamically, compared to the control algorithm, the algorithm in this embodiment has the highest output signal SINR under different input signal SINRs. The average SINR of the output signal under different output signal SINR values ​​is 11.66 dB. Compared with the best-performing control algorithm, the average SINR of the output signal in this embodiment is improved by 28.21 dB under different input signal SINR values.

[0082] It can be seen that the embodiment aims at the problem of spatial domain interference suppression in the scene of fast time-varying DOA, large dynamic SINR and unknown number of interferences, the method combines multiple spatial domain filtering algorithms, analyzes the similarity between the output signal of the spatial domain filtering algorithm and the communication signal based on feature matching through the SVDD mathematical tool, solves the hyper-sphere distinguishing the communication signal and the interference through the Lagrange multiplier method, gives the numerical solution of the output signal SINR by calculating the statistical average value, and finally selects the spatial domain filtering algorithm with the highest numerical solution as the optimal solution among the multiple spatial domain filtering algorithms.

[0083] The simulation results show that, in the high dynamic spatial domain scene, compared with the typical spatial domain filtering algorithm, the output signal SINR of the embodiment method is significantly improved, and the interference suppression performance is obviously enhanced. The embodiment method partially solves the ambiguity problem of the convergence of the spatial domain filtering algorithm in the high dynamic spatial domain scene, and provides a feasible solution for the spatial domain interference suppression in the complex electromagnetic scene of mountain and forest.

[0084] The spatial domain interference suppression method based on support vector data description provided by the application adopts the support vector data description mathematical tool to solve the numerical solution of the output signal SINR of multiple spatial domain filtering algorithms, realizes the optimal decision of multiple spatial domain filtering algorithms, and thus implements interference suppression. The simulation results of the application show that: in the scene of random switching of the number of interferences and high dynamic change of the interference wave arrival angle, the average value of the output signal SINR under different input signal SINR is 11.66dB. Compared with the classical spatial domain filtering algorithm, the application improves by 28.21dB. Therefore, the interference suppression algorithm can effectively improve the interference suppression performance in the complex electromagnetic scene of mountain and forest.

[0085] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiment of the application.

[0086] As Figure 4 shown, the application also provides an electronic device, which includes a display module 103, a memory 102, a processor 101 and a computer program stored in the memory and executable on the processor 101, and the processor 101 implements the steps of the spatial domain interference suppression method based on support vector data description when executing the program.

[0087] In embodiments of the application, the electronic device includes, but is not limited to, a laptop computer, a desktop computer, a workstation, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections, and relationships, and their functions, are shown as examples only and are not meant to limit implementations of the applications described and / or claimed in this document.

[0088] In embodiments of the application, the processor 101 can be implemented by using at least one of an application specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, a microcontroller, a microprocessor, an electronic unit designed to perform the functions described herein, and in some cases, the implementation can be implemented in a controller. For software implementation, the embodiments of processes or functions can be implemented with separate software modules, with the separate software modules allowing one or more functions or operations to be performed, and the software code can be implemented by a software application (or program) written in any suitable programming language. The software code can be stored in memory and executed by a controller.

[0089] The display module 103 is configured to display information input by a user or information provided to the user. The display module 103 can include a display panel, which can be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0090] The memory 102 can be configured to store software programs and various data. The memory 102 can include a high-speed random access memory, and can further include a non-volatile memory such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state memory device.

[0091] The application also provides a storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the steps of the method for spatial domain interference suppression based on support vector data description.

[0092] The storage medium can be any available medium that can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code means in the form of instructions or data structures and that can be accessed by a general-purpose or special-purpose computer, or a general-purpose or special-purpose processor. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or other

[0093] In this document, the terms "computer-readable medium" or "computer- readable media" is intended to include all medium that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or other

[0094] The foregoing description of the disclosed embodiments enables a person skilled in the art to implement or use the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Accordingly, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A spatial interference suppression method based on support vector data description, characterized in that, The methods include: S101: A receiving array composed of multiple receiving antennas is used to receive signals, and the discrete received signal waveform and signal representation of the first array element are analyzed. S102: The received signal is processed using an adaptive spatial filtering algorithm to obtain the output signal representation of the adaptive array, and the SINR representation of the output signal based on the adaptive spatial filtering algorithm is obtained. S103: By performing a discrete Fourier transform on the discrete waveform amplitude value of the adaptive array output signal and obtaining the modulus value, a real-number frequency domain representation of the SVDD algorithm input is obtained; S104: Construct a training dataset with full target sample feature representation, determine the objective function, minimize the objective function under constraints, introduce the Lagrange multiplier to obtain the cost function, and obtain new constraints by setting the partial derivatives of the variables to zero, and maximize the objective function with respect to the Lagrange multiplier; S105: Apply the radial basis function as the kernel function, obtain the Lagrange multiplier by optimizing the objective function, determine the signal state of the signal sample based on the Lagrange multiplier, and calculate the reciprocal of the Euclidean distance between the signal sample and the center of the hypersphere in the low-dimensional feature space to represent the numerical solution of the output signal SINR of the adaptive spatial filtering algorithm. S106: Calculate the numerical solution of the output signal SINR of various spatial filtering algorithms, find the spatial filtering algorithm corresponding to the highest numerical solution, and determine the output signal of the optimal adaptive spatial filtering algorithm.

2. The spatial interference suppression method based on support vector data description according to claim 1, characterized in that, In step S101, the receiving array is a uniform linear array; The receiving arrays are at the same angle Observing a far-field target, the discrete received signal waveform of the first element of the receiving array is as follows: In the formula, N The number of sampling points for the received signal. Let be the complex amplitude of the signal, and ; From a spatial perspective There is a communication signal, in space angle exist Q If there are interference signals, the signal received by the receiving array is represented as follows: The receiving guidance vector in the formula is: ； This indicates the transpose operation. M The number of antennas in the array, and The operating frequency of the communication system. At the speed of light, For the spacing between receiving array elements, The complex amplitude of the communication signal. For the first q The complex amplitude of the interference signal, This indicates that the mean is 0 and the variance is . Gaussian white noise.

3. The spatial interference suppression method based on support vector data description according to claim 1, characterized in that, The output signal of the adaptive array in step S102 is represented as follows: In the formula, For the weight vector, This indicates the conjugate transpose operation.

4. The spatial interference suppression method based on support vector data description according to claim 1, characterized in that, The output signal of the adaptive array in step S102 is represented as follows: 。 5. The spatial interference suppression method based on support vector data description according to claim 1, characterized in that, The specific methods for performing discrete Fourier transform on the discrete waveform amplitude values ​​in step S103 include: The discrete waveform amplitude value of the adaptive array output signal is expressed as: In the formula, N The number of sampling points for the received signal, and ; The complex frequency domain representation of the signal is obtained by using the discrete Fourier transform. By calculating the modulus, the real-number frequency domain representation of the input to the adapted SVDD algorithm is obtained as follows: 。 6. The spatial interference suppression method based on support vector data description according to claim 1, characterized in that, Step S104 constructs the training dataset with full target sample feature representation as follows: ， in, Indicates the first The frequency domain real number representation of a signal sample Let be the total number of signal samples, then the objective function can be expressed as: in, It is the hypersphere radius. It is a punishment factor; Step S104 yields the new constraint as follows The objective function that maximizes the Lagrange multiplier is in, K It is defined as a kernel matrix. 。 7. The spatial interference suppression method based on support vector data description according to claim 6, characterized in that, For signal samples The signal state is Among them, distance Represented as radius is in, These are the support vectors on the hypersphere, if If the signal sample falls within the hypersphere, then the signal sample falls outside the hypersphere and is identified as an anomaly.

8. The spatial interference suppression method based on support vector data description according to claim 1, characterized in that, Step S106 uses the following formula to express the numerical solution of the output signal SINR of the adaptive spatial filtering algorithm. In the formula, This represents the number of output signal samples. The spatial filtering algorithm corresponding to the numerical solution of the highest output signal SINR is as follows: In the formula, This represents the total number of spatial filtering algorithms. The output signal of the optimal adaptive spatial filtering algorithm is represented as follows: In the formula, This represents the weight vector of the optimal spatial filtering algorithm. This indicates the conjugate transpose operation.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the spatial interference suppression method based on support vector data description as described in any one of claims 1 to 8.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the spatial interference suppression method based on support vector data description as described in any one of claims 1 to 8.

Images