Near-field source localization method and system

Abstract

The application provides a near-field signal source positioning method and system. The method comprises the following steps: creating a fourth-order accumulation matrix containing only DOA information based on the DOA and the received signal model in a near-field environment; performing eigenvalue decomposition on the fourth-order accumulation matrix to obtain a strong signal subspace; incorporating the strong signal subspace into a noise subspace to form a first extended noise subspace; determining the DOA of a weak signal source by using the MUSIC algorithm and the first extended noise subspace; creating a covariance matrix of the received signal; performing eigenvalue decomposition on the covariance matrix and expanding the noise subspace to construct a two-dimensional spatial spectrum; inputting the DOA of the weak signal source into the two-dimensional spatial spectrum for one-dimensional search to obtain the positioning information of the weak signal source, wherein the positioning information comprises the distance of the weak signal source, so that the accuracy of positioning the near-field signal source in a scenario where strong and weak signals coexist can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of signal processing, in particular to a near-field signal source positioning method and system. BACKGROUND

[0002] Signal source positioning as a key technology in array signal processing has wide applications in sonar, radar, electronic countermeasures, biomedicine and other fields. According to the distance of the signal source to the receiving antenna array, it can be divided into far-field source positioning and near-field source positioning. The far-field signal source refers to the signal source located in the far-field region of the array, that is, r >> 2D 2 / λ, where r is the distance of the signal source to the reference element, D is the array aperture, and λ is the signal wavelength. When the distance of the signal source to the antenna array satisfies 0.62(D 3 / λ) 1 / 2 ≤r≤2D 2 / λ, the signal source is called a near-field signal source. The distance of the far-field signal source can be regarded as infinite, and the propagation in space can be regarded as a plane wave, so the signal source positioning can be realized by only estimating the DOA (Direction of Arrival). The wave front of the near-field signal source is regarded as a spherical wave, and the positioning of the signal source needs two parameters of DOA and the distance of the signal source to determine.

[0003] Strong and weak signals coexist, which often occurs in near-field signal source positioning. When there is a strong interference signal source near the weak signal source, the spatial spectrum peak of the weak signal is often very low by using the classical DOA estimation algorithm such as two-dimensional MUSIC and MUSIC method based on high-order statistics, which is easily covered by the spectrum peak of the adjacent strong interference signal to cause estimation deviation, or replaced by other pseudo-peak to cause estimation loss. Therefore, at present, in the scene where strong and weak signals coexist, the positioning of the near-field signal source has the problem of inaccurate positioning. SUMMARY

[0004] In view of this, the purpose of the embodiments of the present application is to provide a near-field signal source positioning method and system, which can improve the problem of inaccurate positioning of the near-field signal source in the scene where strong and weak signals coexist.

[0005] To achieve the above technical purpose, the technical scheme adopted by the present application is as follows:

[0006] In a first aspect, the embodiments of the present application provide a near-field signal source positioning method, which comprises:

[0007] S10, creating a receiving signal model based on a receiving array for receiving signals and a signal source with interference signals;

[0008] S20, creating a fourth-order accumulation quantity matrix containing only DOA information based on the DOA in the near-field environment and the receiving signal model;

[0009] S30, eigenvalue decomposition is performed on the fourth-order accumulation matrix to obtain a strong signal subspace;

[0010] S40, the strong signal subspace is incorporated into a noise subspace to form a first extended noise subspace;

[0011] S50, a DOA of a weak signal source is determined by using a MUSIC algorithm and the first extended noise subspace;

[0012] S60, a covariance matrix of a received signal is created;

[0013] S70, eigenvalue decomposition is performed on the covariance matrix, and a noise subspace is expanded to construct a two-dimensional spatial spectrum;

[0014] S80, the DOA of the weak signal source is input into the two-dimensional spatial spectrum for one-dimensional search to obtain positioning information of the weak signal source, the positioning information including a distance of the weak signal source.

[0015] With reference to the first aspect, in some optional embodiments, the step S10 comprises:

[0016] The total number of antennas of a receiving array and the total number of signal sources in an environment are acquired, the total number of antennas of the receiving array is represented as 2M+1, and the antennas of the receiving array are uniformly linearly distributed, are used to receive narrowband near-field signals, and the total number of signal sources is represented as P+J, wherein M is an integer greater than or equal to 0, P is the number of weak signal sources, J is the number of interference signal sources, and the output signal x m (t) is represented as:

[0017]

[0018] wherein s k (t) refers to the kth narrowband near-field signal source, t refers to a time, e refers to a natural constant, j refers to an imaginary unit, τ mk represents the time delay of the kth signal on the mth antenna relative to a reference array element, n m (t) represents additive Gaussian noise, and r m is defined as the distance from the signal source to the mth antenna, r k is the distance from the signal source to the reference array element, and based on the cosine theorem, the following equation is obtained:

[0019]

[0020] In the equation, r k represents the distance from the signal source to the reference array element, d refers to the antenna spacing, and θ k represents the DOA corresponding to the kth signal, and the transmission model of the near-field signal source is a spherical wave, and the time delay τmk is expressed as:

[0021]

[0022] where λ represents the wavelength of the signal source, and r m is substituted into τ mk , and we have:

[0023]

[0024] The formula (4) is expanded into a second-order Taylor expansion, τ mk is converted to:

[0025] τ mk ≈w k m+φ k m 2 (5)

[0026] where:

[0027]

[0028] Based on the output signal x m (t) and the τ mk obtained by the second-order Taylor expansion, a received signal model is created and expressed in matrix form:

[0029] x(t) = A(θ, r)s(t) + n(t) (8)

[0030] where:

[0031]

[0032] In combination with the first aspect, in some optional embodiments, the step S20 comprises:

[0033] The fourth-order cumulant is defined as:

[0034]

[0035] where y m (t), y p (t), y q * (t) represent the output signals of the mth, nth, pth, and qth antennas, respectively; * represents conjugation, represents the fourth-order cumulant of the kth narrowband near-field signal source s k (t), and has:

[0036]

[0037] In the formula, m, n, p, q ∈ [-M, M], m-n and p-q are located in [-2N, 2N], let m = -n and p = -q, based on the received signal model, the formula (10) corresponding to the fourth-order accumulation quantity is converted into:

[0038]

[0039] Based on the formula (12), the fourth-order accumulation quantity matrix is created by selecting specified elements, and the specified elements of the accumulation quantity matrix are given by the following formula:

[0040]

[0041] The fourth-order accumulation quantity matrix C1 is represented by the following formula:

[0042]

[0043] The fourth-order accumulation quantity matrix removes the distance information and only contains the DOA information of the signal, wherein H represents the matrix conjugate transpose, A1(θ) and C S The formula is as follows:

[0044] A1(θ) = [a1(θ1), …, a1(θ k ), …, a1(θ P+J )] (15)

[0045]

[0046] In the formula,

[0047] In combination with the first aspect, in some optional embodiments, the step S30 comprises:

[0048] Eigenvalue decomposition is performed on the fourth-order accumulation quantity matrix:

[0049]

[0050] In the formula, λ i is an eigenvalue, and λ1≥...≥λ J ≥λ J+1 ≥...λ J+P ≥λ 2M+1 , e i is an eigenvector corresponding to λ i , and the subspace of the total signal is represented as E JS = span(e1,...,e J , e J+1 ,...e J+P ), wherein the eigenvectors (e1,...,e J ) are determined by the strong interference signal, and (e J+1,...e J+P The total signal subspace is determined by the weak signal, and the subspace of the total signal includes the strong signal subspace.

[0051] In conjunction with the first aspect, in some alternative implementations, step S40 includes:

[0052] The strong signal subspace is incorporated into the noise subspace, and the noise subspace E is composed of eigenvectors corresponding to 2M+1-PJ eigenvalues. N =span(e J+P+1 ,...,e 2M+1 And construct the first extended noise subspace as follows:

[0053] E JN =span(e1,...,e J ,e J+P+1 ,...,e 2M+1 (18)

[0054] The following transformation is applied to the steering vector of the signal:

[0055]

[0056] Where b(θ) refers to the transformed guiding vector, I refers to the identity matrix, and a1(θ) refers to the guiding vector. For matrix E J The conjugate transpose of E J =span(e1,...,e J The strong signal subspace and the noise subspace are orthogonal. Let θ i For the weak signal angle, we get:

[0057]

[0058] The steering vector of the transformed signal is orthogonal to the first extended noise subspace.

[0059] In conjunction with the first aspect, in some alternative implementations, step S50 includes:

[0060] The DOA of a weak signal source can be determined using the following formula:

[0061]

[0062] In the formula, θ is the DOA of the weak signal source.

[0063] In conjunction with the first aspect, in some alternative implementations, step S60 includes:

[0064] Construct the covariance matrix R of the received signal x :

[0065] R x = E [x(t)x H (t)] = AR s A H + σ 2 I (22)

[0066] A denotes an array flow matrix, R s denotes a covariance matrix of the signal, and σ 2 is noise power.

[0067] With reference to the first aspect, in some optional embodiments, the step S70 comprises:

[0068] Eigenvalue decomposition is performed on the covariance matrix of the received signal, and an extended noise subspace is constructed, to construct a two-dimensional spatial spectrum as follows:

[0069]

[0070] wherein U JN is a second extended noise subspace, and d(θ, r) is a steering vector of the transformed signal.

[0071] In a second aspect, the embodiments of the present application further provide a near-field source positioning system, which comprises a processor and a memory coupled with each other, and the memory stores a computer program, when the computer program is executed by the processor, the near-field source positioning system executes the method described above.

[0072] The application with the technical scheme has the following advantages:

[0073] In the technical scheme provided in the present application, a fourth-order accumulation matrix containing only DOA information is constructed, and then eigenvalue decomposition is performed on the fourth-order accumulation matrix to obtain a strong signal subspace, and the strong signal subspace is incorporated into a noise subspace to form an extended noise subspace. Based on the extended noise subspace, the DOA of a weak signal source can be effectively estimated by using the MUSIC algorithm. Secondly, a covariance matrix of the received signal is constructed, and another extended noise subspace is constructed to obtain a two-dimensional spatial spectrum. The estimated DOA of the weak signal source is sequentially brought into the two-dimensional spatial spectrum to perform one-dimensional search to obtain the distance of the corresponding signal source. In this way, by incorporating the characteristic vector of the strong interference signal into the noise subspace, the spectral peak of the strong interference signal can be effectively eliminated, and then the positioning of the weak signal source is performed, thereby being conducive to improving the accuracy of the positioning of the weak signal source. BRIEF DESCRIPTION OF DRAWINGS

[0074] The application can be further illustrated by the non-limiting embodiments shown in the accompanying drawings. It should be understood that the following drawings only show certain embodiments of the application and therefore should not be considered as limiting the scope, and other related drawings can also be obtained from the drawings without creative labor.

[0075] Figure 1 The flowchart of the near-field source positioning method provided by the embodiments of the application is shown.

[0076] Figure 2 The spatial spectrum contrast simulation diagram of the conventional high-order accumulation quantity and the method provided by the embodiments of the application is shown.

[0077] Figure 3 The distance contrast simulation diagram of the conventional second-order statistical quantity method and the method provided by the embodiments of the application is shown.

[0078] Figure 4 The contrast diagram of the DOA root mean square error and the signal-to-noise ratio relationship of the conventional method and the method provided by the embodiments of the application is shown.

[0079] Figure 5 The contrast diagram of the distance root mean square error and the signal-to-noise ratio relationship of the conventional method and the method provided by the embodiments of the application is shown. DETAILED DESCRIPTION

[0080] The application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that in the drawings or description, similar or identical parts are denoted by the same reference numerals, and the implementation modes not shown or described in the drawings are known to those skilled in the art. In the description of the application, the terms "first", "second", etc. are only used for differentiation and cannot be understood as indicating or implying relative importance.

[0081] Please refer to Figure 1 The embodiments of the application provide a near-field source positioning method. The method can estimate DOA and distance by constructing fourth-order accumulation quantity matrix and covariance matrix, avoid high operation two-dimensional spectrum peak search, secondly, perform eigenvalue decomposition on the fourth-order accumulation quantity matrix and the covariance matrix, construct first and second extended noise subspaces, and effectively suppress strong interference by using the traditional MUSIC algorithm, so that the weak signal source can be accurately positioned in the near-field environment where strong and weak signals coexist. The near-field source positioning method can include the following steps:

[0082] S10, creating a received signal model based on a receiving array for receiving a signal and a signal source with an interference signal;

[0083] S20, creating a fourth-order accumulation matrix containing only DOA information based on the DOA in the near-field environment and the received signal model;

[0084] S30, performing eigenvalue decomposition on the fourth-order accumulation matrix to obtain a strong signal subspace;

[0085] S40, incorporating the strong signal subspace into a noise subspace to form a first extended noise subspace;

[0086] S50, determining the DOA of a weak signal source using the MUSIC algorithm and the first extended noise subspace;

[0087] S60, creating a covariance matrix of the received signal;

[0088] S70, performing eigenvalue decomposition on the covariance matrix and expanding the noise subspace to construct a two-dimensional spatial spectrum;

[0089] S80, inputting the DOA of the weak signal source into the two-dimensional spatial spectrum for one-dimensional search to obtain the positioning information of the weak signal source, the positioning information including the distance of the weak signal source.

[0090] The steps of the near-field source positioning method will be described in detail as follows:

[0091] In this embodiment, step S10 of creating a received signal model can include:

[0092] The total number of antennas of the receiving array and the total number of signal sources in the environment are obtained, the receiving array can include multiple antennas and can be used to receive signals emitted by signal sources, the signal source is a device that emits signals in the environment, the total number of antennas of the receiving array is represented as 2M+1, and the total number of signal sources is represented as P+J, where M is an integer greater than or equal to 0, P is the number of weak signal sources, J is the number of interference signal sources, and the output signal x m (t) is represented as:

[0093]

[0094] where s k (t) refers to the kth narrow-band near-field signal source, t refers to the time, e refers to a natural constant, j refers to an imaginary unit, τ mk represents the time delay of the kth signal relative to the reference element on the mth antenna, the reference element is determined in a conventional manner, n m (t) represents additive Gaussian noise, and r m is defined as the distance from the signal source to the mth antenna, and r kwhere r denotes the distance from the source to the reference element, d denotes the distance between the antennas, and θ denotes the DOA of the kth signal.

[0095]

[0096] where λ denotes the wavelength of the source, and r k denotes the distance from the source to the reference element, d denotes the distance between the antennas, and θ k denotes the DOA of the kth signal. The transmission model of the near-field source is a spherical wave, and the time delay τ mk denotes:

[0097]

[0098] where λ denotes the wavelength of the source, and r m is substituted into τ mk , and the following is obtained:

[0099]

[0100] The second-order Taylor expansion of formula (4) is performed, and τ mk can be approximated as:

[0101] τ mk ≈w k m+φ k m 2 (5)

[0102] where:

[0103]

[0104] As can be seen from formula (6) and formula (7), w k is only related to the DOA of the near-field source, and φ k is related to both the DOA of the near-field source and the distance of the near-field source, and therefore, based on the output signal x m (t) and formulas (5) to (7), a received signal model can be created and represented in matrix form as follows:

[0105] x(t) = A(θ, r)s(t) + n(t) (8)

[0106] where:

[0107]

[0108] In this embodiment, based on the DOA in the near-field environment, the received signal model, and the fourth-order accumulation matrix containing only DOA information is created in step S20, which can include:

[0109] The fourth-order accumulation is defined as:

[0110]

[0111] where y m (t), y p (t), y q * (t) represent the output signals of the mth, nth, pth, qth antennas respectively; * represents conjugate; represents the fourth-order accumulation of the kth narrowband near-field signal source s k (t), which is defined as:

[0112]

[0113] where m, n, p, q ∈ [-M, M], m-n and p-q are located in [-2N, 2N], in order to retain w k and remove φ k , it is necessary to ensure that (m 2 -n 2 )-(p 2 -q 2 ) = 0 and (m-n)-(p-q)≠0, for convenience, let m =-n and p =-q, p =-q, based on the received signal model, the fourth-order accumulation corresponding to formula (10) can be converted to:

[0114]

[0115] Based on formula (12), a fourth-order accumulation matrix is created by selecting specified elements, and a cumulant matrix is constructed, the specified elements of the cumulant matrix are given by:

[0116]

[0117] The specified elements can be flexibly selected according to actual conditions, and a fourth-order accumulation matrix is created by selecting appropriate elements, which removes the distance information and only contains the DOA information of the signal, the fourth-order accumulation matrix C1 is represented by:

[0118]

[0119] where H represents the matrix conjugate transpose, B(θ) and C S The formula is as follows:

[0120] A1(θ) = [a1(θ1), …, a1(θ k ), …, a1(θ P+J )] (15)

[0121]

[0122] where

[0123] In the embodiment, S30, eigenvalue decomposition is performed on the fourth-order cumulant matrix to obtain a strong signal subspace, which can include:

[0124] Eigenvalue decomposition is performed on the fourth-order cumulant matrix:

[0125]

[0126] In the formula, λ i is an eigenvalue, and λ1≥...≥λ J ≥λ J+1 ≥...λ J+P ≥λ 2M+1 , e i is an eigenvalue corresponding to λ i According to the subspace theory, it is known that the eigenvalue of the signal and the signal direction vector span the same subspace, and the subspace of the total signal (including the interference signal and the weak signal) is expressed as E JS =span(e1,...,e J ,e J+1 ,...e J+P ), wherein the eigenvalue (e1,...,e J ) is determined by the strong interference signal, and (e J+1 ,...e J+P ) is determined by the weak signal, the subspace of the total signal includes the strong signal subspace, and the noise subspace E N =span(e J+P+1 ,...,e 2M+1 ) is composed of 2M+1-P-J eigenvalues corresponding to the eigenvalue.

[0127] In the embodiment, S40, the strong signal subspace is included in the noise subspace to form a first extended noise subspace, which can include:

[0128] The strong signal subspace is included in the noise subspace, and the noise subspace E N =span(e J+P+1 ,...,e 2M+1 ) is composed of 2M+1-P-J eigenvalues corresponding to the eigenvalue, and the strong interference signal eigenvalue is included in the noise subspace, which can effectively eliminate the spectrum peak of the strong interference signal, and therefore the first extended noise subspace is constructed as:

[0129] E JN =span(e1,...,e J ,e J+P+1 ,,e 2M+1 ) (18)

[0130] Generally, the weak signal steering vector and the interference signal steering vector are not orthogonal, so in order to ensure the correct formation of the weak signal peak, the following transformation is made to the steering vector of the signal:

[0131]

[0132] where b(θ) is the transformed steering vector, I is a unit matrix, a1(θ) is the steering vector of the signal, is the conjugate transpose of the matrix E J , E J = span(e1, …, e J ), the strong signal subspace is orthogonal to the noise subspace Let θ i be the angle of the weak signal, and the following is obtained:

[0133]

[0134] where the transformed steering vector of the signal is orthogonal to the first extended noise subspace.

[0135] In this embodiment, since the transformed steering vector of the signal is orthogonal to the first extended noise subspace, the DOA of the weak signal can be estimated using the MUSIC spectrum, and the two-dimensional spectrum peak search can be avoided. Specifically, S50, using the MUSIC algorithm and the first extended noise subspace, determining the DOA representing the weak signal source, can include:

[0136] The DOA of the weak signal source is determined by the following formula:

[0137]

[0138] In the formula, P MUSIC-JN (θ) is the spatial spectrum based on θ, which is a function related to θ; the relationship between P MUSIC-JN (θ) and the parameter θ can be understood as the relationship between the dependent variable and the independent variable, and the θ corresponding to the spectrum peak of the spatial spectrum is the DOA. θ is the DOA of the weak signal source.

[0139] In this embodiment, S60, creating a covariance matrix of the received signal, can include:

[0140] The covariance matrix of the received signal is constructed as follows:

[0141] R x = E[x(t)x H (t)] = AR s A H +σ 2 I (22)

[0142] A is the array flow matrix, and R scovariance matrix of the received signal, σ 2 is the noise power.

[0143] In this embodiment, S70, the covariance matrix is eigenvalue decomposition and expansion of noise subspace, and the two-dimensional spatial spectrum is constructed, which can include:

[0144] The covariance matrix of the received signal is eigenvalue decomposition, and the noise subspace is expanded to construct a two-dimensional spatial spectrum as follows:

[0145]

[0146] Where, P MUSIC-JN (θ, r) is a two-dimensional spatial spectrum based on θ and r, U JN is the second extended noise subspace, and d(θ, r) is the steering vector of the transformed signal.

[0147] Based on formula (21), the DOA estimation value of the weak signal source can be obtained, which can be denoted as {θ k ,k=1,...,P}。

[0148] In step S80, the obtained DOA estimation value is brought into formula (23), and the distance r of the weak signal source can be estimated one by one. Wherein, the positioning information of the weak signal source can include the DOA and distance of the weak signal source relative to the receiving array (i.e. the signal collection point), so that the positioning of the near-field weak signal source can be realized accurately. Wherein, formula (23) can be rewritten as:

[0149]

[0150] In this embodiment, by sequentially bringing the estimated DOA into the constructed distance spatial spectrum (i.e. two-dimensional spatial spectrum), the distance of the corresponding weak signal source can be estimated, so that the two-dimensional spectrum peak search can be avoided, and the parameters are automatically paired, and the precise positioning of the weak signal source can be realized.

[0151] Please refer to Figure 2, the spatial spectrum of the conventional high-order cumulant and the method provided in the present application. A uniform linear array with 15 elements is selected, the element spacing d = λ / 4, and the signal sampling snapshot number is 2000. Two weak signal sources are located at [20°, 3λ] and [34°, 5λ], and the signal-to-noise ratio is 36.5 dB. Two weak signal sources are located at [14°, 1λ] and [40, 7λ], and the signal-to-noise ratio is 50 dB. It can be known from the experiment that the pseudo-peak caused by the interference source in the spatial spectrum of the conventional method is very obvious, and the spectrum peak of the weak signal source is almost indistinguishable, which is easy to cause estimation deviation. The algorithm provided in the present application effectively suppresses the strong interference, and the two spectrum peaks of the weak signal are very obvious. Therefore, when the strong interference exists in the space, the conventional method is difficult to accurately estimate, and the strong signal steering vector is included in the noise subspace in the present application, the strong interference is effectively suppressed, and the weak signal source can be accurately estimated.

[0152] Please refer to Figure 3 , the distance comparison chart of the conventional second-order statistic method and the algorithm in the present application. It can be known from the simulation experiment that the conventional method produces spectrum peaks at the distances 3λ and 5λ of the strong signal, and there is a weak spectrum peak at the distance 7λ of the weak signal source and almost no spectrum peak at the distance 1λ. This is because the distance estimation needs the prior DOA estimation value, and the conventional method cannot well estimate the DOA of the weak signal, so the distance estimation distortion occurs. The method in the present application produces very sharp spectrum peaks at the distances 1λ and 7λ of the two weak signal sources, which shows that the method in the present application effectively suppresses the strong interference and can well estimate the distance of the weak signal.

[0153] Please refer to Figure 4 , the comparison chart of the root mean square error of DOA and the signal-to-noise ratio of the conventional method and the method in the present application. The signal-to-noise ratio of the strong interference source is 30 dB, and the other parameters are the same as above, and 300 Monte Carlo experiments are performed. It can be seen from the above chart that with the increase of the signal-to-noise ratio of the weak signal source, the RMSE of the conventional method and the method in the present application gradually decreases. When the power difference between the strong and weak signals is large, the RMSE of the method in the present application is much smaller than that of the conventional method, and as the signal-to-noise ratio approaches the strong interference source, the method in the present application always maintains better performance. The experiment shows that the method provided in the present application can effectively improve the DOA estimation accuracy of the weak signal.

[0154] Please refer to Figure 5 , the comparison chart of the root mean square error of distance and the signal-to-noise ratio of the conventional method and the method in the present application. It can be seen that the RMSE of the two methods decreases with the increase of the signal-to-noise ratio, and the method in the present application always maintains superior performance. The experiment shows that the algorithm in the present application effectively improves the distance estimation accuracy of the weak signal in the strong interference background.

[0155] In the method provided in the application, a fourth-order accumulation matrix containing only DOA information is constructed, and then eigenvalue decomposition is performed on the fourth-order accumulation matrix to obtain an approximate strong signal subspace, which is then included in a noise subspace to form an extended noise subspace. Based on the extended noise subspace, the DOA of a weak signal source can be effectively estimated by using the MUSIC algorithm. Secondly, a covariance matrix of the received signal is constructed, and an extended noise subspace is constructed in the same way. The estimated DOA is sequentially brought into a two-dimensional range spectrum for one-dimensional search to obtain the corresponding range, without the need for parameter pairing.

[0156] Based on the above design, the idea of the extended noise subspace is applied to the near-field strong interference scene, which can effectively make up for the blank of parameter estimation research when strong and weak signals coexist in this field. Simulation experiments prove that the method proposed in the application can effectively estimate the positioning information of a weak signal source, and the resolution and accuracy are higher than those of the conventional mixed-order accumulation algorithm, and the computational complexity is comparable to that of the conventional algorithm.

[0157] The embodiment of the application also provides a near-field signal source positioning system, which can include a processor and a memory. The memory stores a computer program, and when the computer program is executed by the processor, the near-field signal source positioning system can execute the corresponding steps in the above-mentioned near-field signal source positioning method.

[0158] In the embodiment, the processor can be an integrated circuit chip with signal processing capability. For example, the processor can be a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, and can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the application.

[0159] The memory can be, but is not limited to, a random access memory, a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, etc. In the embodiment, the memory can be used to store preset numbers, etc. Of course, the memory can also be used to store programs, and the processor executes the programs after receiving execution instructions.

[0160] It should be noted that the skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the near-field source positioning system described above can refer to the corresponding process of each step in the foregoing method, and will not be described in detail here.

[0161] The embodiment of the present application also provides a computer readable storage medium. The computer readable storage medium stores a computer program. When the computer program runs on a computer, the computer executes the near-field source positioning method described in the foregoing embodiment.

[0162] Through the description of the foregoing embodiments, those skilled in the art can clearly understand that the present application can be implemented by hardware, or can be implemented by means of software and a necessary general hardware platform. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.), and includes a plurality of instructions for causing a computer device (which can be a personal computer, a near-field source positioning system, or a network device, etc.) to execute the methods described in various implementation scenarios of the present application.

[0163] In the embodiments provided by the present application, it should be understood that the disclosed system and method can also be implemented in other ways. The system and method embodiments described above are only schematic. For example, the flowcharts and block diagrams in the drawings show the possible implementation architectures, functions and operation of the system, method and computer program product according to the embodiments of the present application. In this regard, each block in the flowcharts and block diagrams can represent a module, a program segment or a part of code, which contains one or more executable instructions for implementing the specified logical functions. It should also be noted that each block in the block diagrams and / or flowcharts, and the combination of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system, or can be implemented by a combination of special-purpose hardware and computer instructions. In addition, the functional modules in the various embodiments of the present application can be integrated together to form a separate part, or each module can exist independently, or two or more modules can be integrated to form a separate part.

[0164] The above only describes the embodiments of the present application and is not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method of locating a near-field source, the method comprising: The method comprises: S10, creating a received signal model based on a receiving array for receiving signals and a signal source with interference signals; S20, creating a fourth-order cumulant matrix containing only DOA information based on a direction of arrival (DOA) in a near-field environment and the received signal model; S30, performing eigenvalue decomposition on the fourth-order cumulant matrix to obtain a strong signal subspace; S40, including the strong signal subspace into a noise subspace to form a first extended noise subspace; S50, determining a DOA representing a weak signal source by using a MUSIC algorithm and the first extended noise subspace; S60, creating a covariance matrix of the received signals; S70, performing eigenvalue decomposition on the covariance matrix and expanding the noise subspace to construct a two-dimensional spatial spectrum; S80, inputting the DOA of the weak signal source into the two-dimensional spatial spectrum for one-dimensional search to obtain positioning information of the weak signal source, the positioning information including a distance of the weak signal source.

2. The method of claim 1, wherein, Step S10 comprises: The total number of antennas of a receiving array and the total number of signal sources in an environment are acquired, the total number of antennas of the receiving array is represented as 2M+1, the antennas of the receiving array are uniformly linearly distributed, are used for receiving narrow-band near-field signals, and the total number of signal sources is represented as P+J, wherein M is an integer greater than or equal to 0, P is the number of weak signal sources, and J is the number of interference signal sources; an output signal x m (t) is represented as: where s k (t) denotes the kth narrowband near-field signal source, t denotes time, e denotes the natural constant, j denotes the imaginary unit, τ mk represents the time delay of the kth signal on the mth antenna relative to the reference element, n m (t) represents additive Gaussian noise, and r m is defined as the distance from the signal source to the mth antenna, r k is the distance from the signal source to the reference element, and based on the cosine theorem, the following is obtained: In the formula, r k represents the distance from the source to the reference element, d indicates the distance between antennas, θ k represents the DOA corresponding to the kth signal, the transmission model of the near-field source is a spherical wave, and the time delay τ mk represents: where λ represents the wavelength of the source, and r m Substituting τ mk into the equation, we get: Expanding equation (4) to second order Taylor, τ mk Transforms to: τ mk ≈w k m+φ k m 2 (5) wherein: Based on the output signal x m (t) and τ obtained from the second order Taylor expansion mk A model of the received signal is created and represented in matrix form: x(t)=A(θ,r)s(t)+n(t) (8) wherein:

3. The method of claim 2, wherein, Step S20 comprises: The fourth-order cumulant is defined as: where y m (t), y p (t), y q * (t) represent the output signals of the mth, nth, pth, qthantennas respectively; * represents conjugation, represents the fourth-order accumulation of the kth narrow-band near-field signal source s k (t). wherein, m, n, p, q ∈ [-M, M], m-n and p-q are located in [-2N, 2N], let m=-n and p=-q, based on the received signal model, the fourth-order cumulant corresponding to formula (10) is converted into: Based on formula (12), a fourth-order cumulant matrix is created by selecting specified elements, the specified elements of the cumulant matrix are given by: The fourth-order cumulant matrix C1 is represented by: The fourth-order cumulant matrix removes the range information and only contains the DOA information of the signal, where H represents the matrix conjugate transpose, A1(θ) and C S The formula is as follows: A1(θ) = [a1(θ1),..., a1(θ k ),..., a1(θ P+J )] (15) In the formulae, 4. The method of claim 3, wherein, Step S30 comprises: Performing eigenvalue decomposition on the fourth-order cumulant matrix: In the formula, λ i Let λ be an eigenvalue, and λ1≥...≥λ J ≥λ J+1 ≥...λ J+P ≥λ 2M+1 e i For λ i The corresponding eigenvectors, the subspace representation of the total signal are E JS =span(e1,...,e J ,e J+1 ,…,e J+P ), where the feature vectors (e1,...,e J ) is determined by strong interference signals, (e J+1 ,...e J+P The total signal subspace is determined by the weak signal, and the subspace of the total signal includes the strong signal subspace.

5. The method of claim 4, wherein, Step S40 comprises: The strong signal subspace is incorporated into the noise subspace, and the noise subspace E is composed of eigenvectors corresponding to 2M+1-P-J eigenvalues N = span(e J+P+1 ,...,e 2M+1 ), and the first extended noise subspace is constructed as: E JN = span(e1,...,e J ,e J+P+1 ,...,e 2M+1 ) (18) Performing the following transformation on a steering vector of the signal: where b(0) denotes the transformed steering vector, I denotes the identity matrix, a1(0) denotes the steering vector, is the conjugate transpose of the matrix E J J = span(e1,..., e J ), the strong signal subspace is orthogonal to the noise subspace Let 0 i denote the weak signal angle, we have:​ wherein, the steering vector of the signal after transformation is orthogonal to the first extended noise subspace.

6. The method of claim 5, wherein, Step S50 comprises: The DOA of the weak signal source is determined by the following formula: In the formula, P MUSIC-JN (0) is a spatial spectrum based on 0, which is the DOA of the weak signal source.

7. The method of claim 6, wherein, Step S60 comprises: constructing a covariance matrix R of the received signal x : R x = E [x(t)x H (t)] = AR s A H + σ 2 I (22) A denotes an array flow pattern matrix, R s denotes the covariance matrix of the signals, σ 2 is the noise power.

8. The method of claim 7, wherein, Step S70 comprises: Performing eigenvalue decomposition on the covariance matrix of the received signals and expanding the noise subspace to construct a two-dimensional spatial spectrum as follows: where U JN is the second extended noise subspace, and d(θ, r) is the steering vector of the transformed signal.

9. A near-field source localization system, comprising: The near-field source positioning system comprises a processor and a memory coupled with each other, the memory stores a computer program, when the computer program is executed by the processor, the near-field source positioning system executes the method according to any one of claims 1-8.

Images