阅读说明：本技术 一种基于时延差范数加权的宽带信号高分辨方位估计方法 (Broadband signal high-resolution direction estimation method based on time delay difference norm weighting ) 是由 毛卫宁 钱进 于 2021-08-27 设计创作，主要内容包括：本发明公开了一种基于时延差范数加权的宽带信号高分辨方位估计方法,该方法通过对各子带进行波束形成估计信源方位,统计各子带的方位估计值计算得到时延差范数,以时延差范数对各子带波束输出功率进行加权,综合各子带的波束输出功率得到宽带信号的波束输出功率。本发明以宽带常规波束形成算法为基础,通过时延差范数加权实现宽带信号的高分辨方位估计,解决了现有宽带信号高分辨方位估计方法计算量大、算法复杂度高和稳健性差的问题,且可以分辨相干源。(The invention discloses a broadband signal high-resolution direction estimation method based on time delay difference norm weighting. The method is based on the conventional broadband beam forming algorithm, realizes the high-resolution azimuth estimation of the broadband signal by time delay difference norm weighting, solves the problems of large calculated amount, high algorithm complexity and poor robustness of the traditional broadband signal high-resolution azimuth estimation method, and can distinguish the coherent source.)

1. A broadband signal high-resolution direction estimation method based on time delay difference norm weighting is characterized by comprising the following steps:

step S1, acquiring a broadband multi-path array receiving signal;

step S2, carrying out Fourier transform on the broadband multi-path array received signal, and decomposing the broadband multi-path array received signal into J sub-bands;

step S3, performing beam forming on the array received signal of each subband to obtain the beam output power of each subband, where the beam output power of the jth subband is:

P_j(θ)＝a^H(f_j,θ)R(f_j)a(f_j,θ)，j＝1,2,…,J

in the formula, P_j(theta) is the beam output power of the jth sub-band, a (f)_jθ) is the jth sub-band frequency point f_jA guide vector of R (f)_j)＝X(f_j)X^H(f_j) Is a frequency point f_jArray signal covariance matrix of (f)_j) For discrete Fourier transform of array received signal at frequency point f_jThe value of (a) is (b),(·)^Hrepresents conjugate transpose, θ represents scan orientation;

step S4 for determining the beam output power P of each sub-band_jAzimuth theta corresponding to maximum value of (theta)_i,jWherein, i is 1,2, …, N is the number of the information sources, J is 1,2, …, J;

step S5, counting the azimuth estimation value theta of each sub-band and different information sources_i,jAnd calculating the average time delay difference of the ith signal source signal of each sub-band to two adjacent array elementsBased on the average delay differenceObtaining the time delay difference norm delta_iWherein the delay spread norm δ_iComprises the following steps:

in the formula, tau is the time delay difference of two adjacent array elements, and | | · | | represents a norm;

step S6, according to the reciprocal delta of the time delay difference norm calculated in the step S5_iAnd weighting the beam output power of each sub-band, and synthesizing the beam output power of each sub-band to obtain the beam output power of the broadband signal.

2. The method for high-resolution azimuth estimation of broadband signals based on delay difference norm weighting according to claim 1, wherein in the step S1, the expression of the broadband multi-path array received signals is:

in the formula, x_m(t) broadband multipath received signal, s, for the mth array element_i(t) is a broadband signal sent by the ith signal source, N is the number of signal sources,the time delay of the ith sound ray sent by the ith information source received by the mth array element relative to the reference array element is expressed byIndicating the amplitude L of the ith sound ray from the ith source to the mth array element_miRepresenting the number of sound rays received by the m array element and emitted by the i information source, n_mAnd (t) represents zero mean Gaussian white noise received by the mth array element, and M is the number of the received array elements.

3. The method according to claim 1, wherein in step S4, except for the multipath coherent source, the signal bands of the N sources have non-overlapping sub-bands; considering the influence of the channel, the frequency selectivity of the multi-path channel enables some frequency point signals to be strengthened and some frequency point signals to be weakened, and the direction estimation results of the coherent source and the incoherent source in different frequency bands are influenced.

Technical Field

The invention relates to the technical field of broadband signal high-resolution azimuth estimation, in particular to a broadband signal high-resolution azimuth estimation method based on time delay difference norm weighting.

Background

The high-resolution orientation estimation of broadband signals is always a hotspot and difficult problem of array signal processing, and the solution methods are roughly divided into two types: incoherent Signal Subspace Method (ISSM), and Coherent Signal Subspace Method (CSSM). The incoherent signal subspace algorithm (ISSM) decomposes the broadband signal on a frequency domain, calculates a spatial spectrum for each decomposed narrow-band component, then accumulates the spatial spectrums of all the narrow-band components to obtain the spatial spectrum of the broadband signal, and then carries out azimuth estimation. The method has large calculation amount, cannot distinguish coherent sources, and has reduced azimuth estimation performance under the condition of low signal-to-noise ratio; coherent Signal Subspace algorithm (CSSM, Coherent Signal Subspace Method) focuses broadband Signal frequency components to a reference frequency by means of a focus transform, calculates a narrowband spatial spectrum at the reference frequency point, and then performs azimuth estimation. The method can distinguish the coherent source, but when the focusing transformation matrix is obtained, the direction of the signal source needs to be estimated, and the estimated value has a large influence on the final direction measurement precision. For a large array with a large array element number, when the two methods are adopted to realize broadband high-resolution azimuth estimation, because the dimension of a covariance matrix is high and the calculation amount is large, the conventional solution is to decompose the large array into a plurality of sub-arrays, calculate a broadband signal spatial spectrum for each sub-array, and synthesize the results of the sub-arrays to realize the high-resolution azimuth estimation of the large array, but the processing of the sub-arrays loses the degree of freedom and reduces the spatial resolution.

Disclosure of Invention

In view of this, the present invention provides a method for high-resolution azimuth estimation of a wideband signal based on delay difference norm weighting, which can achieve high-resolution azimuth estimation of the wideband signal, reduce sidelobes of beam output, and improve robustness of azimuth estimation, while maintaining the unchanged degree of freedom of array processing.

In order to achieve the purpose, the invention adopts the following technical scheme:

a broadband signal high-resolution direction estimation method based on time delay difference norm weighting comprises the following steps:

step S1, acquiring a broadband multi-path array receiving signal;

step S2, carrying out Fourier transform on the broadband multi-path array received signal, and decomposing the broadband multi-path array received signal into J sub-bands;

step S3, performing beam forming on the array received signal of each subband to obtain the beam output power of each subband, where the beam output power of the jth subband is:

P_j(θ)＝a^H(f_j,θ)R(f_j)a(f_j,θ)，j＝1,2,…,J

formula (II)In, P_j(theta) is the beam output power of the jth sub-band, a (f)_jθ) is the jth sub-band frequency point f_jA guide vector of R (f)_j)＝X(f_j)X^H(f_j) Is a frequency point f_jArray signal covariance matrix of (f)_j) For discrete Fourier transform of array received signal at frequency point f_jValue of (c) (. 1)^HDenotes conjugate transpose, and θ denotes scanning azimuth.

Step S4 for determining the beam output power P of each sub-band_jAzimuth theta corresponding to maximum value of (theta)_i,jWherein, i is 1,2, …, N is the number of the information sources, J is 1,2, …, J;

step S5, counting the azimuth estimation value theta of each sub-band and different information sources_i,jAnd calculating the average time delay difference of the ith signal source signal of each sub-band to two adjacent array elementsBased on the average delay differenceObtaining the time delay difference norm delta_iWherein the delay spread norm δ_iComprises the following steps:

i＝1,2,…,N

in the formula, tau is the time delay difference of two adjacent array elements, and | | · | | represents a norm;

step S6, according to the reciprocal delta of the time delay difference norm calculated in the step S5_iAnd weighting the beam output power of each sub-band, and synthesizing the beam output power of each sub-band to obtain the beam output power of the broadband signal.

Further, in step S1, the expression of the broadband multi-path array received signal is:

in the formula, x_m(t) broadband multipath received signal, s, for the mth array element_i(t) is a broadband signal sent by the ith signal source, N is the number of signal sources,the time delay of the ith sound ray sent by the ith information source received by the mth array element relative to the reference array element is expressed byIndicating the amplitude L of the ith sound ray from the ith source to the mth array element_miRepresenting the number of sound rays received by the m array element and emitted by the i information source, n_mAnd (t) represents zero mean Gaussian white noise received by the mth array element, and M is the number of the received array elements.

Further, in step S4, except for the multipath coherent source, the signal bands of the N sources have non-overlapping sub-bands; considering the influence of the channel, the frequency selectivity of the multi-path channel enables some frequency point signals to be strengthened and some frequency point signals to be weakened, and the direction estimation results of the coherent source and the incoherent source in different frequency bands are influenced.

The invention has the beneficial effects that:

the invention is based on the conventional beam forming algorithm of the broadband, and utilizes the frequency selectivity of non-overlapping sub-bands and multi-path channels in different signal source frequency domains to realize the high-resolution azimuth estimation of the broadband signals by the time delay difference norm weighting. Compared with the common conventional beam forming method, the method has the advantages that the azimuth resolution is greatly improved, the beam output side lobe is reduced, and the weak target detection is facilitated; compared with a broadband high-resolution azimuth estimation method, the method does not need matrix inversion and eigenvalue decomposition operation, greatly reduces the calculated amount and algorithm complexity, has the same azimuth estimation robustness as the conventional beam forming method, and can distinguish coherent sources.

Drawings

Fig. 1 is a beam output power diagram in embodiment 1.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Example 1

Referring to fig. 1, in the present embodiment, a wideband signal high-resolution direction estimation method based on delay difference norm weighting is provided, where a beam is formed for each subband to estimate an information source direction, a direction estimation value of each subband is counted to calculate an average delay difference between two adjacent array elements, so as to obtain a delay difference norm, a beam output power of each subband is weighted by an inverse of the delay difference norm, and the beam output power of each subband is synthesized to obtain a beam output power of a wideband signal.

Specifically, the wideband signal high-resolution azimuth estimation method provided in this embodiment is based on a wideband conventional beamforming algorithm, and utilizes non-overlapping subbands in different signal source frequency domains and frequency selectivity of a multipath propagation channel to implement high-resolution azimuth estimation of a wideband signal through delay difference norm weighting, so as to solve the problems of large calculation amount, high algorithm complexity, and poor robustness of the existing wideband signal high-resolution azimuth estimation method.

Specifically, the present embodiment takes a group of testing calculations of the received signals of the multi-path wideband array to be tested as an example, to describe more specifically the method provided by the present embodiment, which includes:

step 1, obtaining a broadband multi-path array receiving signal: the sampling frequency is 48kHz, and the snapshot number is 1024; the sound velocity of sea water is 1500m/s, the sound velocity of seabed medium is 1800m/s, and the density is 1.6g/cm³The sea surface is vacuum, the sea depth is 1000 meters, and the depth of the sound source and receiving line array is 30 meters; two far-field incoherent sound sources with frequency bands of 2000Hz-4000Hz and 2500Hz-450 Hz respectively0Hz, and orientations of 54 degrees and 60 degrees, respectively; receiving by a uniform linear array, wherein the number of array elements is 35, and the spacing between the array elements is 0.5 m; signal to noise ratio-10 dB. And generating a broadband multipath array receiving signal by using a bellhop sound field model.

And 2, carrying out Fourier transform on the broadband array receiving signal, and decomposing the broadband signal into J sub-bands.

And 3, performing beam forming on the array receiving signal of each sub-band to obtain the beam output power of each sub-band.

In step 3, the beam output power of the jth sub-band is

P_j(θ)＝a^H(f_j,θ)R(f_j)a(f_j,θ) j＝1,2,…,J

Wherein, P_j(theta) is the beam output power of the jth sub-band, a (f)_jθ) is the jth sub-band frequency point f_jA guide vector of R (f)_j)＝X(f_j)X^H(f_j) For the jth own frequency point f_jArray signal covariance matrix of (f)_j) Discrete Fourier transform of array received signal at j sub-band frequency point f_jValue of (c) (. 1)^HDenotes conjugate transpose, and θ denotes scanning azimuth.

Step 4, calculating the output power P of each sub-band wave beam_jAzimuth theta corresponding to maximum value of (theta)_i,jI is 1,2, …, N is the number of sources, and J is 1,2, …, J.

Step 5, counting the azimuth estimation value theta of each sub-band_i,jAnd calculating the average time delay difference of the ith signal source signal of each sub-band to two adjacent array elementsFurther obtain the time delay difference norm delta_i；

In step 5, the delay spread norm isi is 1,2, …, N, where τ is the time delay difference between two adjacent array elements, | | · | | denotes taking norm for uniform linear array Wherein the content of the first and second substances,for each subband azimuth estimation value theta of ith source signal_i,jAnd (4) counting the average value, wherein d is the array element interval, c is the sound velocity, and theta represents the scanning azimuth.

And 6, weighting the beam output power of each sub-band by the reciprocal of the delay difference norm, and synthesizing the beam output power of each sub-band to obtain the beam output power of the broadband signal.

More specifically, in step 4, considering the influence of the channel, the frequency selectivity of the multipath channel causes some frequency point signals to be strengthened and some frequency point signals to be weakened, and the direction estimation results of the coherent source and the incoherent source in different frequency bands are influenced.

More specifically, in fig. 1, the dotted line is the beam output power calculated by the conventional beam forming method in different directions, and since the direction interval between the two broadband information sources is small and 6 degrees, the conventional beam forming method cannot correctly distinguish the two information sources; the solid line is the wideband beam output power in different directions calculated by the method provided by the embodiment, two information sources with an interval of 6 degrees can be correctly distinguished, and the beam output power sidelobe is lower and is below-60 dB, which is beneficial to the detection of a weak target.

In summary, in the broadband signal high-resolution azimuth estimation method based on delay difference norm weighting provided in this embodiment, based on a broadband conventional beam forming algorithm, high-resolution azimuth estimation of a broadband signal is achieved through delay difference norm weighting by using non-overlapping subbands in frequency domains of different signal sources and frequency selectivity of a multipath channel.

Compared with the common conventional beam forming method, the method provided by the embodiment greatly improves the azimuth resolution, reduces the beam output side lobe, and is beneficial to weak target detection.

Compared with a broadband high-resolution azimuth estimation method, the method provided by the embodiment does not need matrix inversion and eigenvalue decomposition operation, greatly reduces the calculation amount and algorithm complexity, has the same azimuth estimation robustness as that of a conventional beam forming method, and can distinguish coherent sources.

The invention is not described in detail, but is well known to those skilled in the art.

The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.