CN116338568A - MUSIC direction finding method based on underwater information system cross array - Google Patents

Abstract

The invention relates to a MUSIC direction finding method based on a cross array of an underwater information system, and belongs to the technical field of underwater sound direction finding. The invention provides a MUSIC direction finding method based on a cross array of an underwater information system, aiming at the problems of low signal-to-noise ratio and low direction finding precision of underwater signal direction finding reception, which is based on a MUSIC algorithm, and utilizes frequency extraction and spatial spectrum accumulation, the cross array is used for carrying out direction finding processing on signals to obtain the relative position relation between the signals and the cross array, and then a proper linear array in the cross array is selected for carrying out direction finding processing on the signals, so that the direction finding of underwater target sound signals under low signal-to-noise ratio can be effectively realized.

Description

MUSIC direction finding method based on underwater information system cross array

Technical Field

The invention relates to the technical field of underwater acoustic direction finding, in particular to a MUSIC direction finding method based on a cross array of an underwater information system.

Background

The acoustic wave is used as the only carrier for the remote transmission of the current underwater information, and the underwater acoustic direction finding technology becomes a key for human beings to enter deep sea by means of an underwater platform, detect the deep sea and develop the deep sea. The underwater sound direction finding technology is used as an important research direction in the field of underwater sound signal processing and sonar, and limits the regional detection, reconnaissance and striking capability of the underwater platform. Therefore, the direction finding technology for underwater signals is an important point and a hot spot of research both from the academic and engineering points of view.

Spatial spectrum estimation is an important aspect in array signal processing. The space spectrum estimation direction-finding technology can realize multi-target direction-finding (including coherent signals and incoherent signals) at the same time, has no special constraint condition on array elements and array arrangement, brings great flexibility to the design of the array elements and the antenna, has high direction-finding precision under the condition of low signal-to-noise ratio, and can be completely used for direction-finding of a radiation source in a complex environment.

The cross array is suitable for the direction finding of a remote sound source by virtue of the characteristics of smaller redundancy, smaller blind area and dimension division. Therefore, the MUSIC direction finding method based on the cross array provides a thought for the water sound direction finding technology. However, there is no related technical solution in the prior art.

Disclosure of Invention

In view of the above, the invention provides a MUSIC direction finding method based on a cross array of an underwater information system, which is based on a MUSIC algorithm and combines the technologies of average periodical graph time accumulation, filtering, signal detection, time sliding window and the like to effectively realize the direction finding of underwater target acoustic signals under low signal-to-noise ratio.

In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:

a MUSIC direction finding method based on a cross array of an underwater information system comprises the following steps:

step 1, initializing variables, including average periodic chart integration time length, sliding window duration, sliding window step distance, system sampling rate, total number of sliding windows in the integration time, number of points in the sliding window, number of sliding window step distance points, number of single-packet data points and signal section reading duration;

step 2, according to the input frequency value, finding out the array element combination under the corresponding frequency band on the cross array, reading the array element signal with the determined time length from the original sampling signal buffer memory, and generating a signal to be processed;

step 3, carrying out Fourier transform and average periodical graph time accumulation on the generated signal to be processed to acquire a signal frequency spectrum and a periodical spectrum;

step 4, detecting the signal processed in the step 3, and if the current array element signal detects the underwater sound signal and other array element signals also detect the underwater sound signal, executing the steps 6-9; otherwise, executing the step 5;

step 5, updating the initial position value of the current array element signal, storing the signal of the step length of the sliding window of the head part of the current array element signal, and returning to the step 2;

step 6, saving the initial position value of the current array element signal, and intercepting the signal segment;

step 7, carrying out sliding window Fourier transform on the intercepted array element signals, and extracting frequency spectrum data under corresponding frequency points;

step 8, calculating a covariance matrix of the frequency spectrum data, carrying out eigenvalue decomposition on the covariance matrix, determining a signal subspace and a noise subspace, and calculating spatial spectrum data;

step 9, accumulating the spatial spectrum data, and searching the peak value of the accumulated spatial spectrum data to find out the angle corresponding to the peak value;

and step 10, selecting a linear array with a proper position according to the relative positions of the angle judgment signal and the two arrays in the cross array, and then using array elements with corresponding frequency bands on the linear array to conduct direction finding.

Further, the cross array of the underwater information system is divided into 33 array elements, wherein the 33 array elements are divided into 5 frequency bands of 0-1875Hz,1875-3750Hz,3750-7500Hz,7500-15000Hz and 15000-35000Hz, and each frequency band corresponds to different array element combinations;

the specific mode of the step 2 is as follows:

selecting corresponding array element combination according to the input frequency value f0channel；

Setting data durationt_dataAnd calculating the data length dataLen by combining the system sampling rate, setting a data initial position t0_data, and reading a signal with the dataLen length from an original sampling signal buffer as a signal to be processed.

Further, the specific mode of the step 7 is as follows:

step 701, setting a sliding window length t_sw, a sampling rate fs, a sliding window number tsw_n and a bandwidth bw, and generating a total frequency axis f_axis and a extracted frequency axis fb_n;

step 702, directly performing sliding window FFT on the signal to be detected, performing FFT computation on the signal with t_sw length in the sliding window, sliding the dt length after completion, and continuing to perform FFT on the signal with t_sw length until FFT computation on the signal with t_ap length is completed, thereby obtaining Nwap sliding window FFT results, nwap=t_ap/dt;

in step 703, the spectral data x_f at the corresponding frequency bin is extracted.

Further, the specific manner of step 9 is as follows:

and accumulating the frequency spectrum data calculated under each frequency point according to the extracted frequency axis, and searching the peak value of the accumulated spatial spectrum data Pmusic to search out the azimuth angle corresponding to the peak value.

In step 10, the direction finding is performed by using the array elements of the corresponding frequency band on the linear array, and the specific manner is as follows:

storing signal initial position values of linear array elements, and intercepting signal segments;

performing sliding window Fourier transform on the intercepted array element signals, and extracting spectrum data under corresponding frequency points;

solving a covariance matrix of the frequency spectrum data, carrying out eigenvalue decomposition on the covariance matrix, determining a signal subspace and a noise subspace, and calculating spatial spectrum data;

and accumulating the spatial spectrum data, carrying out peak search on the accumulated spatial spectrum data, finding out an angle corresponding to the peak value, and completing direction finding.

Compared with the prior art, the invention has the following beneficial effects:

1. the invention adopts the MUSIC direction finding method to carry out direction finding processing on the cross array element signals based on the form of the cross array of the underwater information system, thereby reducing the influence of the array on the accuracy of the direction finding algorithm.

2. The invention utilizes the frequency extraction and space spectrum accumulation method to realize the noise reduction treatment of the array element signals, improves the signal receiving signal to noise ratio and improves the accuracy of the direction finding treatment azimuth estimation of the array element signals.

3. According to the invention, the cross array is used for carrying out direction finding processing on the signals to obtain the relative position relation between the signals and the cross array, and then the proper linear array in the cross array is selected for carrying out direction finding processing on the signals, so that the direction finding accuracy is improved.

Drawings

FIG. 1 is a flow chart of a method for MUSIC direction finding based on a cross array of an underwater information system in an embodiment of the present invention;

FIG. 2 is a flowchart showing a specific step of frequency extraction and spatial spectrum accumulation for the array signals in FIG. 1;

FIG. 3 is a flowchart showing a specific embodiment of the method in FIG. 1 for generating a signal to be processed by reading an array element signal of a determined duration from an original sampled signal buffer;

fig. 4 is a diagram of the result of cross array MUSIC direction finding processing using actual data.

Description of the embodiments

The invention is described in further detail below with reference to the drawings and the detailed description.

A MUSIC direction finding method based on a cross array of an underwater information system comprises the following steps:

(1) Initializing variables including parameters such as average periodical chart integration time length, sliding window duration, sliding window step distance, system sampling rate, total number of sliding windows in the integration time, number of points in the sliding window, sliding window step distance number, number of points of single-packet data, reading signal section duration and the like;

(2) According to the input frequency value, finding out an array element combination under a corresponding frequency band on the cross array, and reading an array element signal with a determined duration from an original sampling signal buffer memory to generate a signal to be processed;

(3) Carrying out Fourier change and average periodical graph time accumulation on the generated signal to be processed to acquire a signal frequency spectrum and a periodical spectrum;

(4) Performing signal detection on the processed signal, and if the current array element signal detects the underwater sound signal and other array element signals also detect the underwater sound signal, executing the steps 6-9 to perform direction finding; otherwise, executing the step 5;

(5) Updating the initial position value of the current array element signal, storing the signal of the step length of the sliding window of the head of the current array element signal, and returning to the execution step (2);

(6) Storing the initial position value of the current array element signal, intercepting a signal segment, and executing the step (7);

(7) Performing sliding window Fourier transform on the intercepted array element signals, and extracting spectrum data under corresponding frequency points;

(8) Solving a covariance matrix of the frequency spectrum data, carrying out eigenvalue decomposition on the covariance matrix, determining a signal subspace and a noise subspace, and calculating spatial spectrum data;

(9) And accumulating the spatial spectrum data, and searching the peak value of the accumulated spatial spectrum data to find out the angle corresponding to the peak value.

(10) And selecting a linear array with a proper position according to the relative positions of the measured angle judgment signal and the two arrays of the cross array, and further carrying out direction finding processing according to the modes from step 6 to step 9.

The following is a more specific example:

a MUSIC direction finding method based on a cross array of an underwater information system comprises 33 array elements on the cross array of the underwater information system, wherein the array elements are divided into 5 frequency bands of 0-1875Hz,1875-3750Hz,3750-7500Hz,7500-15000Hz and 15000-35000Hz, and each frequency band corresponds to different array element combinations. As shown in fig. 1-3, the method comprises the steps of:

step 1: initializing variables including parameters such as average periodical chart integration time length, sliding window duration, sliding window step distance, system sampling rate, total number of sliding windows in the integration time, number of points in the sliding window, sliding window step distance number, number of points of single-packet data, reading signal section duration and the like;

step 2:

(201) According to frequencyf0Selecting corresponding array element combinationchannel；

(202) Setting data durationt_dataCalculating data length in combination with system sampling ratedataLenSetting the initial position of datat0_dataReading from an original sampled signal bufferdataLenA signal of length.

Step 3: read array element combined signal sectionsignalSequentially reading the signals of each array elementtmpdataPerformingnfftAnd (5) FFT calculation of the number of points and a periodic spectrogram.

Step 4: performing signal detection on the processed signal, if the current array element signal detects the underwater sound signal and other array element signals also detect the underwater sound signal, executing the steps 6-9, otherwise, executing the step (5);

step 5: updating the initial position value of the current array element signal, storing the signal of the step length of the sliding window of the head of the current array element signal, and returning to the execution step (2).

Step 6: and storing the initial position value of the current array element signal, and intercepting the signal segment.

Step 7:

(701) Setting a sliding window lengtht_swSampling ratefsSliding window counttsw_nBandwidth ofbwGenerating a total frequency axisf_ axisExtracted frequency axisfb_n；

(702) Directly carrying out sliding window FFT on a signal to be detected, and taking the length of the sliding windowt_swFFT calculation is carried out on the length signal, and sliding is carried out after completiondtLength, continue takingt_swThe length signal is subjected to FFT until completiont_apFFT calculation of length signal to obtainNwapSliding window FFT results, nwap=t_ap/dt;

(703) Extracting frequency spectrum data under corresponding frequency pointsX_f。

Step 8: using the spectral data obtainedX_fObtaining an estimated covariance correlation matrixRxPerforming feature space decomposition on the covariance correlation matrix to obtain a signal subspace and a noise subspace,

noise spaceUArray flow patternaCalculating spatial spectrum dataazimuth_t；

Step 9: according to the extracted frequency axis, accumulating the calculated frequency spectrum data under each frequency point, and accumulating the accumulated spatial spectrum dataPmusicCarrying out peak search to search out azimuth angles corresponding to the peaks;

step 10: step 6-step 9, primarily judging the azimuth of the underwater signal by utilizing the cross array MUSIC direction finding; and judging the relative relation between the signal and the two linear arrays forming the cross array according to the azimuth, selecting the linear array with proper position, and then carrying out direction finding processing again by utilizing the array elements of the corresponding frequency band on the linear array.

Fig. 4 shows experimental data of the method, and from the graph, the method can stably measure the direction of signals of 1kHz and 10kHz at 7 km and 10 km under water, and has smaller direction measurement error and higher accuracy.

In a word, the method of the invention carries out direction finding processing on signals by using a cross array to obtain the relative position relation between the signals and the cross array, and then selects a proper linear array in the cross array to carry out direction finding processing on the signals, thereby improving the direction finding accuracy, reducing the influence of the array type on the accuracy of a direction finding algorithm and being an important improvement on the prior art.

Claims (5)

1. The MUSIC direction finding method based on the underwater information system cross array is characterized by comprising the following steps of:

step 1, initializing variables, including average periodic chart integration time length, sliding window duration, sliding window step distance, system sampling rate, total number of sliding windows in the integration time, number of points in the sliding window, number of sliding window step distance points, number of single-packet data points and signal section reading duration;

step 2, according to the input frequency value, finding out the array element combination under the corresponding frequency band on the cross array, reading the array element signal with the determined time length from the original sampling signal buffer memory, and generating a signal to be processed;

step 3, carrying out Fourier transform and average periodical graph time accumulation on the generated signal to be processed to acquire a signal frequency spectrum and a periodical spectrum;

step 4, detecting the signal processed in the step 3, and if the current array element signal detects the underwater sound signal and other array element signals also detect the underwater sound signal, executing the steps 6-9; otherwise, executing the step 5;

step 5, updating the initial position value of the current array element signal, storing the signal of the step length of the sliding window of the head part of the current array element signal, and returning to the step 2;

step 6, saving the initial position value of the current array element signal, and intercepting the signal segment;

step 7, carrying out sliding window Fourier transform on the intercepted array element signals, and extracting frequency spectrum data under corresponding frequency points;

step 8, calculating a covariance matrix of the frequency spectrum data, carrying out eigenvalue decomposition on the covariance matrix, determining a signal subspace and a noise subspace, and calculating spatial spectrum data;

step 9, accumulating the spatial spectrum data, and searching the peak value of the accumulated spatial spectrum data to find out the angle corresponding to the peak value;

and step 10, selecting a linear array with a proper position according to the relative positions of the angle judgment signal and the two arrays in the cross array, and then using array elements with corresponding frequency bands on the linear array to conduct direction finding.

2. The method for MUSIC direction finding based on the cross array of the underwater information system according to claim 1, wherein the cross array of the underwater information system is divided into 33 array elements, and the 33 array elements are divided into 5 frequency bands of 0-1875Hz,1875-3750Hz,3750-7500Hz,7500-15000Hz and 15000-35000Hz, and each frequency band corresponds to different array element combinations;

the specific mode of the step 2 is as follows:

selecting corresponding array element combination according to the input frequency value f0channel；

Setting data durationt_dataAnd calculating the data length dataLen by combining the system sampling rate, setting a data initial position t0_data, and reading a signal with the dataLen length from an original sampling signal buffer as a signal to be processed.

3. The MUSIC direction finding method based on the underwater information system cross array according to claim 2, wherein the specific mode of the step 7 is as follows:

step 701, setting a sliding window length t_sw, a sampling rate fs, a sliding window number tsw_n and a bandwidth bw, and generating a total frequency axis f_axis and a extracted frequency axis fb_n;

step 702, directly performing sliding window FFT on the signal to be detected, performing FFT computation on the signal with t_sw length in the sliding window, sliding the dt length after completion, and continuing to perform FFT on the signal with t_sw length until FFT computation on the signal with t_ap length is completed, thereby obtaining Nwap sliding window FFT results, nwap=t_ap/dt;

in step 703, the spectral data x_f at the corresponding frequency bin is extracted.

4. The MUSIC direction finding method based on the underwater information system cross array according to claim 3, wherein the specific mode of the step 9 is as follows:

and accumulating the frequency spectrum data calculated under each frequency point according to the extracted frequency axis, and searching the peak value of the accumulated spatial spectrum data Pmusic to search out the azimuth angle corresponding to the peak value.

5. The method for detecting the MUSIC based on the cross array of the underwater information system as claimed in claim 4, wherein in the step 10, the array elements of the corresponding frequency band on the linear array are used for detecting the MUSIC, and the specific mode is as follows:

storing signal initial position values of linear array elements, and intercepting signal segments;

performing sliding window Fourier transform on the intercepted array element signals, and extracting spectrum data under corresponding frequency points;

solving a covariance matrix of the frequency spectrum data, carrying out eigenvalue decomposition on the covariance matrix, determining a signal subspace and a noise subspace, and calculating spatial spectrum data;

and accumulating the spatial spectrum data, carrying out peak search on the accumulated spatial spectrum data, finding out an angle corresponding to the peak value, and completing direction finding.

Images