Autocorrelation structure analysis method, system and device for underwater acoustic target recognition

Abstract

The present invention discloses an autocorrelation structure analysis method, system and device for underwater acoustic target recognition, belonging to the technical field of underwater target recognition. The autocorrelation structure analysis method comprises the following steps: S1, modeling underwater acoustic signals based on the sound line theory; S2, extracting the delay difference structure of the target through autocorrelation analysis, so as to classify the multi-path delay structure into multi-path delay structure A, multi-path delay structure B and multi-path delay structure C; S3, analyzing the sound source depth, receiving depth and the relative distance between the sound source and the receiving, so as to select one of the multi-path delay structure A, multi-path delay structure B and multi-path delay structure C for target recognition. By classifying the multi-path delay structure into multi-path delay structures A, B and C and analyzing the applicable conditions of the multi-path delay structures A, B and C, it is beneficial to perform target recognition based on the multi-path delay structures A, B and C in the future.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater target recognition, and in particular to an autocorrelation structure analysis method, system and device for underwater acoustic target recognition. Background Art

[0002] Traditional underwater acoustic target recognition and classification methods include feature extraction and classifier design, which are mainly based on the unique feature information of the target. It is required that there is extractable information in the target raw data that can accurately and simply indicate the target state and identity. The existing feature extraction methods for underwater acoustic target recognition mainly include three types: methods based on the physical structure characteristics of the ship, based on the time-frequency features with statistical significance, and based on the features of the sound field structure. The method based on the physical structure characteristics of the ship detects the target by extracting features with clear physical significance, such as the number of propeller blades and propeller speed, and then identifies the target category through experience or database, with high false alarm rate and false detection rate; the target recognition method based on the time-frequency features with statistical significance is based on mainstream target detection frameworks such as RCNN and Yolo to extract the target's spectral center, spectral shape, spectral fluctuation, narrowband and broadband spectral features, cepstrum features, Mel cepstrum parameters, pitch frequency parameters and other statistical time-frequency domain features, but due to the time-varying characteristics of the acoustic channel and the instability of the statistical characteristics of the same target, the target recognition accuracy is low. Summary of the invention

[0003] An object of the present invention is to provide an autocorrelation structure analysis method, system and device for underwater acoustic target recognition, wherein the autocorrelation structure analysis method adopts sound source characteristics based on sound field structure, and utilizes the autocorrelation structure analysis method to extract target depth information, thereby achieving the goal of target recognition with robustness and high accuracy.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is: an autocorrelation structure analysis method for underwater acoustic target recognition, wherein the autocorrelation structure analysis method comprises the following steps:

[0005] S1, modeling of underwater acoustic signals based on sound ray theory;

[0006] S2, autocorrelation analysis extracts the target delay difference structure for classifying the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C;

[0007] S3, analyzing the sound source depth, the receiving depth, and the relative distance between the sound source and the receiving device to select one of the multi-path delay structures A, B, and C for target identification.

[0008] As a further improvement of the technical solution of the present invention, in step S2, the autocorrelation function of the received signal obtained by autocorrelation analysis is:

[0009] ,parameter is the autocorrelation function of the sound source signal, and the parameters is the autocorrelation function of the noise, the received signal The autocorrelation function exist A peak value appears when the peak value is extracted, and the delay value is estimated based on the peak value, wherein the eigenvalue line that is reflected by the sea surface only once is the eigenvalue line S, the eigenvalue line that is reflected by the seabed only once is the eigenvalue line B, the eigenvalue line that is reflected by the sea surface once and then by the seabed once is the eigenvalue line SB, and the eigenvalue line that is reflected by the seabed once and then by the sea surface once is the eigenvalue line BS. The multipath delay structure is composed of two groups of delay differences, namely, the eigenvalue line SB-B and the eigenvalue line SBS-BS. The multipath delay structure B is composed of one group of delay differences, namely, the eigenvalue line SD. The multipath delay structure C is composed of four groups of delay differences, namely, the eigenvalue line BS-B, the eigenvalue line BS-SB, the eigenvalue line SBS-B, and the eigenvalue line SBS-SB.

[0010] As a further improvement of the technical solution of the present invention, in step S3, the environment, track and transceiver configuration of the channel multipath delay structure are analyzed respectively to select one of the multipath delay structures A, multipath delay structure B and multipath delay structure C for target identification.

[0011] As a further improvement of the technical solution of the present invention, before step S3, the autocorrelation structure analysis method includes the steps of propagation loss calculation and signal amplitude attenuation analysis to determine whether multi-path delay structure A, multi-path delay structure B and multi-path delay structure C are applicable for target identification.

[0012] As a further improvement of the technical solution of the present invention, in step S3, when the target depth is greater than 280m and less than 500m, the multi-path delay structure A is applied for target identification; when the target depth is greater than 500m, the multi-path delay structure B is applied for target identification.

[0013] According to another aspect of the present invention, the present invention further provides an autocorrelation structure analysis system for underwater acoustic target recognition, comprising:

[0014] A modeling unit, used for modeling underwater acoustic signals based on sound ray theory;

[0015] A classification unit, used for extracting the target delay difference structure through autocorrelation analysis, so as to classify the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C;

[0016] The selection unit is used to analyze the sound source depth, the receiving depth and the relative distance between the sound source and the receiving, so as to select one of the multi-path delay structure A, the multi-path delay structure B and the multi-path delay structure C for target recognition.

[0017] As a further improvement of the technical solution of the present invention, the autocorrelation function of the received signal obtained by autocorrelation analysis is:

[0018] ,parameter is the autocorrelation function of the sound source signal, and the parameters is the autocorrelation function of the noise, the received signal The autocorrelation function exist A peak value appears when the peak value is extracted, and the delay value is estimated based on the peak value, wherein the eigenvalue line that is reflected by the sea surface only once is the eigenvalue line S, the eigenvalue line that is reflected by the seabed only once is the eigenvalue line B, the eigenvalue line that is reflected by the sea surface once and then by the seabed once is the eigenvalue line SB, and the eigenvalue line that is reflected by the seabed once and then by the sea surface once is the eigenvalue line BS. The multipath delay structure is composed of two groups of delay differences, namely, the eigenvalue line SB-B and the eigenvalue line SBS-BS. The multipath delay structure B is composed of one group of delay differences, namely, the eigenvalue line SD. The multipath delay structure C is composed of four groups of delay differences, namely, the eigenvalue line BS-B, the eigenvalue line BS-SB, the eigenvalue line SBS-B, and the eigenvalue line SBS-SB.

[0019] As a further improvement of the technical solution of the present invention, the selection unit analyzes the environment, track and transceiver configuration of the channel multipath delay structure respectively to select one of the multipath delay structure A, multipath delay structure B and multipath delay structure C for target identification.

[0020] As a further improvement of the technical solution of the present invention, the autocorrelation structure analysis system includes a calculation unit for calculating propagation loss and signal amplitude attenuation analysis to determine whether multi-path delay structure A, multi-path delay structure B and multi-path delay structure C are applicable for target identification.

[0021] According to another aspect of the present invention, the present invention further provides a computing device, comprising a processor and a memory coupled to the processor, wherein the memory is used to store program instructions, and the processor is used to execute the program instructions. When the program instructions are executed, the computing device executes an autocorrelation structure analysis method, wherein the autocorrelation structure analysis method comprises the following steps:

[0022] Conduct underwater acoustic signal modeling based on acoustic ray theory;

[0023] The autocorrelation analysis extracts the target delay difference structure for classifying the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C;

[0024] The sound source depth, the receiving depth and the relative distance between the sound source and the receiving are analyzed to select one of the multipath delay structures A, B and C for target identification.

[0025] Compared with the prior art, the present invention has at least the following beneficial effects:

[0026] 1. The present invention classifies the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C and analyzes the applicable conditions of the multipath delay structure A, multipath delay structure B and multipath delay structure C, which is beneficial to subsequent target identification based on the multipath delay structure A, multipath delay structure B and multipath delay structure C;

[0027] 2. The present invention can determine whether multi-path delay structure A, multi-path delay structure B and multi-path delay structure C are suitable for target identification by calculating propagation loss and analyzing signal amplitude attenuation.

[0028] Other effective effects of the present invention will be further disclosed in the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a process diagram of an autocorrelation structure analysis method according to a preferred embodiment of the present invention.

[0030] Figure 2 It is a schematic diagram of the intrinsic sound rays of the deep sea according to the present invention.

[0031] Figure 3 It is a schematic diagram of the delay difference of the multi-path delay structure of the present invention.

[0032] Figure 4 It is a schematic diagram of the autocorrelation analysis of the water surface target of the present invention.

[0033] Figure 5 It is a schematic diagram of the autocorrelation analysis of the underwater target of the present invention.

[0034] Figure 6 It is a schematic diagram of summer propagation loss in summer cross-section situation of the present invention.

[0035] Figure 7 It is a schematic diagram of the multi-path delay structure analysis of the present invention when the sea depth is 1000m.

[0036] Figure 8 It is a schematic diagram of the effect of sea depth on different time delay structures of the present invention.

[0037] FIG9( a ) is a schematic diagram showing the influence of relative distance on the multi-path time structure of a surface target and an underwater target according to the present invention when the relative distance is 5 km.

[0038] FIG9( b ) is a schematic diagram showing the influence of relative distance on the multi-path time structure of the surface target and the underwater target when the relative distance is 10 km according to the present invention.

[0039] FIG. 10 ( a ) is a schematic diagram showing the relationship between the limit sound source depth and the relative distance of the multipath delay structure A of the present invention.

[0040] FIG10( b ) is a schematic diagram showing the relationship between the limit sound source depth and the relative distance of the multipath delay structure B of the present invention.

[0041] Fig.11 It is a schematic diagram of the shortest and longest distance limits of the multi-path delay structure of the present invention.

[0042] FIG. 12 ( a ) is a schematic diagram showing the influence of the receiving depth on the multipath delay structure of the present invention when the receiving depth is 100 m.

[0043] FIG12( b ) is a schematic diagram showing the influence of the receiving depth on the multipath delay structure of the present invention when the receiving depth is 400 m.

[0044] FIG. 13( a ) is a schematic diagram showing the relationship between the limit sound source depth and the receiving depth of the multipath delay structure A of the present invention.

[0045] FIG13( b ) is a schematic diagram showing the relationship between the limit sound source depth and the receiving depth of the multipath delay structure B of the present invention.

[0046] Fig.14 It is a flow chart of an autocorrelation structure analysis method according to a preferred embodiment of the present invention.

[0047] Fig.15 It is a block diagram of an autocorrelation structure analysis system according to a preferred embodiment of the present invention.

[0048] Fig.16 It is a block diagram of a computing device according to a preferred embodiment of the present invention.

[0049] In the figure:

[0050] 10. Autocorrelation structure analysis system; 11. Modeling unit; 12. Classification unit; 13. Selection unit; 14. Calculation unit;

[0051] 20. Computing device; 21. Processor; 22. Memory; 23. Input device; 24. Output device. DETAILED DESCRIPTION

[0052] Before describing in detail any embodiment of the present invention, it should be understood that the present invention is not limited in its application to the construction and arrangement details of the components described below or illustrated in the following figures. The present invention can have other embodiments and can be practiced or carried out in various ways. In addition, it should be understood that the words and terms used here are for descriptive purposes and should not be considered restrictive. The use of "including" or "having" and its variations herein is intended to cover the items and their equivalents and additional items displayed below. Unless otherwise specified or limited, the terms "install", "connect", "support" and "couple" and their variations are widely used and cover direct installation and indirect installation, connection, support and connection. In addition, "connect" and "couple" are not limited to physical or mechanical connections or connections.

[0053] Furthermore, on the first aspect, in the disclosure of the present invention, the orientations or positional relationships indicated by terms such as “longitudinal”, “transverse”, “upper”, “lower”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside” and “outside” are based on the orientations or positional relationships shown in the accompanying drawings, which are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore the above terms cannot be understood as limitations on the present invention; on the second aspect, the term “one” should be understood as “at least one” or “one or more”, that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term “one” cannot be understood as a limitation on the quantity.

[0054] Exemplary Methods

[0055] The present invention provides an autocorrelation structure analysis method for underwater acoustic target recognition, which is used to analyze what kind of time delay structure difference should be used for underwater acoustic target recognition under current transmitting and receiving conditions.

[0056] Figure 1 The process of the autocorrelation structure analysis method of the present invention is shown, wherein the autocorrelation structure analysis method includes the following steps: S110, performing underwater acoustic channel modeling based on sound line theory; S120, extracting the target's delay difference structure through autocorrelation analysis; S130, calculating propagation loss and analyzing signal amplitude attenuation; S140, performing environmental analysis on the channel multi-path delay structure; S150, performing track analysis on the channel multi-path delay structure; S160, performing transceiver configuration analysis on the channel multi-path delay structure.

[0057] Specifically, refer to the attached Figure 1 In step S110, underwater acoustic channel modeling is performed based on sound ray theory.

[0058] In a specific example of the autocorrelation analysis structure of the present invention, the underwater acoustic channel is modeled using the BELLHOP model based on the sound line theory. In the specific application scenario of the autocorrelation structure analysis method of the present invention, the sound source position is the target position, and the receiving point position is the sonar position. The signal at the transceiver position is represented as the addition of several impulse functions, thereby obtaining the channel transfer function, which is used as the basic function of target recognition and the starting point for calculating the action parameters of the target recognition algorithm, wherein the transceiver position is the sonar position and the sound source position, that is, the signal at the sonar position and the sound source position can be represented as the addition of several impulse functions.

[0059] It should be noted that underwater targets move in the deep sea, while surface targets are closer to the sea surface. The different spatial geometric positions of the targets and the sonar have significantly different depths of the excited sound sources, and the corresponding delay structures also have huge differences. In a multi-path environment, the sound source position and the receiving point position are fixed, assuming that The eigenvalues ​​of the sound rays reach the receiving point, and the channel transfer function from the sound source to the receiving point can be expressed as , where the parameters Indicates the depth of the sound source, parameter Indicates the depth of the receiving point, parameter Indicates the distance between the sound source and the receiving point, parameter It is The amplitude of the eigenwave, parameter Indicates Each impulse function represents a different eigenwave, which represents a propagation path of the signal.

[0060] The channel transfer function can be expressed as the addition of several impulse functions. In a specific example of the autocorrelation structure analysis method of the present invention, taking the arrival time of the direct sound as a reference, the underwater acoustic channel model can be expressed as a vector: , under the underwater acoustic channel model, the received signal at the receiving point can be expressed as: .

[0061] The coordinate system of the deep sea eigenwave is established as Figure 2 As shown, the seabed depth parameter is , the sound speed parameter is , the vertical depth parameter between the sound source and the sea surface is , sonar and sea surface vertical depth parameters , is the horizontal distance parameter between the transmitter and the receiver (i.e., the distance parameter between the sound source and the sonar is ), where the distance parameter between the sound source and the sonar is , the distance parameter between the sound source and the image sonar is , the distance parameter between the image sound source and the sonar is , the distance parameter between the image source and the image sonar is According to the geometric relationship, we can conclude that:

[0062] ;

[0063] ;

[0064] ;

[0065] .

[0066] Assuming that the sea surface is a free interface, the time delay difference of each eigenwave can be expressed as:

[0067] ;

[0068] ;

[0069] ;

[0070] ;

[0071] ;

[0072] .

[0073] In the deep sea reflection zone environment, according to The approximation of can simplify the delay difference to:

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] ;

[0079] .

[0080] It should be noted that, in a specific example of the autocorrelation structure analysis method of the present invention, it is stipulated that if the intrinsic sound line passes through the sea surface reflection only once, it is represented by the symbol S; if it passes through the seabed reflection only once, it is represented by the symbol B; if it passes through the sea surface reflection once and then the seabed reflection once, it is represented by the symbol SB; if it passes through the seabed reflection once and then the sea surface reflection once, it is represented by the symbol BS, wherein the symbol R represents the intrinsic sound line.

[0081] Due to the significant difference between surface targets and underwater targets, the depth of the exciting sound source is significantly different. Therefore, the depth information of the unknown target indicates whether the target is a surface target or an underwater target. It is implicit in the relative time delay difference structure, so the autocorrelation structure analysis method of the present invention needs to analyze the differences in the time delay difference structure.

[0082] Reference Figure 1 In step S120, the target delay difference structure is extracted by autocorrelation analysis.

[0083] In a specific example of the autocorrelation structure analysis method of the present invention, the autocorrelation structure analysis method of the present invention is used to extract the time delay difference structure of a preset target during target identification, and the system impulse response function and the time delay difference curve under given input conditions are calculated, thereby utilizing the difference analysis of the time delay structure to identify and classify surface targets and underwater targets.

[0084] The core of the autocorrelation structure analysis method of the present invention is to measure the similarity between the signal and itself under different time delays to reveal the change pattern of the signal on the time axis. The sound source depth information of the target is implicit in the delay difference structure, so it is necessary to use the correlation delay estimation method to extract the delay difference structure of the channel. Based on the delay differences formed by different paths, including their relative delays and energy contributions, the autocorrelation structure analysis method of the present invention can construct a delay structure image of the underwater acoustic channel.

[0085] Assume signal With noise Uncorrelated, autocorrelation analysis to obtain the received signal The autocorrelation function of is: , where the parameters is the autocorrelation function of the sound source signal, and the parameters is the autocorrelation function of the noise. Analysis shows that under multipath conditions, the received signal The autocorrelation function exist A peak value occurs when the delay value is estimated based on the peak value after the peak value is extracted.

[0086] It should be noted that, in a specific example of the autocorrelation structure analysis method of the present invention, as described above, it is specified that if the eigenvalue line passes through only one sea surface reflection, it is represented by the symbol S, if it passes through only one seabed reflection, it is represented by the symbol B, if it passes through one sea surface reflection and then one seabed reflection, it is represented by the symbol SB, and if it passes through one seabed reflection and then one sea surface reflection, it is represented by the symbol BS. In a specific example of the autocorrelation structure analysis method of the present invention, as Figure 3 As shown, the multipath delay structure A consists of two groups of delay differences, namely, the intrinsic sound line SB-B and the intrinsic sound line SBS-BS; the multipath delay structure B consists of one group of delay differences, namely, the intrinsic sound line SD; and the multipath delay structure C consists of four groups of delay differences, namely, the intrinsic sound line BS-B, the intrinsic sound line BS-SB, the intrinsic sound line SBS-B, and the intrinsic sound line SBS-SB.

[0087] Under simulation conditions, the autocorrelation diagram of surface targets and underwater targets is shown in the figure below. Figure 4 and Figure 5 It should be noted that the present invention Figure 4 and Figure 5 The multipath delay structure diagram shown in the figure is only used as an example to illustrate the content of the autocorrelation structure analysis method of the present invention, and does not mean that every setting of the transmission and reception configuration relationship will form a multipath delay structure with exactly the same delay value. Therefore, the present invention Figure 4 and Figure 5 The multipath delay structure diagram shown in the figure does not limit the content and scope of the autocorrelation structure analysis method of the present invention.

[0088] Reference Figure 1 , the step S130, propagation loss calculation and signal amplitude attenuation analysis.

[0089] Due to the existence of the shadow area of ​​the underwater acoustic channel, the direct sound in the applicable scenario of the autocorrelation structure analysis method of the present invention experiences greater propagation loss, and the reflected sound waves passing through the seabed may maintain higher energy due to the energy focusing or smaller attenuation of the reflection process. Therefore, the autocorrelation structure analysis method of the present invention needs to calculate the propagation loss of the four intrinsic sound lines that experience the seabed reflection, and analyze the amplitude attenuation of the multi-path delay structure formed between each two.

[0090] In a specific example of the autocorrelation structure analysis method of the present invention, the receiving depth (Rd) of the BELLHOP model is set at 150m, the sound velocity adopts the Munk profile, and the propagation loss in the area with an ocean depth of 1600m is as follows: Figure 6 shown.

[0091] The calculation of propagation loss gives the attenuation of the propagation energy of a certain frequency of sound waves with distance under different transmission and reception configurations (i.e., different target locations and sonar locations). In a specific example of the autocorrelation structure analysis method of the present invention, when the sound wave with a frequency of 100 Hz is at a source depth of 20 m, the propagation loss at a distance of 5 km is about 30 dB when the receiving depth is 100 m, the propagation loss at a distance of 5 km is about 10 dB when the receiving depth is 200 m, the propagation loss at a distance of 5 km is about 10 dB when the receiving depth is 500 m, and the propagation loss at a distance of 10 km is about 10 dB when the receiving depth is 800 m, and the propagation loss at a distance of 17 km is about 10 dB.

[0092] When the sound source depth is 20m and the relative distance between the target and the sonar is 5km, the absolute time delay and amplitude of each eigenvalue line arriving at the receiving point in turn, the attenuation amplitude calculation of the surface target is shown in Table 1, and the attenuation amplitude calculation of the underwater target is shown in Table 2.

[0093] Table 1 Analysis of multi-path delay structure of surface target at 5 km distance

[0094]

[0095] The delay difference of the intrinsic sound line BS-B is 0.105s, and the delay difference of the intrinsic sound line SBS-SB is 0.107s. As the depth of the sound source increases, the delay difference changes slightly. The multipath delay structure is a delay structure shared by surface targets and underwater targets. The delay difference of the intrinsic sound line SB-B is 0.013s, and the delay difference of the intrinsic sound line SBS-BS is 0.015s. This delay difference cannot be reflected in the autocorrelation image of the surface target, that is, the multipath delay structure A cannot be used in the process of surface target recognition.

[0096] Table 2 Analysis of multi-path delay structure of underwater target at 5 km distance

[0097]

[0098] The delay difference of the intrinsic sound line BS-B is 0.105s, and the delay difference of the intrinsic sound line SBS-SB is 0.109s. As the depth of the sound source increases, the delay difference changes slightly. The multipath delay structure is a delay structure shared by surface targets and underwater targets. The delay difference of the intrinsic sound line SB-B is 0.068s, and the delay difference of the intrinsic sound line SBS-BS is 0.072s. The delay difference cannot be reflected in the autocorrelation image of the surface target, but the autocorrelation image of the underwater target can extract the delay difference. That is, the multipath delay structure A can be used in the process of underwater target recognition.

[0099] It should be noted that the analysis of the specific example of the autocorrelation structure analysis method of the present invention only illustrates how to perform propagation loss calculation and amplitude attenuation calculation. The input parameters used are all variable parameters, which can be changed accordingly according to needs in actual application. After the input parameters are changed, the calculation and analysis methods are as shown in the above specific example of the autocorrelation structure analysis method of the present invention. The propagation loss calculation and signal amplitude attenuation analysis performed in step S130 can calculate the actual applicable distance and conditions of the autocorrelation structure analysis method of the present invention.

[0100] Reference Figure 1 , the step S140, performs environmental analysis on the channel multipath delay structure.

[0101] The autocorrelation structure analysis method of the present invention classifies the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C. Under different applicable conditions, different multipath delay structures are used for different target recognition tasks. Therefore, when performing target recognition, the autocorrelation structure analysis method of the present invention needs to use environmental analysis to provide a basis for target recognition in a specific environment.

[0102] Specifically, in a certain ocean environment, since the sound speed profile is seasonal, given a constant sound speed of 1512 m / s, a relative distance of 5 km, a receiving depth of 100 m, and a sea depth of 1000 m, the multipath delay structure is calculated as follows: Figure 7 As shown in the figure, the analysis gives the relationship between the time delay difference caused by the sound path difference of different eigenwaves and the depth of the sound source. Increase, the main lobe width decreases, generally take , so when the signal processing bandwidth When the frequency is greater than 100Hz, the minimum delay difference can be separated is 10ms.

[0103] The influence of sea depth conditions on multipath delay structure is quantitatively analyzed. Figure 8 As shown. Combined Figure 7 and Figure 8The calculation and analysis show that: (1) The ocean depth does not affect the intrinsic sound line SD, but only affects the four sound lines reflected by the seabed. The multipath delay structure A is composed of two groups of delay differences, the intrinsic sound line SB-B and the intrinsic sound line SBS-BS. The increase in ocean depth will cause a new multipath delay structure A to appear in the autocorrelation image of the underwater target, but the autocorrelation image of the surface target cannot extract the multipath delay structure A. Therefore, the increase in ocean depth will increase the difference in the autocorrelation images of the surface target and the underwater target, which is conducive to target classification; (2) The analysis found that the ocean depth has the greatest impact on the delay difference between the intrinsic sound lines DB. When the sea depth increases from 1000m to 3000m, the delay difference increases from 0.2s to 1.8s; (3) The delay difference caused by the intrinsic sound line SD can only be used as a reference for target identification when the sound source depth is below 280m.

[0104] It should be noted that the above specific analysis of the autocorrelation structure analysis method of the present invention only illustrates how to calculate the applicability of the sea depth conditions. The input sound velocity profile and sea depth conditions are adjusted according to the specific environment. The calculation and analysis methods are as shown in the present invention to calculate the applicable depth conditions of the autocorrelation structure analysis method of the present invention.

[0105] Reference Figure 1 , the step S150, performs track analysis on the channel multipath delay structure.

[0106] The autocorrelation structure analysis method of the present invention classifies multipath delay structures into multipath delay structure A, multipath delay structure B and multipath delay structure C. Under different applicable conditions, different delay structures are used for target recognition tasks. Therefore, the autocorrelation structure analysis method uses track analysis to provide a basis for target recognition in a specific environment.

[0107] In a specific example of the autocorrelation structure analysis method of the present invention, due to the signal bandwidth Increase, the main lobe width decreases, generally take , so when the signal processing bandwidth When it is greater than 100Hz, the minimum separable delay difference is 10ms. The influence of the relative distance between the target and the sonar on the multipath delay structure is quantitatively analyzed. The simulation is based on a constant sound speed of 1512m / s, a sea depth of 1600m, and a receiving depth of 150m. Figure 9 (a) shows the multipath delay structure at different sound source depths when the relative distance is 5km, and Figure 9 (b) shows the multipath delay structure at different sound source depths when the relative distance is 10km.

[0108] The analysis shows that: (1) When the relative distance between the transmitter and the receiver is 5 km, when the sound source depth changes, the delay difference of the intrinsic sound line BD remains unchanged. When the sound source depth exceeds 10 m, a new multipath delay structure A will appear. The new multipath delay structure A is generated by the delay differences of the intrinsic sound line SB-B and the intrinsic sound line SBS-BS. When the sound source depth exceeds 250 m, a new multipath delay structure B will appear. The new multipath delay structure B is generated by the delay difference of the intrinsic sound line SD. (2) When the relative distance between the transmitter and the receiver is 10 km, when the sound source depth changes, the delay difference of the intrinsic sound line BD remains unchanged. When the sound source depth exceeds 25 m, a new multipath delay structure A will appear. The new multipath delay structure A is generated by the delay differences of the intrinsic sound line SB-B and the intrinsic sound line SBS-BS. When the sound source depth exceeds 500 m, a new multipath delay structure B will appear. The new multipath delay structure B is generated by the delay difference of the intrinsic sound line SD.

[0109] Therefore, there is a limit sound source depth for the emergence of a new multipath time delay structure. Once this depth is exceeded, a new multipath time delay structure will appear, and this limit sound source depth is related to the relative distance between the target and the sonar.

[0110] The autocorrelation structure analysis method of the present invention performs a limit relative distance test of the multi-path delay structure A as shown in FIG10 (a). When the separable minimum delay difference is 10 ms, the limit sound source depth of the multi-path delay structure A is linearly related to the relative distance between the target and the sonar. When the relative distance is 1 km, a new multi-path delay structure A will appear when the sound source depth exceeds 7.8 m. When the relative distance is 10 km, the limit sound source depth is 23.8 m to 25.9 m. The farther the relative distance is, the greater the limit depth is, and the more unfavorable it is for target classification. When the separable minimum delay difference is 5 ms, the limit sound source depth of the multi-path delay structure A is also linearly related to the relative distance between the target and the sonar. When the relative distance is 1 km, a new multi-path delay structure A will appear when the sound source depth exceeds 3.9 m. When the relative distance is 10 km, the limit sound source depth is 11.8 m to 12.9 m.

[0111] The autocorrelation structure analysis method of the present invention performs a relative distance test on the multi-path delay structure B as shown in FIG10 (b). When the separable minimum delay difference is 10 ms, the maximum sound source depth at which the multi-path delay structure B appears is positively correlated with the relative distance between the target and the sonar. When the relative distance is 1 km, the multi-path delay structure B will appear when the sound source depth exceeds 50 m, and when the relative distance is 10 km, the maximum sound source depth is 504 m. When the separable minimum delay difference is 5 ms, when the relative distance is 1 km, the multi-path delay structure B will appear when the sound source depth exceeds 25 m. Similarly, when the relative distance is 10 km, the maximum sound source depth is 252 m.

[0112] Based on the calculation and analysis of the above cases, we can get Fig.11 .like Fig.11 As shown, as the distance increases, the delay difference will gradually decrease to the point where it cannot be distinguished and extracted. The autocorrelation structure analysis method of the present invention requires that the surface target and the underwater target have different multi-path delay structures, and the delay difference can be extracted. Therefore, combined with theoretical simulation calculations, the surface target and underwater target classification have delay structure characteristics and the observable distance range is limited to 2.5km at the closest distance and 12.5km at the farthest distance.

[0113] It is worth mentioning that the analysis of the autocorrelation structure analysis method of the present invention only illustrates how to calculate the applicable limit distance of the algorithm. In actual application, it is necessary to adjust the input environment and receiving and sending conditions according to specific experiments. The calculation and analysis methods are as shown in the autocorrelation structure analysis method of the present invention to calculate the applicable limit distance conditions of the algorithm.

[0114] Reference Figure 1 In step S160, the transceiver configuration is analyzed for the channel multipath delay structure.

[0115] The autocorrelation structure analysis method of the present invention classifies the delay structure into multi-path delay structure A, multi-path delay structure B and multi-path delay structure C. Under different applicable conditions, different delay structures are used for target recognition tasks. Therefore, it is necessary to adopt what kind of transceiver configuration conditions to provide a basis for target recognition in a specific environment.

[0116] The specific case simulation is based on a constant sound speed of 1512m / s, a sea depth of 1600m, and a relative distance of 5km. Figure 12 (a) shows the multipath delay structure of different sound source depths at a receiving depth of 100m, and Figure 12 (b) shows the multipath delay structure of different sound source depths at a receiving depth of 400m. Keeping other environmental parameters unchanged, observe the impact of the receiving depth on the multipath delay structure of surface targets and underwater targets. Since the multipath delay structure C exists at any receiving depth, the discussion on the receiving depth limit only considers the multipath delay structure A and the multipath delay structure B.

[0117] Since the signal bandwidth Increase, the main lobe width decreases, generally take , so when the signal processing bandwidth When the frequency is greater than 100Hz, the minimum separable delay difference is 10ms. The influence of the receiving depth on the multipath delay structure is quantitatively analyzed: (1) When the sound source depth changes, the inter-cluster delay difference structure remains unchanged, but the intrinsic sound line S-BS increases significantly with the increase of the receiving depth. This characteristic is not suitable for target identification, but can be used to calibrate the receiving depth range; (2) The multipath delay structure A (generated by the intrinsic sound line B-SB and the intrinsic sound line BS-SBS) is less affected by the receiving depth; (3) The multipath delay structure B (generated by the direct sound intrinsic sound line DS) is greatly affected by the receiving depth. When the receiving depth increases from 100m to 400m, the extreme sound source depth decreases from 400m to 100m. Therefore, the greater the receiving depth, the more conducive it is to use the multipath delay structure B to identify surface targets and underwater targets.

[0118] There is a limit sound source depth when a new multipath delay structure appears. Once this depth is exceeded, a new multipath delay structure will appear, and this limit sound source depth is related to the receiving depth.

[0119] Therefore, the test of the receiving depth and the limit sound source depth of the multipath delay structure can obtain Figure 12 (a) and Figure 12 (b). The relationship between the sound source depth of the multipath delay structure A and the limit receiving depth is shown in Figure 12 (a), and the analysis is as follows: (1) When the separable minimum delay difference is 10ms, the limit sound source depth of the multipath delay structure A is linearly related to the relative distance between the sound source and the sonar. When the receiving depth is 100m, the limit sound source depth of the multipath delay structure A is 13.7m-14.3m, and when the receiving depth is 1000m, the limit sound source depth is 11.7m-18.7m; (2) When the separable minimum delay difference is 5ms, when the receiving depth is 100m, the limit sound source depth of the multipath delay structure A is 6.8m-7.1m. When the receiving depth is 1000m, the limit sound source depth is 5.8m-9.3m.

[0120] Therefore, the analysis shows that the change in receiving depth has little effect on the overall multipath delay structure A, that is, due to the spatial geometric structure, the change in receiving depth has little effect on the emergence of multipath delay structure A.

[0121] As shown in Figure 12 (b), the simulation analysis of the appearance of multipath delay structure B is as follows: (1) When the separable minimum delay difference is 10 ms, the limit sound source depth of multipath delay structure B is negatively correlated with the relative distance between the sound source and the sonar. When the receiving depth is 100 m, the limit sound source depth of multipath delay structure B is 379 m, and when the receiving depth is 1000 m, the limit sound source depth is 38 m; (2) When the separable minimum delay difference is 5 ms, when the receiving depth is 100 m, the limit sound source depth of multipath delay structure B is 189 m. When the receiving depth is 1000 m, the limit sound source depth is 19 m.

[0122] Therefore, the receiving depth has a great impact on the multipath delay structure B. When the receiving depth increases from 100m to 1000m, the limit sound source depth of the multipath delay structure B decreases from 379m to 38m, and from 189m to 19m. The increase in receiving depth is conducive to the use of multipath delay structure B for surface target and underwater target identification.

[0123] In summary, when the sonar is placed at a shallow depth, it is more suitable to use multipath delay structure A for surface target identification, and when the sonar is placed at the seabed depth, multipath delay structure B can be used for underwater target identification. Multipath delay structure C can be used for surface and underwater target identification at any receiver depth.

[0124] In other words, refer to the Fig.14 The present invention provides an autocorrelation structure analysis method for underwater acoustic target recognition, wherein the autocorrelation structure analysis method comprises the following steps:

[0125] S1, modeling of underwater acoustic signals based on sound ray theory;

[0126] S2, autocorrelation analysis extracts the target delay difference structure for classifying the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C;

[0127] S3, analyzing the sound source depth, the receiving depth, and the relative distance between the sound source and the receiving, so as to select one of the multipath delay structure A, the multipath delay structure B, and the multipath delay structure C for target identification. By classifying the multipath delay structure into the multipath delay structure A, the multipath delay structure B, and the multipath delay structure C and analyzing the applicable conditions of the multipath delay structure A, the multipath delay structure B, and the multipath delay structure C, it is beneficial to perform target identification based on the multipath delay structure A, the multipath delay structure B, and the multipath delay structure C in the subsequent process.

[0128] Preferably, in step S3, the environment, track and transceiver configuration of the channel multipath delay structure are analyzed respectively to select one of the multipath delay structure A, multipath delay structure B and multipath delay structure C for target identification. Figure 1 In step S140, the influence of the environment on the multi-path delay structure of the channel is analyzed, and in step S150, the influence of the track on the multi-path delay structure is analyzed. When the target depth is greater than 280m and less than 500m, the multi-path delay structure A is applied for target identification, and when the target depth is greater than 500m, the multi-path delay structure B is applied for target identification. In step S160, the influence of the transceiver configuration on the channel multi-path structure is analyzed.

[0129] Exemplary Systems

[0130] Reference Fig.15 Another aspect of the present invention provides an autocorrelation structure analysis system 10 for underwater acoustic target recognition, wherein the autocorrelation structure analysis system 10 includes a modeling unit 11, a classification unit 12 and a selection unit 13, wherein the modeling unit 11 is used to perform underwater acoustic signal modeling based on the sound line theory; the classification unit 12 is used to extract the delay difference structure of the target through autocorrelation analysis, so as to classify the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C; the selection unit 13 is used to analyze the sound source depth, the receiving depth and the relative distance between the sound source and the receiving, so as to select one of the multipath delay structure A, multipath delay structure B and multipath delay structure C for target recognition; by classifying the multipath delay structure into the multipath delay structure A, multipath delay structure B and multipath delay structure C and analyzing the applicable conditions of the multipath delay structure A, multipath delay structure B and multipath delay structure C, it is beneficial to perform target recognition based on the multipath delay structure A, multipath delay structure B and multipath delay structure C in the subsequent process.

[0131] In a specific example of the autocorrelation structure analysis system 10 of the present invention, the autocorrelation function of the received signal obtained by autocorrelation analysis is:

[0132] ,parameter is the autocorrelation function of the sound source signal, and the parameters is the autocorrelation function of the noise, the received signal The autocorrelation function exist A peak value appears when the peak value is extracted, and the delay value is estimated based on the peak value, wherein the eigenvalue line that is reflected by the sea surface only once is the eigenvalue line S, the eigenvalue line that is reflected by the seabed only once is the eigenvalue line B, the eigenvalue line that is reflected by the sea surface once and then by the seabed once is the eigenvalue line SB, and the eigenvalue line that is reflected by the seabed once and then by the sea surface once is the eigenvalue line BS. The multipath delay structure is composed of two groups of delay differences, namely, the eigenvalue line SB-B and the eigenvalue line SBS-BS. The multipath delay structure B is composed of one group of delay differences, namely, the eigenvalue line SD. The multipath delay structure C is composed of four groups of delay differences, namely, the eigenvalue line BS-B, the eigenvalue line BS-SB, the eigenvalue line SBS-B, and the eigenvalue line SBS-SB.

[0133] In a specific example of the autocorrelation structure analysis system 10 of the present invention, the selection unit 13 analyzes the environment, track and transceiver configuration of the channel multipath delay structure respectively to select one of the multipath delay structure A, multipath delay structure B and multipath delay structure C for target identification.

[0134] Continue to refer to the attached Fig.15The autocorrelation structure analysis system 10 further includes a calculation unit 14 for calculating propagation loss and signal amplitude attenuation analysis to determine whether multi-path delay structure A, multi-path delay structure B and multi-path delay structure C are applicable for target identification.

[0135] Exemplary Devices

[0136] like Fig.16 As shown, a computing device 20 according to a preferred embodiment of the present invention includes one or more processors 21 and one or more memories 22 connected to the processors 21 .

[0137] The processor 21 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the computing device 20 to perform desired functions.

[0138] The memory 22 may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 21 may run the program instructions to implement the autocorrelation structure analysis method of the present invention described above and / or other desired functions. Specifically, when the processor 21 executes the program instructions stored in the memory 22, the computing device 20 executes the autocorrelation structure analysis method, wherein the autocorrelation structure analysis method includes the following steps:

[0139] Conduct underwater acoustic signal modeling based on acoustic ray theory;

[0140] The autocorrelation analysis extracts the target delay difference structure for classifying the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C;

[0141] The sound source depth, the receiving depth and the relative distance between the sound source and the receiving are analyzed to select one of the multipath delay structures A, B and C for target identification.

[0142] In a specific example of the computing device 20 of the present invention, the computing device 20 may also include an input device 23 and an output device 24, and these components are interconnected through a bus system and / or other forms of connection mechanisms (not shown). The input device 23 may include, for example, a keyboard, a mouse, etc. The output device 24 may output various information to the outside. The output device 24 may include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, etc.

[0143] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

Claims

1. An autocorrelation structure analysis method for underwater acoustic target recognition, characterized in that: The autocorrelation structure analysis method comprises the following steps: S1, modeling of underwater acoustic signals based on sound ray theory; S2, autocorrelation analysis extracts the target delay difference structure, which is used to classify the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C; S3, analyzing the sound source depth, the receiving depth, and the relative distance between the sound source and the receiving device to select one of the multipath delay structure A, the multipath delay structure B, and the multipath delay structure C for target recognition; In step S2, the autocorrelation function of the received signal obtained by autocorrelation analysis is: ,parameter is the autocorrelation function of the sound source signal, and the parameters is the autocorrelation function of the noise, and the parameters Indicates The amplitude of the eigenwave, parameter Indicates the number of eigenwaves, parameter Indicates The transmission delay of the eigenwave line, the received signal The autocorrelation function exist A peak value appears when the peak value is extracted, and the delay value is estimated based on the peak value, wherein the eigenacoustic line that has only been reflected by the sea surface once is the eigenacoustic line S, the eigenacoustic line that has only been reflected by the seabed once is the eigenacoustic line B, the direct eigenacoustic line is the eigenacoustic line D, the eigenacoustic line that first passes through the sea surface reflection once and then through the seabed reflection once is the eigenacoustic line SB, and the eigenacoustic line that first passes through the seabed reflection once and then through the sea surface reflection once is the eigenacoustic line BS. The multipath delay structure A is composed of two groups of delay differences, namely, the eigenacoustic line SB-B and the eigenacoustic line SBS-BS. The multipath delay structure B is composed of one group of delay differences, namely, the eigenacoustic line SD. The multipath delay structure C is composed of four groups of delay differences, namely, the eigenacoustic line BS-B, the eigenacoustic line BS-SB, the eigenacoustic line SBS-B, and the eigenacoustic line SBS-SB.

2. The autocorrelation structure analysis method according to claim 1, characterized in that: In step S3, the environment, track and transceiver configuration of the channel multipath delay structure are analyzed respectively to select one of the multipath delay structure A, multipath delay structure B and multipath delay structure C for target identification.

3. The autocorrelation structure analysis method according to claim 1, characterized in that: Before step S3, the autocorrelation structure analysis method includes the steps of propagation loss calculation and signal amplitude attenuation analysis, which are used to determine whether multipath delay structure A, multipath delay structure B and multipath delay structure C are applicable for target identification.

4. The autocorrelation structure analysis method according to claim 1, characterized in that: In step S3, when the target depth is greater than 280m and less than 500m, multi-path delay structure A is applied to identify the target, and when the target depth is greater than 500m, multi-path delay structure B is applied to identify the target.

5. The autocorrelation structure analysis system for underwater acoustic target recognition is characterized by: include: A modeling unit, used for modeling underwater acoustic signals based on sound ray theory; A classification unit, used for extracting the target delay difference structure through autocorrelation analysis, and for classifying the multipath delay structure into multipath delay structure A, multipath delay structure B and multipath delay structure C; A selection unit is used to analyze the sound source depth, the receiving depth and the relative distance between the sound source and the receiving, so as to select one of the multi-path delay structure A, the multi-path delay structure B and the multi-path delay structure C for target recognition; The autocorrelation function of the received signal obtained by autocorrelation analysis is: ,parameter is the autocorrelation function of the sound source signal, and the parameters is the autocorrelation function of the noise, and the parameters Indicates The amplitude of the eigenwave, parameter Indicates the number of eigenwaves, parameter Indicates The transmission delay of the eigenwave line, the received signal The autocorrelation function exist A peak value appears when the peak value is extracted, and the delay value is estimated based on the peak value, wherein the eigenacoustic line that has only been reflected by the sea surface once is the eigenacoustic line S, the eigenacoustic line that has only been reflected by the seabed once is the eigenacoustic line B, the direct eigenacoustic line is the eigenacoustic line D, the eigenacoustic line that first passes through the sea surface reflection once and then through the seabed reflection once is the eigenacoustic line SB, and the eigenacoustic line that first passes through the seabed reflection once and then through the sea surface reflection once is the eigenacoustic line BS. The multipath delay structure A is composed of two groups of delay differences, namely, the eigenacoustic line SB-B and the eigenacoustic line SBS-BS. The multipath delay structure B is composed of one group of delay differences, namely, the eigenacoustic line SD. The multipath delay structure C is composed of four groups of delay differences, namely, the eigenacoustic line BS-B, the eigenacoustic line BS-SB, the eigenacoustic line SBS-B, and the eigenacoustic line SBS-SB.

6. The autocorrelation structure analysis system according to claim 5, characterized in that: The selection unit analyzes the environment, track and transceiver configuration of the channel multipath delay structure respectively to select one of the multipath delay structure A, multipath delay structure B and multipath delay structure C for target identification.

7. The autocorrelation structure analysis system according to claim 5, characterized in that: The autocorrelation structure analysis system includes a calculation unit for calculating propagation loss and signal amplitude attenuation analysis, and for determining whether multipath delay structure A, multipath delay structure B and multipath delay structure C are applicable for target identification.

8. A computing device, characterized in that It comprises a processor and a memory coupled to the processor, wherein the memory is used to store program instructions, and the processor is used to execute the program instructions. When the program instructions are executed, the computing device executes the autocorrelation structure analysis method as described in any one of claims 1 to 4.

Images