Multi-channel adaptive sonar positioning and communication integration method for complex underwater acoustic environment

Abstract

The application discloses a multi-channel adaptive sonar positioning and communication integrated method for complex underwater acoustic environment, and relates to the field of positioning and communication, and comprises the following steps: step 1, deploying a plurality of underwater listening sensors at different positions of a water area to be monitored, and acquiring performance parameters and position coordinates of each sensor in real time; step 2, based on the performance parameters and position coordinates of each underwater listening sensor, defining an initial data trust boundary for each underwater listening sensor as an effective monitoring range of the underwater listening sensor in the water area; step 3, performing spatial superposition analysis on the initial data trust boundaries of all underwater listening sensors; detecting and marking a monitored conflict area, further predicting an abnormal edge diffusion amplitude through a pre-trained correction model, solving the multipath interference and conflict problem, realizing accurate correction of conflict data, and then improving model consistency and positioning accuracy, and improving the anti-interference ability of the system in a complex environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of positioning communication technology, in particular to a multi-channel adaptive sonar positioning and communication integration method for complex underwater acoustic environment. BACKGROUND

[0002] The complex underwater acoustic environment such as shallow sea area, deep sea valley or nearshore complex terrain area has environmental characteristics such as multipath effect, strong background noise, non-uniform medium distribution and time-varying channel characteristics, which further leads to challenges in positioning accuracy and communication reliability of traditional sonar system, and higher requirements for high-precision positioning and stable data transmission are put forward in the fields of modern marine resource exploration, underwater unmanned vehicle navigation and underwater network communication, especially under dynamic and complex hydrological conditions, the high integration of positioning and communication functions is required to reduce the complexity of equipment and spectrum resource occupation.

[0003] However, the traditional sonar system is usually deployed based on the preset sensor coverage range, and lacks the adaptability to the dynamic changes of hydrological environment, when the water flow, temperature-salinity gradient or terrain changes, the fixed coverage range may cause monitoring blind area or data redundancy, when multiple sensors work cooperatively, the data collected by adjacent sensors is easy to have spatial feature conflict due to multipath effect or environmental interference, and the traditional method lacks effective conflict detection and correction mechanism, which leads to distortion of three-dimensional hydrological model or increase of positioning error, the traditional sonar system usually adopts static deployment strategy, once the sensor position is determined, it is difficult to dynamically adjust according to the environmental changes or data conflict, which leads to insufficient fault tolerance of the system in complex underwater acoustic environment. SUMMARY

[0004] (I) Technical problems to be solved

[0005] In view of the above-mentioned shortcomings of the prior art, the present application provides a multi-channel adaptive sonar positioning and communication integration method for complex underwater acoustic environment, which can effectively solve the problems of the prior art.

[0006] (II) Technical scheme

[0007] In order to achieve the above purpose, the present application is realized by the following technical scheme:

[0008] The present application discloses a multi-channel adaptive sonar positioning and communication integration method for complex underwater acoustic environment, comprising the following steps:

[0009] Step 1: deploying a plurality of hydrophone sensors at different positions of the water area to be monitored, and acquiring the performance parameters and position coordinates of each sensor in real time;

[0010] Step 2: Based on the performance parameters and location coordinates of each hydrophone sensor, an initial data trust boundary is drawn for each hydrophone sensor as its effective monitoring range in the water area;

[0011] Step 3: Perform spatial superposition analysis on the initial data trust boundaries of all hydrophone sensors, adjust the sensor deployment location until all data trust boundaries cover the entire water area to be monitored, and set the adjacent data trust boundaries to allow overlapping areas;

[0012] Step 4: Extract the standard edge features from the original sensor data of a single hydrophone sensor, construct a local three-dimensional hydrological model, extract the feature data of adjacent hydrophone sensors and convert it into three-dimensional edge data, and perform spatial alignment verification with the existing model. Compare adjacent sensor data in the allowed overlapping area. If there is a spatial feature conflict, mark it as an abnormal edge feature;

[0013] Step 5: Input the abnormal edge feature into the pre-trained correction model, which is generated based on historical hydrological data training. The correction model predicts the diffusion amplitude of the abnormal edge in three-dimensional space, and corrects the three-dimensional data edge of the conflict area according to the prediction result;

[0014] Step 6: Verify the reasonableness of the corrected data. If the correction is effective, generate the orientation adjustment instructions for the associated hydrophone sensor based on the correction amount and the corresponding data trust boundary;

[0015] Step 7: Based on the corrected three-dimensional hydrological model, select the least disturbed sound wave transmission path, embed communication coding in the positioning data, and perform same-channel transmission of positioning signals and communication signals.

[0016] Further, the hydrophone sensor in step 1 uses a multi-channel wideband transducer array, and its performance parameters include signal-to-noise ratio, frequency response range, and directivity angle. When deployed, the sensor pitch angle is dynamically adjusted by the underwater robot arm to match the terrain.

[0017] Further, the calculation process of the data trust boundary radius in step 2 includes:

[0018] Obtain the transmission sound source intensity level, directivity pattern gain parameter, frequency-dependent sound energy attenuation coefficient, sound wave propagation path distance, and underwater environmental noise spectrum distribution of the current hydrophone sensor, and calculate the sound wave propagation attenuation factor;

[0019] Multiply the transmission sound source intensity level, directivity pattern gain parameter, and sound wave propagation attenuation factor to obtain the joint gain value, and calculate the environmental interference factor;

[0020] Divide the joint gain value by the environmental interference factor to obtain the preliminary effective monitoring radius;

[0021] The preliminary effective monitoring radius is multiplied by a preset signal-to-noise ratio threshold adjustment coefficient to finally determine the data trust edge domain radius range of the sensor.

[0022] Further, the standard edge feature extraction process in step 4 is: performing frequency domain wavelet denoising on the original sonar signal, extracting time-frequency energy mutation points through Hilbert-Huang transform, and mapping the mutation points into three-dimensional space edge vectors by using DBSCAN clustering algorithm.

[0023] Further, the allowed overlapping area of adjacent data trust edge domains in step 4 needs to meet the minimum overlapping area requirement, and the determination rule of the minimum overlapping area is:

[0024] The larger data trust edge domain radius of the two adjacent hydrophone sensors is selected as the reference value;

[0025] According to the depth type of the water area, the environmental complexity weight coefficient is selected: when the water depth is less than a set critical value, the shallow water coefficient is used, and when the water depth is greater than the critical value, the deep water coefficient is used;

[0026] The reference value is multiplied by the corresponding environmental complexity weight coefficient to obtain the core parameter;

[0027] Finally, it is determined that the minimum overlapping area should be greater than or equal to a numerical range of one-half of the core parameter.

[0028] Further, in step 4, the feature data acquisition process of the adjacent hydrophone sensors is: using the time difference and phase difference of the received signals of each hydrophone sensor to calculate the spatial correlation and generate three-dimensional edge data; the conversion process uses orthogonal frequency division multiplexing technology to decompose the multi-channel signal into multiple sub-carrier channels.

[0029] Further, the correction model in step 5 adopts a double-branch graph convolutional neural network, and a historical hydrological database containing several terrains is constructed in advance, wherein:

[0030] The first branch inputs the three-dimensional coordinates and acoustic impedance values of the abnormal edge, and outputs the diffusion gradient tensor;

[0031] The second branch inputs the sound propagation loss matrix of similar terrains in the historical hydrological database, and outputs the terrain distortion compensation parameter;

[0032] The two branch results are fused to generate an edge correction vector.

[0033] Further, the edge correction vector adopts an adaptive interpolation method for calculation, and the calculation formula is:

[0034] ;

[0035] In the formula, an acoustic characteristic value of a three-dimensional space point after correction represented by coordinates (x, y, z), a neighboring effective data node represented by coordinates (x, y, z), a total number of neighboring hydrophone sensors participating in correction represented by coordinates (x, y, z), a total number of effective nodes participating in correction calculation represented by coordinates (x, y, z), an acoustic characteristic value of a kth node originally measured represented by coordinates (x, y, z), a correction vector of a kth node output by a correction model represented by coordinates (x, y, z), a coherence decay function of a sound field represented by coordinates (x, y, z), a three-dimensional coordinate vector (x, y, z) of a point to be corrected represented by coordinates (x, y, z), a three-dimensional coordinate vector (x, y, z) of a kth node represented by coordinates (x, y, z), ), a coherence length of a sound field represented by coordinates (x, y, z), an Euclidean distance between a point to be corrected and a kth node represented by coordinates (x, y, z).

[0036] Further, the generation process of the orientation adjustment instruction in step 6 comprises:

[0037] calculating a maximum spatial position deviation between the corrected three-dimensional edge data and the original data;

[0038] triggering an orientation adjustment mechanism when the maximum spatial position deviation exceeds a preset dynamic adjustment threshold;

[0039] calculating the pose error of the hydrophone based on the correction amount of the conflict area and the boundary range of the corresponding data trust edge, and solving the orientation adjustment amount that minimizes the data conflict through the mapping relationship between the spatial pose parameters of the hydrophone and the monitoring data;

[0040] generating a three-dimensional adjustment instruction including horizontal orientation angle, pitch angle and depth offset, and driving the underwater actuator to adjust the spatial orientation and position of the target hydrophone.

[0041] Further, the communication encoding process in step 7 comprises:

[0042] embedding CRC check bits in the linear frequency modulation segment of the positioning pulse;

[0043] dividing the signal guard interval into multiple orthogonal time slots according to the delay time generated by the multipath effect of the underwater acoustic channel;

[0044] calculating the bearable capacity of the current channel in real time, analyzing the energy distribution of the multipath channel through the received signal strength, estimating the maximum reliable transmission rate based on the channel bandwidth and the ambient noise level, and dynamically adjusting the modulation order and code rate of the communication data packet according to the calculation result.

[0045] (Three) beneficial effects

[0046] Compared with the known prior art, the technical scheme provided by the application has the following beneficial effects:

[0047] 1. By dynamically planning data trust edge domain and actively adjusting sensor deployment, the traditional fixed range limitation is overcome, the monitored water area is effectively covered and the blind area is reduced, the conflict area is detected and marked by extracting standard edge features and constructing a three-dimensional hydrological model, verifying the adjacent sensor data trust edge domain, further predicting the abnormal edge diffusion amplitude by a pre-trained correction model, solving the multipath interference and conflict problem, realizing accurate correction of conflict data, and further improving the model consistency and positioning accuracy.

[0048] 2. By generating sensor orientation adjustment instructions based on the rationality verification of the corrected data, selecting the least interference sound wave transmission path based on the three-dimensional hydrological model, embedding communication codes in the positioning signal to realize co-channel transmission of positioning and communication, improving the spectrum utilization and system integration, and forming an adaptive cycle.

[0049] 3. By constructing a three-dimensional hydrological model with multiple sensors and combining the diffusion amplitude prediction of the correction model, the sound field characteristics in complex environments are accurately represented, and frequency domain wavelet denoising is introduced in feature extraction to effectively separate target signals and interference, thereby improving the fault tolerance of positioning and communication, reducing the complexity and power consumption of the device, and improving the anti-interference ability of the system in complex environments. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical schemes in the embodiments of the present application or the prior art, the drawings needed in the description of the embodiments or the prior art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0051] Figure 1 The flowchart of the present application. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0053] The present application will be further described below in combination with the embodiments.

[0054] The multi-channel adaptive sonar positioning and communication integrated method for complex underwater acoustic environment of the embodiment, as shown in the figure, includes the following steps: Figure 1

[0055] Step 1: Deploy several hydrophone sensors at different positions in the water area to be monitored, and obtain the performance parameters and position coordinates of each sensor in real time; the hydrophone sensor adopts a multi-channel wideband transducer array, and its performance parameters include signal-to-noise ratio, frequency response range and directivity angle; when deploying, the sensor pitch angle is dynamically adjusted through the underwater robot arm to match the terrain undulation.

[0056] Step 2: Based on the performance parameters and position coordinates of each hydrophone sensor, the initial data trust boundary of each hydrophone sensor is drawn to serve as the effective monitoring range of the hydrophone sensor in the water area.

[0057] Step 3: Perform spatial superposition analysis on the initial data trust boundary of all hydrophone sensors, and adjust the sensor deployment position until all data trust boundaries cover the entire water area to be monitored, and the adjacent data trust boundaries are set to allow overlapping areas.

[0058] Step 4: Extract the standard edge features of the original sensing data of a single hydrophone sensor, and the standard edge feature extraction process is as follows: perform frequency domain wavelet denoising on the original sonar signal, extract the time-frequency energy mutation point through Hilbert-Huang transform, and map the mutation point to a three-dimensional edge vector using DBSCAN clustering algorithm; construct a local three-dimensional hydrological model, extract the feature data of adjacent hydrophone sensors and convert it into three-dimensional edge data, and verify it with the existing model, compare the data of adjacent sensors in the allowed overlapping area, and if there is a spatial feature conflict, mark it as an abnormal edge feature; the construction of the three-dimensional hydrological model combines the sound field modeling technology, uses the matching field processing method, generates a high-precision three-dimensional hydrological environment representation based on the sound speed profile and seabed topography data of the marine environment, to support subsequent positioning and communication signal processing.

[0059] The allowed overlapping area of adjacent data trust boundaries needs to meet the minimum overlapping area requirement, and the determination rule of the minimum overlapping area is as follows:

[0060] Select the larger data trust boundary radius of the two adjacent hydrophone sensors as the reference value;

[0061] ​According to the type of water depth, the environmental complexity weight coefficient is selected: when the water depth is less than a set critical value, the shallow water coefficient is used, and when the water depth is greater than the critical value, the deep water coefficient is used; when the deployment water depth is less than a preset depth threshold, a typical value is 50 meters, the deployment water depth is defined as a shallow water area, and the environmental complexity weight coefficient is 0.3; when the deployment water depth is greater than or equal to the depth threshold, the deployment water depth is defined as a deep water area, and the environmental complexity weight coefficient is 0.6;

[0062] The reference value is multiplied by the corresponding environmental complexity weight coefficient to obtain a core parameter.

[0063] Finally, it is determined that the minimum overlap area should be greater than or equal to a numerical range of one half of the core parameter.

[0064] The feature data acquisition process of adjacent hydrophone sensors is as follows: the spatial correlation is calculated by using the time difference and phase difference of the signals received by each hydrophone sensor. Specifically, for each pair of adjacent hydrophone sensors, the system selects their signal sequences one by one, calculates the cross-correlation function of the two signals within a specified time window, identifies the maximum correlation peak value between the signals, and the time difference corresponding to the correlation peak value reflects the propagation time delay of the target signal, while the phase difference provides the spatial direction information of the signal. Based on the cross-correlation calculation result, the system extracts the spatial correlation features of each sensor signal, including the relative direction, distance estimation and signal intensity distribution of the target signal. Through multi-channel cross-correlation, the system can comprehensively extract the spatial position information of the target from the signals of multiple sensors, enhance the fault tolerance to multipath effect and low signal-to-noise ratio in complex underwater acoustic environment, and generate three-dimensional edge data; the transformation process uses orthogonal frequency division multiplexing technology to decompose the multi-channel signal into multiple sub-carrier channels, thereby enhancing the fault tolerance of the data in the complex underwater acoustic environment

[0065] Step 5: input the abnormal edge features into a pre-trained correction model, the correction model is generated based on historical hydrological data training, the correction model predicts the diffusion amplitude of the abnormal edge in the three-dimensional space, and the three-dimensional data edge of the conflict area is corrected according to the prediction result; the correction model adopts a double-branch graph convolutional neural network, and a historical hydrological database containing several terrains is pre-constructed, wherein:

[0066] The first branch inputs the three-dimensional coordinates and acoustic impedance value of the abnormal edge, and outputs the diffusion gradient tensor;

[0067] The second branch inputs the sound propagation loss matrix of similar terrains in the historical hydrological database, and outputs the terrain distortion compensation parameter;

[0068] The results of the two branches are fused to generate an edge correction vector.

[0069] Step 6: Verify the rationality of the corrected data, if the correction is valid, generate the bearing adjustment instruction of the associated hydrophone sensor according to the correction amount and the corresponding data trust edge domain, dynamically move the sensor position according to the generated adjustment bearing, and optimize the process by combining real-time channel state information, dynamically adjusting the signal transmission and reception parameters through adaptive modulation and beamforming technology, to improve the positioning accuracy and communication throughput.

[0070] Step 7: Based on the corrected three-dimensional hydrological model, select the least disturbed sound wave transmission path, embed communication coding in the positioning data, and perform co-channel transmission of positioning signals and communication signals;

[0071] The communication coding process is:

[0072] Embed CRC check bits in the linear frequency modulation segment of the positioning pulse;

[0073] Using different delay paths generated by the multipath effect of the underwater acoustic channel, the signal guard interval is divided into multiple orthogonal time slots according to the delay time, and each time slot independently transmits differential phase shift keying modulated data, that is, information is carried through the phase difference between adjacent symbols rather than the absolute phase value, improving the anti-interference ability in a multipath environment;

[0074] Real-time calculation of the current channel's carrying capacity, analysis of the energy distribution of the multipath channel through the received signal strength, estimation of the maximum reliable transmission rate based on the channel bandwidth and environmental noise level, dynamic adjustment of the modulation order and code rate of the communication data packet according to the calculation results, improvement of the transmission rate when the channel quality is good, and enhancement of the error correction capability when the channel quality is poor, embedding check codes in the inherent frequency variation process of the linear frequency modulation positioning pulse to realize the gap filling of the communication data to the positioning signal spectrum without additional frequency band resources, the natural time slots formed by different delay paths are used as parallel data transmission channels, and the orthogonal time slot design avoids multi-path symbol crosstalk, providing strong error correction protection for important bits at the coding level, adapting to burst errors in underwater acoustic channels, and realizing stable communication in complex underwater acoustic conditions.

[0075] Compared with the prior art, through the dynamic adjustment of the sensor deployment strategy and the trust edge domain superposition mechanism based on the environmental complexity weight, seamless coverage of water area monitoring in complex terrain is realized, the dual-branch graph convolution network is used to fuse acoustic impedance and historical terrain data to improve the correction accuracy of abnormal edge features, and the multipath effect is used to construct orthogonal time slots, adaptive modulation communication data is embedded in the positioning signal frequency band, and co-channel fusion transmission of positioning and communication is realized, which improves the positioning accuracy and improves the spectrum utilization and communication fault tolerance in a multi-channel environment.

[0076] In other aspects, the embodiment provides a data trust edge domain radius calculation process, specifically including:

[0077] The current hydrophone sensor's emitted sound source intensity level, directional mode gain parameter, frequency-related sound energy attenuation coefficient, sound wave propagation path distance, and underwater environmental noise spectrum distribution are obtained. The sound wave propagation attenuation factor is calculated by using the natural constant of approximately 2.718 as the base and the product of the negative underwater acoustic energy attenuation coefficient and the propagation distance as the exponent.

[0078] The combined gain value is obtained by multiplying the intensity level of the emitted sound source, the gain parameter of the directional mode, and the sound wave propagation attenuation factor. The environmental interference factor is calculated by multiplying the propagation path distance by itself twice to obtain the squared distance value, and then multiplying the squared distance value by the underwater environmental noise spectrum distribution.

[0079] Divide the combined gain value by the environmental interference factor to obtain the preliminary effective monitoring radius;

[0080] The initial effective monitoring radius is multiplied by the preset signal-to-noise ratio threshold adjustment coefficient to finally determine the data trust boundary radius range of the sensor.

[0081] Compared with existing technologies, by comprehensively considering sound source intensity, directivity gain, frequency attenuation, propagation distance and environmental noise parameters, the natural exponential attenuation model is used to accurately calculate the sound wave attenuation factor. The product of distance cube and environmental noise is introduced as an interference factor. Combined with dynamic adjustment of signal-to-noise ratio threshold, a trust boundary radius calculation model with strong environmental adaptability is constructed, which greatly improves the accuracy of underwater monitoring range assessment.

[0082] This embodiment provides a calculation process for the edge correction vector using an adaptive interpolation method, and the calculation formula is as follows:

[0083] ;

[0084] In the formula, This represents the acoustic eigenvalue of a three-dimensional point in space after correction of coordinates (x, y, z). Representing adjacent valid data nodes, This represents the total number of adjacent hydrophone sensors involved in the correction. This represents the total number of valid nodes participating in the correction calculation. This represents the acoustic feature value originally measured at the k-th node. This represents the correction vector at the k-th node of the correction model output. Represents the coherent attenuation function of the sound field. The three-dimensional coordinate vector (x, y, z) represents the point to be corrected, i.e., the target location that needs to be corrected within the conflict area. The three-dimensional coordinate vector representing the k-th node ( ), that is, the spatial location of the effective reference node. representing a coherence length of the sound field, representing the Euclidean distance between the to-be-corrected point and the kth node.

[0085] By weighted interpolation constrained by the coherence of the sound field, the discrete correction vector output by the correction model is continuously diffused to the entire conflict area under the premise of preserving the topology of the original data , overcoming the phase distortion problem of traditional interpolation methods in the underwater acoustic multipath environment.

[0086] The embodiment provides a generation process of a sensor orientation adjustment instruction, including:

[0087] Calculate the maximum spatial position deviation of the corrected three-dimensional edge data and the original data;

[0088] When the maximum spatial position deviation exceeds a preset dynamic adjustment threshold, trigger the orientation adjustment mechanism, and the threshold is set to one quarter of the wavelength of the sound wave by default;

[0089] According to the correction amount of the conflict area and the boundary range of the corresponding data trust edge domain, the underwater acoustic sensor pose error is reversely calculated, and through the mapping relationship between the underwater acoustic sensor spatial pose parameter and the monitoring data, the orientation adjustment amount that minimizes the data conflict is solved;

[0090] Generate a three-dimensional adjustment instruction including the horizontal orientation angle, the pitch angle and the depth offset, and drive the underwater execution mechanism to adjust the spatial orientation and position of the target underwater acoustic sensor.

[0091] By taking the correction result of the data conflict as the detection basis for the sensor deployment defects, compared with the traditional fixed deployment scheme, the sensor offset problem caused by the flow impact can be dynamically adapted, the dynamic adjustment threshold is bound with the wavelength of the sound wave, so that normal data fluctuations caused by minor environmental disturbances cannot trigger the adjustment, signal distortion caused by real hardware displacement can be captured in time, and the geometric correlation between the orientation angle, the pitch angle and the depth parameter of the underwater acoustic sensor and the three-dimensional hydrological model is utilized, the position offset of the boundary point of the data trust edge domain directly reflects the sensor installation angle deviation, and the spatial distribution of the correction vector in the conflict area indicates the sinking or floating amount of the sensor

[0092] In summary, the present application effectively eliminates the monitoring blind area by dynamically planning the data trust edge domain and optimizing the sensor deployment, adapts to the change of the complex underwater acoustic environment, constructs a three-dimensional hydrological model by multiple sensors, accurately corrects the data conflict in combination with the pre-trained correction model, improves the positioning accuracy, dynamically adjusts the sensor orientation according to the corrected data, and enhances the system fault tolerance.

[0093] The positioning and communication co-channel transmission is jointly modulated, the spectrum occupation and the device complexity are reduced, and the communication reliability is improved, the method realizes high-precision positioning and stable communication in a complex underwater acoustic environment, and greatly improves the system integration and environmental adaptability.

[0094] The above examples are only used to illustrate the technical solutions of the present application, but not to limit it; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part of the technical features; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environments, characterized in that, Includes the following steps: Step 1: Deploy several hydrophone sensors at different locations in the water area to be monitored, and acquire the performance parameters and location coordinates of each sensor in real time; Step 2: Based on the performance parameters and location coordinates of each hydrophone, define an initial data trust boundary for each hydrophone to serve as the effective monitoring range of the hydrophone in the water area. Step 3: Perform spatial overlay analysis on the initial data trust boundaries of all hydrophone sensors, adjust the sensor deployment positions until all data trust boundaries cover the entire water area to be monitored, and set overlapping areas for adjacent data trust boundaries. Step 4: Extract standard edge features from the raw sensing data of a single hydrophone sensor to obtain a three-dimensional spatial edge vector; extract feature data from adjacent hydrophone sensors and convert them into three-dimensional edge data for constructing a local three-dimensional hydrological model; perform spatial alignment verification between the three-dimensional spatial edge vector and the local three-dimensional hydrological model; compare adjacent sensor data within the allowed overlapping area; if there is a spatial feature conflict, mark it as an abnormal edge feature. The process of acquiring feature data from adjacent hydrophones is as follows: by using the arrival time difference and phase difference of the signals received by each hydrophone, spatial correlation is calculated to generate three-dimensional edge data; Step 5: Input the abnormal edge features into the pre-trained correction model. The correction model is trained and generated based on historical hydrological data. The correction model predicts the diffusion range of the abnormal edge in three-dimensional space and corrects the three-dimensional data edge of the conflict area according to the diffusion range. Step 6: Verify the rationality of the corrected data. If the correction is effective, generate the orientation adjustment command of the associated hydrophone sensor based on the correction amount and the corresponding data trust boundary. Step 7: Based on the corrected 3D data edges, obtain the corrected 3D hydrological model, select the acoustic wave transmission path with the least interference, embed communication codes in the positioning data, and transmit positioning signals and communication signals in the same channel.

2. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environments according to claim 1, characterized in that, The hydrophone sensor in step 1 uses a multi-channel wideband transducer array. Its performance parameters include signal-to-noise ratio, frequency response range, and directivity angle. During deployment, the sensor's pitch angle is dynamically adjusted by an underwater robotic arm to match the terrain undulations.

3. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environments according to claim 1, characterized in that, The standard edge feature extraction process in step 4 is as follows: frequency domain wavelet denoising is performed on the original sonar signal, time-frequency domain energy mutation points are extracted by Hilbert-Huang transform, and the mutation points are mapped to three-dimensional spatial edge vectors using the DBSCAN clustering algorithm.

4. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environments according to claim 1, characterized in that, In step 3, the allowed overlapping areas of adjacent data trust boundaries must meet the minimum overlap area requirement. The rule for determining this minimum overlap area is as follows: The larger data trust boundary radius between two adjacent hydrophone sensors is selected as the benchmark value; The environmental complexity weighting coefficient is selected based on the water depth type: when the water depth is less than the set critical value, the shallow water coefficient is used; when the water depth is greater than the critical value, the deep water coefficient is used. Multiply the benchmark value by the corresponding environmental complexity weight coefficient to obtain the core parameters; Ultimately, it was determined that the minimum overlap area should be greater than or equal to half the value range of this core parameter.

5. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environments according to claim 1, characterized in that, In step 5, the correction model employs a dual-branch graph convolutional neural network, and a historical hydrological database containing several terrain features is pre-constructed, wherein: The first branch takes the three-dimensional coordinates and acoustic impedance value of the abnormal edge as input and outputs the diffusion gradient tensor. The second branch takes the acoustic propagation loss matrix of similar terrain in the historical hydrological database as input and outputs the terrain distortion compensation parameters. The results from the two branches are merged to generate an edge correction vector.

6. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environments according to claim 5, characterized in that, The edge correction vector is calculated using an adaptive interpolation method, and its calculation formula is as follows: ； In the formula, This represents the acoustic eigenvalue of a three-dimensional point in space after correction of coordinates (x, y, z). Representing adjacent valid data nodes, This represents the total number of adjacent hydrophone sensors involved in the correction. This represents the total number of valid nodes participating in the correction calculation. This represents the acoustic feature value originally measured at the k-th node. This represents the correction vector at the k-th node of the correction model output. Represents the coherent attenuation function of the sound field. The three-dimensional coordinate vector (x, y, z) represents the point to be corrected. The three-dimensional coordinate vector representing the k-th node ( ), Represents the coherence length of the sound field. This represents the Euclidean distance between the point to be corrected and the k-th node.

7. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environments according to claim 1, characterized in that, The process of generating the position adjustment command in step 6 includes: Calculate the maximum spatial deviation between the corrected 3D edge data and the original data; When the maximum value of the spatial position deviation exceeds the preset dynamic adjustment threshold, the orientation adjustment mechanism is triggered; Based on the correction amount of the conflict area and the boundary range of the corresponding data trust domain, the hydrophone pose error is calculated in reverse. By using the mapping relationship between the spatial pose parameters of the hydrophone and the monitoring data, the orientation adjustment amount that minimizes data conflict is solved. Generate three-dimensional adjustment commands including horizontal azimuth, pitch angle and depth offset to drive the underwater actuator to adjust the spatial orientation and position of the target hydrophone.

8. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environments according to claim 1, characterized in that, The communication encoding process in step 7 is as follows: Embed a CRC check bit in the linear frequency modulation band of the positioning pulse; By utilizing the different delay paths generated by the multipath effect of the underwater acoustic channel, the signal protection interval is divided into multiple orthogonal time slots according to the delay time. The system calculates the current channel's carrying capacity in real time, analyzes the energy distribution of the multipath channel by receiving signal strength, estimates the maximum reliable transmission rate based on channel bandwidth and environmental noise level, and dynamically adjusts the modulation order and code rate of communication data packets based on the calculation results.

Images