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

Abstract

The invention discloses a multi-channel self-adaptive sonar positioning and communication integration method for a complex underwater acoustic environment, and relates to the field of positioning communication, and the method comprises the following steps: 1, deploying a plurality of hydrophone sensors at different positions of a water area to be monitored, and obtaining the performance parameters and position coordinates of each sensor in real time; 2, on the basis of the performance parameters and the position coordinates of the hydrophone sensors, an initial data trust side domain is delimited for each hydrophone sensor to serve as an effective monitoring range of the hydrophone sensors in a water area; 3, performing spatial overlay analysis on the initial data trust side domains of all the hydrophone sensors; a monitored conflict area is detected and marked, the abnormally-changed edge diffusion amplitude is further predicted through a pre-training correction model, the multipath interference and conflict problems are solved, accurate correction of conflict data is achieved, the model consistency and positioning accuracy are improved, and the anti-interference capability of a system in a complex environment is improved.

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. Modern marine resource exploration, underwater unmanned vehicle navigation, underwater network communication and other fields have higher requirements for high-precision positioning and stable data transmission, 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, and 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 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.

[0005] (II) Technical scheme In order to achieve the above object, the present application is realized by the following technical scheme: The present application discloses a multi-channel adaptive sonar positioning and communication integration method for complex underwater acoustic environment, comprising the following steps: 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; 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 as the effective monitoring range of the hydrophone sensor in the water area; Step 3: Perform spatial superposition analysis on the initial data trust edge of all hydrophone sensors, adjust the sensor deployment position until all data trust edges cover the entire water area to be monitored, and set adjacent data trust edges to allow overlapping areas; Step 4: Perform standard edge feature extraction on the original sensor data of a single hydrophone, construct a local three-dimensional hydrological model, extract feature data of adjacent hydrophones and convert it to 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; 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; Step 6: Verify the reasonableness of the corrected data. If the correction is effective, generate orientation adjustment instructions for the associated hydrophone based on the correction amount and the corresponding data trust edge; 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.

[0006] 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.

[0007] Further, the calculation process of the data trust edge radius in step 2 includes: 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, and calculate the sound wave propagation attenuation factor; 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; Divide the joint gain value by the environmental interference factor to obtain the preliminary effective monitoring radius; Multiply the preliminary effective monitoring radius by the preset signal-to-noise ratio threshold adjustment coefficient to finally determine the data trust edge radius range of the sensor.

[0008] Further, the standard edge feature extraction process in step 4 is: Perform frequency domain wavelet denoising on the original sonar signal, extract time-frequency energy mutation points through Hilbert-Huang transform, and map the mutation points to three-dimensional space edge vectors using DBSCAN clustering algorithm.

[0009] Further, the allowed overlap area of the adjacent data trust edge domain in step 4 needs to meet the minimum overlap area requirement, and the determination rule of the minimum overlap area is: The larger data trust edge domain radius of the two adjacent hydrophone sensors is selected as the reference value; According to the depth type of the water area, the environmental complexity weight coefficient is selected: when the water depth is less than the 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; The reference value is multiplied by the corresponding environmental complexity weight coefficient to obtain the core parameter; 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.

[0010] Further, in step 4, the feature data acquisition process of the adjacent hydrophone sensor is: using the time difference and phase difference of the received signal of each hydrophone sensor, calculating the spatial correlation, and generating three-dimensional edge data; the conversion process uses orthogonal frequency division multiplexing technology to decompose the multi-channel signal into multiple sub-carrier channels.

[0011] Further, in step 5, the correction model adopts a double-branch graph convolutional neural network, and a historical hydrology database containing several terrains is constructed in advance, wherein: The first branch inputs the three-dimensional coordinates and acoustic impedance value of the abnormal edge, and outputs the diffusion gradient tensor; The second branch inputs the sound propagation loss matrix of the similar terrain in the historical hydrology database, and outputs the terrain distortion compensation parameter; The fusion of the two branch results generates an edge correction vector.

[0012] Further, the edge correction vector adopts an adaptive interpolation method for calculation, and the calculation formula is: ; In the formula, represents the corrected three-dimensional spatial point acoustic feature value of the coordinates (x, y, z), represents the adjacent effective data node, represents the total number of adjacent hydrophone sensors participating in the correction, represents the total number of effective nodes participating in the correction calculation, represents the original measured acoustic feature value of the kth node, represents the correction vector of the kth node output by the correction model, represents the coherence attenuation function of the sound field, represents the three-dimensional coordinate vector (x, y, z) of the point to be corrected, represents the three-dimensional coordinate vector of the kth node, ), represents the coherence length of the sound field, representing the Euclidean distance between the to-be-corrected point and the kth node.

[0013] Further, the step 6 of generating the orientation adjustment instruction comprises: calculating a maximum spatial position deviation between the corrected three-dimensional edge data and the original data; triggering the orientation adjustment mechanism when the maximum spatial position deviation exceeds a preset dynamic adjustment threshold; 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 domain, 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; generating a three-dimensional adjustment instruction including the horizontal azimuth, the pitch angle and the depth offset, and driving the underwater actuator to adjust the spatial orientation and position of the target hydrophone.

[0014] Further, the communication encoding process in step 7 comprises: embedding CRC check bits in the linear frequency modulation segment of the positioning pulse; dividing the signal guard interval into multiple orthogonal time slots according to the delay time generated by the different delay paths of the underwater acoustic channel; 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.

[0015] (Three) beneficial effects Compared with the known prior art, the technical scheme provided by the present application has the following beneficial effects: 1. By dynamically planning the data trust edge domain and actively adjusting the sensor deployment, the limitations of traditional fixed range are overcome, the monitored water area is effectively covered and the blind area is reduced, the conflict area is detected and marked by extracting the standard edge features and constructing the three-dimensional hydrological model, verifying the data trust edge domain of the adjacent sensor, further predicting the abnormal edge diffusion amplitude by the pre-trained correction model, solving the multipath interference and conflict problem, realizing the accurate correction of the conflict data, and further improving the model consistency and positioning accuracy.

[0016] 2. By generating sensor orientation adjustment instructions based on the rationality verification of the corrected data, selecting the least disturbed sound wave transmission path based on the three-dimensional hydrological model, embedding communication encoding in the positioning signal, realizing the same channel transmission of positioning and communication, improving the spectrum utilization and system integration, and forming an adaptive cycle.

[0017] 3. The method can accurately represent the sound field characteristics in complex environment by constructing a three-dimensional hydrological model through multi-sensor cooperation and combining with the diffusion range prediction of the correction model, can effectively separate the target signal and interference by introducing frequency domain wavelet noise reduction in feature extraction, thereby improving the fault tolerance of positioning and communication, reducing the complexity and power consumption of the equipment, and improving the anti-interference ability of the system in complex environment. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creating any creative labor.

[0019] Figure 1 The flowchart of the present application. DETAILED DESCRIPTION

[0020] 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 the following with reference to 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 creating any creative labor belong to the scope of protection of the present application.

[0021] The present application will be further described in the following with reference to the embodiments.

[0022] The multi-channel adaptive sonar positioning and communication integrated method for complex underwater acoustic environment of the present embodiment, as shown in Figure 1 The method comprises the following steps: 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; the hydrophone sensor adopts a multi-channel wideband transducer array, and the performance parameters include signal-to-noise ratio, frequency response range and directivity angle; when deploying, the sensor pitch angle is dynamically adjusted by underwater robot arm to match the terrain undulation.

[0023] 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.

[0024] Step 3: spatial superposition analysis is performed on the initial data trust boundary of all hydrophone sensors, and the sensor deployment position is adjusted until all data trust boundaries cover the entire water area to be monitored, and the adjacent data trust boundaries are set to allow overlapping area.

[0025] Step 4: Standard edge feature extraction is performed on the raw sensing data of a single hydrophone sensor, and the standard edge feature extraction process is as follows: wavelet denoising is performed on the original sonar signal in the frequency domain, time-frequency energy mutation points are extracted through Hilbert-Huang transformation, and DBSCAN clustering algorithm is used to map the mutation points into three-dimensional space edge vectors; a local three-dimensional hydrological model is constructed, the feature data of adjacent hydrophone sensors is extracted and converted into three-dimensional edge data, and the existing model is verified for spatial alignment, the adjacent sensor data is compared in the allowed overlapping area, and if there is a spatial feature conflict, it is marked as an abnormal edge feature; the construction of the three-dimensional hydrological model combines sound field modeling technology, uses matching field processing method, generates high-precision three-dimensional hydrological environment representation based on sound velocity profile and seabed topography data of marine environment, to support subsequent positioning and communication signal processing.

[0026] The allowed overlapping area of the adjacent data trust edge domain needs to meet the minimum overlapping area requirement, and the determination rule of the minimum overlapping area is as follows: The larger data trust edge domain radius of the two adjacent hydrophone sensors is selected as the reference value; The environmental complexity weight coefficient is selected according to the type of water depth: when the water depth is less than the 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 the pre-set depth threshold, the typical value is 50 meters, it is defined as a shallow water area, and the environmental complexity weight coefficient is 0.3 at this time; when the deployment water depth is greater than or equal to the depth threshold, it is defined as a deep water area, and the environmental complexity weight coefficient is 0.6 at this time; The reference value is multiplied by the corresponding environmental complexity weight coefficient to obtain the core parameter; Finally, it is determined that the minimum overlapping area should be greater than or equal to the numerical range of one-half of the core parameter.

[0027] The process of acquiring characteristic data from adjacent hydrophone sensors is as follows: Spatial correlation is calculated using the arrival 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 between the signals, and the time difference corresponding to the correlation peak reflects the propagation delay of the target signal, while the phase difference provides the spatial direction information of the signal. Based on the cross-correlation calculation results, the system extracts the spatial correlation features of each sensor signal, including the relative orientation of the target signal, distance estimation, and signal intensity distribution. Through multi-channel cross-correlation, the system can comprehensively extract the spatial location information of the target from the signals of multiple sensors, enhancing the fault tolerance to multipath effects and low signal-to-noise ratios in complex underwater acoustic environments, and generating three-dimensional edge data. The conversion process uses orthogonal frequency division multiplexing (OFDM) technology to decompose the multi-channel signal into multiple subcarrier channels, enhancing the fault tolerance of the data in complex underwater acoustic environments. 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 amplitude of the abnormal edge in three-dimensional space and corrects the three-dimensional data edges of the conflict area based on the prediction results. The correction model uses a two-branch graph convolutional neural network and pre-constructs a historical hydrological database containing several terrain features, including: 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.

[0028] Step 6: Verify the rationality of the corrected data. If the correction is effective, generate the azimuth adjustment command for the associated hydrophone sensor based on the correction amount and the corresponding data trust boundary. Dynamically move the sensor position according to the generated adjustment azimuth. The optimization process combines real-time channel state information and uses adaptive modulation and beamforming technology to dynamically adjust the signal transmission and reception parameters to improve positioning accuracy and communication throughput.

[0029] Step 7: Based on the corrected three-dimensional hydrological model, select the acoustic wave transmission path with the least interference, embed communication codes in the positioning data, and transmit the positioning signal and communication signal in the same channel. The communication encoding process 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 guard interval is divided into multiple orthogonal time slots according to the delay time. Each time slot independently transmits differential phase shift keying modulation data, that is, information is carried through the phase difference of adjacent symbols rather than the absolute phase value, thereby improving the anti-interference capability in multipath environments. Real-time calculation of the current channel's bearable capacity, analysis of the energy distribution of the multipath channel through the received signal strength, estimation of the maximum reliable transmission rate according to the channel bandwidth and the ambient noise level, dynamic adjustment of the modulation order and code rate of the communication data packet according to the calculation results, increase of the transmission rate when the channel quality is good, and enhancement of the error correction capability when the channel quality is poor, embedding of 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 occupation of frequency band resources, the natural time slots formed by different delay paths are used as parallel data transmission channels, orthogonal time slot design avoids inter-symbol crosstalk in multipath, strong error correction protection is provided to important bits at the coding level, which adapts to the burst error of underwater acoustic channel and realizes stable communication under complex underwater acoustic conditions.

[0030] 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 under complex terrain is realized, the acoustic impedance and historical terrain data are fused by using a double-branch graph convolution network to improve the correction accuracy of abnormal edge features, and the orthogonal time slots are constructed by using the multipath effect, the adaptive modulation communication data is embedded in the positioning signal frequency band to realize the co-channel fusion transmission of positioning and communication, which improves the positioning accuracy and improves the spectrum utilization and communication fault tolerance under multi-channel environment.

[0031] In other aspects, the embodiment provides a data trust edge domain radius calculation process, which specifically includes: The transmission sound source intensity level, the directivity pattern gain parameter, the frequency-dependent sound energy attenuation coefficient, the sound wave propagation path distance, and the underwater ambient noise spectrum distribution of the current hydrophone sensor are obtained, and the sound wave propagation attenuation factor is calculated. The factor is calculated by taking the natural constant about 2.718 as the base number and the product of the negative underwater sound energy attenuation coefficient and the propagation distance as the exponent; The transmission sound source intensity level, the directivity pattern gain parameter and the sound wave propagation attenuation factor are multiplied to obtain a joint gain value, and an environmental interference factor is calculated. The environmental interference factor is obtained by squaring the propagation path distance twice to obtain the distance square value, and then multiplying the distance square value with the underwater ambient noise spectrum distribution; The joint gain value is divided by the environmental interference factor to obtain a preliminary effective monitoring radius; The preliminary effective monitoring radius is multiplied by the preset signal-to-noise ratio threshold adjustment coefficient to finally determine the data trust edge domain radius range of the sensor.

[0032] 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.

[0033] This embodiment provides a calculation process for the edge correction vector using an adaptive interpolation method, and the 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, 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. Represents the coherence length of the sound field. This represents the Euclidean distance between the point to be corrected and the k-th node.

[0034] By using weighted interpolation constrained by sound field coherence, the discrete correction vector output by the correction model is corrected while preserving the original data topology. It continuously diffuses throughout the entire conflict area, overcoming the phase distortion problem of traditional interpolation methods in underwater acoustic multipath environments.

[0035] This embodiment provides a process for generating sensor orientation adjustment commands, including: 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. The threshold is set to one-quarter of the sound wave wavelength by default. 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 instructions including horizontal azimuth angle, pitch angle and depth offset, drive underwater actuators to adjust the spatial orientation and position of the target hydrophone.

[0036] By taking the correction result of data conflict as the basis for detecting sensor deployment defects, compared with the traditional fixed deployment scheme, the dynamic adjustment threshold is bound with the sound wave wavelength, ensuring that normal data fluctuations caused by minor environmental disturbances will not trigger adjustment, and signal distortion caused by real hardware displacement can be captured in time, using the geometric correlation of the azimuth angle, pitch angle and depth parameters of the hydrophone and the three-dimensional hydrological model, the position offset of the data trust edge boundary point directly reflects the sensor installation angle deviation, and the spatial distribution of the correction vector in the conflict area indicates the amount of sensor sinking or floating In summary, the present application effectively eliminates the monitoring blind area by dynamically planning the data trust edge and optimizing the sensor deployment, adapts to changes in complex underwater acoustic environment, constructs a three-dimensional hydrological model with multiple sensors, combines a pre-trained correction model to accurately correct data conflicts, improves positioning accuracy, dynamically adjusts the orientation of the sensor according to the corrected data, and enhances system fault tolerance. The positioning and communication are transmitted in the same channel, are jointly modulated, reduce the spectrum occupation and the equipment complexity, improve the communication reliability, this method realizes high-precision positioning and stable communication in complex underwater acoustic environment, greatly improves the system integration and environmental adaptability.

[0037] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, 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 positioning and communication integration method for complex underwater acoustic environment, characterized in that, The method comprises the following steps: Step 1: deploying a plurality of underwater acoustic sensors at different positions of the water area to be monitored to obtain performance parameters and position coordinates of the sensors in real time; Step 2: based on the performance parameters and position coordinates of the underwater acoustic sensors, an initial data trust boundary is drawn for each underwater acoustic sensor as an effective monitoring range of the underwater acoustic sensor in the water area; Step 3: performing spatial superposition analysis on the initial data trust boundaries of all underwater acoustic sensors to adjust the deployment positions of the sensors until all data trust boundaries cover the entire water area to be monitored, and adjacent data trust boundaries are set to allow overlapping areas; Step 4: performing standard edge feature extraction on the original sensing data of a single underwater acoustic sensor to construct a local three-dimensional hydrological model, extracting feature data of adjacent underwater acoustic sensors and converting the feature data into three-dimensional edge data, and performing spatial alignment verification with the existing model, comparing adjacent sensor data in the allowed overlapping area, and marking abnormal edge features if there is a spatial feature conflict; Step 5: inputting the abnormal edge features into a pre-trained correction model, the correction model being generated based on historical hydrological data training, the correction model predicting the diffusion amplitude of the abnormal edge in the three-dimensional space, and correcting the three-dimensional data edge of the conflict area according to the prediction result; Step 6: verifying the rationality of the corrected data, and if the correction is effective, generating a position adjustment instruction for the associated underwater acoustic sensor according to the correction amount and the corresponding data trust boundary; Step 7: based on the corrected three-dimensional hydrological model, selecting a sound wave transmission path with the least interference, embedding a communication code in the positioning data, and performing same-channel transmission of the positioning signal and the communication signal.

2. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environment according to claim 1, characterized in that, The underwater acoustic sensor in step 1 adopts a multi-channel wideband transducer array, and the performance parameters include signal-to-noise ratio, frequency response range and directivity angle. When deployed, the sensor pitch angle is dynamically adjusted by an underwater robot arm to match the terrain undulations.

3. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environment according to claim 1, characterized in that, The calculation process of the data trust boundary radius in step 2 includes: obtaining the transmission sound source intensity level, the directivity pattern gain parameter, the frequency-dependent sound energy attenuation coefficient, the sound wave propagation path distance and the underwater environmental noise spectrum distribution of the current underwater acoustic sensor, calculating the sound wave propagation attenuation factor; performing multiplication operation on the transmission sound source intensity level, the directivity pattern gain parameter and the sound wave propagation attenuation factor to obtain a joint gain value, and calculating an environmental interference factor; dividing the joint gain value by the environmental interference factor to obtain a preliminary effective monitoring radius; multiplying the preliminary effective monitoring radius by a preset signal-to-noise ratio threshold adjustment coefficient to finally determine the data trust boundary radius range of the sensor.

4. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environment according to claim 1, characterized in that, 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 to three-dimensional space edge vectors by using DBSCAN clustering algorithm.

5. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environment according to claim 1, characterized in that, The allowed overlapping area of adjacent data trust boundaries in step 4 needs to meet the minimum overlapping area requirement, and the determination rule of the minimum overlapping area is: selecting the larger data trust boundary radius of the two adjacent underwater acoustic sensors as the reference value; According to the water depth type, the environmental complexity weight coefficient is selected: when the water depth is less than a set critical value, a shallow water coefficient is used, and when the water depth is greater than the critical value, a deep water coefficient is used; The reference value is multiplied by the corresponding environmental complexity weight coefficient to obtain a core parameter; 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.

6. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environment according to claim 1, characterized in that, In step 4, the feature data acquisition process of adjacent hydrophone sensors is: using the arrival time difference and phase difference of the signals received by each hydrophone sensor, calculating the spatial correlation, and generating three-dimensional edge data; the conversion process uses orthogonal frequency division multiplexing technology to decompose the multi-channel signal into multiple sub-carrier channels.

7. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environment according to claim 1, characterized in that, 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: The first branch inputs the three-dimensional coordinates and acoustic impedance values of the abnormal edge, and outputs the diffusion gradient tensor; The second branch inputs the sound propagation loss matrix of similar terrain in the historical hydrological database, and outputs the terrain distortion compensation parameter; The fusion of the results of the two branches generates an edge correction vector.

8. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environment according to claim 7, characterized in that, The edge correction vector uses an adaptive interpolation method for calculation, and the calculation formula is: ； wherein, represents the acoustic feature value of the three-dimensional space point after correction of the coordinates (x, y, z), represents the adjacent effective data nodes, represents the total number of adjacent hydrophone sensors participating in the correction, represents the total number of effective nodes participating in the correction calculation, represents the acoustic feature value of the kth node originally measured, represents the kth node correction vector output by the correction model, represents the coherence decay function of the sound field, represents the three-dimensional coordinate vector (x, y, z) of the point to be corrected, represents the three-dimensional coordinate vector (x, y, z) of the kth node, ), represents the coherence length of the sound field, represents the Euclidean distance between the point to be corrected and the kth node.

9. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environment according to claim 1, characterized in that, In step 6, the generation process of the orientation adjustment instruction includes: Calculate the maximum spatial position deviation of the corrected three-dimensional edge data and the original data; When the maximum spatial position deviation exceeds the preset dynamic adjustment threshold, the orientation adjustment mechanism is triggered; According to the correction amount of the conflict area and the boundary range of the corresponding data trust domain, the pose error of the hydrophone sensor is inversely calculated, and through the mapping relationship between the spatial pose parameters of the hydrophone sensor and the monitoring data, the orientation adjustment amount that minimizes the data conflict is solved; Generate a three-dimensional adjustment instruction including horizontal orientation angle, pitch angle and depth offset, and drive the underwater actuator to adjust the spatial orientation and position of the target hydrophone sensor.

10. The multi-channel adaptive sonar localization and communication integrated method for complex underwater acoustic environment according to claim 1, characterized in that, The communication encoding process in step 7 is: Embed CRC check bits in the linear frequency modulation segment of the positioning pulse; 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; Real-time calculation of the current channel carrying capacity, analysis of the energy distribution of the multipath channel through the received signal strength, estimation of the maximum reliable transmission rate according to the channel bandwidth and environmental noise level, and dynamic adjustment of the modulation order and code rate of the communication data packet according to the calculation result.

Images