A Sonar Imaging Reverberation Elimination Method and System Based on Manifold Constraint

Abstract

The present invention provides a method and system for sonar imaging reverberation elimination based on manifold constraints, which are applied to the field of imaging data processing. By real-time collecting the time-domain waveform of the echo signal and determining whether irregular reflection peaks are detected, the present invention can quickly identify potential interferences and reflections in the underwater environment. Especially in complex underwater environments, non-target objects (such as seabed obstacles or debris) may cause irregular reflection signals. At the same time, by collecting the reflection paths of the echo signals and predicting the propagation parameters of the reverberation signals, the present invention can accurately model the propagation characteristics of sound waves underwater, especially the seabed reflection signals. And by using the pre-recorded information such as seabed topography, geological characteristics, and sound wave propagation characteristics, the present invention can provide accurate background data for the propagation modeling of reverberation signals.

Description

Technical Field

[0001] The present invention relates to the field of imaging data processing, and particularly to a method and system for sonar imaging reverberation elimination based on manifold constraint. Background Art

[0002] In order to effectively monitor underwater targets, active sonar has become the main detection tool. Active sonar emits sound waves into the underwater environment and receives the echo signals reflected by the targets, thereby obtaining the position, distance, speed, and characteristic information of the targets. Compared with passive sonar that only relies on the noise emitted by the targets themselves for detection, active sonar can actively emit signals, improve the detection ability for stationary or low-noise targets, and is applicable to complex underwater environments.

[0003] However, active sonar often needs to work in an environment where noise and reverberation are complexly intertwined, and the reverberation is formed by the multiple scattering of sonar signals by a large number of scatterers in the ocean. In shallow water areas, the reverberation caused by seabed reflection is the main interference source. Especially, its strong scattering characteristics are similar to the target echo signals, which significantly limits the target detection performance of the active sonar system. Summary of the Invention

[0004] The present invention aims to solve the problem of how to reduce the reverberation caused by seabed reflection and improve the target detectability of active sonar, and provides a method and system for sonar imaging reverberation elimination based on manifold constraint.

[0005] The present invention adopts the following technical means to solve the technical problems:

[0006] The present invention provides a method for sonar imaging reverberation elimination based on manifold constraint, including:

[0007] Collect the time-domain waveform of the echo signal based on the echo signal of the target pre-monitored by the active sonar.

[0008] Judge whether a preset number of irregular reflection peaks are detected in the time-domain waveform.

[0009] If so, collect the reflection path of the echo signal according to the pre-recorded characteristic information, predict the propagation parameters of the reverberation signal based on the reflection path, and dynamically adjust the detection threshold of the active sonar for the target through the reflection path and the propagation parameters, where the characteristic information specifically includes seabed topography, geological characteristics, and acoustic wave characteristics.

[0010] Judge whether the propagation parameters can be modeled and eliminated.

[0011] If possible, based on the preset modeling basic information of the active sonar, obtain the sound speed distribution data of sound waves underwater. According to the sound speed distribution data, generate the multipath propagation parameters of the sound waves. Based on the multipath propagation parameters, model the characteristics to be eliminated for each path. Among them, the modeling basic information specifically includes hard bottom, soft bottom, and rock. The multipath propagation specifically includes the direct path from the sound source to the target and the indirect path after underwater reflection. The characteristics to be eliminated specifically include the time delay characteristics, frequency change, and amplitude attenuation of each path.

[0012] Further, in the step of collecting the reflection path of the echo signal and predicting the propagation parameters of the reverberation signal based on the reflection path, it further includes:

[0013] Based on the preset sound speed profile of the active sonar, calculate the propagation time delay of the reflection path;

[0014] Judge whether the propagation time delay is greater than the preset time delay threshold;

[0015] If so, according to the relative speed when the sound wave propagates, obtain the offset frequency of the echo signal. Based on the coefficient characteristics pre-recorded by the active sonar, calculate the attenuation characteristics of the reverberation signal. Among them, the coefficient characteristics specifically include the sound absorption coefficient of the water body, the reflection coefficient of the seabed, and the attenuation coefficient of the distance.

[0016] Further, before the step of dynamically adjusting the detection threshold of the active sonar for the target through the reflection path and the propagation parameters, it further includes:

[0017] Based on the monitoring type of the target, collect the reflection characteristics of the target. Among them, the monitoring type specifically includes small targets, large targets, and dynamic targets. The reflection characteristics specifically include reflection intensity, shape, size, and material;

[0018] Judge whether the monitoring type matches the reflection characteristics;

[0019] If so, obtain the real-time monitored environmental noise, distinguish the difference data between the noise interference and the target echo from the environmental noise, and according to the difference data, apply the preset Wiener filter to suppress the corresponding noise information. Among them, the noise information specifically includes background noise and underwater environmental noise.

[0020] Further, in the step of obtaining the sound speed distribution data of sound waves underwater and generating the multipath propagation parameters of the sound waves according to the sound speed distribution data, it further includes:

[0021] Based on the distribution law of the sound velocity distribution data in different water layers, identify the influence coefficients of the different water layers on the sound wave, where the influence coefficients specifically include a flow velocity coefficient, a water temperature coefficient, and a salinity coefficient;

[0022] Determine whether the influence coefficients cause a change in the sound wave propagation path;

[0023] If so, based on the depth information of the different water layers, obtain the measured distance between the active sonar and the target, and based on the measured distance, dynamically correct the positioning accuracy of the target underwater. Through the positioning accuracy, adaptively adjust the sonar parameters of the active sonar, where the sonar parameters specifically include transmission power, signal frequency, and receiving sensitivity.

[0024] Further, in the step of determining whether a preset number of irregular reflection peaks are detected in the time-domain waveform, it further includes:

[0025] Based on the position parameters of the irregular reflection peaks in the time-domain waveform, calculate the delay distance of the irregular reflection peaks;

[0026] Determine whether the delay distance matches the actual sound wave propagation characteristics;

[0027] If so, collect the persistence of the irregular reflection peaks, generate the change period of the irregular reflection peaks according to the persistence, identify the variability of a single peak appearing within a preset time length from the change period, and based on the variability, collect the corresponding ambient noise.

[0028] Further, in the step of determining whether the propagation parameters can be modeled and eliminated, it further includes:

[0029] Based on the underwater environment and the target position, perform signal reconstruction on the reverberation signal, where the signal reconstruction specifically includes time-domain reconstruction and frequency-domain reconstruction;

[0030] Determine whether the actual signal of the active sonar reaches the preset matching value of the signal reconstruction;

[0031] If so, calculate the mean square error between the reconstructed signal and the actual signal, obtain the correlation coefficient between the reconstructed signal and the actual signal according to the mean square error, and based on the correlation coefficient, reflect the information sharing degree between the reconstructed signal and the actual signal.

[0032] Further, in the step of collecting the time-domain waveform of the echo signal based on the pre-monitoring of the target by the active sonar, it further includes:

[0033] Identify the noise sources in the echo signal, where the noise sources specifically include background noise, electronic noise, and environmental interference;

[0034] Determine whether the noise sources match the underwater features pre-recorded by the active sonar;

[0035] If not, dynamically adjust the detection parameters of the active sonar based on the noise sources, where the detection parameters specifically include propagation loss, reflection interface, and scattering coefficient.

[0036] The present invention also provides a sonar imaging reverberation cancellation system based on manifold constraint, including:

[0037] An acquisition module, configured to acquire the time-domain waveform of the echo signal based on the echo signal of the target pre-monitored by the active sonar;

[0038] A judgment module, configured to judge whether a preset number of irregular reflection peaks are detected in the time-domain waveform;

[0039] An execution module, configured to, if so, acquire the reflection path of the echo signal according to the pre-recorded characteristic information, predict the propagation parameters of the reverberation signal based on the reflection path, and dynamically adjust the detection threshold of the active sonar for the target through the reflection path and the propagation parameters, where the characteristic information specifically includes seabed topography, geological characteristics, and acoustic wave characteristics;

[0040] A second judgment module, configured to judge whether the propagation parameters can be modeled and eliminated;

[0041] A second execution module, configured to, if so, obtain the sound speed distribution data of the acoustic wave underwater based on the preset modeling basic information of the active sonar, generate the multi-path propagation parameters of the acoustic wave according to the sound speed distribution data, and model the to-be-eliminated features of each path based on the multi-path propagation parameters, where the modeling basic information specifically includes hard bottom, soft bottom, and rock, the multi-path propagation specifically includes the direct path from the sound source to the target and the indirect path after underwater reflection, and the to-be-eliminated features specifically include the time-delay feature, frequency change, and amplitude attenuation of each path.

[0042] Further, the execution module further includes:

[0043] A calculation unit, configured to calculate the propagation time delay of the reflection path based on the preset sound speed profile of the active sonar;

[0044] A judgment unit, configured to judge whether the propagation time delay is greater than a preset time delay threshold;

[0045] An execution unit, which, if so, obtains the offset frequency of the echo signal according to the relative speed during sound wave propagation, and calculates the attenuation characteristics of the reverberation signal based on the coefficient characteristics pre-recorded by the active sonar, where the coefficient characteristics specifically include the sound absorption coefficient of the water body, the reflection coefficient of the seabed, and the attenuation coefficient of the distance.

[0046] Furthermore, it further includes:

[0047] A second acquisition module, which is used to acquire the reflection characteristics of the target based on the monitoring type of the target, where the monitoring type specifically includes small targets, large targets, and dynamic targets, and the reflection characteristics specifically include reflection intensity, shape, size, and material;

[0048] A third judgment module, which is used to judge whether the monitoring type matches the reflection characteristics;

[0049] A third execution module, which, if so, obtains the ambient noise monitored in real time, distinguishes the difference data between the noise interference and the target echo from the ambient noise, and applies a preset Wiener filter to suppress the corresponding noise information according to the difference data, where the noise information specifically includes background noise and underwater ambient noise.

[0050] The present invention provides a method and system for sonar imaging reverberation cancellation based on manifold constraints, having the following beneficial effects:

[0051] By collecting the time-domain waveform of the echo signal in real time and judging whether irregular reflection peaks are detected, the present invention can quickly identify potential interferences and reflections in the underwater environment. Especially in a complex underwater environment, non-target objects (such as seabed obstacles or debris) may cause irregular reflection signals. At the same time, by collecting the reflection path of the echo signal and predicting the propagation parameters of the reverberation signal, the propagation characteristics of sound waves underwater can be accurately modeled, especially the seabed reflection signal. And through the pre-recorded information such as the seabed topography, geological characteristics, and sound wave propagation characteristics, accurate background data can be provided for the propagation modeling of the reverberation signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a sectional view of the overall structure of an embodiment of the method for sonar imaging reverberation cancellation based on manifold constraints of the present invention;

[0053] Figure 2 It is a partial structure diagram of an embodiment of the system for sonar imaging reverberation cancellation based on manifold constraints of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0054] It should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention. The implementation, functional features, and advantages of the present invention will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

[0055] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts belong to the scope of protection of the present invention.

[0056] Refer to the attached Figure 1 , which is a method for eliminating sonar imaging reverberation based on manifold constraints in an embodiment of the present invention, including:

[0057] S1: Based on the echo signal of the target pre-monitored by the active sonar, collect the time-domain waveform of the echo signal;

[0058] S2: Determine whether a preset number of irregular reflection peaks are detected in the time-domain waveform;

[0059] S3: If so, according to the pre-recorded characteristic information, collect the reflection path of the echo signal, and based on the reflection path, predict the propagation parameters of the reverberation signal. Through the reflection path and the propagation parameters, dynamically adjust the detection threshold of the active sonar for the target. Among them, the characteristic information specifically includes seabed topography, geological characteristics, and acoustic wave characteristics;

[0060] S4: Determine whether the propagation parameters can be modeled and eliminated;

[0061] S5: If possible, based on the preset modeling basic information of the active sonar, obtain the sound speed distribution data of the sound wave underwater. According to the sound speed distribution data, generate the multi-path propagation parameters of the sound wave. Based on the multi-path propagation parameters, model the characteristics to be eliminated for each path. Among them, the modeling basic information specifically includes hard bottom, soft bottom, and rock. The multi-path propagation specifically includes the direct path from the sound source to the target and the indirect path after underwater reflection. The characteristics to be eliminated specifically include the time-delay characteristics, frequency change, and amplitude attenuation of each path.

[0062] In this embodiment, the system pre - monitors the echo signals of targets underwater based on active sonar, collects the time - domain waveforms of these echo signals, and then the system determines whether the time - domain waveform detects a preset number of irregular reflection peaks to perform corresponding steps. For example, when the system determines that the time - domain waveform of the echo signal does not detect a preset number of irregular reflection peaks, the system will consider the echo signal to be relatively simple and not significantly interfered by factors such as seabed reflection, stray objects, or multipath propagation. The system will continue to perform target detection according to the normal process, enhance the amplification processing of the target signal, adjust the sensitivity of the sonar system to capture the target echo, further accurately analyze the time - frequency characteristics of the signal to ensure target detection, and dynamically adjust the detection threshold of the system to make it more sensitive to weaker signals, especially in an environment with a low signal - to - noise ratio. For example, when the system determines that the time - domain waveform of the echo signal detects a preset number of irregular reflection peaks, the system will consider that the echo signal is significantly interfered by factors such as seabed reflection. The system will collect the reflection path of the echo signal according to the pre - recorded characteristic information, which specifically includes seabed topography, geological characteristics, and acoustic wave characteristics. Based on different reflection paths, predict the propagation parameters of the reverberation signal, and dynamically adjust the detection threshold of the active sonar for the target through different reflection paths and propagation parameters. By detecting irregular reflection peaks, the system can identify the reverberation signal caused by seabed reflection or other stray reflections. Such reflections may come from changes in seabed topography, geological characteristics (such as soft bottom, hard bottom, rock, etc.) or acoustic wave propagation characteristics (such as changes in water flow, temperature, salinity, etc.). After identifying these reflection interference signals, the system can use the reflection path information to accurately distinguish the target signal from the non - target signal, thereby reducing the impact of reverberation on target detection. At the same time, by analyzing different reflection paths and propagation parameters, the system can dynamically adjust the target detection threshold in real - time to cope with changes in different underwater environments. For example, in an area with strong seabed reflection, the system may appropriately increase the threshold to avoid misjudging these reflection signals as target signals, while in a cleaner environment, the system may lower the threshold to improve the sensitivity of target detection. And using the reflection path and propagation parameter information, the system can identify the characteristics of the reverberation, and then adopt appropriate filtering and denoising techniques to reduce reverberation interference. By analyzing characteristics such as time delay, frequency change, and amplitude attenuation on the propagation path, the system can more accurately identify which signals belong to reverberation, so as to perform effective denoising and signal enhancement. Then the system determines whether the propagation parameters of the reverberation signal can be modeled and eliminated to perform corresponding steps;For example, when the system determines that the propagation parameters of the reverberation signal cannot be modeled and eliminated, the system will consider that the signal propagation path is strongly interfered (such as reflections from other underwater objects, marine biological activities, bubbles, or noise sources, etc.). The reverberation signal may be mixed in the target echo signal, resulting in the system being unable to accurately model the propagation path of these interference factors. The system will introduce a multipath propagation model and ultra-high-resolution echo analysis technology. By further refining the analysis of the reflection path, it tries to distinguish the interference signal from the target echo as much as possible. At the same time, in a complex interference environment, by fusing the data of multiple sonar sensors or combining underwater acoustic image data, it provides more accurate reverberation signal recognition and removal capabilities. For example, using sonar detection information with different frequencies and directions to enhance the signal discrimination; For example, when the system determines that the propagation parameters of the reverberation signal can be modeled and eliminated, at this time the system will consider that the signal propagation path is not significantly strongly interfered and can model the propagation path of the interference factors. The system will be based on the pre-set modeling basic information of the active sonar. The modeling basic information specifically includes hard bottom, soft bottom, and rock. It obtains the sound speed distribution data of the sound wave underwater. According to these sound speed distribution data, it generates the multipath propagation parameters of the sound wave. The multipath propagation specifically includes the direct path from the sound source to the target and the indirect path after underwater reflection. Based on the multipath propagation parameters, it models the characteristics to be eliminated for each path. The characteristics to be eliminated specifically include the time delay characteristics, frequency change, and amplitude attenuation of each path; By combining the sound speed distribution data underwater and the characteristics of different bottom sediments (hard bottom, soft bottom, and rock), the system can more accurately model the path of the sound wave propagating underwater. This accurate path modeling can help the system identify the propagation characteristics of the signal, including the direct path and the indirect reflection path, so as to effectively distinguish reverberation from the target echo. At the same time, by modeling the multipath propagation parameters (such as time delay, frequency change, amplitude attenuation, etc.), the system can identify the reverberation characteristics on each path and perform targeted elimination. Specifically, the system can construct an accurate elimination model by analyzing the propagation characteristics of different paths to reduce the impact of reverberation. And based on the modeling of the multipath propagation parameters, the system can dynamically adjust the threshold of target detection. By identifying the propagation characteristics of different paths, the system can adaptively adjust the threshold to avoid misjudging the target in a complex underwater environment. For example, in an area with more reflection paths or stronger signal attenuation, the system can adjust the threshold to improve the recognition ability of the target signal. And by obtaining and using the sound speed distribution data underwater, the system can better adapt to the dynamic changes of the underwater environment (such as temperature, salinity, and flow velocity changes). These changes will affect the speed and path of the sound wave propagation and the reflection characteristics of the signal. By obtaining the sound speed data in real time and adjusting the modeling parameters, the system can quickly respond to environmental changes and optimize the signal processing process.;

[0063] It should be noted that the reflection path of the echo signal is collected, and based on this reflection path, the propagation parameters of the reverberation signal are predicted. Through this reflection path and the propagation parameters, the detection threshold of the active sonar for the target is dynamically adjusted. The specific example is as follows:

[0064] Suppose a submarine uses an active sonar to detect a target in the ocean, and the environmental conditions are as follows:

[0065] Water depth: 500 meters;

[0066] Seabed sediment: The seabed is soft (silt layer);

[0067] Obstacles: Underwater rocks and reefs;

[0068] Target: A diving enemy submarine;

[0069] First, collect the reflection path of the echo signal. The active sonar equipment on the submarine emits a beam of sound waves and receives the echo signal reflected from the underwater target. At this time, the echo signal will experience the following reflection paths:

[0070] Direct path: The sound wave directly propagates from the sound source (the sonar transmitter of the submarine) to the target (the enemy submarine) and then reflects back to the receiver;

[0071] Indirect path: A part of the sound wave is reflected by the soft seabed and then propagates to the target again and then reflects back; The echo of this path usually has a large time delay because the sound wave needs to pass through the seabed;

[0072] At this step, the system determines the propagation path of the echo signal through the time difference of arrival of the echo signals collected by multiple array receivers; The time delay differences of these paths help the system distinguish between the direct path and the indirect path;

[0073] Then, predict the propagation parameters of the reverberation signal. By analyzing the propagation path of the echo signal, the system predicts the propagation parameters of the reverberation signal based on the following factors:

[0074] Time delay: The echo signal of the indirect path has a large time delay due to reflection by the seabed; By calculating the propagation speed of the sound wave and the path distance, the system can determine the time delay;

[0075] Frequency change: Since the sound wave is reflected by the seabed, especially the soft seabed, it may cause a change in the frequency of the sound wave, especially when there is a significant speed difference between the silt layer on the seabed and the water layer; This frequency shift will affect the characteristics of the echo signal and thus affect the recognition of the target signal;

[0076] Amplitude attenuation: Since sound waves attenuate during propagation due to reasons such as absorption and scattering by water bodies, especially signals reflected by soft bottoms, the attenuation is more significant; the system can estimate the amplitude attenuation of each path by analyzing the signal intensity of different paths.

[0077] Subsequent dynamic adjustment of the target detection threshold. Once the system predicts the propagation parameters of the reverberation signal, the next step is to dynamically adjust the target detection threshold:

[0078] Influence of time delay characteristics: The echo signal of the indirect path has a longer time delay, which means that the arrival time of the echo is different from that of the direct path echo; the system can set an appropriate time delay window based on the time delay information of the echo to distinguish between the effective target echo and interference signals from other paths; if the system detects that the time delay matches that of the target echo, the system will lower the threshold to enhance the detection sensitivity of the target signal.

[0079] Influence of frequency change: If it is predicted that the echo signal frequency shifts due to the reflection of the soft bottom of the seabed, the system will adjust the frequency range to a range that adapts to this change to avoid misjudging the reverberation signal with a large frequency shift as a target signal; the system can set a frequency width threshold to ensure that only echo signals with frequencies matching the target signal are accepted.

[0080] Influence of amplitude attenuation: Since the reflection of the soft bottom usually causes the amplitude attenuation of the echo signal, the system dynamically adjusts the detection threshold by estimating the attenuation degree; if the amplitude attenuation of the echo signal is large, the system will lower the target detection threshold to detect possible weak target echo signals; if the attenuation is small, the threshold will be increased to avoid misjudging weak background noise as a target echo.

[0081] Optimized target detection: By accurately modeling the time delay, frequency change, and amplitude attenuation of multi-path signals, the system successfully identifies the echo signal of the submarine, avoiding the reverberation interference caused by the reflection of the soft bottom of the seabed; after dynamically adjusting the detection threshold, the detection sensitivity of the target signal is effectively improved.

[0082] Reducing false alarms: By adjusting the threshold, the system avoids misjudging weak echoes generated by seabed reflections as target signals, reducing the occurrence of false alarms and ensuring more accurate detection of submarines.

[0083] In summary, by collecting the reflection paths of echo signals, predicting the propagation parameters of reverberation signals, and dynamically adjusting the detection threshold based on these parameters, the system can effectively identify target echoes in a complex underwater environment, reduce the interference of reverberation, and enhance the reliability of target detection.

[0084] It should be added that based on the preset modeling basic information of the active sonar, the sound speed distribution data of sound waves underwater is obtained. According to the sound speed distribution data, the multipath propagation parameters of the sound waves are generated. Based on the multipath propagation parameters, the characteristics to be eliminated for each path are modeled. The specific examples are as follows:

[0085] Suppose a submarine detects underwater obstacles through an active sonar, and the environmental conditions are as follows:

[0086] Water depth: 300 meters;

[0087] Water body: Deep sea area, with relatively stable seawater salinity and temperature;

[0088] Seabed sediment: The seabed is soft (silt layer);

[0089] Obstacle: A large rock (to be avoided);

[0090] First, the sound speed distribution data of sound waves underwater is obtained. When sound waves propagate underwater, the speed is affected by various factors, including water temperature, salinity, and depth. The system needs to obtain accurate sound speed distribution data based on the underwater environmental conditions, mainly including:

[0091] Water layer distribution: The propagation speed of sound waves is different in water layers at different depths. Especially in the deep sea area, changes in water temperature and salinity may lead to significant differences in sound speed;

[0092] Seabed characteristics: The material of the seabed (soft bottom, hard bottom, rock, etc.) also affects the propagation speed of sound waves;

[0093] The system can obtain the sound speed distribution data through a sound speed profile measurement tool (such as a sound speed detector) or from existing ocean data sets. Suppose the sound speed measured at a water depth of 300 meters is 1500 m / s, and the sound speed slightly increases with the increase in depth;

[0094] Then, based on the sound speed distribution data, the multipath propagation parameters of the sound waves are generated. Based on the sound speed distribution data, the system generates the multipath propagation parameters of the sound waves. After the sound waves are emitted from the transmitter, they may propagate to the receiver along different paths, mainly including:

[0095] Direct path: From the sound source directly to the target and then back;

[0096] Indirect path: After the sound waves are reflected by the seabed or underwater obstacles (such as rocks, ships, etc.), they then propagate to the target and are reflected back to the receiver;

[0097] For each propagation path, the system will calculate the following parameters:

[0098] Time delay: The propagation time of each path; the propagation times of different paths are different. The direct path is usually shorter, while the indirect path through reflection has a longer propagation time;

[0099] Frequency change: During the propagation of sound waves, due to different propagation distances and media, the frequency may change; especially after reflection from the seabed, the frequency may be attenuated or shifted;

[0100] Amplitude attenuation: The signal intensity on each path will be weakened due to scattering, absorption, and reflection losses during propagation; the direct path usually has less amplitude attenuation, while the indirect path (especially through the seabed) may have greater attenuation;

[0101] For example, in this example, assume that after the sound wave propagates directly to the target obstacle, through calculation, the propagation time delay of the sound wave is 0.2 seconds, the amplitude attenuation is -3 dB, and the frequency change is small (only ±0.5%); the sound wave first reflects from the seabed and then propagates to the target obstacle, with a time delay of 0.5 seconds, an amplitude attenuation of -8 dB, and a larger frequency change (±5%);

[0102] Subsequently, based on the multipath propagation parameters, model the characteristics to be eliminated for each path. Once the multipath propagation parameters are obtained, the system can model the characteristics to be eliminated for each path based on these parameters, mainly including:

[0103] Time delay characteristic: Since the propagation times of different paths are different, the system needs to adjust the reception time window of the signal according to the time delay characteristics of each path to avoid signal aliasing of different paths; for the echo signal of the indirect path, the system may delay the acquisition to ensure that the direct path signal and the indirect path signal are not misidentified as the same target echo at the same time;

[0104] Frequency change characteristic: Sound waves will have frequency offsets due to different media during propagation, especially when reflecting from the seabed; the system needs to model these frequency changes and correct the frequency during signal processing to avoid signal errors caused by frequency changes;

[0105] Amplitude attenuation characteristic: The amplitude attenuation of the signal can be corrected through the built-in propagation model of the system; the system compensates the received echo signal according to the attenuation coefficients of different paths to restore a more realistic target signal; for the indirect path signal reflected from the seabed, the system can perform gain compensation according to the attenuation model;

[0106] Finally, dynamically eliminate reverberation and optimize target signal detection. By modeling the characteristics to be eliminated for each path, the system can effectively eliminate the interference of reverberation on the target signal and enhance the clarity of the signal;

[0107] Time delay correction: The system will perform signal time delay compensation according to the time delay characteristics of the path to ensure that signals from different paths are correctly aligned;

[0108] Frequency correction: Through frequency correction, the system will remove the frequency offset caused by seabed reflection, making the frequencies of the target signal and the reverberation signal more consistent and avoiding false alarms caused by frequency drift;

[0109] Amplitude compensation: The system compensates the signals of different paths according to the attenuation characteristics of the propagation path to improve the detectability of weak signals;

[0110] As a result, the accuracy of target detection can be improved: By dynamically eliminating the reverberation signals caused by multipath propagation, the system can effectively and clearly identify the echo signals of target obstacles, avoiding interference from seabed reflection or other obstacles; eliminating signal errors caused by time delay, frequency change, and amplitude attenuation, ensuring that the target echo signals are clearly distinguishable;

[0111] In summary, the system can efficiently optimize the processing process of echo signals through the multipath propagation parameters generated based on the sound speed distribution data and modeling the characteristics to be eliminated for each path by these parameters, thereby improving the accuracy and stability of target detection in a complex underwater environment; this method is particularly applicable to environments with large changes in seabed sediment and many obstacles.

[0112] In this embodiment, in step S3 of collecting the reflection path of the echo signal and predicting the propagation parameters of the reverberation signal according to the reflection path, it further includes:

[0113] S31: Calculate the propagation time delay of the reflection path based on the preset sound speed profile of the active sonar;

[0114] S32: Determine whether the propagation time delay is greater than the preset time delay threshold;

[0115] S33: If so, obtain the offset frequency of the echo signal according to the relative speed during sound wave propagation, and calculate the attenuation characteristics of the reverberation signal according to the coefficient characteristics pre-recorded by the active sonar, where the coefficient characteristics specifically include the sound absorption coefficient of the water body, the reflection coefficient of the seabed, and the attenuation coefficient of the distance.

[0116] In this embodiment, the system calculates the propagation delay of the reflection path based on the pre-set sound speed profile of the active sonar, and then the system determines whether the propagation delay is greater than the pre-set delay threshold to execute corresponding steps. For example, when the system determines that the propagation delay of the reflection path is not greater than the pre-set delay threshold, the system will consider that the signal has not been reflected by a complex underwater environment (such as the seabed, rocks or obstacles, etc.), so no complex multipath effect or severe reverberation will be generated. The system will appropriately lower the detection threshold to ensure that even if the target signal is weak (due to the short distance and low signal intensity), it can be detected in time. At the same time, the decision threshold for false alarms is increased to reduce the chance of false alarms. For example, the system may consider that within this delay range, the target is relatively reliable, so it may no longer pay attention to interference sources (such as noise, clutter, etc.), and maintain the current target monitoring state, and be ready to enter a more complex monitoring mode at any time to cope with subsequent possible complex situations (such as target position changes, seabed changes, etc.). For example, when the system determines that the propagation delay of the reflection path is greater than the pre-set delay threshold, at this time the system will consider that the signal has been reflected by a complex underwater environment. The system will obtain the offset frequency of the echo signal according to the relative speed during sound wave propagation, and calculate the attenuation characteristics of the reverberation signal based on the coefficient characteristics pre-recorded by the active sonar. The coefficient characteristics specifically include the absorption coefficient of the water body, the reflection coefficient of the seabed, and the attenuation coefficient of the distance. By judging that the propagation delay is greater than the set threshold, the system can accurately distinguish whether the signal has passed through a complex underwater reflection environment. These reflected signals usually become more complex due to the influence of multipath propagation, seabed reflection, rocks and other obstacles, generating characteristics such as reverberation and delay. The system can identify these characteristics and make corresponding optimization processing. At the same time, by calculating the attenuation characteristics of the reverberation signal, the system can identify and eliminate the interference of factors such as multipath effect and seabed reflection on the signal, thereby improving the detection accuracy of the target. Especially in a complex environment, the system can accurately eliminate the errors caused by the underwater environment, improve the signal quality, thereby avoiding false target recognition or false alarms. And by relying on the pre-recorded coefficient characteristics (such as absorption coefficient, reflection coefficient, etc.), the system can dynamically adapt to the changes in different underwater environments. The water quality, seabed geological characteristics and obstacle distribution in different regions may cause changes in signal propagation characteristics. The system can adaptively adjust the signal processing strategy by calculating and adjusting the attenuation characteristics, so as to ensure the efficient operation of the system in various environments.

[0117] It should be noted that obtaining the offset frequency of the echo signal according to the relative speed during sound wave propagation and calculating the attenuation characteristics of the reverberation signal based on the coefficient characteristics pre-recorded by the active sonar are specifically exemplified as follows:

[0118] Suppose an active sonar is used to detect a target (such as a submarine) in the ocean. The submarine has relative motion with respect to the sonar transmitter, and during the signal propagation, the sound wave passes through a complex seabed environment, resulting in reflection and attenuation. The following are the specific steps to calculate the reverberation signal attenuation based on the above characteristics:

[0119] First, obtain the echo signal. The system receives the echo signal returned from the submarine and detects the offset frequency of the signal through frequency analysis. Suppose the original frequency of the echo is 10 kHz, and the system calculates a frequency offset of 200 Hz, which indicates that the submarine has a certain motion (relative velocity) with respect to the sonar transmitter;

[0120] The frequency offset (Δf) can be calculated by the following formula:

[0121]

[0122] where, is the relative velocity between the sound source and the target, is the sound speed in water, is the original frequency of the sound wave;

[0123] Through calculation,

[0124] This shows that the relative velocity of the submarine is 10 m / s, resulting in an offset of 66.67 Hz in the echo frequency;

[0125] Calculate the attenuation characteristics: Suppose the absorption coefficient of the water body is known to be 0.02 dB / m, the reflection coefficient of the seabed is 0.8 (hard bottom), and the target is 1000 meters away from the sound source. According to the attenuation coefficient based on the distance (assumed to be 0.05 dB / m), the signal attenuation can be calculated:

[0126] Absorption attenuation = 0.02 dB / m × 1000 m = 20 dB;

[0127] Reflection attenuation = (1 - 0.8) × 10 dB = 2 dB;

[0128] Total attenuation = 20 dB + 2 dB = 22 dB;

[0129] Modify the reverberation signal. The system corrects the signal intensity based on the attenuation characteristics (22 dB) to make it closer to the echo signal of the real target. The attenuation characteristics help the system identify which parts of the signal are strongly absorbed and attenuated, and thus determine where to strengthen or compensate in terms of frequency and path;

[0130] In summary, by obtaining the offset frequency of the echo signal and calculating the attenuation characteristics of the reverberation signal based on the coefficient characteristics, the system can accurately identify the signal propagation path and the signal attenuation factors, and at the same time dynamically adjust the signal processing strategy. For example, it can enhance weak signals, weaken interference signals, and improve the detection accuracy to ensure the effective extraction of the target signal.

[0131] In this embodiment, before step S3 of dynamically adjusting the detection threshold of the active sonar for the target through the reflection path and the propagation parameters, the following steps are further included:

[0132] S301: Based on the monitoring type of the target, collect the reflection characteristics of the target, where the monitoring type specifically includes small targets, large targets, and dynamic targets, and the reflection characteristics specifically include reflection intensity, shape, size, and material;

[0133] S302: Determine whether the monitoring type matches the reflection characteristics;

[0134] S303: If so, obtain the real-time monitored environmental noise, distinguish the difference data between the noise interference and the target echo from the environmental noise, and apply the preset Wiener filter to suppress the corresponding noise information according to the difference data, where the noise information specifically includes background noise and underwater environmental noise.

[0135] In this embodiment, the system monitors the reflection characteristics of a target based on the monitoring type of the target. The monitoring types specifically include small targets, large targets, and dynamic targets. The reflection characteristics specifically include reflection intensity, shape, size, and material. Then, the system determines whether the monitoring type of the target matches the reflection characteristics of the target to execute corresponding steps. For example, when the system determines that the monitoring type of the target does not match the reflection characteristics of the target, the system will consider that the reflection characteristics of the target do not match the expected monitoring type, which may be because the physical properties, shape, or material characteristics of the target do not match the set monitoring model. The system will automatically switch to a more suitable monitoring mode based on the reflection characteristics of the current target. For example, if it was originally set to monitor large targets, but a small object is detected, the system can switch to a monitoring type more suitable for small targets, adjust the detection sensitivity and algorithm parameters. At the same time, if the target is a non-metal or sound-absorbing material, the system can increase or adjust parameters such as the sound absorption coefficient and reflection coefficient to make the model more conform to the reflection characteristics of the actual target. For irregular or special-shaped targets, the system can perform shape analysis and adjust the shape recognition model of the echo signal to adapt to the reflection mode of irregular targets. In addition to sonar echo signals, the system can perform data fusion through other sensors (such as radar, infrared, or optical sensors) to help improve the accuracy of target recognition. Especially in a complex environment, when the sonar echo signal cannot clearly identify the target, the information of other sensors (such as the thermal radiation of the target, changes in the reflection waveform, etc.) can be combined for supplementation to more accurately infer the characteristics and types of the target. For example, when the system determines that the monitoring type of the target can match the reflection characteristics of the target, at this time, the system will consider that the reflection characteristics of the target match the expected monitoring type. The system will obtain the real-time monitored environmental noise, and distinguish the difference data between the noise interference and the target echo from these environmental noises. According to different difference data, the preset Wiener filter is applied to suppress the corresponding noise information. The noise information specifically includes background noise and underwater environmental noise. By obtaining the environmental noise information in real time and distinguishing the difference between the noise and the echo, the system can accurately identify the noise interference source and effectively reduce the impact of these interferences on the target echo, which is particularly important for the detection of dynamic targets and targets in a complex environment. At the same time, by dynamically applying the Wiener filter, the system can adjust the noise suppression strategy according to different environmental noise characteristics. For example, the characteristics of background noise and underwater environmental noise are different, and the system can adjust the intensity and method of noise suppression according to the real-time environmental monitoring data to ensure the optimization of the noise suppression effect. And after the target echo signal passes through noise suppression, the system can perform further target recognition, tracking, classification, etc. based on the cleaner signal. The improvement of the signal quality provides a more accurate basis for subsequent decisions, which helps to improve the working efficiency and accuracy of the entire system.

[0136] It should be noted that the real-time monitored environmental noise is obtained, and the difference data between the noise interference and the target echo is distinguished from the environmental noise. According to the difference data, the preset Wiener filter is applied to suppress the corresponding noise information. The specific example is as follows:

[0137] Suppose the system is monitoring a submarine underwater; the system emits sound waves through an active sonar and receives the echo signal reflected by the submarine; underwater, the target echo signal is usually interfered by environmental noise (such as ocean currents, ship activities, seabed reflections, etc.), which affects the signal quality and reduces the accuracy of target detection;

[0138] First, collect environmental noise. The system first collects environmental noise signals through sonar sensors; the noise may come from different sources, such as:

[0139] Ship noise: low-frequency noise, usually occurring between 20 Hz and 500 Hz, may be caused by the mechanical equipment of ships or submarines on the water surface;

[0140] Ocean current noise: high-frequency noise, usually occurring between 2 kHz and 5 kHz, is caused by the interaction between water currents and seabed structures;

[0141] Suppose the system collects a 10-second signal segment in real time, which contains the submarine echo signal and ocean current and ship noise;

[0142] Then, distinguish the difference between the noise and the target echo signal. The system distinguishes the noise and the submarine echo by analyzing the spectral characteristics of the signal:

[0143] The submarine echo signal usually shows the characteristic of stable frequency, and may have obvious time delay and amplitude attenuation; for example, the reflected echo of the submarine may be mainly concentrated between 1 kHz and 3 kHz, and as the distance increases, its amplitude will gradually attenuate;

[0144] Ship noise is mainly low-frequency (20 Hz to 500 Hz) and shows strong randomness;

[0145] Ocean current noise is usually high-frequency (2 kHz to 5 kHz) noise, with a relatively wide spectrum and usually no obvious periodicity;

[0146] Through these characteristics, the system can distinguish the spectral differences between the submarine echo signal and the noise; for example, the time delay of the submarine echo signal is about 3 seconds, while the time delay of the noise is random and irregular;

[0147] Subsequently, apply Wiener filtering. After clarifying the spectral differences between the noise and the echo signal, the system starts to apply Wiener filtering to suppress the noise; the Wiener filter optimizes its coefficients based on the power spectral density of the signal and the noise, thereby eliminating the noise;

[0148] Step 1: Estimate the power spectral density (PSD) of the noise. The system first estimates the power spectral density of the noise. Assume that the system uses the signals in the past period to estimate the statistical characteristics (i.e., the power spectrum) of the noise. Assume that:

[0149] The power spectrum of vessel noise is concentrated between 20 Hz and 500 Hz;

[0150] The power spectrum of ocean current noise is concentrated between 2 kHz and 5 kHz;

[0151] Assume that the system can obtain such a noise power spectrum through the noise model:

[0152] Low-frequency noise (20 Hz - 500 Hz): The power spectrum is P_noise_low = 0.5;

[0153] High-frequency noise (2 kHz - 5 kHz): The power spectrum is P_noise_high = 0.2;

[0154] Step 2: Estimate the power spectrum of the target echo. The system then estimates the power spectrum of the submarine echo signal. Assume that the frequency spectrum of the submarine echo signal is concentrated between 1 kHz and 3 kHz, and the time-delay characteristic of the signal is relatively clear. As the reflection distance increases, the signal will show a certain frequency attenuation (for example, in the high-frequency band of 3 kHz, the amplitude of the echo signal will attenuate more than that at 1 kHz); The system estimates the power spectrum of the submarine echo signal: P_echo = 1 in the range of 1 kHz to 3 kHz;

[0155] Step 3: Optimize the Wiener filter. According to the estimated signal and noise power spectra above, the Wiener filter will be optimized; The goal of the Wiener filter is to minimize the mean square error between the signal and the noise, that is, by calculating the gain function of the filter, to retain the submarine echo signal to the greatest extent while suppressing the noise;

[0156] For example, for the echo signal in the range of 1 kHz to 3 kHz, the Wiener filter will give a higher gain. Assume the gain is G_echo = 1 (i.e., the signal remains unchanged), while for the ocean current noise in the range of 2 kHz to 5 kHz, the filter will reduce the gain. Assume the gain is G_noise_high = 0.1;

[0157] Step 4: Filter and enhance the echo signal. According to the optimized filter coefficients, the system will filter the original signal, remove the noise part, and retain the target echo signal at the same time;

[0158] For the echo signal in the range of 1 kHz to 3 kHz, the system will completely retain the signals in these frequency bands;

[0159] For the noise in the ranges of 20 Hz to 500 Hz and 2 kHz to 5 kHz, the system will suppress the signal intensity in these frequency bands;

[0160] After Wiener filtering, the target echo signal becomes clearer and the noise part is effectively removed; for example, the frequency attenuation characteristics of the submarine echo signal are retained, and the time delay and frequency changes can also be more clearly reflected; the noise part (such as the noise from ships and ocean currents) is effectively suppressed, especially in the low-frequency and high-frequency bands, and the influence of the noise is greatly reduced;

[0161] Finally, detection is performed after signal enhancement, and the enhanced signal will greatly improve the signal-to-noise ratio of the target echo; the system can more accurately extract information such as the position, speed, and reflection intensity of the submarine; for example:

[0162] The reflection intensity and position of the submarine can be accurately estimated;

[0163] The system can track the movement trajectory of the submarine, further improving the accuracy of target detection and tracking;

[0164] In summary, through Wiener filtering, the system can effectively remove underwater noise and enhance the target echo signal; in this example, Wiener filtering effectively removes the low-frequency noise of ships and the high-frequency noise of ocean currents, making the echo signal of the submarine clearer; in this way, the system can more accurately detect and track the submarine, significantly improving the accuracy and reliability of target detection; this noise suppression method is very important for target monitoring in complex underwater environments and can greatly improve the performance of active sonar.

[0165] In this embodiment, in step S5 of obtaining the sound speed distribution data of the sound wave underwater and generating the multipath propagation parameters of the sound wave according to the sound speed distribution data, it further includes:

[0166] S51: Based on the distribution law of the sound speed distribution data in different water layers, identify the influence coefficients of the different water layers on the sound wave, where the influence coefficients specifically include a flow velocity coefficient, a water temperature coefficient, and a salinity coefficient;

[0167] S52: Determine whether the influence coefficients cause changes in the sound wave propagation path;

[0168] S53: If so, according to the depth information of the different water layers, obtain the measured distance between the active sonar and the target, and based on the measured distance, dynamically correct the positioning accuracy of the target underwater, and adaptively adjust the sonar parameters of the active sonar through the positioning accuracy, where the sonar parameters specifically include transmission power, signal frequency, and receiving sensitivity.

[0169] In this embodiment, based on the distribution law of the sound velocity distribution data in different water layers, the system identifies the influence coefficients of different water layers on sound waves. The influence coefficients specifically include the flow velocity coefficient, the water temperature coefficient, and the salinity coefficient. Then, the system determines whether these influence coefficients cause changes in the sound wave propagation path to execute corresponding steps. For example, when the system determines that the influence coefficients of different water layers on sound waves do not cause changes in the sound wave propagation path, the system will consider that parameter changes such as the flow velocity, water temperature, and salinity in the water layer do not significantly change the sound wave propagation mode. The system will continue to use the current propagation model for target detection without the need to adjust the sound wave propagation path, that is, the system can continue to execute conventional tasks such as echo signal processing, target recognition, and tracking, while saving computing resources without the need for additional path correction or data calibration, and directly enter target detection and tracking, thereby improving real-time performance. And when no significant path changes are found, the system can ensure the monitoring stability under the existing environmental conditions, without the need to react frequently to environmental changes, and can maintain the stable operation of the system, focusing on target detection and echo analysis. For example, when the system determines that the influence coefficients of different water layers on sound waves cause changes in the sound wave propagation path, at this time, the system will consider that the water layer significantly changes the sound wave propagation mode. The system will obtain the measured distance between the active sonar and the target based on the depth information of different water layers. Based on these measured distances, the system will dynamically correct the positioning accuracy of the target underwater. Through different positioning accuracies, the sonar parameters of the active sonar will be adaptively adjusted. The sonar parameters specifically include the transmission power, the signal frequency, and the receiving sensitivity. By adaptively adjusting the transmission power, signal frequency, and receiving sensitivity of the active sonar according to the positioning accuracy, the system can optimize the sonar performance, provide the best signal quality under different water layer conditions. At the same time, according to the actual position of the target and the positioning accuracy, the system can optimize the transmission power and receiving sensitivity of the sound wave, thereby improving the signal-to-noise ratio of the echo signal and reducing the interference of environmental noise on the signal, which is crucial for target detection and recognition, especially in the case of strong background noise and complex underwater environments. And dynamically adjusting the sonar parameters can not only improve the positioning accuracy and detection performance, but also save unnecessary power and computing resources. For example, in an area where the underwater signal attenuation is small, the system can appropriately reduce the transmission power to reduce energy consumption, while in an area where the signal attenuation is large, the system will enhance the signal to ensure reliable detection of the target.

[0170] In step S2 of determining whether a preset number of irregular reflection peaks are detected in the time-domain waveform in this embodiment, it further includes:

[0171] S21: Calculate the delay distance of the irregular reflection peak based on the position parameter of the irregular reflection peak in the time-domain waveform;

[0172] S22: Determine whether the delay distance matches the actual sound wave propagation characteristics;

[0173] S23: If so, collect the persistence of the irregular reflection peak, generate the change period of the irregular reflection peak according to the persistence, identify the variability of the occurrence of a single peak within a preset time duration from the change period, and collect the corresponding environmental noise based on the variability.

[0174] In this embodiment, the system calculates the delay distance of the irregular reflection peak based on the position parameter of the irregular reflection peak in the time-domain waveform, and then the system determines whether the delay distance matches the actual acoustic wave propagation characteristics to perform corresponding steps; for example, when the system determines that the delay distance of the irregular reflection peak cannot match the actual acoustic wave propagation characteristics, the system will consider that the reflection path of the echo signal may be affected by external interference or environmental factors, resulting in the propagation characteristics of the reflected signal not conforming to the expectation. The system will correct the sound speed distribution according to the real-time data obtained by the underwater sensor, thereby adjusting the calculation of the acoustic wave propagation delay, avoiding errors caused by environmental changes, and at the same time checking the working state of the active sonar device to see if there are device failures or configuration errors that cause the delay of the reflection peak not to match the theoretical value. If the device is operating normally but the delay still does not match, check whether parameters such as the transmission power, frequency, and reception sensitivity of the sonar signal need to be adjusted. And if the calculation of the delay of the irregular reflection peak continuously fails to match the actual propagation characteristics, the system can enable the target tracking function, combine the motion state and historical trajectory of the target, speculate on the target position and propagation path, and correct the delay calculation; for example, when the system determines that the delay distance of the irregular reflection peak can match the actual acoustic wave propagation characteristics, at this time the system will consider that the reflection path of the echo signal has not been affected by external interference, etc. The system will collect the persistence of the irregular reflection peak, generate the change period of the irregular reflection peak according to the persistence, identify the variability of the occurrence of a single peak within a preset time duration from the change period, and collect the corresponding environmental noise based on different variabilities; by confirming that the delay of the irregular reflection peak is consistent with the actual propagation characteristics, the system can ensure that the reflection path of the echo signal has not been interfered by external factors, which is crucial for accurately judging the target position and the acoustic wave propagation path. At the same time, analyzing the persistence, change period, and variability of the irregular reflection peak helps the system identify possible regular features in the echo signal. Based on these features, the system can determine whether the echo signal has continuous periodic or non-periodic fluctuations. And by monitoring the change period of the irregular reflection peak, the system can identify which peak changes are related to environmental noise and which are related to target reflections. The system can capture the change characteristics related to environmental noise based on the differences identified in the change period. And by accurately analyzing the change period of the irregular reflection peak and its matching with the target echo, the system can more clearly distinguish the difference between the target signal and environmental noise, thereby improving the accuracy and response speed of target positioning.

[0175] In this embodiment, in step S4 of determining whether the propagation parameters can be modeled and eliminated, the following steps are further included:

[0176] S41: Based on the underwater environment and the target position, perform signal reconstruction on the reverberation signal, where the signal reconstruction specifically includes time-domain reconstruction and frequency-domain reconstruction;

[0177] S42: Determine whether the actual signal of the active sonar reaches the preset matching value of the signal reconstruction;

[0178] S43: If so, calculate the mean square error between the reconstructed signal and the actual signal, obtain the correlation coefficient between the reconstructed signal and the actual signal according to the mean square error, and reflect the information sharing degree between the reconstructed signal and the actual signal based on the correlation coefficient.

[0179] In this embodiment, the system reconstructs the reverberation signal based on the underwater environment and the target position. The signal reconstruction specifically includes time-domain reconstruction and frequency-domain reconstruction. Then, the system determines whether the actual signal of the active sonar reaches the preset matching value of the signal reconstruction to execute corresponding steps. For example, when the system determines that the actual signal of the active sonar does not reach the preset matching value of the signal reconstruction, the system will consider that there are some unexpected deviations in the signal reconstruction process, which may be due to errors in the signal processing process or external interference. The system will re-collect the environmental parameters, dynamically adjust the propagation model of the sonar, and at the same time, based on the new environmental data, such as the underwater sound speed profile and the propagation path, adjust the time-domain or frequency-domain parameters in the signal reconstruction in real time for adaptive correction, and based on the real-time echo signal, re-estimate the propagation parameters such as the reflection path, propagation delay, and attenuation coefficient, so as to optimize the signal reconstruction process. For example, when the system determines that the actual signal of the active sonar reaches the preset matching value of the signal reconstruction, at this time, the system will consider that the signal reconstruction process meets the expectations. The system will calculate the mean square error between the reconstructed signal and the actual signal, and according to this mean square error, obtain the correlation coefficient between the reconstructed signal and the actual signal. Based on these correlation coefficients, the information sharing degree between the reconstructed signal and the actual signal is reflected. By calculating the mean square error (MSE) between the reconstructed signal and the actual signal, the system can quantify the accuracy of the signal reconstruction. The smaller the mean square error, the smaller the gap between the signal reconstruction and the actual echo signal, and the more accurately the system can restore the propagation characteristics of the signal. At the same time, by obtaining the correlation coefficient between the reconstructed signal and the actual signal, the system can quantify the linear correlation between them. The closer the correlation coefficient is to 1, the more consistent the reconstructed signal and the actual signal are in characteristics, and the higher the information sharing degree. And when the signal reconstruction process meets the expectations, and both the mean square error and the correlation coefficient are within the ideal range, the system can determine that the reconstruction process is effective and reliable, which provides a solid foundation for subsequent signal analysis, target detection, noise cancellation, etc. According to the correlation between the reconstructed signal and the actual signal, the system can further adjust and optimize the parameters in the signal processing flow (such as transmission power, frequency, receiving sensitivity, etc.). If the correlation is found to be low, the system will timely adjust the operation mode of the sonar and optimize the signal processing algorithm to better match the actual signal characteristics.

[0180] It should be noted that calculating the mean square error between the reconstructed signal and the actual signal, obtaining the correlation coefficient between the reconstructed signal and the actual signal according to the mean square error, and reflecting the information sharing degree between the reconstructed signal and the actual signal based on the correlation coefficient are specifically exemplified as follows:

[0181] Suppose in an underwater environment, the system uses sonar to send signals and receive echo signals. After certain signal processing, the system reconstructs the echo signals to obtain a reconstructed signal. Now, it is necessary to compare the similarity between the reconstructed signal and the actually received echo signal to evaluate the quality of signal reconstruction;

[0182] First, calculate the mean square error (MSE). The mean square error (Mean Square Error, MSE) is a criterion for measuring the difference between two signals (the actual signal and the reconstructed signal). It calculates the differences between the actual signal and the reconstructed signal at each sampling point, and then takes the average of the sum of the squares of these differences. The formula is as follows:

[0183]

[0184] where, is the total number of sampling points of the signal;

[0185] is the value of the i-th sampling point of the actual signal;

[0186] is the value of the i-th sampling point of the reconstructed signal;

[0187] Suppose the values of the first sample points of the actual signal and the reconstructed signal are respectively:

[0188] Actual signal: ;

[0189] Reconstructed signal: ;

[0190] If the signal length is 5, the actual signal and the reconstructed signal are respectively:

[0191] Actual signal: [10, 15, 20, 25, 30];

[0192] Reconstructed signal: [8, 16, 19, 26, 29];

[0193] Then the calculation of the mean square error is:

[0194]

[0195] Then use the correlation coefficient formula to calculate the correlation between these two signals. The correlation coefficient is an index for measuring the degree of linear correlation between two signals. Its value ranges from [-1, 1]. The closer the value is to 1, the more highly correlated the two signals are. The closer it is to -1, the more completely negatively correlated the two signals are. And when it is close to 0, the two signals are irrelevant. The calculation formula for the correlation coefficient is:

[0196]

[0197] Among them, and are the means of the actual signal and the reconstructed signal; the value of the correlation coefficient r reflects the similarity between the reconstructed signal and the actual signal. When r is close to 1, it indicates that the two are highly consistent and the reconstruction effect is good. When the correlation coefficient is negative or close to zero, it indicates that the similarity between the reconstructed signal and the actual signal is low, and the signal reconstruction algorithm may need to be adjusted;

[0198] Assume that the means of the actual signal and the reconstructed signal are respectively:

[0199]

[0200]

[0201] Substitute the above data into the formula to calculate the correlation coefficient r:

[0202]

[0203] Assume that the calculated correlation coefficient is r = 0.98. According to the results, the mean square error MSE = 1.6 and the correlation coefficient r = 0.98, the following conclusions can be drawn: the mean square error is small, indicating that the difference between the reconstructed signal and the actual signal is small and the reconstruction effect is good, while the correlation coefficient is high (close to 1), indicating that the reconstructed signal and the actual signal are highly consistent and the information sharing degree is high, and the system can effectively restore the characteristics of the echo signal.

[0204] In this embodiment,

[0205] In this embodiment, in step S1 of collecting the time-domain waveform of the echo signal based on the echo signal of the active sonar for pre-monitoring the target, it further includes:

[0206] S11: Identify the noise sources in the echo signal, where the noise sources specifically include background noise, electronic noise, and environmental interference;

[0207] S12: Determine whether the noise sources match the underwater characteristics pre-recorded by the active sonar;

[0208] S13: If not, dynamically adjust the detection parameters of the active sonar based on the noise sources, where the detection parameters specifically include propagation loss, reflection interface, and scattering coefficient.

[0209] In this embodiment, the system identifies the sources of noise in the echo signal. The sources of noise specifically include background noise, electronic noise, and environmental interference. Then, the system determines whether these sources of noise match the underwater characteristics pre-recorded by the active sonar to execute corresponding steps. For example, when the system determines that the sources of noise in the echo signal can match the underwater characteristics pre-recorded by the active sonar, the system will consider that the noise in the echo signal may be known environmental noise, and can take preset noise suppression measures for processing based on the existing underwater characteristics and noise characteristics. The system will identify the sources of noise in the echo signal (such as background noise, electronic noise, environmental interference, etc.), and determine its source by comparing with the pre-recorded underwater noise characteristics. For example, if the identified noise matches the preset underwater ocean current noise or equipment electronic noise of the system, the system can infer that the noise does not come from the target echo itself, but is environmental noise. At the same time, according to the frequency range, amplitude characteristics, and time-domain characteristics of the noise, corresponding filtering algorithms are used to remove or reduce noise interference to ensure that the target echo signal is purer. And if the noise matches the known characteristics, the system can also dynamically adjust signal processing parameters (such as the sampling rate of the signal, the window size of the filter, the gain, etc.) to make the signal processing process more accurate. For example, when the system determines that the sources of noise in the echo signal cannot match the underwater characteristics pre-recorded by the active sonar, the system will consider that the noise in the echo signal is other unknown noise. The system will dynamically adjust the detection parameters of the active sonar based on the sources of noise. The detection parameters specifically include propagation loss, reflection interface, and scattering coefficient. By identifying unknown noise sources and timely adjusting sonar detection parameters, the system can cope with newly emerging environmental changes, reduce the interference of unknown noise sources on the signal, and improve the noise recognition and processing ability. At the same time, dynamically adjusting detection parameters (such as propagation loss, reflection interface, and scattering coefficient) can help the system adapt to different environmental conditions and target characteristics in real time, optimize the quality of the received echo signal, thereby improving the accuracy of target detection. And adjusting the working parameters according to the real-time changes of noise sources enhances the adaptability of the active sonar in complex or dynamic underwater environments. For example, in complex underwater environments, factors such as marine organisms and changing ocean currents will introduce unknown noise. By adjusting the detection parameters of the system, the sonar equipment can work more reliably. And by adapting to different noise sources and dynamically adjusting detection parameters, the system can reduce noise interference, ensure the extraction of more target information, and improve detection sensitivity and recognition accuracy, especially in variable or complex underwater environments.

[0210] Reference appendix Figure 2 , which is a system for eliminating sonar imaging reverberation based on manifold constraint in an embodiment of the present invention, includes:

[0211] An acquisition module 10, configured to acquire the time-domain waveform of the echo signal based on the echo signal of the target pre-monitored by the active sonar;

[0212] A judgment module 20, configured to judge whether a preset number of irregular reflection peaks are detected in the time-domain waveform;

[0213] An execution module 30, configured to, if so, collect the reflection path of the echo signal according to pre-recorded characteristic information, predict the propagation parameters of the reverberation signal based on the reflection path, and dynamically adjust the detection threshold of the active sonar for the target through the reflection path and the propagation parameters, wherein the characteristic information specifically includes seabed topography, geological characteristics, and acoustic wave characteristics;

[0214] A second judgment module 40, configured to judge whether the propagation parameters can be modeled and eliminated;

[0215] A second execution module 50, configured to, if possible, obtain the sound speed distribution data of the acoustic wave underwater based on the preset modeling basic information of the active sonar, generate the multi-path propagation parameters of the acoustic wave according to the sound speed distribution data, and model the characteristics to be eliminated for each path based on the multi-path propagation parameters, wherein the modeling basic information specifically includes hard bottom, soft bottom, and rock, the multi-path propagation specifically includes the direct path from the sound source to the target and the indirect path after underwater reflection, and the characteristics to be eliminated specifically include the time delay characteristics, frequency change, and amplitude attenuation of each path.

[0216] In this embodiment, the acquisition module 10 pre - monitors the echo signals of the target underwater based on an active sonar, acquires the time - domain waveforms of these echo signals, and then the judgment module 20 determines whether the time - domain waveform detects a preset number of irregular reflection peaks to execute corresponding steps; for example, when the system determines that the time - domain waveform of the echo signal does not detect a preset number of irregular reflection peaks, the system will consider the echo signal to be relatively simple and not significantly interfered by factors such as seabed reflection, stray objects, or multipath propagation. The system will continue to perform target detection according to the normal process, enhance the amplification processing of the target signal, adjust the sensitivity of the sonar system to capture the target echo, further precisely analyze the time - frequency characteristics of the signal to ensure the detection of the target, and dynamically adjust the detection threshold of the system to make it more sensitive to weaker signals, especially in an environment with a low signal - to - noise ratio; for example, when the system determines that the time - domain waveform of the echo signal detects a preset number of irregular reflection peaks, at this time the execution module 30 will consider that the echo signal is significantly interfered by factors such as seabed reflection. The system will collect the reflection path of the echo signal according to the pre - recorded characteristic information, which specifically includes seabed topography, geological characteristics, and acoustic wave characteristics, predict the propagation parameters of the reverberation signal based on different reflection paths, and dynamically adjust the detection threshold of the active sonar for the target through different reflection paths and propagation parameters; by detecting irregular reflection peaks, the system can identify the reverberation signal caused by seabed reflection or other stray reflections. Such reflections may come from changes in seabed topography, geological characteristics (such as soft bottom, hard bottom, rock, etc.) or acoustic wave propagation characteristics (such as water flow, temperature, salinity changes, etc.). After identifying these reflection interference signals, the system can use the reflection path information to accurately distinguish the target signal from the non - target signal, thereby reducing the impact of reverberation on target detection. At the same time, by analyzing different reflection paths and propagation parameters, the system can dynamically adjust the target detection threshold in real - time to cope with changes in different underwater environments. For example, in an area with strong seabed reflection, the system may appropriately increase the threshold to avoid misjudging these reflection signals as target signals, while in a relatively clean environment, the system may lower the threshold to improve the sensitivity of target detection. And using the reflection path and propagation parameter information, the system can identify the characteristics of the reverberation, and then adopt appropriate filtering and denoising techniques to reduce reverberation interference. By analyzing characteristics such as time delay, frequency change, and amplitude attenuation on the propagation path, the system can more accurately identify which signals belong to the reverberation, so as to perform effective denoising and signal enhancement; then the second judgment module 40 determines whether the propagation parameters of the reverberation signal can be modeled and eliminated to execute corresponding steps;For example, when the system determines that the propagation parameters of the reverberation signal cannot be modeled and eliminated, the system will consider that the signal propagation path is strongly interfered (such as reflections from other underwater objects, marine biological activities, bubbles or noise sources, etc.). The reverberation signal may be mixed in the target echo signal, resulting in the system being unable to accurately model the propagation path of these interference factors. The system will introduce a multipath propagation model and ultra-high resolution echo analysis technology. By further refining the analysis of the reflection path, it tries to distinguish the interference signal from the target echo as much as possible. At the same time, in a complex interference environment, by fusing the data of multiple sonar sensors or combining underwater acoustic image data, it provides a more accurate reverberation signal recognition and removal ability. For example, using sonar detection information at different frequencies and directions to enhance the signal discrimination; for example, when the system determines that the propagation parameters of the reverberation signal can be modeled and eliminated, at this time, the second execution module 50 will consider that the signal propagation path is not significantly strongly interfered and the propagation path of the interference factor can be modeled. The system will obtain the sound speed distribution data of the sound wave underwater based on the pre-set modeling basic information of the active sonar. The modeling basic information specifically includes hard bottom, soft bottom and rock. According to these sound speed distribution data, the multipath propagation parameters of the sound wave are generated. The multipath propagation specifically includes the direct path from the sound source to the target and the indirect path after underwater reflection. Based on the multipath propagation parameters, the characteristics to be eliminated for each path are modeled. The characteristics to be eliminated specifically include the time delay characteristics, frequency change and amplitude attenuation of each path; by combining the sound speed distribution data underwater and the characteristics of different bottom types (hard bottom, soft bottom and rock), the system can more accurately model the propagation path of the sound wave underwater. This accurate path modeling can help the system identify the propagation characteristics of the signal, including the direct path and the indirect reflection path, so as to effectively distinguish the reverberation from the target echo. At the same time, by modeling the multipath propagation parameters (such as time delay, frequency change, amplitude attenuation, etc.), the system can identify the reverberation characteristics on each path and perform targeted elimination. Specifically, the system can construct an accurate elimination model by analyzing the propagation characteristics of different paths to reduce the influence of reverberation. And based on the modeling of the multipath propagation parameters, the system can dynamically adjust the threshold of target detection. By identifying the propagation characteristics of different paths, the system can adaptively adjust the threshold to avoid misjudging the target in a complex underwater environment. For example, in an area with more reflection paths or stronger signal attenuation, the system can adjust the threshold to improve the recognition ability of the target signal. And by obtaining and using the sound speed distribution data underwater, the system can better adapt to the dynamic changes of the underwater environment (such as temperature, salinity, flow velocity changes). These changes will affect the speed and path of sound wave propagation and the reflection characteristics of the signal. By obtaining the sound speed data in real time and adjusting the modeling parameters, the system can quickly respond to environmental changes and optimize the signal processing process.;

[0217] In this embodiment, the execution module further includes:

[0218] A calculation unit, configured to calculate the propagation delay of the reflection path based on the preset sound velocity profile of the active sonar;

[0219] A judgment unit, configured to judge whether the propagation delay is greater than a preset delay threshold;

[0220] An execution unit, configured to, if so, obtain the offset frequency of the echo signal according to the relative velocity during the propagation of the sound wave, and calculate the attenuation characteristic of the reverberation signal based on the coefficient characteristics pre-recorded by the active sonar, where the coefficient characteristics specifically include the sound absorption coefficient of the water body, the reflection coefficient of the seabed, and the attenuation coefficient of the distance.

[0221] In this embodiment, the system calculates the propagation delay of the reflection path based on the pre-set sound speed profile of the active sonar, and then the system determines whether the propagation delay is greater than the pre-set delay threshold to execute corresponding steps. For example, when the system determines that the propagation delay of the reflection path is not greater than the pre-set delay threshold, the system will consider that the signal has not been reflected by a complex underwater environment (such as the seabed, rocks or obstacles, etc.), so no complex multipath effect or severe reverberation will be generated. The system will appropriately lower the detection threshold to ensure that even if the target signal is weak (due to a short distance and low signal intensity), it can be detected in time. At the same time, the decision threshold for false alarms is increased to reduce the chance of false alarms. For example, the system may consider that within this delay range, the target is relatively reliable, so it may no longer pay attention to interference sources (such as noise, clutter, etc.), and maintain the current target monitoring state, and be ready to enter a more complex monitoring mode at any time to cope with possible subsequent complex situations (such as changes in target position, seabed changes, etc.). For example, when the system determines that the propagation delay of the reflection path is greater than the pre-set delay threshold, at this time the system will consider that the signal has been reflected by a complex underwater environment. The system will obtain the offset frequency of the echo signal according to the relative speed during sound wave propagation, and calculate the attenuation characteristics of the reverberation signal based on the coefficient characteristics pre-recorded by the active sonar. The coefficient characteristics specifically include the absorption coefficient of the water body, the reflection coefficient of the seabed, and the attenuation coefficient of the distance. By determining that the propagation delay is greater than the set threshold, the system can accurately distinguish whether the signal has passed through a complex underwater reflection environment. These reflected signals usually become more complex due to the influence of multipath propagation, seabed reflection, and obstacles such as rocks, generating characteristics such as reverberation and delay. The system can identify these characteristics and make corresponding optimization processes. At the same time, by calculating the attenuation characteristics of the reverberation signal, the system can identify and eliminate the interference of factors such as multipath effect and seabed reflection on the signal, thereby improving the detection accuracy of the target. Especially in a complex environment, the system can accurately eliminate errors caused by the underwater environment, improve the signal quality, thus avoiding false target recognition or false alarms. And by relying on the pre-recorded coefficient characteristics (such as absorption coefficient, reflection coefficient, etc.), the system can dynamically adapt to changes in different underwater environments. The water quality, seabed geological characteristics, and obstacle distribution in different regions may cause changes in signal propagation characteristics. The system can adaptively adjust the signal processing strategy by calculating and adjusting the attenuation characteristics, so as to ensure the efficient operation of the system in various environments.

[0222] In this embodiment, it further includes:

[0223] A second acquisition module, configured to acquire the reflection characteristics of the target based on the monitoring type of the target, where the monitoring type specifically includes small targets, large targets, and dynamic targets, and the reflection characteristics specifically include reflection intensity, shape, size, and material;

[0224] A third judgment module, configured to judge whether the monitoring type matches the reflection characteristic;

[0225] A third execution module, configured to, if so, obtain the ambient noise monitored in real time, distinguish the difference data between the noise interference and the target echo from the ambient noise, and apply a preset Wiener filter to suppress the corresponding noise information according to the difference data, where the noise information specifically includes background noise and underwater ambient noise.

[0226] In this embodiment, the system monitors the target based on the monitoring type of the target. The monitoring types specifically include small targets, large targets, and dynamic targets, and collects the reflection characteristics of the target. The reflection characteristics specifically include reflection intensity, shape, size, and material. Then, the system determines whether the monitoring type of the target matches the reflection characteristics of the target to execute corresponding steps. For example, when the system determines that the monitoring type of the target cannot match the reflection characteristics of the target, the system will consider that the reflection characteristics of the target do not match the expected monitoring type, which may be because the physical attributes, shape, or material characteristics of the target do not match the set monitoring model. The system will automatically switch to a more suitable monitoring mode based on the reflection characteristics of the current target. For example, if it was originally set to monitor large targets, but a small object is detected, the system can switch to a monitoring type more suitable for small targets, adjust the detection sensitivity and algorithm parameters. At the same time, if the target is a non-metal or sound-absorbing material, the system can increase or adjust parameters such as the sound absorption coefficient and reflection coefficient to make the model more conform to the reflection characteristics of the actual target. For targets with irregular or special shapes, the system can perform shape analysis and adjust the shape recognition model of the echo signal to adapt to the reflection mode of irregular targets. In addition to sonar echo signals, the system can perform data fusion through other sensors (such as radar, infrared, or optical sensors) to help improve the accuracy of target recognition. Especially in complex environments, when the sonar echo signal cannot clearly identify the target, the information of other sensors (such as the thermal radiation of the target, changes in the reflection waveform, etc.) is combined for supplementation, so as to more accurately infer the characteristics and types of the target. For example, when the system determines that the monitoring type of the target can match the reflection characteristics of the target, at this time, the system will consider that the reflection characteristics of the target match the expected monitoring type. The system will obtain the real-time monitored environmental noise, and distinguish the differential data between the noise interference and the target echo from these environmental noises. According to different differential data, the preset Wiener filter is applied to suppress the corresponding noise information. The noise information specifically includes background noise and underwater environmental noise. By obtaining the environmental noise information in real time and distinguishing the difference between the noise and the echo, the system can accurately identify the noise interference source and effectively reduce the influence of these interferences on the target echo, which is particularly important for the detection of dynamic targets and targets in complex environments. At the same time, by dynamically applying the Wiener filter, the system can adjust the noise suppression strategy according to different environmental noise characteristics. For example, the characteristics of background noise and underwater environmental noise are different, and the system can adjust the intensity and method of noise suppression according to the real-time environmental monitoring data to ensure the optimization of the noise suppression effect. And after the target echo signal passes through noise suppression, the system can perform further target recognition, tracking, classification, etc. based on the cleaner signal. The improvement of the signal quality provides a more accurate basis for subsequent decision-making, which helps to improve the working efficiency and accuracy of the entire system.

[0227] In this embodiment, the second execution module further includes:

[0228] An identification unit, configured to identify the influence coefficients of different water layers on the acoustic wave based on the distribution law of the sound velocity distribution data in different water layers, wherein the influence coefficients specifically include a flow velocity coefficient, a water temperature coefficient, and a salinity coefficient;

[0229] A second judgment unit, configured to judge whether the influence coefficients cause a change in the acoustic wave propagation path;

[0230] A second execution unit, configured to, if so, obtain the measured distance between the active sonar and the target according to the depth information of different water layers, and dynamically correct the positioning accuracy of the target underwater based on the measured distance, and adaptively adjust the sonar parameters of the active sonar through the positioning accuracy, wherein the sonar parameters specifically include transmission power, signal frequency, and receiving sensitivity.

[0231] In this embodiment, based on the distribution law of the sound velocity distribution data in different water layers, the system identifies the influence coefficients of different water layers on sound waves. The influence coefficients specifically include the flow velocity coefficient, the water temperature coefficient, and the salinity coefficient. Then, the system determines whether these influence coefficients cause changes in the sound wave propagation path to execute corresponding steps. For example, when the system determines that the influence coefficients of different water layers on sound waves do not cause changes in the sound wave propagation path, the system will consider that parameter changes such as the flow velocity, water temperature, and salinity in the water layer do not significantly change the sound wave propagation mode. The system will continue to use the current propagation model for target detection without the need to adjust the sound wave propagation path, that is, the system can continue to execute conventional tasks such as echo signal processing, target recognition, and tracking, while saving computing resources and without the need for additional path correction or data calibration. It directly enters target detection and tracking, thereby improving real-time performance. And in the case where no significant path changes are found, the system can ensure monitoring stability under the existing environmental conditions, without the need to respond frequently to environmental changes, and can maintain the stable operation of the system, focusing on target detection and echo analysis. For example, when the system determines that the influence coefficients of different water layers on sound waves cause changes in the sound wave propagation path, at this time, the system will consider that the water layer significantly changes the sound wave propagation mode. The system will obtain the measured distance between the active sonar and the target based on the depth information of different water layers. Based on these measured distances, it will dynamically correct the positioning accuracy of the target underwater. Through different positioning accuracies, it will adaptively adjust the sonar parameters of the active sonar. The sonar parameters specifically include the transmission power, the signal frequency, and the receiving sensitivity. By adaptively adjusting the transmission power, signal frequency, and receiving sensitivity of the active sonar according to the positioning accuracy, the system can optimize the sonar performance and provide the best signal quality under different water layer conditions. At the same time, according to the actual position of the target and the positioning accuracy, the system can optimize the transmission power and receiving sensitivity of the sound wave, thereby improving the signal-to-noise ratio of the echo signal and reducing the interference of environmental noise on the signal, which is crucial for target detection and recognition, especially in the case of strong background noise and complex underwater environments. And dynamically adjusting the sonar parameters can not only improve the positioning accuracy and detection performance, but also save unnecessary power and computing resources. For example, in an area where the underwater signal attenuation is small, the system can appropriately reduce the transmission power to reduce energy consumption, while in an area where the signal attenuation is large, the system will enhance the signal to ensure reliable detection of the target.

[0232] In this embodiment, the judgment module further includes:

[0233] A second calculation unit for calculating the delay distance of the irregular reflection peak based on the position parameter of the irregular reflection peak in the time domain waveform;

[0234] A third judgment unit for judging whether the delay distance matches the actual sound wave propagation characteristics;

[0235] A third execution unit, which, if so, collects the persistence of the irregular reflection peak, generates a change period of the irregular reflection peak according to the persistence, identifies the variability of the occurrence of a single peak within a preset time duration from the change period, and collects the corresponding ambient noise based on the variability.

[0236] In this embodiment, the system calculates the delay distance of the irregular reflection peak based on the position parameter of the irregular reflection peak in the time-domain waveform, and then the system determines whether the delay distance matches the actual acoustic wave propagation characteristics to execute corresponding steps. For example, when the system determines that the delay distance of the irregular reflection peak cannot match the actual acoustic wave propagation characteristics, the system will consider that the reflection path of the echo signal may be affected by external interference or environmental factors, resulting in the propagation characteristics of the reflected signal not conforming to the expectation. The system will correct the sound speed distribution according to the real-time data obtained by the underwater sensor, so as to adjust the calculation of the acoustic wave propagation delay, avoid errors caused by environmental changes, and at the same time check the working state of the active sonar device to see if there are equipment failures or configuration errors that cause the delay of the reflection peak not to match the theoretical value. If the device is operating normally but the delay still does not match, check whether parameters such as the transmission power, frequency, and receiving sensitivity of the sonar signal need to be adjusted. And if the calculation of the delay of the irregular reflection peak continuously fails to match the actual propagation characteristics, the system can enable the target tracking function, combine the motion state and historical trajectory of the target, speculate on the target position and propagation path, and correct the delay calculation. For example, when the system determines that the delay distance of the irregular reflection peak can match the actual acoustic wave propagation characteristics, at this time the system will consider that the reflection path of the echo signal has not been affected by external interference, etc. The system will collect the persistence of the irregular reflection peak, generate a change period of the irregular reflection peak according to the persistence, identify the variability of the occurrence of a single peak within a preset time duration from the change period, and collect the corresponding ambient noise based on different variabilities. By confirming that the delay of the irregular reflection peak is consistent with the actual propagation characteristics, the system can ensure that the reflection path of the echo signal has not been interfered by external factors, which is crucial for accurately judging the target position and the acoustic wave propagation path. At the same time, analyzing the persistence, change period, and variability of the irregular reflection peak helps the system identify possible regular characteristics in the echo signal. According to these characteristics, the system can determine whether the echo signal has continuous periodic or non-periodic fluctuations. And by monitoring the change period of the irregular reflection peak, the system can identify which peak changes are related to ambient noise and which are related to target reflections. The system can capture the change characteristics related to ambient noise based on the differences identified in the change period. And by accurately analyzing the change period of the irregular reflection peak and its matching with the target echo, the system can more clearly distinguish the difference between the target signal and the ambient noise, thereby improving the accuracy and response speed of target positioning.

[0237] In this embodiment, the second judgment module further includes:

[0238] A reconstruction unit, configured to perform signal reconstruction on the reverberation signal based on the underwater environment and the target position, where the signal reconstruction specifically includes time-domain reconstruction and frequency-domain reconstruction;

[0239] A fourth judgment unit, configured to judge whether the actual signal of the active sonar reaches a preset matching value of the signal reconstruction;

[0240] A fourth execution unit, configured to, if so, calculate the mean square error between the reconstructed signal and the actual signal, obtain the correlation coefficient between the reconstructed signal and the actual signal according to the mean square error, and reflect the information sharing degree between the reconstructed signal and the actual signal based on the correlation coefficient.

[0241] In this embodiment, the system reconstructs the reverberation signal based on the underwater environment and the target position. The signal reconstruction specifically includes time-domain reconstruction and frequency-domain reconstruction. Then, the system determines whether the actual signal of the active sonar reaches the preset matching value of the signal reconstruction to execute corresponding steps. For example, when the system determines that the actual signal of the active sonar does not reach the preset matching value of the signal reconstruction, the system will consider that there are some unexpected deviations in the signal reconstruction process, which may be due to errors in the signal processing process or external interference. The system will re-collect the environmental parameters, dynamically adjust the propagation model of the sonar, and at the same time, based on the new environmental data, such as the underwater sound speed profile and the propagation path, adjust the time-domain or frequency-domain parameters in the signal reconstruction in real time for adaptive correction, and re-estimate the propagation parameters such as the reflection path, propagation delay, and attenuation coefficient based on the real-time echo signal, so as to optimize the signal reconstruction process. For example, when the system determines that the actual signal of the active sonar reaches the preset matching value of the signal reconstruction, the system will consider that the signal reconstruction process meets the expectations. The system will calculate the mean square error between the reconstructed signal and the actual signal, and obtain the correlation coefficient between the reconstructed signal and the actual signal based on this mean square error. Based on these correlation coefficients, the information sharing degree between the reconstructed signal and the actual signal can be reflected. By calculating the mean square error (MSE) between the reconstructed signal and the actual signal, the system can quantify the accuracy of the signal reconstruction. The smaller the mean square error, the smaller the gap between the signal reconstruction and the actual echo signal, and the more accurately the system can restore the propagation characteristics of the signal. At the same time, by obtaining the correlation coefficient between the reconstructed signal and the actual signal, the system can quantify their linear correlation. The closer the correlation coefficient is to 1, the more consistent the reconstructed signal and the actual signal are in characteristics, and the higher the information sharing degree. And when the signal reconstruction process meets the expectations, and both the mean square error and the correlation coefficient are within the ideal range, the system can determine that the reconstruction process is effective and reliable, which provides a solid foundation for subsequent signal analysis, target detection, noise cancellation, etc. According to the correlation between the reconstructed signal and the actual signal, the system can further adjust and optimize the parameters in the signal processing flow (such as transmit power, frequency, receive sensitivity, etc.). If the correlation is found to be low, the system will adjust the operation mode of the sonar in a timely manner and optimize the signal processing algorithm to better match the actual signal characteristics.

[0242] In this embodiment, the acquisition module further includes:

[0243] A second recognition unit for recognizing the noise source in the echo signal, where the noise source specifically includes background noise, electronic noise, and environmental interference;

[0244] A fifth judgment unit for judging whether the noise source matches the underwater characteristics pre-recorded by the active sonar;

[0245] A fifth execution unit, configured to, if the result is negative, dynamically adjust detection parameters of the active sonar based on the noise source, where the detection parameters specifically include propagation loss, reflection interface, and scattering coefficient.

[0246] In this embodiment, the system identifies the noise source in the echo signal. The noise source specifically includes background noise, electronic noise, and environmental interference. Then the system determines whether these noise sources match the underwater characteristics pre-recorded by the active sonar to execute corresponding steps. For example, when the system determines that the noise source in the echo signal can match the underwater characteristics pre-recorded by the active sonar, the system will consider that the noise in the echo signal may be known environmental noise, and can take preset noise suppression measures for processing based on the existing underwater characteristics and noise characteristics. The system will identify the noise source in the echo signal (such as background noise, electronic noise, environmental interference, etc.), and determine its source by comparing with the pre-recorded underwater noise characteristics. For example, if the identified noise matches the preset underwater sea current noise or equipment electronic noise of the system, the system can infer that the noise does not come from the target echo itself, but is environmental noise. At the same time, according to the frequency range, amplitude characteristics, and time-domain characteristics of the noise, corresponding filtering algorithms are used to remove or reduce noise interference to ensure that the target echo signal is purer. And if the noise matches the known characteristics, the system can also dynamically adjust signal processing parameters (such as the sampling rate of the signal, the window size of the filter, the gain, etc.) to make the signal processing process more accurate. For example, when the system determines that the noise source in the echo signal cannot match the underwater characteristics pre-recorded by the active sonar, the system will consider that the noise in the echo signal is other unknown noise. The system will dynamically adjust the detection parameters of the active sonar based on the noise source. The detection parameters specifically include propagation loss, reflection interface, and scattering coefficient. By identifying unknown noise sources and timely adjusting sonar detection parameters, the system can cope with newly emerging environmental changes, reduce the interference of unknown noise sources on the signal, and improve the noise recognition and processing ability. At the same time, dynamically adjusting detection parameters (such as propagation loss, reflection interface, and scattering coefficient) can help the system adapt to different environmental conditions and target characteristics in real time, optimize the quality of the received echo signal, thereby improving the accuracy of target detection. And adjusting the working parameters according to the real-time noise source changes enhances the adaptability of the active sonar in complex or dynamic underwater environments. For example, in complex underwater environments, factors such as marine organisms and changing sea currents will introduce unknown noise. By adjusting the detection parameters of the system, the sonar device can work more reliably. And by adapting to different noise sources and dynamically adjusting detection parameters, the system can reduce noise interference, ensure the extraction of more target information, and improve detection sensitivity and recognition accuracy, especially in variable or complex underwater environments.

[0247] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A sonar imaging reverberation elimination method based on manifold constraints, characterized in that: The following steps are involved: Based on the echo signal of the target pre-monitored by the active sonar, the time domain waveform of the echo signal is collected; Determining whether a preset number of irregular reflection peaks are detected in the time domain waveform; If yes, then according to the pre-recorded characteristic information, the reflection path of the echo signal is collected, and according to the reflection path, the propagation parameters of the reverberation signal are predicted, and the detection threshold of the active sonar for the target is dynamically adjusted through the reflection path and the propagation parameters, wherein the characteristic information specifically includes seabed topography, geological characteristics and acoustic wave characteristics; Determining whether the propagation parameters can be modeled and eliminated; If possible, then based on the preset basic modeling information of the active sonar, the sound speed distribution data of the sound wave underwater is obtained, and according to the sound speed distribution data, the multipath propagation parameters of the sound wave are generated, and according to the multipath propagation parameters, the features to be eliminated of each path are modeled, wherein the basic modeling information specifically includes hard bottom, soft bottom and rock, the multipath propagation specifically includes the direct path from the sound source to the target and the indirect path after underwater reflection, and the features to be eliminated specifically include the time delay characteristics, frequency changes and amplitude attenuation of each path.

2. The sonar imaging reverberation elimination method based on manifold constraint according to claim 1 is characterized in that: The step of collecting the reflection path of the echo signal and predicting the propagation parameters of the reverberation signal according to the reflection path further includes: Calculating the propagation delay of the reflection path based on the sound speed profile preset by the active sonar; Determining whether the propagation delay is greater than a preset delay threshold; If so, the offset frequency of the echo signal is obtained according to the relative speed of sound wave propagation, and the attenuation characteristics of the reverberation signal are calculated according to the coefficient characteristics pre-recorded by the active sonar, wherein the coefficient characteristics specifically include the sound absorption coefficient of the water body, the reflection coefficient of the seabed and the attenuation coefficient of the distance.

3. The sonar imaging reverberation elimination method based on manifold constraint according to claim 1, characterized in that: Before the step of dynamically adjusting the detection threshold of the active sonar to the target through the reflection path and the propagation parameter, the method further includes: Based on the monitoring type of the target, collecting the reflection characteristics of the target, wherein the monitoring type specifically includes small targets, large targets and dynamic targets, and the reflection characteristics specifically include reflection intensity, shape, size and material; determining whether the monitoring type matches the reflection characteristic; If so, obtain the environmental noise monitored in real time, distinguish the difference data between the noise interference and the target echo from the environmental noise, and apply the preset Wiener filter to suppress the corresponding noise information based on the difference data, wherein the noise information specifically includes background noise and underwater environmental noise.

4. The sonar imaging reverberation elimination method based on manifold constraint according to claim 1, characterized in that: The step of obtaining the sound velocity distribution data of the sound wave underwater and generating the multipath propagation parameters of the sound wave according to the sound velocity distribution data further includes: Based on the distribution law of the sound velocity distribution data in different water layers, identifying the influence coefficients of the different water layers on the sound waves, wherein the influence coefficients specifically include flow velocity coefficient, water temperature coefficient and salinity coefficient; Determining whether the influence coefficient causes a change in the sound wave propagation path; If so, the measured distance between the active sonar and the target is obtained according to the depth information of the different water layers, and the positioning accuracy of the target underwater is dynamically corrected according to the measured distance. The sonar parameters of the active sonar are adaptively adjusted through the positioning accuracy, wherein the sonar parameters specifically include transmission power, signal frequency and receiving sensitivity.

5. The sonar imaging reverberation elimination method based on manifold constraint according to claim 1, characterized in that: The step of determining whether a preset number of irregular reflection peaks are detected in the time domain waveform further includes: Calculating the time delay distance of the irregular reflection peak based on the position parameter of the irregular reflection peak in the time domain waveform; Determining whether the time delay distance matches actual sound wave propagation characteristics; If so, collect the persistence of the irregular reflection peak, generate a change cycle of the irregular reflection peak based on the persistence, identify the variability of a single peak within a preset time period from the change cycle, and collect the corresponding environmental noise based on the variability.

6. The sonar imaging reverberation elimination method based on manifold constraint according to claim 1, characterized in that: The step of determining whether the propagation parameters can be modeled and eliminated also includes: Based on the underwater environment and the target position, reconstructing the reverberation signal, wherein the signal reconstruction specifically includes time domain reconstruction and frequency domain reconstruction; Determining whether the actual signal of the active sonar reaches a preset matching value of the signal reconstruction; If so, the mean square error between the reconstructed signal and the actual signal is calculated, and the correlation coefficient between the reconstructed signal and the actual signal is obtained according to the mean square error. The information sharing degree between the reconstructed signal and the actual signal is reflected according to the correlation coefficient.

7. The sonar imaging reverberation elimination method based on manifold constraints according to claim 1 is characterized in that: The step of collecting the time domain waveform of the echo signal based on the active sonar pre-monitoring of the target also includes: Identifying noise sources in the echo signal, wherein the noise sources specifically include background noise, electronic noise and environmental interference; Determining whether the noise source matches the underwater features pre-recorded by the active sonar; If not, the detection parameters of the active sonar are dynamically adjusted based on the noise source, wherein the detection parameters specifically include propagation loss, reflection interface and scattering coefficient.

8. A sonar imaging reverberation elimination system based on manifold constraints, characterized in that: include: An acquisition module, used for acquiring a time domain waveform of an echo signal pre-monitored by active sonar to a target; A judging module, used for judging whether a preset number of irregular reflection peaks are detected in the time domain waveform; an execution module, configured to, if yes, collect the reflection path of the echo signal according to the pre-recorded characteristic information, predict the propagation parameters of the reverberation signal according to the reflection path, and dynamically adjust the detection threshold of the active sonar for the target through the reflection path and the propagation parameters, wherein the characteristic information specifically includes seabed topography, geological characteristics and acoustic wave characteristics; A second judgment module is used to judge whether the propagation parameters can be modeled and eliminated; The second execution module is used for, if possible, obtaining the sound speed distribution data of the sound wave underwater based on the basic modeling information preset by the active sonar, generating the multipath propagation parameters of the sound wave according to the sound speed distribution data, and modeling the features to be eliminated of each path according to the multipath propagation parameters, wherein the basic modeling information specifically includes hard bottom, soft bottom and rock, the multipath propagation specifically includes the direct path from the sound source to the target and the indirect path after underwater reflection, and the features to be eliminated specifically include the time delay characteristics, frequency change and amplitude attenuation of each path.

9. The sonar imaging reverberation elimination system based on manifold constraints according to claim 8, characterized in that: The execution module also includes: A calculation unit, configured to calculate the propagation delay of the reflection path based on the sound speed profile preset by the active sonar; A judging unit, configured to judge whether the propagation delay is greater than a preset delay threshold; The execution unit is used to obtain the offset frequency of the echo signal according to the relative speed of the sound wave propagation, and calculate the attenuation characteristics of the reverberation signal according to the coefficient characteristics pre-recorded by the active sonar, wherein the coefficient characteristics specifically include the sound absorption coefficient of the water body, the reflection coefficient of the seabed and the attenuation coefficient of the distance.

10. The sonar imaging reverberation elimination system based on manifold constraints according to claim 8, characterized in that: Also includes: A second acquisition module is used to acquire the reflection characteristics of the target based on the monitoring type of the target, wherein the monitoring type specifically includes small targets, large targets and dynamic targets, and the reflection characteristics specifically include reflection intensity, shape, size and material; A third judgment module is used to judge whether the monitoring type matches the reflection characteristic; The third execution module is used to obtain the environmental noise monitored in real time, distinguish the difference data between the noise interference and the target echo from the environmental noise, and apply a preset Wiener filter to suppress the corresponding noise information according to the difference data, wherein the noise information specifically includes background noise and underwater environmental noise.

Images