A forward-looking sonar coherent echo rapid simulation method, device and medium

Abstract

This application relates to a method, apparatus, and medium for rapid simulation of forward-looking sonar coherent echoes. The method includes: acquiring a surface model and system parameters of a three-dimensional scene; performing visibility determination and propagation path solving based on the surface model and system parameters to generate an arrival set; constructing a sparse impulse response tensor and generating an upsampled domain coherent echo based on the arrival set and an upsampled time base; downsampling and aligning the upsampled domain coherent echo to the device sampling rate to obtain a coherent echo at the device sampling rate; performing simulation processing on the coherent echo at the device sampling rate and inputting it into a backend processing chain to output imaging results.

Description

Technical Field

[0001] This application relates to the field of sonar echo simulation technology, and in particular to a method, apparatus, equipment and storage medium for rapid simulation of forward-looking sonar coherent echoes. Background Technology

[0002] Existing forward-looking sonar echo simulation methods often face the challenge of balancing accuracy and efficiency when dealing with complex 3D scenes. On one hand, some methods primarily output "imaging results" or "parametric results such as intensity and distance"; others output "beam-level timing data"; and still others focus on injecting analog signals into receiver channels via digital-to-analog conversion for integration. These outputs or injections are often not strictly "coherent time-domain echoes at the array element / channel level," lacking unified constraints regarding time delay alignment, phase preservation, inter-channel consistency, and alignment with the device's preset sampling rate. Consequently, when R&D evaluation requires verifying link indicators such as matched filtering, pulse compression, coherent beamforming, and sidelobes, grating lobes, and ambiguous structures using the same input benchmark, the input data may not be entirely consistent with the actual processing chain requirements. On the other hand, under high-frequency close-range imaging conditions, echoes mainly originate from the visible surfaces of scatterers irradiated by sound waves, and the occlusion relationship can be represented as shadowed areas in the image. If the 3D solid scene is directly discretized into scattered scatterers or features are superimposed only in the image domain, without introducing visibility-based occlusion determination and back surface suppression in the echo generation stage, the occluded or back areas may still be "incorrectly counted as echo contributions", making it difficult to form shadow representation consistent with occlusion.

[0003] Therefore, it is necessary to provide a method, device, equipment, and medium for rapid simulation of forward-looking sonar coherent echoes to achieve rapid calculation of forward-looking sonar echo simulations. Summary of the Invention

[0004] In view of the above, this application provides a method, apparatus, device and storage medium for rapid simulation of forward-looking sonar coherent echo, the purpose of which is to solve the above-mentioned technical problems.

[0005] In a first aspect, this application provides a fast simulation method for forward-looking sonar coherent echo, the method comprising: Obtain the surface model and system parameters of the 3D scene; Based on the surface model and the system parameters, visibility determination and propagation path solving are performed to generate an arrival set; Based on the arrival set and the upsampling time base, a sparse impulse response tensor is constructed and an upsampling domain coherent echo is generated; wherein, the upsampling time base is determined according to the device sampling rate in the system parameters; The coherent echo in the upsampling domain is downsampled and aligned to the device sampling rate to obtain the coherent echo at the device sampling rate; The coherent echo at the sampling rate of the device is simulated and input into the back-end processing chain to output the imaging result.

[0006] In some embodiments, the step of performing visibility determination and propagation path solving based on the surface model and the system parameters to generate an arrival set includes: For each emission source, a set of sound rays is generated within the emission direction range; For each sound ray, the intersection of the ray tracing algorithm with the triangular facets of the surface model is calculated, the intersection points are determined and occlusion is assessed, and the visible intersection points are identified. For each sound ray, the propagation path is solved at its corresponding visible intersection point to generate the arrival set.

[0007] In some embodiments, the step of using a ray tracing algorithm to find the intersection of the triangular facets of the surface model, calculating the intersection points, determining occlusion, and identifying the nearest visible intersection point includes: Calculate the geometric properties of the triangular facet; the geometric properties include the normal vector, the centroid of the facet, and the boundary information of the facet. Determine the acoustic parameters of the sound ray; the acoustic parameters include emission pointing vector, sound ray resolution, and directivity gain; Based on the geometric properties and the acoustic parameters, the ray tracing algorithm is used to perform intersection calculations to determine the set of intersection points; Based on the set of intersection points, the effective range is determined and the occlusion is determined to identify the nearest visible intersection point.

[0008] In some embodiments, the step of solving the propagation path for each sound ray at its corresponding visible intersection point to generate an arrival set includes: For each sound ray, at the visible intersection, calculate the propagation length, propagation delay, propagation loss, directivity weight, and scattering weight, and form the arrival set.

[0009] In some embodiments, constructing a sparse impulse response tensor and generating an upsampled domain coherent echo based on the arrival set and the upsampled time base includes: Map the propagation delay of each arriving element in the arrival set to a time index under the upsampled time base; The complex weight of each arriving element is written to the corresponding position of the sparse impulse response tensor, and the complex weights of multiple arriving elements mapped to the same time index are coherently accumulated to generate the upsampled domain coherent echo. The sparse impulse response tensor is stored in a sparse storage format.

[0010] In some embodiments, downsampling and aligning the upsampled domain coherent echo to the device sampling rate to obtain the coherent echo at the device sampling rate includes: The coherent echo in the upsampled domain is low-pass filtered; Based on the coherent echo in the upsampled domain after low-pass filtering, downsampling is performed according to a preset sampling multiple to obtain the coherent echo at the sampling rate of the device.

[0011] In some embodiments, the step of simulating the coherent echo at the device sampling rate and inputting it into a back-end processing chain to output imaging results includes: Environmental noise is applied to the coherent echo at the sampling rate of the device; For coherent echoes after the application of environmental noise, carrier motion is added through resampling to form an equivalent Doppler effect; The resampled coherent echo is input into the back-end processing chain.

[0012] Secondly, this application provides a rapid simulation device for forward-looking sonar coherent echo, the rapid simulation device for forward-looking sonar coherent echo includes: The acquisition module is used to acquire the surface model and system parameters of the 3D scene; The first generation module is used to perform visibility determination and propagation path solving based on the surface model and the system parameters to generate an arrival set; The second generation module is used to construct a sparse impulse response tensor and generate an upsampled domain coherent echo based on the arrival set and the upsampled time base; wherein the upsampled time base is determined according to the device sampling rate in the system parameters; An alignment module is used to downsample and align the upsampled coherent echo to the device sampling rate to obtain the coherent echo at the device sampling rate. The output module is used to simulate the coherent echo at the sampling rate of the device and input it into the back-end processing chain to output the imaging results.

[0013] Thirdly, this application provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in a memory, it implements the steps of the fast simulation method for forward-looking sonar coherent echo as described in any embodiment of the first aspect.

[0014] Fourthly, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the forward-looking sonar coherent echo rapid simulation method as described in any embodiment of the first aspect.

[0015] The technical solutions provided in this application have the following advantages compared with the prior art: 1. A unified generation and docking method for channel-level coherent echoes: Under complex 3D scenes and controllable physical conditions, channel-level coherent echo data is generated that maintains propagation delay and phase coherence information and can be docked with the device's preset sampling conditions. This data can be used to directly input end-to-end back-end processing chains such as matched filtering, pulse compression, and coherent beamforming, supporting unified evaluation. 2. Introducing surface irradiation constraints and occlusion shadow representation on the echo generation side: Based on visibility determination and occlusion processing in triangular patch scenes, the echoes mainly originate from visible surfaces irradiated by sound waves. A shadow region consistent with the occlusion relationship is formed in the imaging output, suppressing unexpected echo characteristics generated by back surfaces or occluded areas, thereby improving the physical consistency and interpretability of simulation results. 3. Introducing reusable intermediate representations to reduce redundant computations and improve iterative evaluation efficiency: The propagation / scattering solution results are organized into reusable intermediate representations such as "arrival sets" and "sparse impulse response tensors," enabling the reuse of propagation and scattering results during frequent iterations of waveforms, systems, array geometry, and processing parameters. This reduces redundant computations and improves the efficiency of multi-scheme comparison and parameter scanning. 4. Achieving rapid echo synthesis and scaling in large-scale array elements and complex scenarios: Utilizing sparse structures, batch computation organization, and parallelization strategies reduces the growth rate of computational and storage / bandwidth overhead, improving echo synthesis throughput and the scalability of array element size / array geometry, meeting the rapid evaluation needs of the engineering R&D phase. It provides a fundamental capability for channel-level coherent echo simulation and end-to-end evaluation that balances physical consistency, controllability, reusability, and computational efficiency, improving the efficiency, verifiability, and comparability of forward-looking sonar system R&D evaluation. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1This is a flowchart illustrating a preferred embodiment of the rapid simulation method for forward-looking sonar coherent echo in this application. Figure 2 This is a flowchart of the fast simulation method for forward-looking sonar channel-level coherent echo in this application; Figure 3 This is a block diagram illustrating the principle of surface irradiation constraint and shadow restoration based on ray tracing in this application; Figure 4 A comprehensive diagram illustrating the data structure and generation rules for reaching elements; Figure 5 A diagram illustrating the construction and sparse storage of the sparse impulse response tensor in the upsampled domain; Figure 6 Here is a block diagram of the waveform upsampling-upsampling domain synthesis-downsampling alignment mechanism; Figure 7 Block diagram for coherent echo synthesis and acceleration in the upsampling domain channel; Figure 8 Block diagram for channel echo emulation processing; Figure 9 Diagram of the end-to-end processing chain connection; Figure 10 For use of the forward-looking sonar images obtained in this application; Figure 11 This is a schematic diagram of a preferred embodiment of the forward-looking sonar coherent echo rapid simulation device of this application. Figure 12 This is a schematic diagram of a preferred embodiment of the electronic device of this application; The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that the use of terms such as "first" and "second" in this application is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of those features. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed in this application.

[0021] Reference Figure 1 The diagram shown is a flowchart illustrating an embodiment of the forward-looking sonar coherent echo rapid simulation method of this application. The method is executed by an electronic device, which can be implemented by a software system and / or a hardware system. This forward-looking sonar coherent echo rapid simulation method includes: Step 101: Obtain the surface model and system parameters of the 3D scene.

[0022] A 3D scene is a three-dimensional spatial representation of a simulation environment. A 3D scene can include elements such as objects, terrain, and water bodies, used to simulate the medium and reflective interfaces for sound wave propagation.

[0023] A surface model is a geometric representation of the surface of an object in a 3D scene. It can consist of meshes, patches, or point clouds and describes the shape, location, and orientation of the surface. For example, a triangular mesh model can represent the undulating surface of an ocean floor.

[0024] System parameters are the technical parameters of a forward-looking sonar system, including the device sampling rate, sound frequency, transmitted signal waveform, receiver array geometry, and sound velocity.

[0025] In some embodiments, a pre-built 3D scene surface model data file can be read from a storage medium or a user interface, and a parameter configuration file of the sonar system can also be read. The surface model data file may include vertex coordinates, patch connectivity, and normal vector information; the system parameter configuration file may include device sampling rate values, a list of transmitted signal parameters, and receiver array position coordinates.

[0026] Step 102: Based on the surface model and the system parameters, perform visibility determination and propagation path solving to generate the arrival set.

[0027] Visibility determination is the process of determining whether the propagation path of sound waves between a sonar transmitter and receiver is obstructed by objects in the scene. For example, for the wavefront of a sonar wave, a ray is emitted from the transmitter location toward the center of each facet of the surface model. If the ray does not intersect with any other facet before reaching that facet, then that facet is visible.

[0028] Path determination is the process of calculating the geometric and physical properties of the propagation path of a sound wave from the transmitter to a surface point and then reflected to the receiver. For example, for a visible surface point, the straight-line distance from the transmitter to that point is calculated, and then the straight-line distance from that point to the receiver is added to obtain the total path length. The propagation time is then calculated based on the speed of sound.

[0029] The arrival set is a collection of propagation path information corresponding to all visible surface points, with each element including path delay, amplitude attenuation, phase change, and orientation angle.

[0030] In some embodiments, based on the transmitter and receiver positions in the system parameters, each facet or point of the surface model can be traversed, and visibility can be determined using ray casting or shadow testing methods, eliminating points obscured by other facets. For each visible point, the propagation path of the sound wave from the transmitter to the point and then to the receiver can be calculated according to the principles of geometric acoustics. The path length is equal to the distance from the transmitter to the point plus the distance from the point to the receiver, the propagation time is equal to the path length divided by the speed of sound, and the amplitude attenuation is determined by the propagation distance and the reflection coefficient. The propagation path information of all visible points is summarized to form an arrival set.

[0031] In some embodiments, the step of performing visibility determination and propagation path solving to generate an arrival set based on the surface model and the system parameters includes: generating a set of sound ray within the emission pointing range for each emission source; for each sound ray, using a ray tracing algorithm to find the intersection with the triangular facets of the surface model, calculating the intersection point and performing occlusion determination to determine the visible intersection point; and solving the propagation path for each sound ray at its corresponding visible intersection point to generate an arrival set.

[0032] The transmitter is the physical unit in a forward-looking sonar system that emits acoustic signals; it can be a transducer or a transducer array.

[0033] The emission direction range is the spatial angular range within which the sound wave energy is concentrated and radiated when the sonar emits sound waves.

[0034] A ray set is a discrete set of sound rays emitted from the source location in different directions within the emission direction range. Each ray represents a main path along which the sound wave may propagate, used to approximate the wavefront propagation of the sound wave.

[0035] Ray tracing is a computer graphics algorithm that simulates the propagation of light or waves by tracing the interaction between rays emanating from a source and geometric objects in a scene to simulate physical processes.

[0036] A triangular facet is the basic geometric unit that makes up a surface model; it is a polygon defined by three vertices and a normal vector.

[0037] The intersection point is the three-dimensional spatial coordinate of the point where the sound ray intersects with the triangular facet.

[0038] Obstruction determination refers to the process of determining whether there are other triangular facets blocking the straight path from the emission source to the intersection point, that is, determining whether the intersection point is directly visible to the emission source.

[0039] It can be seen that the intersection point is determined by occlusion, that is, there are no other triangular facets occluding the path from the emission source to the intersection point.

[0040] The propagation path solution is the process of calculating the geometric and physical properties of the complete propagation path of a sound wave from the source to the visible intersection point, and then from the visible intersection point to each receiving array element.

[0041] The arrival set is the collection of propagation path information corresponding to all visible intersections. Each sound ray may contribute one or more propagation path information entries (corresponding to multiple receiver elements). Each entry may include propagation time, complex amplitude (including phase), and possible direction information.

[0042] In some embodiments, the step of solving the propagation path for each sound ray at its corresponding visible intersection point to generate an arrival set includes: for each sound ray, calculating the propagation length, propagation delay, propagation loss, directivity weight, and scattering weight at the visible intersection point, and forming the arrival set.

[0043] In some embodiments, the step of using a ray tracing algorithm to find the intersection of the triangular facets of the surface model, calculating the intersection points, determining occlusion, and identifying the nearest visible intersection point may include the following operations: S11, calculate the geometric properties of the triangular facet; the geometric properties include the normal vector, the centroid of the facet, and the boundary information of the facet.

[0044] The normal vector is a unit vector perpendicular to the plane containing the triangular facet.

[0045] The centroid of a facet is the geometric center of the triangular facet, that is, the point corresponding to the arithmetic mean of the coordinates of its three vertices.

[0046] Patch boundary information is the bounding volume information used to quickly describe the approximate extent of a triangular patch in space.

[0047] S12, determine the acoustic parameters of the sound ray; the acoustic parameters include emission pointing vector, acoustic resolution, and directional gain.

[0048] A ray parameter is a set of parameters that describe the characteristics of a ray emitted from a source. These parameters may include, for example, its direction, angular resolution, and gain value in that direction.

[0049] The emission pointing vector is a unit vector that describes the direction in which the sound ray originates from the emission source.

[0050] Sound ray resolution is the angular interval between adjacent sound rays within the emission direction range.

[0051] Directivity gain is the relative change in signal intensity of a sound source in different radiation directions.

[0052] S13, Based on the geometric properties and the acoustic parameters, the ray tracing algorithm is used to perform intersection calculations to determine the set of intersection points.

[0053] Ray tracing algorithms are algorithms that determine visibility and calculate intersection points by simulating the intersection of rays with scene geometry. For example, algorithms used to calculate the intersection of sound rays with triangular faces.

[0054] The intersection set is the collection of all valid intersection points obtained after a sound ray intersects with one or more triangular facets in the surface model. Information for each intersection point may include its coordinates, the index of the corresponding triangular facet, and the distance from the origin of the ray to the intersection point.

[0055] In some embodiments, intersection calculations can be performed using ray tracing algorithms based on geometric properties and ray parameters to determine the set of intersection points. For example, the currently processed ray can be considered as a ray extending infinitely from the emission source location along the emission pointing vector direction. A spatial acceleration structure (such as a bounding box hierarchy BVH) is constructed using the boundary information of triangular facets (such as axially aligned bounding boxes) to quickly filter out a subset of triangular facets that may intersect with the current ray. For each candidate triangular facet selected, a precise intersection calculation is performed using a ray-triangle intersection algorithm (such as the Möller-Trumbore algorithm). If the algorithm determines an intersection, the coordinates of the intersection point and the distance t from the ray origin to the intersection point are calculated. The set of intersection points for the ray is obtained by combining the information of all calculated valid intersection points (where t > 0), including the intersection point coordinates, distance t, and the corresponding triangular facet index.

[0056] S14, based on the set of intersection points, determine the effective range and the occlusion, and determine the nearest visible intersection point.

[0057] Effective range determination is the process of determining whether an intersection point is within the effective propagation distance or time range based on acoustic or system constraints.

[0058] The occlusion determination is the process of determining whether there are other triangular facets blocking the straight path from the emission source to a certain intersection point (i.e. whether the intersection point is occluded by other facets).

[0059] The nearest visible intersection is the intersection in the intersection set that is determined by both the effective range and the occlusion, and is the closest to the emission source.

[0060] In some embodiments, all intersection points in the intersection point set can be sorted in ascending order according to their distance from the emission source. The sorted intersection point set is then traversed starting from the nearest intersection point. For the currently traversed intersection point, a valid range determination is first performed: it is checked whether the distance t of the intersection point is less than or equal to the system's preset maximum valid distance (or the corresponding time). If it exceeds this, the intersection point and any subsequent more distant intersection points are invalid, and the traversal is terminated. If the intersection point is within the valid range, an occlusion determination is performed: a test ray is emitted from the emission source with the current intersection point as the target. This test ray performs a fast intersection test with all other faces in the surface model except the triangular facet to which the current intersection point belongs (or possible faces selected by the acceleration structure), checking whether there are other intersection points before reaching the current intersection point at a distance (within a threshold slightly less than t). If any other intersection points exist, the current intersection point is occluded; if not, the current intersection point is not occluded. The first intersection point that passes both the valid range determination and the occlusion determination is determined as the nearest visible intersection point. If no matching point is found after traversing all intersection points, then the sound ray has no nearest visible intersection point.

[0061] Step 103: Based on the arrival set and the upsampling time base, construct a sparse impulse response tensor and generate an upsampling domain coherent echo; wherein the upsampling time base is determined according to the device sampling rate in the system parameters.

[0062] The upsampling time base is a discrete-time series with higher time resolution, where the time interval is smaller than the device sampling interval, used for signal representation in the upsampling domain. The upsampling time base is determined based on the device sampling rate in the system parameters, for example, by interpolation using integer multiples.

[0063] The sparse impulse response tensor is a multidimensional array that represents the impulse response of a sonar system at sparse time points. "Sparse" means that most of the array elements are zero, and non-zero elements only appear at the positions corresponding to the arrival time.

[0064] Upsampled domain coherent echo is a coherent echo signal represented on an upsampled time base, i.e., a high time resolution complex sequence of echoes, including amplitude and phase information.

[0065] In some embodiments, the time sequence of the upsampling time base can be determined by interpolation or integer multiple upsampling based on the device sampling rate in the system parameters; the path delay is mapped to the nearest time index of the upsampling time base based on the path delay to each element in the set, and the complex amplitude contribution of the path is accumulated at the corresponding position of the sparse impulse response tensor; the sparse impulse response tensor is convolved with the discrete representation of the transmitted signal to generate an upsampling domain coherent echo.

[0066] In some embodiments, constructing a sparse impulse response tensor and generating an upsampled domain coherent echo based on the arrival set and the upsampled time base includes: mapping the propagation delay of each arrival element in the arrival set to a time index under the upsampled time base; writing the complex weight of each arrival element into the corresponding position of the sparse impulse response tensor, and coherently accumulating the complex weights of multiple arrival elements mapped to the same time index to generate the upsampled domain coherent echo; wherein the sparse impulse response tensor is stored in a sparse storage format.

[0067] The arrival set is a collection that summarizes information about all valid sound wave propagation paths. Each element in the set is called an arrival element, and it may include information such as propagation delay, complex weight, ray identifier, and receiver array element identifier.

[0068] Propagation delay is the time delay experienced by a sound wave from its emission to its reflection by the scene and reception by a specific receiving array element.

[0069] The time index is the sequential position number of each discrete time point in the upsampled time base, and can be an integer starting from 0.

[0070] In some embodiments, the sampling interval of the upsampling time base can be determined, that is, the time difference between adjacent time points of the upsampling time base. For each arriving element in the arrival set, the value of its propagation delay is read, and the propagation delay is divided by the sampling interval of the upsampling time base to obtain the discrete time index corresponding to the propagation delay on the upsampling time base.

[0071] Complex weights are factors in the arriving elements that contribute to the complex amplitude of the final received signal; they can also be called complex coefficients.

[0072] Coherent summation is the operation of adding multiple complex numbers. In acoustics, since sound waves are coherent signals, when contributions from different paths are superimposed at the receiving point, their respective phases (i.e., the arguments of the complex numbers) must be considered. Therefore, it is a complex number addition, not an amplitude addition.

[0073] Sparse storage is a computer storage method used to efficiently store sparse data structures, such as matrices or tensors where most elements are zero. It only stores the values ​​of non-zero elements and their indexes, thus saving memory space.

[0074] Constructing the Sparse Impulse Response Tensor: Initialize a logically all-zero sparse impulse response tensor, whose dimensions are determined by the number of receiver elements, the upsampling time base length, and the number of transmission events. Iterate through each arriving element in the arrival set. For the current arriving element, read its receiver element identifier, complex weight, and the calculated upsampling time base time index. Locate the specific position in the sparse impulse response tensor determined by the receiver element identifier and time index. Since multiple arriving elements may map to the same receiver element and time index (e.g., paths from different points in the scene but with the same propagation delay), coherent accumulation is required. Therefore, perform a complex addition on the complex weight of the current arriving element with the existing value (initially zero) in the sparse impulse response tensor at that position, and write the result back to that position. After iterating through all arriving elements, the non-zero positions in the sparse impulse response tensor store the coherent accumulation result of all path contributions.

[0075] Generating the upsampled domain coherent echo: The sparse impulse response tensor can be viewed as the impulse response of the system at discrete time points. To obtain the received signal, this impulse response needs to be convolved with the transmitted signal. For each receive element channel, the corresponding sparse impulse response sequence (i.e., a slice of the tensor in the dimension of that receive element) is convolved with the discretized transmitted signal sequence. Since the impulse response is sparse, the convolution calculation can be optimized, only calculating the contribution at non-zero elements. This operation is performed on all receive element channels, and the final output is the upsampled domain coherent echo.

[0076] Step 104: The coherent echo in the upsampling domain is downsampled and aligned to the device sampling rate to obtain the coherent echo at the device sampling rate.

[0077] Downsampling is the process of reducing the signal sampling rate, converting a high-sampling-rate signal into a low-sampling-rate signal through filtering and decimation. The device sampling rate is the sampling frequency at which the sonar equipment actually acquires signals; it is one of the system parameters.

[0078] Coherent echoes are echo signals that retain phase information and are in complex form.

[0079] In some embodiments, downsampling and aligning the upsampled domain coherent echo to the device sampling rate to obtain the coherent echo at the device sampling rate includes: performing low-pass filtering on the upsampled domain coherent echo; and downsampling based on the low-pass filtered upsampled domain coherent echo according to a preset sampling multiple to obtain the coherent echo at the device sampling rate.

[0080] Low-pass filtering is a signal processing operation that allows frequency components in a signal below a certain cutoff frequency to pass through, while attenuating or blocking frequency components above that cutoff frequency.

[0081] In some embodiments, the required cutoff frequency of the low-pass filter can be determined based on the device sampling rate in the system parameters. According to the Nyquist sampling theorem, to avoid spectral aliasing during subsequent downsampling, the filter's cutoff frequency can be set to half the device sampling rate. A digital low-pass filter that meets this cutoff frequency requirement is designed, such as a finite impulse response (FIR) filter or an infinite impulse response (IR) filter. Each channel of the coherent echo in the upsampling domain (i.e., the signal sequence corresponding to each receiving element) is taken as input and filtered by this low-pass filter.

[0082] The preset sampling multiple is the ratio between the sampling rate of the upsampling time base and the device sampling rate, and is a positive integer.

[0083] Step 105: Simulate the coherent echo at the sampling rate of the device and input it into the back-end processing chain to output the imaging result.

[0084] Simulation processing involves adding noise, interference, or distortion to echo signals to mimic the non-ideal characteristics of real sonar systems. For example, adding Gaussian white noise to coherent echoes simulates ambient noise, and adding amplitude fluctuations simulates channel fading.

[0085] The back-end processing chain is a sequence of algorithm modules in a sonar system that perform subsequent processing on the echo signal, and may include beamforming, matched filtering, imaging algorithms, etc.

[0086] The imaging result is a sonar image generated after a back-end processing chain, which can be a two-dimensional or three-dimensional reflection intensity distribution map. For example, a grayscale image, where pixel values ​​represent the acoustic reflection intensity at various locations in the underwater scene.

[0087] In some embodiments, the step of simulating the coherent echo at the device sampling rate and inputting it into the back-end processing chain to output imaging results includes: applying environmental noise to the coherent echo at the device sampling rate; adding carrier motion to the coherent echo after applying environmental noise by resampling to form an equivalent Doppler effect; and inputting the resampled coherent echo into the back-end processing chain.

[0088] Environmental noise is a random interference signal that exists in a real underwater acoustic environment and is independent of the target's reflected signal. Its purpose is to make the simulated echo more closely resemble the actual measurement data. Environmental noise can be modeled as additive white Gaussian noise.

[0089] In some embodiments, the signal-to-noise ratio (SNR) value can be set according to simulation requirements or system parameters, and the power of the required added environmental noise can be calculated based on the SNR and the average power of the coherent echo signal at the device sampling rate.

[0090] Resampling is the process of changing the sampling time of a signal. For example, non-uniform resampling means that the new sampling times are no longer at equal intervals.

[0091] The motion of the carrier refers to the movement of the sonar-equipped vehicle (such as an autonomous underwater vehicle or a towed vehicle), including parameters such as velocity and acceleration. The relative motion between the carrier and the scene is the cause of the Doppler effect. The equivalent Doppler effect is a phenomenon where the frequency of the received echo signal shifts due to the relative radial motion between the sound source, receiver, and reflecting target. This frequency shift effect can be equivalently simulated by resampling to change the time scale of the signal.

[0092] Reference Figure 2 The diagram shown is a flowchart of the fast simulation method for coherent echo at the forward-looking sonar channel level in this application.

[0093] Step 201: Initialize scene and system parameters.

[0094] Read the 3D scene surface model (e.g., triangular facet model), set the array geometry (array element coordinates, integrated or separate transceiver relationship, array element directivity parameters), transmission system parameters (center frequency, bandwidth, pulse width, encoding method, etc.), device preset sampling parameters (device sampling rate, number of sampling points, frame rate, etc.), and set environmental parameters (sound velocity, absorption coefficient or equivalent absorption model parameters), noise parameters, and carrier motion parameters.

[0095] 1) Calculation of geometric properties of patches.

[0096] 1.1 Definition of triangular facets.

[0097] Let the triangular face be defined by three vertices: .

[0098] 1.2 Calculation of normal vector.

[0099] Normal vector of a triangular face Calculated using the cross product of the two edges e1 and e2: (1) (2) Normalized normal vector n: (3) in, ( For numerical tolerance, it is usually taken as ).

[0100] 1.3 Calculation of the centroid of the facet.

[0101] The centroid (baric center) C of the triangular facet is the arithmetic mean of its three vertices. (3) 1.4 Boundary Information.

[0102] The area A of the triangular facet: (4) 2) Voice generation.

[0103] 2.1 Emission pointing vector.

[0104] Let the location of the launch source be The emission is directed towards the unit vector d. tx From azimuth and pitch angle definition: (5) 2.2 Sound resolution.

[0105] Audio resolution is determined by configurable parameters. Control. For the pointed range The number of vocal liner notes is : (6) No. Direction angle of a sound ray : (7) 2.3 Directivity Gain.

[0106] The emission directivity uses a cosine power model. : (8) in, The angle between the direction of the sound ray and the direction of emission. For Heaviside step function, It is a directional index.

[0107] Step 202: Determine the upsampling rate and establish a unified time base.

[0108] Based on the device's preset sampling rate f s Set the upsampling factor U≥2, and establish the upsampling rate f. s ^'=Uf s A unified time base is used for arrival set discretization, sparse impulse response tensor construction, and echo synthesis.

[0109] Step 203, Visibility determination of surface irradiation constraint.

[0110] For each emission index, a set of sound rays is generated within its pointing range. The intersection of the sound rays and the triangular facets is calculated and occlusion is determined to identify the visible surface irradiated by the sound waves. No arrival records are generated for the back surface or the occluded surface, thus forming a shadow area in the imaging output that is consistent with the occlusion relationship.

[0111] 3) Sound line-surface intersection and occlusion determination.

[0112] 3.1 Möller-Trumbore Ray-Triangle Intersection Algorithm.

[0113] Let the ray parameter equation be: : (9) in, As the starting point of the ray, Let be the normalized direction vector, t be the ray parameter, and the length of ray R(t) be determined by t, which is t unit vectors. The question of whether the ray intersects the plane depends on the modulus of the ray. In other words, we need to see if there is a specific t, 0≤t≤∞, such that point P = R(t) lies on the plane.

[0114] Step 1: Calculate the determinant ( , As an edge, (where 10e-6 is a local minimum constant).

[0115] (10) (11) like If the ray is parallel to the triangle, then there is no intersection.

[0116] Step 2: Calculate the centroid coordinates .

[0117] (12) (13) like or If the intersection point is outside the triangle, then the intersection point is outside the triangle.

[0118] Step 3: Calculate the centroid coordinates .

[0119] (14) (15) like or If the intersection point is outside the triangle, then the intersection point is outside the triangle.

[0120] Step 4: Calculate the ray parameters .

[0121] (16) like Then there exists a valid intersection point, and the position of the intersection point P is: (17) 3.2 Determination of the nearest intersection point.

[0122] For a set of intersection points of a sound ray and multiple surfaces The nearest intersection point is: (18) corresponding intersection points These are the visible intersection points.

[0123] 3.3 Determination of the effective range.

[0124] The intersection point must meet the following conditions: (19) in, and These are the proximal and distal cutoff distances, respectively.

[0125] 4) Back surface suppression.

[0126] Assuming from the source Pointing to the intersection The direction vector is: (20) The normal vector of the surface is (Normalized). The condition for determining the direction of divergence is: (twenty one) That is, the angle between the normal vector and the incident direction is greater than or equal to , indicating that the faceplate faces away from the emission source.

[0127] 4.2 Orientation constraints.

[0128] Complete conditions for back surface suppression, combined with directional gain. for: (twenty two) in, Gain threshold (usually set to) If the above conditions are met, arrival records will not be generated or their contribution will be suppressed.

[0129] 4.3 Back-end orientation determination at the receiving end.

[0130] Similarly, for the receiving end Receive direction vector D RP for: (twenty three) The receive directionality constraint is: (twenty four) in, To receive the pointer vector, For receiving directional gain functions.

[0131] 5) Shadow formation.

[0132] 5.1 Line segment occlusion determination.

[0133] For from point Time The occlusion determination function for the line segment is: (25) in: This is the normalized direction vector; The length of the line segment; This is the set of all triangular faces in the scene; Ensure the intersection point is inside the line segment (excluding endpoints). 5.2 Detection of obstruction from the source to the intersection.

[0134] For a single jump path The visibility from the emission source S to the intersection point P is: (26) in, For including intersections Triangular facets need to be excluded from occlusion detection to avoid self-intersection.

[0135] 5.3 Obstruction detection from the intersection point to the receiving end.

[0136] Visibility V from intersection point P to receiver R PR for: (27) 5.4 Complete visibility condition.

[0137] The complete visibility condition for a single bounce path is: (28) in, Use the Heaviside function to ensure the facets are oriented correctly.

[0138] 6) Comprehensive judgment process.

[0139] 6.1 Arrival record generation conditions.

[0140] For triangular facets sampling points on To generate an arrival record, the following conditions must be met simultaneously: Geometric conditions: (33) Orientation conditions: (34) Directional conditions: (35) Obstruction conditions: (36) 6.2 Shaded areas are formed.

[0141] If any of the above conditions are not met, no arrival record is generated. In subsequent coherent echo synthesis and imaging output, this region naturally forms a shadow (no echo contribution).

[0142] The above mechanism avoids the "lack of visibility" problem caused by simply discretizing three-dimensional entities into scattered points, thereby improving the physical consistency and imaging interpretability of the echo generation side from the source.

[0143] Step 204: Solving the propagation path and generating the arrival set.

[0144] For each valid sound ray, the propagation length, propagation delay, propagation loss, directivity weight, scattering weight, etc. are calculated at the visible intersection point, and an arrival set is formed; optionally, the single reflection path is extended and multi-path arrival elements are recorded.

[0145] 1) Reach the element data structure.

[0146] Let the arriving element be It contains the following fields: (37) in, : Emission index (scalar or vector, identifying the emission source); : Receive index (scalar, identifying the receive channel); Path type / Number of reflections (1 = direct path / single scattering); : Propagation length (meters); Propagation delay (seconds); : Amplitude weight (linear amplitude); Phase term (radians); Complex coefficients; : Location of the scattering point (three-dimensional coordinates).

[0147] 2) Calculation of propagation length.

[0148] For path : (38) in: (39) (40) 3) Calculation of propagation delay.

[0149] Propagation delay Determined by path length and speed of sound: (41) in, The speed of sound (usually taken as...) m / s, marine environment).

[0150] 4) Calculation of amplitude weight.

[0151] Amplitude weight Calculated by combining multiple physical elements: (42) 4.1 Reflection Coefficient .

[0152] The reflection coefficient depends on the reflectivity of the reflecting surface: (43) in, The reflectivity of the reflecting surface ( ).

[0153] 4.2 BRDF Scattering Weights .

[0154] The scattering weights of the BRDF (Two-Way Reflectance Distribution Function) depend on the surface scattering characteristics: 4.2.1 Energy Conservation BRDF Model: (44) Wherein, the power domain BRDF is: (45) Specular reflection power (normalized Phong model): (46) Diffuse reflection power (Lambert model): (47) Roughness suppression factor (Kirchhoff approximation): (48) in, Scattering coefficient ( ); Mirror Index ( ); : Angle of incidence (the angle between the normal vector and the incident direction); Roughness standard deviation (meters); :wavelength( , (Center frequency).

[0155] 4.3 Geometric Attenuation

[0156] Geometric attenuation uses a two-path propagation model: (49) in, The geometric decay exponent,

[0157] 4.4 Dielectric Absorption Attenuation

[0158] The dielectric absorption is modeled using Thorp (amplitude domain): (50) in, The amplitude-domain absorption coefficient (Np / m) is calculated using Thorp's formula: (51) (52) in, Frequency (kHz) This represents the total path length.

[0159] 4.5 Emit Directivity Gain .

[0160] The emission directivity uses a cosine power model: (53) in, The angle between the launch direction and the incident direction; ; It is the Heaviside step function; This is the launch directionality index.

[0161] 4.6 Receiver Directivity Gain .

[0162] The receiver directivity uses the same cosine power model: (54) in, The angle between the receiving direction and the emission direction; ; To receive the directional index.

[0163] 4.8 Full Amplitude Weighting Formula.

[0164] Based on the above factors, the magnitude weights are as follows: (55) 5) Phase term calculation.

[0165] The phase term is related to the propagation delay and the center frequency: (56) in, The center frequency (Hz).

[0166] 6) Calculation of complex coefficients.

[0167] 6.1 Basic complex coefficient formula.

[0168] Complex coefficients consist of amplitude and phase: (57) in, It is the imaginary unit.

[0169] 6.2 Real part and imaginary part.

[0170] (58) in: (59) (60) 6.3 Amplitude and Phase Extraction.

[0171] The amplitude and phase can be extracted from the complex coefficients: (61) (62) Step 205: Construct a sparse impulse response tensor at an upsampling rate.

[0172] Map the propagation delay τ in the arrival set to an upsampling time index. = The complex coefficients of each arriving element are written to the tensor location; the multipath contributions mapped to the same index location are coherently accumulated to form an upsampled domain sparse impulse response tensor, which is then stored in a sparse storage format.

[0173] This application uses an upsampling sampling rate. A sparse impulse response tensor is constructed to improve the time delay discretization accuracy and maintain a unified time base with the upsampled domain echo synthesis.

[0174] 1) Discretization rule: For each arriving element, map the propagation delay to an upsampled time series index. .

[0175] (63) 2) Coherent writing and aggregation: Writing the complex coefficients of the arriving elements into the tensor location ( If multiple elements are mapped to the same index, they are coherently accumulated. : (64) 3) Sparse storage: due to Typically sparsely distributed along the timeline, this application employs a sparse storage format to reduce memory / GPU memory usage and improve parallel access efficiency. The sparse storage is implemented using a coordinate sparse format (COO).

[0176] Step 206: Upsample / generate the transmitted waveform and synthesize the channel-level coherent echo in the upsampled domain.

[0177] upsampling sampling rate The transmitted waveform is generated directly, or the transmitted waveform at the device's sampling rate is upsampled using zero-plugging and low-pass filtering (or equivalent polyphase filter) to obtain the upsampled transmitted waveform. Then, the transmitted waveform is convolved with the sparse impulse response tensor in the upsampled domain and accumulated according to the transmission index to generate the upsampled coherent echo for each receiving channel. Convolution can be implemented using time-domain sparse accumulation or frequency-domain batch fast convolution (Fast Fourier Transform (FFT) / Inverse Fast Fourier Transform (IFFT)) to improve throughput.

[0178] The basic form of upsampled domain echo is: (65) in, For the upsampled domain transmitted waveform, This represents the impulse response in the upsampled domain.

[0179] To improve speed, this application can employ frequency domain batch fast convolution: perform Fast Fourier Transform (FFT) on the transmitted waveform and impulse response, multiply and accumulate them in the frequency domain, and then perform Inverse Fast Fourier Transform (IFFT) to obtain the time domain echo; batch processing is used for multiple channels to improve throughput.

[0180] Step 207: Low-pass filtering and sampling / downsampling are aligned to the device sampling rate.

[0181] Low-pass filtering is performed on the upsampled domain echo, and sampling / downsampling is performed by multiple U to obtain the channel-level coherent echo at the device's preset sampling rate, so that the output is consistent with the device's sampling conditions, which facilitates reuse and comparative evaluation in the back-end processing chain.

[0182] To achieve more accurate time delay alignment on the echo generation side, this application adopts a consistent time base mechanism of "tensor upsampling construction + waveform synchronous upsampling + upsampling domain synthesis + downsampling alignment".

[0183] 1) Waveform upsampling: Upsample the device's sampling rate The transmitted waveform is upsampled to zero through zero-placing and low-pass filtering. or directly in The transmitted waveform is generated below.

[0184] 2) Upsampling domain echo synthesis: In Convolution and coherent synthesis are performed at lower sampling rates to avoid being limited by coarse time-delay quantization at the device sampling rate.

[0185] 3) Echo downsampling alignment: The echo in the upsampled domain is first low-pass filtered to suppress aliasing, and then sampled / downsampled to the device sampling rate by a multiple U to ensure that the output can be directly input into the device with a consistent processing chain.

[0186] Step 208, Realization Processing: Superimposed wideband additive noise from the environment and added carrier Doppler.

[0187] At the device sampling rate, ambient broadband additive noise is superimposed on the echo of each channel, with noise intensity or power spectral density being configurable parameters; and a time scaling factor is calculated based on the carrier velocity or radial velocity to resample the channel echoes to incorporate the equivalent Doppler effect, ensuring that the signal is more consistent with motion conditions in terms of frequency and time domain variations.

[0188] After downsampling to the device sampling rate, this application performs realistic processing on the channel echo: 1) Broadband additive noise superposition: Broadband additive noise is superimposed on the echo of each channel, and the noise intensity is a configurable parameter.

[0189] 2) Carrier Doppler effect (resampling implementation): Based on the carrier's motion speed, the time scaling relationship is calculated, and the channel echo is resampled to form an equivalent Doppler effect, improving the consistency with the actual received signal under motion conditions.

[0190] Step 209: Backend processing chain and output.

[0191] The channel-level coherent echo is input into the back-end processing chain, where downconversion (optional), time gain compensation (TVG), matched filtering / pulse compression are performed, and digital beamforming (DBF) imaging processing is carried out to output two-dimensional / three-dimensional imaging results. At the same time, resource indicators such as runtime, memory usage, and video memory usage are statistically analyzed and output.

[0192] The channel-level coherent echo generated in this application can be directly used for backend end-to-end processing chain verification. The backend processing chain may include: Downconversion (optional); Time gain compensation (TVG); Matched filtering / pulse compression; Digital beamforming (DBF) imaging processing (any form of imaging beamforming algorithm can be used); Output two-dimensional and / or three-dimensional imaging results.

[0193] The output metrics of this application include: Imaging results: Two-dimensional / three-dimensional imaging results; Resource metrics: runtime, memory usage, and video memory usage (average and peak values ​​can be calculated as optional implementation).

[0194] In this embodiment, the triangular patch scene is first read and the array and system parameters are configured to establish an upsampling time base. Then, visible surfaces that can be irradiated by sound waves are determined by ray tracing and occlusion is handled so that shadow areas are naturally reflected in the echo generation stage. The propagation path is then solved in parallel to form the arrival set, and a sparse impulse response tensor is constructed at the upsampling sampling rate. The transmitted waveform and impulse response are coherently synthesized in the upsampling domain to obtain a multi-channel echo, which is then aligned to the device sampling rate through low-pass filtering and downsampling. Finally, environmental noise is superimposed and the carrier Doppler effect is added to form a channel-level coherent echo that can be directly input into the back-end matched filtering, pulse compression, and imaging beamforming processing. At the same time, two-dimensional / three-dimensional imaging results and resource indicators such as running time, memory usage, and video memory usage are output.

[0195] Reference Figure 3 The diagram shown illustrates the principle of surface irradiance constraint and shadow restoration based on ray tracing in this application. It demonstrates the process of using triangular facets and ray tracing to simulate the "primary irradiance scattering surface" of a high-frequency forward-looking sonar. For each emission direction, a ray is generated. After intersecting with the triangular facet, occlusion is determined by nearest-intersection reduction. Between different geometries, only unoccupied facets that meet the validity conditions are considered irradiable surfaces and arrival records are generated. On the same geometry surface, arrival records are not generated for back-facing or occluded facets. Since the shadow region "has no arrival contribution" during the echo generation stage, the subsequent imaging output will naturally form a shadow consistent with the occlusion relationship, avoiding unexpected image features such as "back echoes" that may occur when simply discretizing a 3D entity into scattered points.

[0196] Reference Figure 4 The diagram shown is a comprehensive illustration of the data structure and generation rules for reaching elements. Figure 4 The entire process of generating the arrival element data structure is demonstrated, logically divided into three parts: physical propagation geometry, generation rule calculation engine, and final data structure. First, the physical path is constructed from the upper emission source, scattering point, and receiving point, and distance and angle parameters are extracted. Then, the data flows into the calculation engine in the middle, and is processed by the delay calculation module, the multi-dimensional amplitude weight combination module (integrating geometric attenuation, medium absorption, and directional gain), and the phase calculation module, and amplitude threshold filtering is performed. Finally, the calculated key information such as delay, complex coefficients, and physical index is encapsulated in the standardized arrival element data structure at the bottom, realizing the mapping from physical scene to simulation calculation data.

[0197] Reference Figure 5 The diagram shows the construction and sparse storage of the sparse impulse response tensor in the upsampling domain. First, a unified time base is established with a higher upsampling rate, mapping the propagation delay of each arriving element to an upsampling time index. Then, the complex coefficients of the arriving elements are written to the corresponding index positions, and multiple arrivals mapped to the same index position are coherently accumulated to maintain phase consistency. Since the impulse response is typically sparsely distributed along the time axis, this application employs sparse storage to store the tensor, thereby reducing memory / GPU memory usage and improving access efficiency during batch synthesis.

[0198] Reference Figure 6 The diagram shows the waveform upsampling—upsampling domain synthesis—downsampling alignment mechanism. The impulse response tensor is constructed at the upsampling rate, and the transmitted waveform is also generated or upsampled at the same upsampling rate, ensuring that echo synthesis is completed on a unified fine-grained time base, thereby improving time delay alignment accuracy and preserving coherent information. After echo synthesis, aliasing is suppressed by low-pass filtering, and then downsampled to the device's preset sampling rate by the upsampling factor, allowing the output echo to be directly connected to the actual device and back-end processing chain, facilitating scheme comparison and reuse.

[0199] Reference Figure 7 The diagram shows the block diagram for the upsampled domain channel coherent echo synthesis and acceleration implementation. The essence of echo synthesis is the convolution of the transmitted waveform and the channel impulse response, summed according to the transmission index. To improve speed, this application employs batch fast convolution, using FFT to transform the convolution into frequency domain multiplication and batch processing multiple channels, thus improving throughput. Both implementations output multi-channel coherent echoes in the upsampled domain for subsequent downsampling alignment and fidelity processing.

[0200] Reference Figure 8The diagram shows the channel echo realignment processing block. After channel-level echo generation, realignment processing is performed to improve consistency with actual operating conditions. First, broadband additive environmental noise is superimposed at the device sampling rate; the noise intensity can be configured to cover different sea states or background conditions. Then, the time scaling relationship is calculated based on the carrier velocity or radial velocity, and the echo is resampled to form an equivalent Doppler effect, making the signal frequency shift and time-domain variation more consistent with the conditions of the moving platform. The realigned echo can be directly input into the back-end processing chain for end-to-end evaluation.

[0201] Reference Figure 9 The diagram shows the end-to-end processing chain connection. The echo can undergo typical matched filtering and pulse compression processing before entering the imaging beamforming module to obtain two-dimensional or three-dimensional imaging results. The imaging beamforming algorithm can be implemented in different ways according to R&D needs; there is no limitation on the specific algorithm type. The key is to provide a unified channel-level coherent input, allowing different algorithms or parameter schemes to be compared and evaluated under a consistent input benchmark. Simultaneously, resource metrics such as runtime, memory usage, and GPU memory usage can be statistically analyzed during the processing chain execution to assess implementation costs and deployability.

[0202] Reference Figure 10 The image shown is a forward-looking sonar image obtained using this application. The shadows of the forward-looking sonar are restored, making the sonar image more realistic. Furthermore, the channel-level coherent echo output by this application can be processed using various beamforming algorithms, providing more realistic physical information restoration and better supporting the iteration of forward-looking sonar arrays and algorithms.

[0203] Reference Figure 11 The diagram shown is a functional module schematic of the forward-looking sonar coherent echo rapid simulation device 100 of this application.

[0204] The forward-looking sonar coherent echo rapid simulation device 100 described in this application is installed in an electronic device. Depending on its function, the forward-looking sonar coherent echo rapid simulation device 100 includes an acquisition module 110, a first generation module 120, a second generation module 130, an alignment module 140, and an output module 150. These modules can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, and are stored in the memory of the electronic device.

[0205] In this embodiment, the functions of each module / unit are as follows: Module 110 is used to acquire the surface model and system parameters of the 3D scene; The first generation module 120 is used to perform visibility determination and propagation path solving based on the surface model and the system parameters to generate an arrival set; The second generation module 130 is used to construct a sparse impulse response tensor and generate an upsampled domain coherent echo based on the arrival set and the upsampled time base; wherein the upsampled time base is determined according to the device sampling rate in the system parameters; Alignment module 140 is used to downsample and align the upsampled domain coherent echo to the device sampling rate to obtain the coherent echo at the device sampling rate; The output module 150 is used to simulate the coherent echo at the sampling rate of the device and input it into the back-end processing chain to output the imaging result.

[0206] The specific implementation of the forward-looking sonar coherent echo rapid simulation device in this application is largely the same as the specific implementation of the aforementioned forward-looking sonar coherent echo rapid simulation method, and will not be repeated here.

[0207] Reference Figure 12 The diagram shown is a schematic representation of a preferred embodiment of the electronic device of this application.

[0208] The electronic device includes a processor 111, a communication interface 112, a memory 113, and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 communicate with each other through the communication bus 114. The memory 113 is used to store computer programs, such as a forward-looking sonar coherent echo rapid simulation program. In some embodiments, the processor 111 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 111 can be used to control the overall operation of the electronic device, such as performing data interaction or communication-related control and processing. In this embodiment, the processor 111 is used to run program code stored in the memory 113 or process data.

[0209] The communication interface 112 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The communication interface 112 may also be used to establish a communication connection between the electronic device and other electronic devices.

[0210] The memory 113 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 113 may be an internal storage unit of the electronic device, such as the hard disk or memory of the electronic device. In other embodiments, the memory 113 may also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. of the electronic device. Of course, the memory 113 may include both internal storage units and external storage devices of the electronic device. In this embodiment, the memory 113 can be used to store the operating system and various computer programs installed on the electronic device, such as the program code of a forward-looking sonar coherent echo rapid simulation program. In addition, the memory 113 can also be used to temporarily store various types of data that have been output or will be output.

[0211] Figure 12 Only an electronic device with components 111-114 is shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0212] In one embodiment of this application, when the processor 111 executes the program stored in the memory 113, it implements the fast simulation method for forward-looking sonar coherent echo provided in any of the foregoing method embodiments, including: Obtain the surface model and system parameters of the 3D scene; Based on the surface model and the system parameters, visibility determination and propagation path solving are performed to generate an arrival set; Based on the arrival set and the upsampling time base, a sparse impulse response tensor is constructed and an upsampling domain coherent echo is generated; wherein, the upsampling time base is determined according to the device sampling rate in the system parameters; The coherent echo in the upsampling domain is downsampled and aligned to the device sampling rate to obtain the coherent echo at the device sampling rate; The coherent echo at the sampling rate of the device is simulated and input into the back-end processing chain to output the imaging result.

[0213] For a detailed explanation of the above steps, please refer to the above. Figure 1A flowchart illustrating an embodiment of the rapid simulation method for forward-looking sonar coherent echo.

[0214] Furthermore, this application also proposes a computer-readable storage medium that is both non-volatile and volatile. This computer-readable storage medium is any one or any combination of several of the following: hard disk, multimedia card, SD card, flash memory card, SMC, read-only memory (ROM), erasable programmable read-only memory (EPROM), portable compact disc read-only memory (CD-ROM), USB memory, etc. The computer-readable storage medium includes a data storage area and a program storage area. The program storage area stores a forward-looking sonar coherent echo fast simulation program. When executed by a processor, the forward-looking sonar coherent echo fast simulation program performs the following operations: Obtain the surface model and system parameters of the 3D scene; Based on the surface model and the system parameters, visibility determination and propagation path solving are performed to generate an arrival set; Based on the arrival set and the upsampling time base, a sparse impulse response tensor is constructed and an upsampling domain coherent echo is generated; wherein, the upsampling time base is determined according to the device sampling rate in the system parameters; The coherent echo in the upsampling domain is downsampled and aligned to the device sampling rate to obtain the coherent echo at the device sampling rate; The coherent echo at the sampling rate of the device is simulated and input into the back-end processing chain to output the imaging result.

[0215] The specific implementation of the computer-readable storage medium in this application is largely the same as the specific implementation of the aforementioned forward-looking sonar coherent echo rapid simulation method, and will not be repeated here.

[0216] It should be noted that the sequence numbers of the embodiments in this application are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, apparatus, article, or method. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0217] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware simulation platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0218] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A fast simulation method for coherent echoes from forward-looking sonar, characterized in that, The method includes: Obtain the surface model and system parameters of the 3D scene; Based on the surface model and the system parameters, visibility determination and propagation path solving are performed to generate an arrival set; Based on the arrival set and the upsampling time base, a sparse impulse response tensor is constructed and an upsampling domain coherent echo is generated; wherein, the upsampling time base is determined according to the device sampling rate in the system parameters; The coherent echo in the upsampling domain is downsampled and aligned to the device sampling rate to obtain the coherent echo at the device sampling rate; The coherent echo at the sampling rate of the device is simulated and input into the back-end processing chain to output the imaging result.

2. The fast simulation method for forward-looking sonar coherent echo as described in claim 1, characterized in that, The process of performing visibility determination and propagation path solving based on the surface model and the system parameters to generate an arrival set includes: For each emission source, a set of sound rays is generated within the emission direction range; For each sound ray, the intersection of the ray tracing algorithm with the triangular facets of the surface model is calculated, the intersection points are determined and occlusion is assessed, and the visible intersection points are identified. For each sound ray, the propagation path is solved at its corresponding visible intersection point to generate the arrival set.

3. The fast simulation method for forward-looking sonar coherent echo as described in claim 2, characterized in that, The process of using a ray tracing algorithm to find the intersection of the triangles of the surface model, calculating the intersection points, determining occlusion, and identifying the nearest visible intersection point includes: Calculate the geometric properties of the triangular facet; the geometric properties include the normal vector, the centroid of the facet, and the boundary information of the facet. Determine the acoustic parameters of the sound ray; the acoustic parameters include emission pointing vector, sound ray resolution, and directivity gain; Based on the geometric properties and the acoustic parameters, the ray tracing algorithm is used to perform intersection calculations to determine the set of intersection points; Based on the set of intersection points, the effective range is determined and the occlusion is determined to identify the nearest visible intersection point.

4. The fast simulation method for forward-looking sonar coherent echo as described in claim 2, characterized in that, The step of solving the propagation path for each sound ray at its corresponding visible intersection point to generate an arrival set includes: For each sound ray, at the visible intersection, calculate the propagation length, propagation delay, propagation loss, directivity weight, and scattering weight, and form the arrival set.

5. The fast simulation method for forward-looking sonar coherent echo as described in claim 1, characterized in that, The step of constructing a sparse impulse response tensor and generating an upsampled domain coherent echo based on the arrival set and the upsampled time base includes: Map the propagation delay of each arriving element in the arrival set to a time index under the upsampled time base; The complex weight of each arriving element is written to the corresponding position of the sparse impulse response tensor, and the complex weights of multiple arriving elements mapped to the same time index are coherently accumulated to generate the upsampled domain coherent echo. The sparse impulse response tensor is stored in a sparse storage format.

6. The fast simulation method for forward-looking sonar coherent echo as described in claim 5, characterized in that, The step of downsampling and aligning the upsampled domain coherent echo to the device sampling rate to obtain the coherent echo at the device sampling rate includes: The coherent echo in the upsampled domain is low-pass filtered; Based on the coherent echo in the upsampled domain after low-pass filtering, downsampling is performed according to a preset sampling multiple to obtain the coherent echo at the sampling rate of the device.

7. The fast simulation method for forward-looking sonar coherent echo as described in claim 1, characterized in that, The process of simulating the coherent echo at the sampling rate of the device and inputting it into the back-end processing chain to output the imaging result includes: Environmental noise is applied to the coherent echo at the sampling rate of the device; For coherent echoes after the application of environmental noise, carrier motion is added through resampling to form an equivalent Doppler effect; The resampled coherent echo is input into the back-end processing chain.

8. A rapid simulation device for coherent echo of forward-looking sonar, characterized in that, The device includes: The acquisition module is used to acquire the surface model and system parameters of the 3D scene; The first generation module is used to perform visibility determination and propagation path solving based on the surface model and the system parameters to generate an arrival set; The second generation module is used to construct a sparse impulse response tensor and generate an upsampled domain coherent echo based on the arrival set and the upsampled time base; wherein the upsampled time base is determined according to the device sampling rate in the system parameters; An alignment module is used to downsample and align the upsampled coherent echo to the device sampling rate to obtain the coherent echo at the device sampling rate. The output module is used to simulate the coherent echo at the sampling rate of the device and input it into the back-end processing chain to output the imaging results.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor, when executing a program stored in memory, implements the fast simulation method for forward-looking sonar coherent echo as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the fast simulation method for forward-looking sonar coherent echo as described in any one of claims 1 to 7.

Images