CPU + GPU heterogeneous acceleration-based chaff clutter simulation method and system

Abstract

The invention relates to the technical field of radar, and discloses a chaff clutter simulation method and system based on CPU + GPU heterogeneous acceleration, and the method comprises the steps: obtaining a chaff dynamic model according to chaff spiral descending motion characteristics, and simulating the motion trail and attitude change of a chaff; according to the chaff dynamic model, carrying out simulation calculation on the motion of chaff cloud in an actual environment to obtain an attitude angle of each chaff and a radar line-of-sight distance under the slow time of the radar; according to the chaff electromagnetic scattering characteristics, calculating the radar scattering sectional area of each chaff by using an electromagnetic scattering theoretical model; based on a radar distance equation, a modulation function is obtained by combining the radar line-of-sight distance and the radar scattering sectional area of each chaff, convolution is carried out on a baseband transmitting signal and the modulation function, and finally radar clutter baseband echo data of the chaff are generated. According to the chaff clutter modeling and simulation method, the chaff clutter modeling and simulation efficiency, precision and applicability can be improved, and theoretical and technical support is provided for chaff clutter echo simulation.

Description

Technical Field

[0001] The present invention belongs to the field of radar technology and relates to a chaff clutter simulation method and system based on CPU+GPU heterogeneous acceleration. Background Art

[0002] In the detection process, in addition to the target signal, the radar will also receive interference from the enemy (such as chaff shells launched by the enemy). In the air defense and anti-missile system, accurate chaff clutter data can provide data support for the anti-interference capability test of the radar system, and provide a basis for algorithm optimization, improvement of filtering technology and improvement of radar performance. Relying on field experimental testing of chaff clutter data not only requires a lot of manpower, material and financial resources, but also the measured data is difficult to cover a variety of scenarios, so chaff clutter simulation is of great significance.

[0003] Since the clutter signals received by radars often have a high degree of temporal and spatial complexity, and the processing process involves some highly complex calculations such as matrix operations and Fourier transforms, traditional CPU processors can only perform serial calculations when processing these tasks. When the number of pulse signals increases, the serial calculation method of the CPU often leads to a significant increase in simulation time, making it difficult to meet real-time processing requirements. GPUs, due to their highly parallel computing architecture, are capable of large-scale parallel computing tasks. Introducing GPU acceleration into radar clutter simulation can not only significantly improve the processing speed of clutter simulation algorithms, but also effectively cope with large-scale data operations, meeting the real-time and high-precision requirements of high-bandwidth, large-data-volume radar systems.

[0004] The existing chaff modeling methods mainly include statistical models, dynamic modeling methods and electromagnetic scattering theory models. Statistical model: By assuming the distribution characteristics of the chaff group, the probability distribution function is used to describe the position and scattering characteristics of the chaff in space, but the dynamic behavior of a single chaff is usually ignored. It is suitable for roughly describing large-scale chaff targets, with high computational efficiency but limited accuracy. Dynamic modeling method: The chaff is regarded as an independent moving target, and its motion trajectory and posture changes are simulated by dynamic equations, which can accurately describe the dynamic characteristics of a single chaff. Electromagnetic scattering theory model: Methods based on electromagnetic scattering theory, such as physical optics and moment method, can accurately solve the scattering characteristics of chaff to radar signals. The computational complexity of the dynamic modeling method and the electromagnetic scattering theory model is high. When dealing with large-scale chaff targets, the computing resources required are large, and parallel acceleration technology is needed for assistance. When facing complex battlefield environments or real-time radar interference simulation, traditional methods often find it difficult to simultaneously meet the requirements of real-time and high-precision computing. In addition, the current mainstream parallel computing hardware facilities are mostly CPUs and GPUs. The CPU has strong computing power but fewer cores than the GPU, while the GPU has weaker computing power than the CPU but more cores.

[0005] In summary, there are obvious limitations in the existing methods in balancing efficiency and accuracy, and it is difficult for statistical models to describe the dynamic characteristics of chaff. Summary of the Invention

[0006] The purpose of the present invention is to provide a chaff clutter simulation method and system based on CPU+GPU heterogeneous acceleration, which can improve the efficiency, accuracy and applicability of chaff clutter modeling and simulation, and provide theoretical and technical support for chaff clutter echo simulation.

[0007] To solve the above technical problems, the present invention is implemented by the following technical solutions.

[0008] In a first aspect, the present invention provides a chaff clutter simulation method based on CPU+GPU heterogeneous acceleration, including the following steps:

[0009] According to the spiral descent motion characteristics of chaff, a chaff dynamics model is obtained to simulate the motion trajectory and attitude change of chaff;

[0010] According to the chaff dynamics model, the motion of the chaff cloud in the actual environment is simulated and calculated to obtain the attitude angle of each chaff under the radar slow time and the radar line-of-sight distance;

[0011] According to the electromagnetic scattering characteristics of chaff, the radar cross section of each chaff is calculated using the electromagnetic scattering theory model;

[0012] Based on the radar range equation, the modulation function (the echo amplitude size of each range gate) is obtained by combining the radar line-of-sight distance and the radar cross section of each chaff, and the baseband transmitted signal is convolved with the modulation function to finally generate the radar clutter baseband echo data of chaff.

[0013] Combined with the first aspect, further, the step of obtaining a chaff dynamics model according to the spiral descent motion characteristics of chaff to simulate the motion trajectory and attitude change of chaff includes:

[0014] The spatial distribution of each chaff is generated according to the three-dimensional normal distribution, and the chaff attitude angle is generated according to the bimodal normal distribution. The chaff attitude angle includes the attitude pitch angle and the attitude azimuth angle;

[0015] Based on the obtained spatial distribution and chaff attitude angle, the horizontal speed and vertical speed of the chaff are calculated through the attitude pitch angle, and the diffusion model is updated through the attitude pitch angle; the expression of the dynamics model is as follows:

[0016] ;

[0017] Wherein, is the vertical speed of the chaff, is the horizontal speed of the chaff, is the density difference between chaff and air, and respectively represent the length and radius of the chaff, the pitch angle of the chaff attitude;

[0018] The diffusion model is:

[0019] ;

[0020] where, is the helical angular velocity of the chaff rotating around the z-axis, is the attitude azimuth angle, is the initial attitude azimuth angle, is the slow time;

[0021] According to the velocity components of each chaff at each moment on the x-axis and the velocity components on the y-axis, calculate the attitude angle of each chaff and the radar line-of-sight distance under the radar slow time;

[0022] The velocity component formulas of each chaff at each moment on the x-axis and the y-axis are as follows:

[0023] ;

[0024] where, represents the velocity component of the chaff at each moment on the x-axis, represents the velocity component of the chaff at each moment on the y-axis; represents the convected velocity caused by the wind.

[0025] Combined with the first aspect, further, the simulation calculation of the movement of the chaff cloud in the actual environment according to the chaff dynamics model to obtain the attitude angle of each chaff and the radar line-of-sight distance under the radar slow time includes:

[0026] Assign the calculation of the chaff dynamics model to CPU for parallel calculation, and assign the calculation of the electromagnetic scattering theory model to GPU for parallel calculation;

[0027] The CPU calculates the attitude angle of each chaff and the radar line-of-sight distance under the radar slow time;

[0028] Initialize the GPU, allocate video memory space, and perform multi-threading and block division for the two dimensions of the number of chaffs and the number of pulses;

[0029] Use shunt processing to process data, divide the smaller dimension of the number of chaffs and the number of pulses into several blocks, and sequentially pass them into the GPU for calculation;

[0030] The GPU copies the calculated clutter modulation function data and radar clutter baseband echo data from the GPU to the CPU.

[0031] Combined with the first aspect, further, calculating the radar cross section of each chaff according to the electromagnetic scattering characteristics of chaff by using an electromagnetic scattering theory model includes:

[0032]

[0033]

[0034] In the formula, is the Mueller matrix of the scattering of a single chaff, is the normalized Stokes vector at the time of transmitting polarization, is the normalized Stokes vector at the time of receiving polarization; among them, when vertically polarized when horizontally polarized when left-handed circularly polarized when right-handed circularly polarized , is the characteristic impedance of free space, is the scattering impedance of the dipole, is the equivalent length of the dipole, is the pitch angle of the chaff attitude, is the azimuth angle of the chaff attitude, is the average radar cross section of a single chaff.

[0035] Combined with the first aspect, further, the steps executed by the GPU further include:

[0036] Calculating the antenna transceiver gain according to the antenna pattern;

[0037] Calculating the clutter modulation function of each chaff, and accumulating the clutter modulation functions of all chaffs to obtain the final clutter modulation function;

[0038] Convolving the final clutter modulation function with the baseband transmitted signal to obtain clutter baseband echo data.

[0039] Accelerate the chaff dynamics modeling in parallel through the CPU, calculate the radar cross section and clutter modulation function of each chaff in parallel by using the GPU, and achieve load balancing of the calculation through dynamic task scheduling and optimized data transmission.

[0040] Combined with the first aspect, further, copying the clutter modulation function data and clutter baseband echo data obtained by the GPU into the CPU, and the CPU processes and analyzes the received data, and compares the obtained data with the theoretical data to verify the accuracy and correctness of the simulation data.

[0041] In combination with the first aspect, further, the data is efficiently exchanged through an optimized data transmission technology, including transmitting the clutter modulation function data and clutter baseband echo data that have completed calculations from the GPU to the CPU host side for post-processing; the data transmission ensures load balancing and efficient data exchange through video memory management and memory transmission optimization technologies.

[0042] In a second aspect, the present invention provides a chaff clutter simulation system based on CPU+GPU heterogeneous acceleration, including:

[0043] A memory for storing computer programs;

[0044] A processor for executing the computer program to implement the steps of the above-mentioned chaff clutter simulation method based on CPU+GPU heterogeneous acceleration; the processor includes a CPU and a GPU.

[0045] In combination with the second aspect, further, the simulation system supports parallel computing of multi-core CPUs and multiple GPU units, and adapts to the requirements of large-scale chaff simulation.

[0046] In a third aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned chaff clutter simulation method based on CPU+GPU heterogeneous acceleration are implemented.

[0047] In a fourth aspect, the present invention provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned chaff clutter simulation method based on CPU+GPU heterogeneous acceleration are implemented.

[0048] The chaff clutter simulation of the present invention has high efficiency and accuracy, and solves the problems in the prior art that due to the high computational complexity of radar echoes, low resource utilization rate, and uneven task allocation when facing targets of chaff scattered by airplanes or ships, traditional simulation methods are difficult to meet the requirements of real-time and high-precision modeling.

[0049] The chaff clutter modeling of the present invention has fine-grained description, and solves the problem that traditional simulation methods have a relatively rough description of the dynamic behavior of chaff targets, are difficult to capture the details of their motion characteristics and electromagnetic scattering changes, and require more refined modeling to enhance the accuracy of modeling and the ability to simulate dynamic behavior.

[0050] The CPU+GPU heterogeneous computing of the present invention is more efficient, and solves the problem that the parallel computing method of traditional methods is difficult to fully exploit the computing power of hardware, especially the low computing efficiency when the CPU cooperates with multiple GPUs.

[0051] The task assignment and load balancing in the chaff clutter calculation of the present invention improve the parallel speed and achieve load balancing by reasonable task division and efficient computing resource allocation in the face of a large amount of data to be calculated in radar clutter simulation.

[0052] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0053] (1) The simulation method of the present invention combines two existing clutter simulation methods, namely the dynamic modeling method and the electromagnetic scattering theory model method. Each chaff is regarded as a single target, and the electromagnetic scattering characteristics and related motion parameters of each chaff are calculated separately (the related motion parameters include the vertical velocity of the chaff, the horizontal velocity of the chaff, the pitch angle of the chaff attitude, the azimuth angle of the chaff attitude, the helical angular velocity of the chaff rotating around the Z-axis, and the entrainment velocity caused by the wind). It simulates the chaff clutter motion state and the three electromagnetic characteristics in the real scenario, combines the advantages of the two methods, and significantly improves the accuracy and real-time performance of chaff clutter simulation.

[0054] (2) The heterogeneous computing architecture proposed by the present invention for chaff clutter simulation makes full use of the single-core computing power of the CPU and the parallel computing power of the GPU, significantly improves the computing efficiency, and provides a solution to the problem of low utilization rate of existing parallel simulation computing resources.

[0055] (3) The present invention adopts an optimized data transmission technology to ensure load balancing in computing and reduce the bottleneck of memory transmission; the clutter modulation function and echo data generated during the computing process are efficiently transmitted to the host side through video memory management technology, ensuring efficient data exchange and processing, and further improving the system performance.

[0056] (4) The simulation method and system of the present invention have strong adaptability, consider the influence of different polarization states on electromagnetic scattering, and are applicable to various radar signal analysis and target recognition tasks.

[0057] (5) Through efficient parallel computing and resource optimization, without sacrificing the simulation accuracy, the present invention can significantly reduce the computing time and resource consumption, reduce the computing cost, which is of great significance for industrial applications that require fast and accurate simulation. Description of the Drawings

[0058] Figure 1 It is a schematic diagram of the architecture of the simulation system proposed in Embodiment 1 of the present invention;

[0059] Figure 2 It is a schematic diagram of the spatial distribution of chaff in the chaff dynamics modeling of Embodiment 1 of the present invention;

[0060] Figure 3 It is a schematic diagram of the dynamics modeling of a single chaff in Embodiment 1 of the present invention;

[0061] Figure 4 Schematic diagram for kinetic modeling of multiple chaffs in Embodiment 1 of the present invention;

[0062] Figure 5 Schematic flow diagram for the method of GPU shunt processing data in Embodiment 1 of the present invention;

[0063] Figure 6 Clutter amplitude spectrogram obtained by parallel simulation of chaff clutter using the simulation method of Embodiment 1 of the present invention;

[0064] Figure 7 Clutter power spectrogram obtained by parallel simulation of chaff clutter using the simulation method of Embodiment 1 of the present invention;

[0065] Figure 8 Existing CPU simulation time graph;

[0066] Figure 9 CPU + GPU heterogeneous acceleration simulation time graph using Embodiment 1 of the present invention;

[0067] Figure 10 Acceleration ratio graph of existing CPU simulation and CPU + GPU heterogeneous acceleration simulation using Embodiment 1 of the present invention. Detailed implementation manners

[0068] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. Without conflict, the technical features in the embodiments of the present invention and the embodiments can be combined with each other.

[0069] The term "and / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " generally represents an "or" relationship between the associated objects before and after.

[0070] Embodiment 1

[0071] As Figure 1 shown, a chaff clutter simulation method based on CPU + GPU heterogeneous acceleration of the present invention includes the following steps:

[0072] Steps executed in the CPU are as follows:

[0073] Step S1: Parameter initialization. Generate the radar baseband transmit signal required for simulation. In this embodiment, some common radar signals are used, such as linear frequency modulation signals, frequency modulated continuous wave signals, OFDM signals, etc. Utilizing the delay characteristic of convolution, convolving the signal with the calculated modulation function (composed of the clutter power amplitude of each range cell) can obtain the radar echo signal. Therefore, the present invention supports the simulation of various signals and has strong scalability.

[0074] Step S2: Adopt the dynamic modeling method to obtain the chaff dynamic model and simulate the movement trajectory and attitude change of the chaff. Specifically:

[0075] Step S21: Generate the spatial distribution of each chaff according to the three-dimensional normal distribution, and generate the chaff attitude angle according to the bimodal normal distribution. The chaff attitude angle includes the attitude pitch angle and the attitude azimuth angle.

[0076] Step S22: Based on the obtained spatial distribution and chaff attitude angle, calculate the horizontal velocity and vertical velocity of the chaff through the attitude pitch angle, and update the diffusion model through the attitude pitch angle.

[0077] The spiral descent motion of chaff in the actual environment is a complex non-linear dynamic phenomenon, and its characteristics are jointly determined by multiple factors. The gravitational force causes it to accelerate, while the shape and mass distribution trigger aerodynamic instability, forming a spiral trajectory. Air resistance, as the main external force, is closely related to the instantaneous velocity and windward area of the chaff, and significantly affects its deceleration effect and trajectory stability. During the descent, the air dynamic moment causes a spin motion, further changing the overall trajectory and dynamic characteristics. By synthesizing these influencing factors, a dynamic model can be established to simulate the movement of chaff in the actual scenario.

[0078] Chaff in the air has to go through an immature stage and a mature stage. The simulation mainly focuses on the mature stage. The RCS of the chaff cloud in the mature stage tends to be the maximum under the action of air resistance and has the ability to protect real targets. The movement form of chaff in the mature stage is a spiral motion based on the low Reynolds number flow theory. First, generate the spatial distribution of chaff according to the three-dimensional normal distribution, and then generate the initial chaff attitude angle according to the bimodal normal distribution. After obtaining the spatial distribution and the attitude pitch angle, the horizontal velocity and vertical velocity of the chaff can be calculated according to the formula, as shown below.

[0079]

[0080] Where, is the vertical velocity of the chaff, is the horizontal velocity of the chaff, is the density difference between the chaff and the air, and represent the length and radius of the chaff respectively, is the chaff strip attitude pitch angle. With the above conditions, the diffusion model is used to update the chaff strip attitude azimuth , the expression of the diffusion model is ,in is the spiral angular velocity of the foil strip rotating around the z-axis, plus the drag velocity caused by the wind The formulas for the velocity components of the foil strip on the x-axis and y-axis at each moment are as follows.

[0081]

[0082] in, represents the velocity component of the foil strip at each moment on the x-axis, represents the velocity component of the foil strip on the y-axis at each moment; is the initial attitude azimuth, For slow time.

[0083] The results of the foil strip dynamics modeling are as follows Figures 2 to 4 As shown, Figure 2 is the spatial distribution of foil strips, Figure 3 Modeling the dynamics of a single foil strip, Figure 4 Model the dynamics of multiple foil strips.

[0084] Step S3: According to the relevant foil strip dynamics theory, the movement of the foil strip cloud in the actual environment is simulated and calculated to obtain the attitude angle of each foil strip and the radar line of sight distance under radar slow time. Due to the high computational complexity of the dynamic modeling process, the CPU is used for parallel computing to make full use of the CPU's advantages in processing big data.

[0085] Step S4, storing the results of the CPU parallel calculation, that is, storing the attitude angle of each chaff strip under radar slow time and the radar line of sight distance, wherein the attitude angle of each chaff strip under radar slow time is used to prepare for calculating the RCS of the chaff strip using the electromagnetic scattering theoretical model.

[0086] Step S5, initializing the GPU, allocating video memory space, and dividing threads and blocks according to the two dimensions of the number of foil strips and the number of pulses.

[0087] Step S6: Use the split flow to process the data, such as Figure 5 As shown in the figure, the smaller dimension between the number of foil strips and the number of pulses is selected, and it is divided into multiple blocks, which are sequentially transmitted to the GPU for calculation to reduce the GPU calculation load. Figure 5 It can be seen that the shunting process effectively improves the performance of parallel computing of chaff clutter.

[0088] Step S7: After the calculation in the GPU is completed, copy the clutter modulation function data and the clutter baseband echo data from the GPU to the CPU host. Since the data is shunted, the data transmitted back should also be recombined on the CPU host to obtain the final data.

[0089] Step S8: After the CPU obtains the final data, perform radar data processing and analysis, including pulse compression, constant false alarm rate (CFAR), moving target indication (MTI), moving target detection (MTD), power spectrum verification, and amplitude spectrum verification, etc., and compare with the theoretical data to test the accuracy and precision of the simulation data.

[0090] The steps executed in the GPU are as follows:

[0091] Step 1: Set up the shared memory. The shared memory serves as a high-speed cache inside the GPU thread block, effectively reducing the number of accesses to the global memory and significantly reducing the memory access latency. By loading the key parameters of the chaff (such as the kinetic model parameters of spatial distribution, attitude angle, etc. and related radar parameters) into the shared memory, the threads can quickly access this data, thus accelerating the electromagnetic scattering calculation and radar echo generation process.

[0092] Step 2: Calculate the radar cross section (RCS) of each chaff according to the electromagnetic scattering theory model, considering multiple polarization modes. The specific calculation formula is as follows:

[0093]

[0094]

[0095] In the formula, is the Mueller matrix of the single-chaff scattering, is the normalized Stokes vector at the transmitting polarization, is the normalized Stokes vector at the receiving polarization. Among them, when the vertical polarization , when the horizontal polarization , when the left-handed circular polarization , when the right-handed circular polarization . is the characteristic impedance of free space, is the scattering impedance of the dipole, is the equivalent length of the dipole, is the pitch angle of the chaff attitude, is the azimuth angle of the chaff attitude, is the RCS of a single chaff.

[0096] Step 3: According to the antenna pattern, calculate the antenna transmit-receive gain using the sinc function.

[0097] Step 4: According to the radar range equation, calculate the power intensity at each range unit, that is, the clutter modulation function, and add up the clutter modulation functions of all foil strips corresponding to each pulse to obtain the final clutter modulation function.

[0098] Step 5: Use the time delay characteristics of convolution to convolve the final clutter modulation function with the baseband transmit signal to obtain the clutter baseband echo data.

[0099] Step 6: Copy the clutter modulation function and clutter baseband echo data to the CPU host for the next step of data processing and analysis.

[0100] For the foil clutter signal, the amplitude of the echo signal usually obeys the Rayleigh distribution, while the power spectrum often presents a Gaussian distribution. The parallel simulation results of the foil clutter using the simulation method of this embodiment are as follows: Figure 6 and Figure 7 As shown in the figure. The clutter amplitude spectrum obtained by simulation is consistent with the theoretical Rayleigh distribution model, and the clutter power spectrum is approximately consistent with the Gaussian distribution, which verifies that the simulation results are consistent with the theoretical expectations in terms of amplitude spectrum and power spectrum. This shows that the simulation method proposed in this embodiment meets the requirements of practical applications in terms of precision and accuracy, and can effectively simulate the radar characteristics of chaff clutter. Figures 8 to 10 As shown, Figures 8 to 10 The time and speedup ratio of CPU simulation and CPU+GPU heterogeneous accelerated simulation were compared. The results show that in large-scale computing tasks, the simulation method of this embodiment shows significant performance advantages over the CPU simulation method, and the acceleration effect becomes more significant as the number of pulses and foil strips increase.

[0101] Example 2

[0102] Based on the same inventive concept as Example 1, this example introduces a chaff clutter simulation system based on CPU+GPU heterogeneous acceleration, including: a memory for storing computer programs; a processor for executing computer programs to implement the steps of the above-mentioned chaff clutter simulation method based on CPU+GPU heterogeneous acceleration; the processor includes a CPU and a GPU.

[0103] Example 3

[0104] Based on the same inventive concept as other embodiments, this embodiment introduces a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned foil clutter simulation method based on CPU+GPU heterogeneous acceleration are implemented.

[0105] Example 4

[0106] Based on the same inventive concept as other embodiments, this embodiment introduces a computer program product, including a computer program, which when executed by a processor implements the steps of the above-mentioned chaff clutter simulation method based on CPU+GPU heterogeneous acceleration.

[0107] The simulation method and system of the present invention have a wide range of application prospects, especially in the fields of electronic warfare, radar interference analysis, complex battlefield environment simulation, etc. By efficiently simulating the electromagnetic scattering characteristics and dynamic behaviors of large-scale chaff clutter targets to generate high-precision clutter, it provides near-real training data for radar systems, improving the recognition accuracy and anti-interference ability of radar systems in complex interference environments. In addition, combined with the parallel computing framework proposed in the present invention, it can be applied to real-time radar simulation, target tracking and tactical evaluation, solving the problems of resource waste and computing bottlenecks in traditional computing methods, and having significant engineering application potential.

[0108] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0109] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0110] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0111] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions for implementing the steps specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 steps of the functions specified in one block or a plurality of blocks.

[0112] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit and scope of the present invention as defined by the claims. All of these are within the scope of protection of the present invention.

Claims

1. A chaff clutter simulation method based on CPU+GPU heterogeneous acceleration, characterized in that: The following steps are involved: According to the spiral descent motion characteristics of the foil strip, the foil strip dynamic model is obtained to simulate the motion trajectory and posture change of the foil strip. According to the foil strip dynamics model, the movement of the foil strip cloud in the actual environment is simulated and calculated to obtain the attitude angle of each foil strip and the radar line of sight distance under radar slow time; According to the electromagnetic scattering characteristics of the foil strip, the radar scattering cross-section of each foil strip is calculated using the electromagnetic scattering theoretical model; Based on the radar range equation, the modulation function is obtained by combining the radar line of sight distance and the radar scattering cross-sectional area of ​​each chaff strip. The baseband transmit signal is convolved with the modulation function to finally generate the radar clutter baseband echo data of the chaff strip.

2. The chaff clutter simulation method based on CPU+GPU heterogeneous acceleration according to claim 1, characterized in that: The method of obtaining a foil strip dynamics model based on the spiral descent motion characteristics of the foil strip and simulating the motion trajectory and posture change of the foil strip includes: Generate the spatial distribution of each foil strip according to the three-dimensional normal distribution, and generate the foil strip attitude angle according to the bimodal normal distribution, wherein the foil strip attitude angle includes the attitude pitch angle and the attitude azimuth angle; Based on the obtained spatial distribution and the foil strip attitude angle, the horizontal velocity and vertical velocity of the foil strip are calculated by the attitude pitch angle, and the diffusion model is updated by the attitude pitch angle; the dynamic model expression is as follows: ； in, is the vertical velocity of the foil strip, is the horizontal velocity of the foil strip, is the density difference between foil and air, and represent the length and radius of the foil strip, respectively. Chaff attitude pitch angle; The diffusion model is: ； in, is the helical angular velocity of the foil strip rotating around the z-axis, is the attitude azimuth, is the initial attitude azimuth, For slow time; According to the velocity component of the foil strip on the x-axis and the velocity component on the y-axis at each moment, the attitude angle of each foil strip and the radar line of sight distance in radar slow time are calculated; The formulas for the velocity components of the foil strip at each moment on the x-axis and y-axis are as follows: ； in, represents the velocity component of the foil strip at each moment on the x-axis, represents the velocity component of the foil strip on the y-axis at each moment; Indicates the drag speed caused by wind.

3. The chaff clutter simulation method based on CPU+GPU heterogeneous acceleration according to claim 1, characterized in that: According to the foil strip dynamics model, the movement of the foil strip cloud in the actual environment is simulated and calculated to obtain the attitude angle of each foil strip and the radar line of sight distance under radar slow time, including: Assign the calculation of the foil strip dynamics model to the CPU parallel computing, and assign the calculation of the electromagnetic scattering theory model to the GPU parallel computing; The CPU calculates the attitude angle of each foil strip and the radar line of sight distance in radar slow time; Initialize the GPU, allocate video memory space, and divide the threads and blocks according to the number of chaff strips and the number of pulses; Use split processing to process data, dividing the smaller dimension between the number of foil strips and the number of pulses into several blocks, and then pass them to the GPU for calculation. The GPU copies the calculated clutter modulation function data and radar clutter baseband echo data from the GPU to the CPU.

4. The chaff clutter simulation method based on CPU+GPU heterogeneous acceleration according to claim 1, characterized in that: The method of calculating the radar scattering cross-section of each foil strip using an electromagnetic scattering theoretical model according to the electromagnetic scattering characteristics of the foil strip comprises: ； ； In the formula, is the Mueller matrix of a single foil strip scattering, is the normalized Stokes vector for transmit polarization, is the normalized Stokes vector at the receiving polarization; where, for vertical polarization , horizontal polarization , left-hand circular polarization , right-hand circular polarization , is the characteristic impedance of free space, is the scattering impedance of the dipole, is the equivalent length of the dipole, is the chaff attitude pitch angle, is the chaff attitude azimuth, is the average radar cross-section of a single foil strip.

5. The chaff clutter simulation method based on CPU+GPU heterogeneous acceleration according to claim 3, characterized in that: The steps performed by the GPU also include: Calculate the antenna receiving and transmitting gain according to the antenna pattern; Calculate the clutter modulation function of each foil strip, accumulate the clutter modulation functions of all foil strips, and obtain the final clutter modulation function; The final clutter modulation function is convolved with the baseband transmit signal to obtain the clutter baseband echo data.

6. The method for simulating chaff clutter based on CPU+GPU heterogeneous acceleration according to claim 5, characterized in that: The clutter modulation function data and clutter baseband echo data obtained by the GPU are copied to the CPU, and the CPU performs radar data processing and analysis on the received data, compares the obtained data with the theoretical data, and verifies the precision and accuracy of the simulation data.

7. A chaff clutter simulation system based on CPU+GPU heterogeneous acceleration, characterized in that: include: Memory for storing computer programs; A processor is used to execute the computer program to implement the steps of the chaff clutter simulation method based on CPU+GPU heterogeneous acceleration according to any one of claims 1 to 6; the processor includes a CPU and a GPU.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the chaff clutter simulation method based on CPU+GPU heterogeneous acceleration described in any one of claims 1 to 6 are implemented.

9. A computer program product, comprising a computer program, characterized in that: When the computer program is executed by a processor, the steps of the chaff clutter simulation method based on CPU+GPU heterogeneous acceleration described in any one of claims 1 to 6 are implemented.

Images