A software high real-time integrated waveform processing method

Abstract

The application discloses a software high real-time integrated waveform processing method, selects a signal with good correlation performance and certain anti-Doppler frequency shift capability as a detection signal, and integrates communication information into the detection signal in different arrangement modes, so that detection and communication are integrated, the problem of complex integrated waveform design and difficult signal generation and processing at the hardware end is solved, a software high real-time integrated waveform processing framework based on Radarlab is provided, multi-thread parallel acceleration and multi-component distributed acceleration are adopted, a single module is processed by multi-thread binding cores to realize parallel calculation, and the calculation and throughput time of the module is reduced, all processing modules are distributed to multiple boards, the performance of different boards is fully utilized, the calculation time is improved, the problems of complex hardware structure, low reusability and long development time of the integrated system are solved, and software high real-time processing of the integrated waveform is realized.

Description

Technical Field

[0001] This invention belongs to the field of radar technology, specifically relating to a software-based, high-real-time integrated waveform processing method. Background Technology

[0002] Integrated detection and communication systems can simultaneously achieve target detection and data communication using limited spectrum and space resources. Currently, the operating modes of these systems can be categorized into four types. Among them, the integrated waveform system is a fully shared system, achieving integrated operation through joint design of shared waveforms. Detection and communication share transceiver systems, antennas, and other hardware platforms, reducing hardware maintenance costs, saving resources, and improving equipment survivability. However, integrated waveform systems are highly complex and difficult to implement and process: 1) Most radars use pulse systems, while communication generally uses continuous wave systems; 2) Radar and communication systems differ in hardware structure and signal processing; 3) The physical entities that build the integrated system cannot perform rapid testing, reconstruction, and iteration. Achieving integrated detection and communication through waveform design is a research hotspot, but existing integrated waveform research focuses primarily on waveform design and performance analysis, lacking research on specific implementations and applications. The few implementation methods available are mostly based on FPGAs, failing to enable rapid iterative applications of the system on other integrated waveforms. Currently, there is still a lack of software-defined integrated systems and waveform processing methods.

[0003] To share time-frequency resources, waveforms such as frequency division multiplexing based on multi-carrier chirp signals, waveforms based on LFM carrier modulation communication information, linear frequency modulation sharing waveforms based on multiphase shift keying, index modulation, and OFDM-LFM modulation have been proposed, providing a broad technical foundation for the realization of integrated systems. However, research on these waveforms has mostly focused on waveform design and performance analysis, lacking research on specific implementations and applications. Therefore, an integrated FPGA implementation method for LoRa signals has been proposed. This method replaces the intra-pulse linear frequency modulation signal with a LoRa signal to complete the communication function, and performs pulse-by-pulse compression to eliminate the influence of intra-pulse modulation on coherence to complete the detection function. A hardware platform is built using an FPGA board (VC707) and AD9361 to complete the hardware implementation of the radar communication integrated algorithm. However, this method cannot achieve rapid iterative application of the system on other integrated waveforms, and currently, there is still a lack of software-defined integrated systems and waveform processing methods.

[0004] The steps of an integrated FPGA implementation method for LoRa signals are as follows:

[0005] (1) Integrated waveform construction: At the signal transmitting end, binary data is grouped and converted into decimal, and combined with the pulse radar waveform, pulsed LoRa modulation is performed to generate LoRa pulses with different initial frequencies. At the radar receiving end, the baseband echo received by the RF front end is processed, mainly including pulse compression, range gate rearrangement, coherent accumulation, and target detection. Unlike single radar signal processing, the pulse compression of the integrated signal is changed to time-domain pulse compression using multiple matched filters. At the communication receiving end, the main functions include signal acquisition and signal extraction. The pulsed LoRa signal is converted into a continuous wave, and decimal information is obtained after LoRa demodulation. After conversion, binary data is obtained.

[0006] (2) Generate integrated waveform: Set waveform parameters, and use the designed waveform parameters in conjunction with the AD9361 development board. First, convert the binary grouped information sequence msg (SF bit binary number) into a decimal value K. At the same time, generate an LFM signal according to the designed bandwidth B and pulse width T. Then, complete the cyclic shift of the LFM signal according to the K value to generate a LoRa modulation signal. Then, complete the pulse generation according to the designed pulse repetition period. In this way, the original LFM signal in the pulse is converted into a LoRa modulation signal, generating an integrated baseband signal, which is then output through the RF terminal.

[0007] (3) Integrated waveform reception and processing:

[0008] Radar signal processing mainly includes pulse compression, range gate rearrangement, coherent accumulation, and target detection. At the communication receiver, the communication information frame structure is processed: the initial pulse is unmodulated (a conventional LFM signal), the communication data stream is passed through a matched filter, peak values ​​are detected to determine the starting position, and the last pulse modulates information known to the receiver to correct the impact of errors caused by the acquisition starting position. Since the received signal is in pulse form, after determining the starting position, the intra-pulse LoRa signal needs to be extracted based on the starting position, then combined for LoRa demodulation.

[0009] The disadvantages of existing technologies are as follows:

[0010] (1) In the existing radar communication integrated working system, the detection and communication of the fully shared system share the same hardware platform such as the transceiver system and antenna, which can reduce hardware maintenance costs, save resources, improve equipment survivability, and achieve integrated functions by relying on waveform design. However, the designed waveforms are extremely complex and it is difficult to realize the waveform generation and signal processing process on the hardware platform.

[0011] (2) Existing integrated waveform implementation methods are based on FPGA implementation, which has the disadvantages of not being able to quickly iterate and upgrade on other integrated waveforms and not being able to reuse them. Summary of the Invention

[0012] To overcome the shortcomings of existing technologies, this invention provides a software-based, high-real-time integrated waveform processing method. It selects signals with good autocorrelation performance and a certain degree of resistance to Doppler shift as detection signals, and integrates communication information into detection signals with different arrangements, achieving integrated detection and communication. This solves the problems of complex existing integrated waveform designs and difficulties in signal generation and processing at the hardware level. Furthermore, it proposes an integrated waveform software processing framework based on RadarLab, employing multi-threaded parallel acceleration and multi-component distributed acceleration. By performing multi-threaded core binding processing on a single module for parallel computation, the computation and throughput time of the module are reduced. By distributing all processing modules across multiple boards, the performance of different boards is fully utilized, improving computation time and solving the problems of complex hardware structure, low reusability, and long development time in integrated systems. This achieves software-based, high-real-time processing of integrated waveforms.

[0013] The technical solution adopted by this invention to solve its technical problem is as follows:

[0014] Step 1: Construct an integrated signal;

[0015] Let the echoes from the two targets be respectively: and The fuzzy function formula is:

[0016]

[0017] in, Indicates time delay. Indicates Doppler frequency shift, * indicates taking the conjugate; when The smaller the value, the higher its resolution;

[0018] The performance evaluation criterion for waveform communication is the peak-to-average power ratio (PAPR), defined as the ratio of the peak power to the average power of the signal. For discrete signals... Its mathematical expression is:

[0019]

[0020] The method for integrating communication information is as follows: For a communication sequence composed of binary 0s and 1s, the number of 0s and 1s is calculated in advance, and "0"s and "1"s are distributed on different frequency bands with a guard bandwidth between the two frequency bands. The step frequency interval of the symbols 0 and 1 is obtained according to the number of symbols. and ;

[0021] The transmitted information is demodulated by the signal frequency at the communication receiving end, and the target is detected by a series of signal processing at the radar receiving end;

[0022] Step 2: Set radar parameters;

[0023] A low repetition rate step frequency signal was selected as the radar detection signal, and the pulse repetition period was [missing information]. The symbol width is T, and the signal bandwidth is... Sampling rate Each CPI consists of N pulses, and the duration of one CPI is 1.6 ms; the minimum range resolution for radar detection is... The velocity resolution unit is ,in At the speed of light, For carrier frequency;

[0024] Step 3: Build a software-based processing project;

[0025] The integrated system operates by having a laptop at one end control an FPGA to generate waveforms. After passing through an up-conversion module, the signal is radiated into space by an antenna. At the other end, the signal in space is received by a command receiver and demodulated by the laptop at the receiving end to obtain communication information. Simultaneously, the antenna at the transmitting end receives the echo signal reflected from the target, which is then converted into a baseband echo signal by a down-conversion module. The transmitting end laptop then processes the radar signal to obtain target detection information.

[0026] The radar signal processing process is divided into data generation, signal processing, data processing, and display / control demonstration. Data generation is completed in Matlab, while signal processing and data processing are implemented in RadarLab. An integrated waveform back-end processing flow is built on the RadarLab software platform. The signal processing flow is divided into a communication processing branch and a radar signal processing branch. The down-converted baseband complex signal is used as input. The transmitted information is directly demodulated in the communication signal processing branch. In the radar signal processing branch, target detection is completed through pulse compression, moving target indication (MTI), moving target detection (MTD), and constant false alarm rate (CFAR). The target detection results are displayed on the display / control interface through point convergence and track processing (TWS).

[0027] Step 4: Multi-threaded parallel acceleration;

[0028] RadarLab defaults to one thread per module. The number of threads can be set within a module using the `omp_set_num_threads(number)` function, allowing FFT and loop operations to be performed concurrently across multiple threads. The pulse compression and MTI modules allocate threads based on the number of pulse repetition periods (PRTs) within a coherent processing time (CPI), with the number of PRTs being an integer multiple of the number of parallel threads. The MTD and CFAR modules allocate threads based on the number of sampling points within a single PRT, with the number of sampling points being an integer multiple of the number of parallel threads.

[0029] The module divides the data to be transmitted into multiple segments, and each segment is transmitted to the next module through a different output port.

[0030] The pulse compression, MTI, MTD, and CFAR modules in radar signal processing are accelerated through multi-threaded parallelism, with the number of threads set via `omp_set_num_threads(number)`. Pulse compression and MTI are divided according to the number of pulses per CPI, using... The MTD and CFAR are processed in parallel, with the MTD and CFAR results divided according to the number of sampling points in a PRT. The results are then processed in parallel with 10 channels, and finally the merging module integrates the CFAR results.

[0031] Step 5: Distributed acceleration of multiple components;

[0032] The signal processing modules in the multi-threaded parallel acceleration are distributed to different boards. The first component of radar signal processing includes a module for reading PRT-level echoes, forming CPI, pulse compression, and MTI. The second component includes a module for MTD, CFAR, and merging. The third component of data processing includes a module for point aggregation and track processing. The components communicate with each other using the Transmission Control Protocol (TCP).

[0033] An electronic device includes: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the electronic device performs the above-described software-based high real-time integrated waveform processing method.

[0034] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned software-based, high-real-time integrated waveform processing method.

[0035] A chip includes a processor for calling and running a computer program from a memory, causing a device equipped with the chip to perform the aforementioned software-based, high real-time integrated waveform processing method.

[0036] A computer program product includes a computer storage medium storing a computer program, the computer program including instructions executable by at least one processor, which, when executed by the at least one processor, implement the aforementioned software-based high real-time integrated waveform processing method.

[0037] The beneficial effects of this invention are as follows:

[0038] (1) Simple waveform design: To address the shortcomings of existing integrated waveforms being complex and difficult to implement on a hardware platform for waveform generation and signal processing, this method optimizes the waveform design. It selects signals with good autocorrelation performance and a certain degree of resistance to Doppler shift as radar detection signals, while integrating communication information into radar detection signals with different arrangements. At the communication receiver, the original signal can be accurately recovered using correlation demodulation. At the radar receiver, detection results can be obtained through pulse compression, coherent accumulation, and constant false alarm rate (CFAR) detection, thus achieving integrated communication and radar. The waveform designed in this method does not require changes to the original radar and communication hardware; only the waveform generation and pulse compression coefficient need to be modified. It has the advantages of simple hardware structure and ease of implementation. Furthermore, testing shows that the communication bit error rate is below 10⁻⁶, the minimum signal-to-noise ratio (SNR) for radar detection is -25 dB, and the target detection results are correct.

[0039] (2) Software-based high real-time implementation and processing method: Existing integrated waveform implementation methods are based on FPGA, which are time-consuming, difficult, and have poor system reusability. This method is based on a software platform and builds a software processing project, which has the advantages of universal interface, reusable and reconfigurable modules, rapid iterative upgrades, and reduced hardware costs. In the signal processing stage, multi-threaded parallel acceleration and multi-component distributed acceleration technology are adopted to achieve high real-time processing in software and meet the real-time processing requirements of the system. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the integrated waveform design of the present invention;

[0041] Figure 2 This is a schematic diagram of the integrated system of the present invention.

[0042] Figure 3 This is a flowchart of the integrated waveform processing of the present invention;

[0043] Figure 4 This is a schematic diagram of the software-based processing engineering and components of the present invention. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0045] This invention selects signals with good autocorrelation performance and a certain degree of resistance to Doppler frequency shift as radar detection signals. Simultaneously, communication information is integrated into radar detection signals with different arrangements. At the communication receiver, the original signal can be accurately recovered using correlation demodulation. At the radar receiver, the detection result can be obtained through pulse compression, coherent accumulation, and constant false alarm rate (CFAR) detection, thus achieving integrated communication and radar. The method for integrating communication information is as follows: For a communication sequence composed of binary 0 / 1, the number of 0 and 1 symbols is calculated in advance, and "0" and "1" are distributed on different frequency bands with a guard bandwidth between them. Based on the number of symbols, the step frequency interval for symbols 0 and 1 can be obtained respectively. and Taking the m-th symbol 1 as an example, its carrier frequency is An integrated waveform back-end processing workflow was built on the RadarLab software platform. By performing multi-threaded core binding processing on a single module, the computations within the module are distributed to different threads for parallel processing, reducing the computation and throughput time of the pulse compression, coherent accumulation, and constant false alarm rate detection modules. By distributing all modules in the entire process across multiple boards, although the multi-component distribution increases the signal transmission time between boards, the computation time is reduced to within a coherent processing interval (CPI). The real-time performance and effectiveness of the proposed method have been verified by the platform's built-in performance testing tools and statistical analysis of timestamps on the boards.

[0046] 1. Construct an integrated signal;

[0047] The criterion for evaluating radar waveform detection performance is the ambiguity function. The most ideal ambiguity function graph is called a "thumbtack-shaped" ambiguity function graph. Assuming the echoes of two targets are, in order... and The fuzzy function formula is:

[0048]

[0049] in, Indicates time delay. Indicates Doppler frequency shift, * indicates taking the conjugate, when The smaller the value, the higher its resolution.

[0050] The performance evaluation criterion for waveform communication is the peak-to-average power ratio (PAPR), defined as the ratio of the signal's peak power to its average power. For discrete signals... In other words, its mathematical expression is:

[0051]

[0052] Combining peak-to-average power ratio and ambiguity function, this method selects signals with good autocorrelation performance and a certain degree of resistance to Doppler frequency shift as radar detection signals. Simultaneously, communication information is integrated into radar detection signals with different arrangements to construct an integrated waveform. The method for integrating communication information is as follows: for communication sequences composed of binary 0 / 1, the number of 0 and 1 symbols is calculated in advance, and "0" and "1" are distributed on different frequency bands with a guard bandwidth between the two bands. Based on the number of symbols, the step frequency interval for symbols 0 and 1 can be obtained respectively. and Taking the m-th symbol 1 as an example, its carrier frequency is At the communication receiver, the transmitted information can be demodulated using the signal frequency; at the radar receiver, the target can be detected through a series of signal processing steps.

[0053] 2. Radar parameters are as follows:

[0054] Low repetition rate step frequency signal is selected as the radar detection signal, pulse repetition period The symbol width is T, and the signal bandwidth is... Sampling rate Each CPI consists of N pulses, and the duration of one CPI is 1.6 ms. The minimum range resolution for radar detection is... The velocity resolution unit is ,in At the speed of light, For carrier frequency.

[0055] 3. Establish a software-based processing system;

[0056] The integrated system operates as follows: A transmitting laptop (1) controls an FPGA to generate waveforms. After up-conversion, the signal is radiated into space by an antenna. At the receiving end (2), the signal in space is received by a command receiver and then demodulated by the laptop on that side to obtain communication information. Simultaneously, the antenna on the transmitting side receives the echo signal reflected from the target. After down-conversion, a baseband echo signal is generated, which is then processed by laptop 1 to obtain target detection information.

[0057] The specific radar signal processing process is divided into data generation, signal processing, and display / control demonstration. Data generation is completed in Matlab, while signal and data processing are implemented in RadarLab. An integrated waveform back-end processing flow is built on the RadarLab software platform. The signal processing flow is divided into a communication processing branch and a radar signal processing branch. The down-converted baseband complex signal is used as input. The transmitted information is directly demodulated in the communication signal processing branch. In the radar signal processing branch, target detection is completed through pulse compression, MTI (Moving Target Indication), MTD (Moving Target Detection), and CFAR (Constant False Alarm Rate Detection). The target detection results are displayed on the display / control interface through point convergence and TWS (Track Processing).

[0058] 4. Multi-threaded parallel acceleration;

[0059] In radar signal processing, the pulse compression, MTI, MTD, and CFAR modules are accelerated through multi-threaded parallel processing. The time consumption of the software modules is mainly in the computation time and throughput time. FFT (Fourier Transform) operations and loop operations occupy most of the computation time, especially the FFT calculations of pulse compression and MTD, which have a lot of data. The module throughput time is related to the amount of data transmitted. Parallel computing is used to reduce computation time, and multi-input / output modules are used to transmit data in segments to reduce throughput time.

[0060] RadarLab defaults to one thread per module, but to reduce computation time, the number of threads can be set within each module using the `omp_set_num_threads(number)` function. This allows FFT and loop operations to be performed concurrently in multiple threads. Since pulse compression performs FFT along the fast time dimension, the pulse compression and MTI modules in this method are divided into threads based on the number of PRTs (pulse repetition cycles) in a CPI (coherent processing time), with the number of PRTs being an integer multiple of the number of parallel threads. MTD operations, on the other hand, perform FFT along the slow time dimension, so the MTD and CFAR modules are divided into threads based on the number of sampling points in a PRT, with the number of sampling points being an integer multiple of the number of parallel threads.

[0061] To reduce module throughput time, the data to be transmitted is divided into multiple segments within the module, and each segment is transmitted to the next module through a different output port. This greatly reduces the amount of data transmitted through each port, thereby saving module throughput time.

[0062] The specific method involves multi-threading and parallel acceleration of the pulse compression, MTI, MTD, and CFAR modules in radar signal processing, with the number of threads set via `omp_set_num_threads(number)`. Pulse compression and MTI are divided according to the number of pulses per CPI, using... The MTD and CFAR are processed in parallel, with MTD and CFAR divided according to the number of sampling points in a PRT. The CFAR results are then integrated by a merging module.

[0063] 5. Multi-component distributed acceleration;

[0064] Due to the limited performance of the circuit boards, placing multiple modules on the same board can lead to conflicts in the available computing cores, affecting computation speed. Distributing the signal processing modules in multi-threaded parallel acceleration across different boards can fully utilize the performance of different boards and reduce computation time. The radar signal processing component 1 includes a module for reading PRT-level echoes, assembling CPI, pulse compression, and MTI; component 2 includes a module for MTD, CFAR, and merging; and component 3 for data processing includes a module for point convergence and track processing. The components communicate with each other using TCP (Transmission Control Protocol).

[0065] Example:

[0066] Figure 1 In a communication sequence consisting of binary 0s and 1s, the "0"s and "1"s are distributed across different frequency bands, with a guard band between them. The step frequency interval for symbols 0 and 1 can be determined based on the number of symbols. and For example, the m-th symbol 1 has a carrier frequency of... At the communication receiver, the information is demodulated based on the frequency of the received signal, and at the radar receiver, traditional radar signal processing is performed to detect the target.

[0067] Figure 2 In this integrated operation, a laptop on the left transmitter controls an FPGA to generate waveforms. After up-conversion, the signal is radiated into space by an antenna. At the receiver, the signal in space is received by a command receiver, and then processed by the laptop on the right for demodulation and other processing to obtain communication information. Simultaneously, the left antenna receives the echo signal reflected from the target, which is then processed by down-conversion and other modules to generate a baseband echo signal. This baseband echo signal is then processed by the left laptop to obtain target detection information.

[0068] Figure 3 The processing flow is divided into data generation, signal processing, data processing, and display / control demonstration. Data generation is completed in Matlab, while signal processing and data processing are implemented in RadarLab. The data generation flow includes target parameter setting, echo generation, and pulse compression coefficient generation; the signal processing flow includes: reading echo / reading pulse compression coefficient pulse compression—MTI—MTD—CFAR; and the data processing flow includes: point convergence—TWS (track processing).

[0069] Figure 4 This diagram illustrates the engineering components in the RadarLab software platform. There are three components: Component 1 (top) has five modules, primarily handling pulse compression and MTI (Mean Transmission Interference); Component 2 (bottom left) has three modules, mainly handling coherent accumulation and constant false alarm rate (CFAR) detection; and Component 3 (bottom right) has five modules, handling data processing and display. These three components complete the integrated waveform back-end processing. It can also be seen that the pc_para (parallel pulse compression), MTI (parallel), MTD_para (parallel coherent accumulation), and CFAR_para (parallel CFAR detection) modules are all multi-input / output modules, indicating that they utilize multi-threaded parallel acceleration.

Claims

1. A software-based, high real-time integrated waveform processing method, characterized in that, Includes the following steps: Step 1: Construct an integrated signal; Let the echoes from the two targets be respectively: and The fuzzy function formula is: ； in, Indicates time delay. Indicates Doppler frequency shift, * indicates taking the conjugate; when The smaller the value, the higher its resolution; The performance evaluation criterion for waveform communication is the peak-to-average power ratio (PAPR), defined as the ratio of the peak power to the average power of the signal. For discrete signals... Its mathematical expression is: ； The method for integrating communication information is as follows: For a communication sequence composed of binary 0s and 1s, the number of 0s and 1s is calculated in advance, and "0"s and "1"s are distributed on different frequency bands with a guard bandwidth between the two frequency bands. The step frequency interval for symbols 0 and 1 is determined based on the number of symbols. and ; The transmitted information is demodulated by the signal frequency at the communication receiving end, and the target is detected by a series of signal processing at the radar receiving end; Step 2: Set radar parameters; A low repetition rate step frequency signal was selected as the radar detection signal, and the pulse repetition period was [missing information]. The symbol width is T, and the signal bandwidth is... Sampling rate Each CPI consists of N pulses, and the duration of one CPI is 1.6 ms; the minimum range resolution for radar detection is... The velocity resolution unit is ,in At the speed of light, For carrier frequency; Step 3: Build a software-based processing project; The integrated system operates by having a laptop at one end control an FPGA to generate waveforms. After passing through an up-conversion module, the signal is radiated into space by an antenna. At the other end, the signal in space is received by a command receiver and demodulated by the laptop at the receiving end to obtain communication information. Simultaneously, the antenna at the transmitting end receives the echo signal reflected from the target, which is then converted into a baseband echo signal by a down-conversion module. The transmitting end laptop then processes the radar signal to obtain target detection information. The radar signal processing process is divided into data generation, signal processing, data processing, and display / control demonstration. Data generation is completed in Matlab, while signal processing and data processing are implemented in RadarLab. An integrated waveform back-end processing flow is built on the RadarLab software platform. The signal processing flow is divided into a communication processing branch and a radar signal processing branch. The down-converted baseband complex signal is used as input. The transmitted information is directly demodulated in the communication signal processing branch. In the radar signal processing branch, target detection is completed through pulse compression, moving target indication (MTI), moving target detection (MTD), and constant false alarm rate (CFAR). The target detection results are displayed on the display / control interface through point convergence and track processing (TWS). Step 4: Multi-threaded parallel acceleration; RadarLab defaults to one thread per module. The number of threads can be set within a module using the `omp_set_num_threads(number)` function, allowing FFT and loop operations to be performed concurrently across multiple threads. The pulse compression and MTI modules allocate threads based on the number of pulse repetition periods (PRTs) within a coherent processing time (CPI), with the number of PRTs being an integer multiple of the number of parallel threads. The MTD and CFAR modules allocate threads based on the number of sampling points within a single PRT, with the number of sampling points being an integer multiple of the number of parallel threads. The module divides the data to be transmitted into multiple segments, and each segment is transmitted to the next module through a different output port. The pulse compression, MTI, MTD, and CFAR modules in radar signal processing are accelerated through multi-threaded parallelism, with the number of threads set via `omp_set_num_threads(number)`. Pulse compression and MTI are divided according to the number of pulses per CPI, using... The MTD and CFAR are processed in parallel, with MTD and CFAR divided according to the number of sampling points in a PRT. The CFAR results are then integrated by the merging module. Step 5: Distributed acceleration of multiple components; The signal processing modules in the multi-threaded parallel acceleration are distributed to different boards. The first component of radar signal processing includes modules for reading PRT-level echoes, forming CPI, pulse compression, and MTI. The second component includes modules for MTD, CFAR, and merging. The third component of data processing includes modules for point aggregation and track processing. The components communicate with each other using the Transmission Control Protocol (TCP).

2. An electronic device, characterized in that, include: Processor and memory; The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to cause the electronic device to perform the method as described in claim 1.

3. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in claim 1.

4. A chip, characterized in that, include: A processor for retrieving and running a computer program from memory, causing a device on which the chip is mounted to perform the method as described in claim 1.

5. A computer program product, characterized in that, The computer program product includes a computer storage medium storing a computer program, the computer program including instructions executable by at least one processor, which, when executed by the at least one processor, implement the method as described in claim 1.

Images