A DSP-based design method for synthetic aperture radar (SAR) imaging components

Abstract

This invention discloses a DSP-based synthetic aperture radar (SAR) imaging component design method. The main idea is to build a processing system for this DSP-based SAR imaging component design method. First, a command parsing framework is built in the signal processing board (DSP). Then, a data flow diagram for the DSP-based SAR imaging component design method is constructed. Finally, the RD algorithm is implemented using the DSP, controlled by the main control board, to process the radar echo data, thereby realizing the software-based implementation of SAR imaging technology and decoupling the SAR imaging algorithm from the hardware. This design method can adapt to different hardware resources.

Description

Technical Field

[0001] This invention belongs to the field of software-based radar, and specifically relates to a design method for a synthetic aperture radar (SAR) imaging component based on DSP (Digital Signal Processor). This method features long-range, high-precision, and flexible deployment under software control, and can be applied to achieve real-time ground imaging in multi-functional airborne payload systems. Background Technology

[0002] As radar requirements continue to evolve, higher goals have been set for radar systems engineering, hardware architecture, software and hardware architecture, and signal processing algorithms. Improving radar system optimization requires, firstly, standardizing software and hardware, decoupling them as much as possible; secondly, the traditional waterfall development model for radar systems urgently needs to be changed; and finally, simplified radar system development methods, efficient prototyping, and proof-of-concept techniques are also crucial conditions for advancing radar system optimization.

[0003] With the increasing penetration of software-defined radio (SDR) concepts into the radar field, the open architecture of software-defined radar has undergone significant development. Software-defined radar, designed for multiple functions and tasks and centered on application-oriented principles, greatly reduces development time and costs, making radar systems easier to upgrade, expand, and replace in both hardware and software.

[0004] Synthetic Aperture Radar (SAR), with its wide range of applications, is a type of radar capable of real-time target imaging, offering advantages such as all-weather, all-day operation, long range, and high precision. However, traditional SAR systems often use customized hardware, meaning that the system's hardware and software suffer from problems such as limited functionality, high coupling between software applications and hardware systems, and a waterfall development model. This leads to challenges such as weak hardware resource versatility, long software development cycles, and difficulties in maintaining and updating functionality. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, the present invention aims to propose a DSP-based synthetic aperture radar (SAR) imaging component design method. This DSP-based SAR imaging component design method can software-ize the SAR imaging component, enabling flexible deployment and control on hardware under the control of the DSP core framework.

[0006] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution.

[0007] A design method for a DSP-based synthetic aperture radar (SAR) imaging component includes the following steps:

[0008] Step 1: Build a processing system for a DSP-based synthetic aperture radar (SAR) imaging component design method;

[0009] Step 2: Build a command parsing framework in the signal processing board DSP;

[0010] Step 3: Construct a data flow diagram for a DSP-based synthetic aperture radar (SAR) imaging component design method;

[0011] Step 4: Use the signal processing board DSP to perform SAR imaging RD (Range Doppler) algorithm processing.

[0012] The processing system consists of a display and control terminal and four boards, namely a DDR memory board, an ARM main control board, an FPGA preprocessing board, and a DSP signal processing board.

[0013] Among them, the storage board DDR is used to store and replay radar echo data, and the radar echo data is replayed to the preprocessing board FPGA through the Aurora interface protocol of the preprocessing board FPGA.

[0014] The main control board ARM is used to run software that sends control commands to the signal processing board DSP and the preprocessing board FPGA. By sending control commands, it controls the preprocessing board FPGA and the signal processing board DSP to complete the functions of starting, stopping, reconstructing and hardware migration of the synthetic aperture radar SAR imaging algorithm.

[0015] The display and control terminal is used to issue control commands to control the main control board ARM. After receiving the control command, the main control board ARM transmits the control command to the preprocessing board FPGA and the signal processing board DSP.

[0016] The preprocessing board FPGA is used to control the flow of radar echo data from the preprocessing board FPGA to the signal processing board DSP via its SRIO signal;

[0017] The signal processing board (DSP) is used to complete the synthetic aperture radar (SAR) imaging RD algorithm processing and upload the results to the display and control terminal for display.

[0018] Compared with existing technologies, the present invention has the following advantages:

[0019] First, this invention realizes the software-based synthetic aperture radar (SAR) imaging technology. The difference between this invention and traditional customized SAR imaging components is that the software can control the SAR imaging algorithm to start, stop, reconstruct, and migrate hardware.

[0020] Second, this invention is decoupled from hardware and can be designed as a radar component that can flexibly deploy synthetic aperture radar (SAR) imaging algorithms to adapt to different hardware resources. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A schematic diagram of the overall framework for a DSP-based synthetic aperture radar (SAR) imaging component design method;

[0023] Figure 2 This is a schematic diagram of the RD algorithm in DSP-based synthetic aperture radar SAR imaging according to the present invention;

[0024] Figure 3 This is a schematic diagram of the overall connection relationship of the DSP-based synthetic aperture radar SAR imaging component design method of the present invention;

[0025] Figure 4 A flowchart illustrating how the DSP signal processing board receives and parses middleware control commands.

[0026] Figure 5 This is a data flow diagram of the DSP-based synthetic aperture radar SAR imaging component of the present invention;

[0027] Figure 6 A schematic diagram of range compression for a DSP-based synthetic aperture radar (SAR) imaging component provided as an example of the present invention;

[0028] Figure 7 This is a schematic diagram of motion compensation for the DSP-based synthetic aperture radar (SAR) imaging component of the present invention.

[0029] Figure 8 This diagram illustrates the relationship between the geographic coordinate system and the radar coordinate system. The ideal heading angle α is the angle between the north axis and the ideal heading, with clockwise rotation being positive; αr is the angle between the north axis and the actual heading, with clockwise rotation being positive; Δα is the angle by which the actual heading deviates from the ideal heading; and the eastward velocity V... e Northbound speed V n The magnitudes are provided by the inertial navigation system, respectively, along the due east and due north directions of the geographic coordinate system.

[0030] Figure 9 This is a schematic diagram of the range migration scenario for the design method of the synthetic aperture radar SAR imaging component based on DSP of the present invention;

[0031] Figure 10 A schematic diagram of azimuth compression of a DSP-based synthetic aperture radar (SAR) imaging component provided as an example of the present invention.

[0032] Figure 11 This diagram illustrates the process by which the DSP (Digital Signal Processing Board) quantizes the azimuth-compressed data of this invention to obtain a pixel map.

[0033] Figure 12 This diagram shows the peripheral and memory resources of the TI TMS320C6678 DSP chip.

[0034] Figure 13 The image shows a simulation of measured data from a DSP-based synthetic aperture radar (SAR) imaging component that does not utilize the present invention.

[0035] Figure 14 The image shows a simulation of measured data from the DSP-based synthetic aperture radar (SAR) imaging component of this invention. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Reference Figure 1 This paper presents an overall framework diagram of a DSP-based synthetic aperture radar (SAR) imaging component design method. First, a command parsing framework is built in the signal processing board DSP. Then, a data flow diagram of the DSP-based SAR imaging component design method is constructed. Finally, the RD algorithm is implemented using the DSP based on the signal processing board, which is controlled by the main control board, to process the radar echo signal, thereby realizing the software-based SAR imaging technology and decoupling the SAR imaging algorithm from the hardware. This design method can adapt to different hardware resources.

[0038] The synthetic aperture radar (SAR) imaging software component in the DSP-based synthetic aperture radar (SAR) imaging component design method of this invention mainly realizes two functions: first, to complete the imaging function of radar echo data while ensuring the real-time processing and imaging accuracy of radar echo data; second, to complete the start-up, reconstruction, hardware migration and other operations in the entire radar echo data processing process under the control of the DSP core framework.

[0039] Start-stop functionality refers to the ability of a component algorithm module to start or stop under the control of the host computer: startup begins computation, and stoppage stops computation. Reconfiguration refers to the ability, under the control of the host computer, for a component running on DSP hardware to be reloaded or replaced with another component. Migration refers to the ability, under the control of the host computer, for a component to be loaded onto different hardware.

[0040] Reference Figure 2 The diagram illustrates the RD algorithm in Synthetic Aperture Radar (SAR) imaging. The RD algorithm in SAR imaging consists of five steps: range compression, motion compensation, migration correction, azimuth compression, and quantization. All five steps are completed within a signal processing system composed of a DSP and an FPGA, including the following steps:

[0041] Step 1: Build a processing system for the DSP-based synthetic aperture radar (SAR) imaging component.

[0042] like Figure 3 The diagram shows the overall connection relationship of a DSP-based synthetic aperture radar (SAR) imaging component design method. The system consists of a display and control terminal and four boards, namely a DDR storage board (or RF front-end board), an ARM main control board, an FPGA preprocessing board, and a DSP signal processing board.

[0043] The storage board is used to store and replay radar echo data; it can also be replaced by an RF front-end board that can acquire echo data in real time. The storage board (or RF front-end board) can replay radar echo data to the preprocessing board via the Aurora interface protocol of the preprocessing board FPGA. Aurora is a scalable, lightweight link layer protocol for moving data between point-to-point serial links.

[0044] The main component of the main control board is an ARM (Acorn RISC Machine) chip. The software running on the ARM sends control commands to the signal processing board and the preprocessing board. The main control board sends control commands through GE (Gigabit Ethernet) to control the preprocessing board and the signal processing board to complete the functions of starting, stopping, reconstructing, and migrating the synthetic aperture radar SAR imaging algorithm.

[0045] The display and control terminal software runs on a PC (Personal Computer). The software on the PC sends control commands to the main control board via GE (Gigabit Ethernet). After receiving the control commands, the main control board transmits the control commands to the preprocessing board and the signal processing board.

[0046] The main component of the preprocessing board is an FPGA (Field-Programmable Gate Array) chip. The SRIO signal of the FPGA controls the flow of radar echo data from the preprocessing board to the signal processing board.

[0047] The main component of the signal processing board is the DSP chip, which completes the main synthetic aperture radar (SAR) imaging RD algorithm processing. The signal processing board uploads the results to the display terminal via GE (Gigabit Ethernet) for result display.

[0048] Step 2: Build a command parsing framework in the signal processing board DSP.

[0049] like Figure 4 The flowchart shown is for the signal processing board DSP receiving and parsing middleware control commands. Figure 4 Other commands include 'stop', 'restructure', and 'switch'.

[0050] 1) Upon startup, the DSP component of the signal processing board first enters a state waiting for control commands;

[0051] 2) When a control command is received, it is parsed. If the parsed command is not 'start' or 'stop', then...

[0052] 'restructure' and 'switch' will exit and wait for control commands again;

[0053] 3) If the received instruction is the character 'stop' from other commands, then the stop operation will be executed, and the system will continue to wait for new control commands;

[0054] 4) If the received instruction is the character 'restructure' from other commands, then perform the reconstruction operation, and after completion, continue to wait for new control commands;

[0055] 5) If the received instruction is the character 'switch' from another command, then perform the migration operation, and after completion, continue to wait for a new control command;

[0056] 6) If the received command is 'start', execute the SAR imaging processing algorithm. After processing one image, check again for any new commands. If so, proceed to step 2) to parse the new control commands. Otherwise, continue processing the next image.

[0057] The specific processing steps are as follows:

[0058] 1) The signal processing board (DSP) allocates a message buffer and a new message flag;

[0059] 2) When the signal processing board DSP starts running for the first time after power-on, it continuously checks the new message flag bit. When the new message flag bit is 1, it means that a new command has arrived.

[0060] 3) If the new message flag is 1, the control command information in the message buffer of the signal processing board DSP is compared with 'start'. If the control command information is the same as the character 'start', the synthetic aperture radar SAR imaging processing algorithm will be executed.

[0061] 4) If the control command information in the message buffer of the DSP on the signal processing board is the character 'stop', then a stop operation will be performed;

[0062] 5) If the control command information in the message buffer of the DSP on the signal processing board is 'restructure', then perform the reconstruction operation;

[0063] 6) If the control command information in the DSP message buffer of the signal processing board is 'switch', then the migration operation is performed;

[0064] 7) After completing the operation, continue to wait for the arrival of new commands;

[0065] 8) If the control command information in the message buffer of the signal processing board DSP is not 'start', 'stop', 'restructure', or 'switch', then skip the parsing of the control command information in the message buffer of the signal processing board DSP and return to step 2) to query the new message flag.

[0066] The 'start' and 'stop' commands control the start and stop of components by using flags in the command post-code.

[0067] The 'restructure' function is achieved by rebooting the DSP to load the FLASH programming code. Specifically, the FLASH can be partitioned according to address, and different functional codes can be programmed into it. The DSP can then read the code from different address locations in the FLASH through software control, thereby loading different functions.

[0068] 'switch' achieves hardware decoupling, allowing loading functions to be implemented on different DSPs under software control. This is mainly achieved by software middleware sending code to the DSP via the network.

[0069] During the execution of the synthetic aperture radar (SAR) imaging processing algorithm, the signal processing board (DSP) does not query newly received control commands in real time. Newly received control commands are first cached in the message buffer of the signal processing board (DSP) and the new message flag is set to 1. Only after the signal processing board (DSP) completes one round of the synthetic aperture radar (SAR) imaging processing algorithm will it query again whether there are new control commands in the message buffer. If there are new control commands, the corresponding new control commands will be parsed and executed according to step 3).

[0070] Step 3: Construct the data flow of the DSP-based synthetic aperture radar (SAR) imaging component.

[0071] Reference Figure 5 This is a data flow diagram of the DSP-based synthetic aperture radar (SAR) imaging component of the present invention.

[0072] 1) First, the preprocessing board FPGA transmits the radar echo data to the signal processing board DSP through the high-speed interface SRIO at the front end. Every time the high-speed interface SRIO at the front end of the preprocessing board FPGA transmits radar echo data for one pulse repetition period (PRT) to the signal processing board DSP, it sends an SRIO doorbell interrupt to the signal processing board DSP.

[0073] 2) After the signal processing board DSP receives the SRIO doorbell interrupt, it starts calling the interrupt response function. The interrupt response function uses the signal processing board DSP's EDMA (Enhanced Direct Memory Access) to transfer the radar echo data of one pulse repetition period (PRT) received from the preprocessing board FPGA and stored in the buffer area located at a fixed address in the signal processing board DSP core to the specified address in the storage board DDR (Double Data Rate) for storage and waiting for processing by the synthetic aperture radar SAR imaging RD algorithm.

[0074] 3) Then, after the memory board DDR receives radar echo data of N pulse repetition periods PRT from the signal processing board DSP buffer, it triggers the signal processing board DSP to transmit the radar echo data of N pulse repetition periods PRT located in the memory board DDR to the core buffer of the signal processing board DSP through its EDMA.

[0075] 4) Further, with the cooperation of the signal processing board DSP and the storage board DDR, the RD algorithm, which includes five modules: distance compression, motion compensation, migration correction, azimuth compression, and quantization, is started. The final result is displayed by the host computer consisting of the main control board ARM and the display terminal.

[0076] Step 4: Use the signal processing board DSP to perform SAR imaging RD algorithm processing.

[0077] Reference Figure 5 The diagram shows the data flow of the DSP-based synthetic aperture radar (SAR) imaging component of this invention. The SAR imaging RD algorithm includes five modules: range compression, motion compensation, migration correction, azimuth compression, and quantization. The final result is displayed by the host computer.

[0078] The processing logic of the range compression module is as follows: 1) The signal processing board DSP first reads N pulse repetition period PRT radar echo data from the storage board DDR through its EDMA to perform range compression module calculations; 2) After the calculation is completed, the range compression data of the N pulse repetition period PRT radar echo data is transmitted back to the storage board DDR buffer through the signal processing board DSP's EDMA to wait for processing by the motion compensation, migration correction, azimuth compression, and quantization processing modules.

[0079] The data flow for the subsequent motion compensation, migration correction, azimuth compression, and quantization modules follows the same pattern as above: The DDR memory board triggers the DSP signal processing board to transmit the N pulse repetition period PRT radar echo result data processed by the previous module in the DDR memory board to the DSP signal processing board via EDMA. The DSP signal processing board then performs the corresponding algorithm calculation for the RD algorithm module. After the calculation is completed, the N pulse repetition period PRT radar echo result data is transmitted back to the peripheral DDR via EDMA for buffering, in preparation for the next module's processing. This process is repeated to complete the calculation of all corresponding modules of the RD algorithm.

[0080] The sub-steps of step 4 are as follows:

[0081] 1) Range compression: The DDR memory board triggers the signal processing board DSP to read the N pulse repetition cycles of PRT radar echo data stored in the DDR memory board via EDMA to perform range compression module calculations.

[0082] like Figure 6 This is a schematic diagram of range compression of a DSP-based synthetic aperture radar (SAR) imaging component provided in this invention example. Range compression of radar echo data is achieved through matched filtering technology, which requires range-direction matched filtering of radar echo data in the range compression algorithm.

[0083] The processing steps are as follows:

[0084] 1a) Perform a Fast Fourier Transform (FFT) on the radar's fundamental frequency echo signal along the range direction;

[0085] 1b) The matching function required for range compression (the matching function is calculated based on the bandwidth and sampling points of the radar transmitted signal) is subjected to Fast Fourier Transform (FFT) and then windowed.

[0086] 1c) Perform conjugate dot product of the results from 1a) and 1b);

[0087] 1d) Perform an inverse fast Fourier transform (IFFT) on the data resulting from 1c) to complete matched filtering and achieve distance compression.

[0088] After range compression of the PRT radar echo data with N pulse repetition cycles, the signal processing board DSP sends the processed range-compressed data to the storage board DDR for storage via EDMA, thus completing the range compression step of the PRT radar echo data with N pulse repetition cycles.

[0089] 2) Motion compensation: The DSP on the memory board triggers the signal processing board via EDMA to read the data from the memory board's DDR.

[0090] The motion compensation module calculates N pulse repetition cycles of PRT radar echo data stored in the system after range compression processing.

[0091] The flowchart of motion compensation is as follows Figure 7 This is a motion compensation diagram illustrating the design method of the synthetic aperture radar (SAR) imaging component based on DSP of the present invention. The overall steps are divided into velocity projection, slant range error calculation, envelope compensation, slant range error recalculation, and phase compensation. First, the northeast-sky velocity (i.e., the due east velocity, due north velocity, and zenith velocity) is projected onto the radar coordinate system velocity, such as... Figure 8 This diagram illustrates the relationship between the geographic coordinate system and the radar coordinate system. The ideal heading angle α is the angle between the north axis and the ideal heading, with clockwise rotation being positive; αr is the angle between the north axis and the actual heading, with clockwise rotation being positive; Δα is the angle by which the actual heading deviates from the ideal heading; and the eastward velocity V... e Northbound speed V n The magnitudes are provided by the inertial navigation system, respectively, along the due east and due north directions of the geographic coordinate system.

[0092] The magnitude of the instantaneous heading angle αr is determined by the east and north velocities (V). e V n Substituting into the following formula, we get:

[0093] αr=atan(V e / V n (1)

[0094] Then calculate the angle by which the actual course deviates from the ideal course (yaw angle Δα):

[0095] Δα=αr-α(2)

[0096] If the aircraft carrying the radar platform flies at an altitude of H, and the distance between the radar and the center of the imaging scene is R, then the downward angle β...

[0097] It is calculated using the following formula:

[0098]

[0099] Then from the east and north speed V e V n Calculate the speed V of the actual heading using the Pythagorean theorem. r size:

[0100]

[0101] Speed ​​V r This represents the speed of the aircraft on the horizontal plane. The speed V... r Projecting these velocities onto the X and Y axes of the radar coordinate system yields velocities Vx and Vy along these two axes. The inertial navigation system's velocity corresponds to the velocity Vz along the Z-axis.

[0102] Vx = V r *cos(Δα)

[0103] Vy = V r *sin(Δα)

[0104] Vz = V u (5)

[0105] Each pulse repetition period PRT has three velocity components. The three velocity components are integrated with respect to slow time to obtain three position components X(tm), Y(tm), and Z(tm). These three position components correspond to the actual positions along the X, Y, and Z axes, respectively.

[0106] The motion error ΔR can be calculated:

[0107]

[0108] in,

[0109] Y2(tm)=∑V r sin(Δα+θ) (7)

[0110] The oblique angle is the angle at which the target deviates from the vertical direction of the aircraft's heading. In the above formula, θ is the angle projected onto the ground by the oblique angle.

[0111] The actual position of the aircraft along the X-axis is X(tm). Taking the average of Vx, we obtain the average velocity of the aircraft along the flight path. Integrating the average velocity over slow time, we obtain the ideal position X of the aircraft along the X-axis. i The position error ΔX of the carrier aircraft is obtained by subtracting the actual position from the ideal position.

[0112] ΔX=X(tm)-X i (8)

[0113] Then according to

[0114]

[0115] The slant distance error term ΔRx is obtained.

[0116] The signal is transformed to the distance frequency domain by performing a Fast Fourier Transform (FFT), multiplied by the compensation function H1, and then transformed to the distance time domain by performing an Inverse Fast Fourier Transform (IFFT) to complete the envelope compensation.

[0117]

[0118] Here, fr represents the frequency axis, and c represents the speed of light. In the compensation function H1, ΔR and ΔRx represent the slant range error of the central target in the scene, meaning that all targets in all range units are compensated uniformly according to the slant range error of the central target, ignoring the spatial variability of the range direction. In practical applications, segmented compensation can be performed in the range direction based on the spatial variability of the data.

[0119] The envelope of the radar echo data is compensated to complete the motion error correction of the radar echo data using inertial navigation information.

[0120] After performing range compensation on the PRT radar echo data with N pulse repetition cycles, the signal processing board DSP sends the processed range compensation data to the storage board DDR for storage via EDMA, thus completing the range compensation step for the PRT radar echo data with N pulse repetition cycles.

[0121] 3) Migration Correction: The DSP of the signal processing board triggers the memory board DDR to read the N pulse repetition cycles of PRT radar echo data stored in the memory board DDR after motion compensation processing via EDMA to perform the migration correction module calculation.

[0122] like Figure 9 The diagram shows a range migration scenario for the design method of the synthetic aperture radar (SAR) imaging component based on DSP of the present invention. The radar carrier flies in a straight line under ideal conditions, and the scenario is divided into two directions: range and azimuth.

[0123] The azimuth direction is the direction of radar carrier movement, the range direction is the direction perpendicular to the azimuth direction, target P is the target point on the ground in the imaging model, the range gate represents the distance represented by each sampling point in the range direction, and the range line is the line connecting the distance between the radar carrier and the same target at different positions. It can be seen that as the radar carrier passes the target point, the straight-line distance between the target and the radar carrier is initially short and then increases. In high-resolution scenarios, the difference in slant range for the same target point from different apertures can exceed the length of a range cell, causing data points contributing to the same target point from different pulses to fall into different range cells, resulting in range migration. To separate synthetic aperture radar (SAR) imaging into range and azimuth dimensions for separate processing, range migration correction is performed on the motion-compensated radar echo signal.

[0124] The migration correction module of the RD imaging algorithm is implementing range migration in the two-dimensional frequency domain: 3a) First, the range-oriented blocks are divided into N pulse repetition period (PRT) radar echo data after motion compensation processing, and a range constant region is set; 3b) Then, based on the return time and aircraft speed corresponding to the pulse echo of each radar echo data in the N pulse repetition period (PRT) radar echo data, the migration amount corresponding to the range unit of each radar echo data is calculated; 3c) The signal is compensated in the frequency domain and multiplied in the frequency domain to obtain the range migration correction data of the N pulse repetition period (PRT) radar echo data after migration correction.

[0125] After performing migration correction on the PRT radar echo data for N pulse repetition cycles, the signal processing board DSP sends the processed migration correction data to the storage board DDR for storage via EDMA, thus completing the migration correction step for the PRT radar echo data for N pulse repetition cycles.

[0126] 4) Azimuth compression: The DSP of the signal processing board triggers the storage board DDR to read the migration-corrected N pulse repetition period PRT radar echo data stored in the storage board DDR through EDMA to perform azimuth compression module calculation.

[0127] like Figure 10 This is a schematic diagram of azimuth compression of a DSP-based synthetic aperture radar (SAR) imaging component provided in this invention example. Similar to the range compression module, the azimuth compression module is also implemented using a matched filtering method.

[0128] The processing steps are as follows:

[0129] 4a) Perform a Fast Fourier Transform (FFT) on the azimuth radar echo data reference matrix to obtain the frequency domain azimuth radar echo data;

[0130] 4b) Apply Fast Fourier Transform (FFT) to the pulse compression coefficients to obtain the pulse compression coefficients in the frequency domain;

[0131] 4c) Multiply the matrices from 4a) and 4b);

[0132] 4d) Perform an inverse fast Fourier transform (IFFT) on the data obtained in 4c) to convert it into time-domain azimuth compressed radar echo data.

[0133] After azimuth compression of the PRT radar echo data with N pulse repetition cycles, the signal processing board DSP sends the processed azimuth-compressed data to the storage board DDR for storage via EDMA, thus completing the azimuth compression step of the PRT radar echo data with N pulse repetition cycles.

[0134] 5) Quantization processing: The DSP of the DDR trigger signal processing board reads the N pulse repetition cycles of PRT radar echo data stored in the DDR of the storage board through EDMA and performs calculations by the quantization processing module.

[0135] The azimuth-compressed radar echo data remains complex, with its real and imaginary parts interleaved and stored in the DSP memory of the signal processing board. The synthetic aperture radar (SAR) image data is pixel data obtained by quantizing the results of the azimuth compression process, with each pixel represented by 8 bits. Complex point data from N pulse repetition periods (PRT) radar echo data are uploaded to the display and control terminal via GE.

[0136] like Figure 11 The diagram shows the process of the signal processing board DSP quantizing the azimuth-compressed data of this invention to obtain a pixel map. The steps of the signal processing board DSP quantizing the azimuth-compressed data to obtain pixels are as follows:

[0137] 5a) The radar echo data after azimuth compression is still complex. In the signal processing board DSP, the modulus operation is performed on each complex point data in the PRT radar echo data of N pulse repetition periods to obtain the modulus value data. The modulus value data of all complex points obtained in 5a) is put back into the peripheral DDR cache through EDMA.

[0138] 5b) In the signal processing board DSP, the radar echo data in each pulse repetition period PRT is taken as the unit, and the mean value of all complex point modulus data included in the radar echo data in each pulse repetition period PRT is calculated according to 5a) to obtain the unit mean value data.

[0139] 5c) The unit mean data obtained in 5b) is placed into the MSM (Multicore Shared Memory) located in the DSP for caching. The maximum value library function of the DSP built into the signal processing board is called to calculate the maximum value of the unit mean data stored in the MSM in the DSP. The modulus data of the N pulse repetition period PRT radar echo data cached in the peripheral DDR is retrieved. Each complex point data is divided by the maximum value of the unit mean data obtained in 5b) to realize the normalization operation of the PRT radar echo data in the N pulse repetition period PRT.

[0140] 5d) Multiply the normalized radar echo data from 5c) by 255, convert it into uint8 data type, and store it in the DSP signal processing board to complete the quantization operation of complex point data of PRT radar echo data with N pulse repetition cycles.

[0141] The effects of the present invention will be further explained below with reference to simulation experiments.

[0142] The simulation uses a TI TMS320C6678DSP chip for the signal processing board. The storage board and storage resources are as follows: Figure 12 This diagram illustrates the peripherals and storage resources of the TI TMS320C6678DSP chip. The TI TMS320C6678DSP integrates a large amount of on-chip memory. In addition to 32KB of L1 program and data cache (buffered data memory), each core has 512KB of purpose-specific memory configured as mapped RAM or cache. The device also integrates 4096KB of multi-core shared memory that can be used as shared L2 SRAM and / or shared L3 SRAM. All L2 memory includes error detection and error correction. For fast access to external memory, this device includes a 64-bit DDR-3 External Memory Interface (EMIF) operating at 1600MHz and featuring ECC DRAM support. This series supports a wide range of high-speed standard interfaces, including Fast I / O 2.0, PCI bus 2.0, and Gigabit Ethernet similar to an integrated Ethernet switch. It also includes I2C, UART, Telecom Serial Interface (TSIP), and a 16-bit EMIF, along with a general-purpose CMOS I / O.

[0143] The present invention relates to memory including L1, L2, and DDR, and peripherals including EDMA and SRIO.

[0144] Simulation experimental parameters:

[0145] Frequency band: 34.5GHz;

[0146] Imaging mode: strip;

[0147] Signal type: Linear frequency modulation;

[0148] Signal bandwidth: 500MHz;

[0149] Original data storage location: storage board

[0150] (1) Simulation of measured data without using the DSP-based synthetic aperture radar SAR imaging component design method of this invention.

[0151] The simulation steps are as follows:

[0152] 1) The radar echo parameters pre-stored on the storage board are played back through the storage board control software, and the FPGA on the storage board sends the data to the FPGA on the preprocessing board through the Arura protocol.

[0153] 2) The FPGA on the preprocessing board forwards the radar echo data to the DSP on the signal processing board through the SRIO interface;

[0154] 3) The DSP performs data processing;

[0155] 4) The DSP uploads the processing results to the host computer via the network, where the results can be observed.

[0156] The processing results are displayed on the host computer as follows: Figure 13 The figure shown is a simulation diagram of measured data for a synthetic aperture radar (SAR) imaging component design method based on DSP of the present invention that was not adopted.

[0157] (2) Simulation of measured data using the DSP-based synthetic aperture radar SAR imaging component design method of the present invention.

[0158] To verify the effectiveness and practicality of this component, the radar imaging component has been tested in the electronic reconnaissance-radar imaging-communication waveform scenario. Through software control, the electronic reconnaissance component is first loaded on the hardware. After the electronic reconnaissance successfully detects the target, it sends back the result and switches to the radar imaging function. The radar imaging performs imaging on the target, and the imaging result is stored in the board's memory device as the data source for the communication waveform. Then, it switches to the communication waveform function and sends the radar imaging result out via radio frequency.

[0159] The parameter settings in this test are the same as those in Simulation 1. The test steps are as follows:

[0160] 1) Deploy the SAR radar component from the main control board's display and control software;

[0161] 2) Start the DSP on the signal processing board and the FPGA on the preprocessing board;

[0162] 3) Send echo data;

[0163] 4) Observe the processing results on the host computer of the display and control terminal.

[0164] Test data results are as follows Figure 14 The image shown is a test result of the experimental data from the design method of the DSP-based synthetic aperture radar SAR imaging component of this invention. The five angular reflections spaced 1 meter apart in the lower left corner indicate that the resolution of this image is 1 meter.

[0165] The test utilizes a TI TMS320C6678DSP to build a processing system for a DSP-based synthetic aperture radar (SAR) imaging component design method. This system processes radar echo data to obtain SAR imaging results from the radar echo data. This invention enables the software-based implementation of SAR imaging components.

[0166] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A design method for a DSP-based synthetic aperture radar (SAR) imaging component, characterized in that, The method comprises the following steps: Step 1, building a processing system of a synthetic aperture radar (SAR) imaging component based on a DSP; Step 2, building a command analysis framework in a signal processing board DSP; Step 3, building a data flow direction of the synthetic aperture radar (SAR) imaging component based on the DSP; Step 4, using the signal processing board DSP to perform SAR imaging RD algorithm processing; In step 1, the processing system comprises a display control terminal and four board cards, and the four board cards are a storage board DDR, a master control board ARM, a preprocessing board FPGA and a signal processing board DSP; The storage board DDR is used for storing and playing back radar echo data, and plays back the radar echo data to the preprocessing board through an Aurora interface protocol of the preprocessing board; The master control board ARM is used for running software for issuing control instructions to the signal processing board DSP and the preprocessing board FPGA, and controls the preprocessing board FPGA and the signal processing board DSP to start, stop, restructure and migrate hardware functions of a function program of a synthetic aperture radar (SAR) imaging algorithm through the issued control instructions; The display control terminal is used for issuing control instructions to control the master control board ARM, and the master control board ARM transmits the control instructions to the preprocessing board FPGA and the signal processing board DSP after receiving the control instructions; The preprocessing board FPGA is used for controlling radar echo data to flow from the preprocessing board FPGA to the signal processing board DSP through an SRIO signal of the preprocessing board FPGA; The signal processing board DSP is used for completing synthetic aperture radar (SAR) imaging RD algorithm processing and uploading a result to the display control terminal for result display; In step 2, the command analysis framework of the signal processing board DSP is as follows: 2a) the signal processing board DSP component enters a waiting control instruction state first; 2b) when a control instruction is received, the control instruction is analyzed, and if the analyzed instruction is not'start','stop','restructure' or'switch', the signal processing board DSP component exits and waits for a new control instruction; 2c) if the received instruction is a character'stop' in other instructions, a stop operation is performed, and the signal processing board DSP component continues to wait for a new control instruction; 2d) if the received instruction is a character'restructure' in other instructions, a restructure operation is performed, and the signal processing board DSP component continues to wait for a new control instruction after the execution; 2e) if the received instruction is a character'switch' in other instructions, a migration operation is performed, and the signal processing board DSP component continues to wait for a new control instruction after the execution; 2f) if the received instruction is'start', a SAR imaging processing algorithm is performed, a result of processing an image is obtained, and the signal processing board DSP component rechecks whether a new instruction arrives, and performs step 2b) to analyze the new control instruction if the new instruction arrives, or continues to process a next image if the new instruction does not arrive; In step 3, the data flow direction framework is as follows: 3a) Firstly, the pre-processing board FPGA transmits the radar echo data to the signal processing board DSP through the high-speed interface SRIO in the front end, and the pre-processing board FPGA sends an SRIO doorbell interrupt to the signal processing board DSP every time it transmits one pulse repetition period PRT of radar echo data to the signal processing board DSP; 3b) After the signal processing board DSP receives the SRIO doorbell interrupt, it starts to call the interrupt response function, which transmits one pulse repetition period PRT of radar echo data received from the pre-processing board FPGA and stored in the cache area at a fixed address in the core of the signal processing board DSP to a specified address in the storage board DDR for SAR imaging RD algorithm processing; 3c) Then, after the storage board DDR receives N pulse repetition periods PRT of radar echo data from the cache area of the signal processing board DSP, the storage board DDR triggers the signal processing board DSP to transmit the N pulse repetition periods PRT of radar echo data in the storage board DDR to the cache in the core of the signal processing board DSP through EDMA; 3d) Further, the signal processing board DSP and the storage board DDR start to launch the RD algorithm including five modules of range compression, motion compensation, migration correction, azimuth compression, and quantization, and the final result is displayed by the host computer composed of the display and control terminal and the main control board ARM.

2. The DSP-based synthetic aperture radar (SAR) imaging component design method of claim 1, wherein, The SAR imaging RD algorithm processing procedure using the signal processing board DSP is as follows: 4a) Range compression: the storage board DDR triggers the signal processing board DSP to read the N pulse repetition periods PRT of radar echo data stored in the storage board DDR through EDMA for range compression module calculation; 4b) Motion compensation: the storage board DDR triggers the signal processing board DSP to read the N pulse repetition periods PRT of radar echo data stored in the storage board DDR through EDMA for motion compensation module calculation after range compression processing; 4c) Migration correction: the storage board DDR triggers the signal processing board DSP to read the N pulse repetition periods PRT of radar echo data stored in the storage board DDR through EDMA for migration correction module calculation after motion compensation processing; 4d) Azimuth compression: the storage board DDR triggers the signal processing board DSP to read the N pulse repetition periods PRT of radar echo data stored in the storage board DDR through EDMA for azimuth compression module calculation after migration correction; 4e) Quantization processing: the storage board DDR triggers the signal processing board DSP to read the N pulse repetition periods PRT of radar echo data stored in the storage board DDR through EDMA for quantization processing module calculation after azimuth compression.

Images