Synthetic aperture radar echo parallel simulation method based on depth cooperation

Abstract

The invention discloses a synthetic aperture radar echo parallel simulation method based on depth cooperation, belonging to the technical field of synthetic aperture radar application. Under the traditional CPU+GPU heterogeneous computing mode, generally CPU refers to operation with high processing logicality, while GPU refers to computing which is used for processing relatively dense data and is suitable for parallel operations. The invention takes an example by a related satellite-borne SAR echo simulation parallel simulation algorithm on the basis of fully researching related articles of SAR rapid echo simulation at home and abroad, and provides a satellite-borne SAR rapid echo simulation algorithm based on SIMD heterogeneous parallel processing, the SAR echo simulation process is accelerated in parallel by virtue of depth cooperation of multi-core vector expanded CPU / many-core GPU, optimization of redundant computation and parallel depth optimization aiming at irregular problem computing in the SAR echo process are performed on the basis. The experimental result proves that compared with the traditional serial computing method, the optimized CPU / GPU heterogeneous cooperation mode has the advantage that the computational efficiency can be improved by 2-3 order of magnitudes.

Description

technical field [0001] The invention relates to a synthetic aperture radar echo parallel simulation method based on depth coordination, and belongs to the technical field of synthetic aperture radar applications. Background technique [0002] Synthetic Aperture Radar (SAR) is a radar that can provide high-resolution imaging effects, because it can obtain higher-resolution radar images under special meteorological conditions, and can effectively penetrate cover and Features such as identifying camouflage are widely used in military and civilian fields. Usually before the synthetic aperture radar is officially put into use, in order to reduce unnecessary cost investment, it is necessary to perform SAR echo simulation as an indispensable prerequisite to ensure the normal operation of the radar. At the same time, before radar imaging, SAR is required. echo data as a strong support. [0003] However, with the development of technology, the technology of synthetic aperture radar...

AI Technical Summary

Problems solved by technology

Acceleration of echo simulation based on signal processing methods has always been a research hotspot at home and abroad. Generally, acceleration is carried out by methods such as two-dimensional frequency domain and mixed domain, but it is difficult to simulate actual system errors; methods based on hardware acceleration are too expensive on the one hand. On the other hand, due to the limitation of storage, it is difficult to simulate the echo data of large scenes and long strips; the method based on high-performance computing can effectively make up for the shortcomings of the first two, and realize Accurate Simulation of Large Scene SAR Echoes

However, the existing GPU-based SAR echo high-performance simulation method seriously underestimates the computing power of the CPU. During the echo simulation process, the CPU only undertakes small calculation steps such as input and output, logic control, etc., and does not participate in high-density computing. The echo calculation process is a waste of computing resources

Images