Accumulation cross-correlation envelope alignment 8-core DSP on-chip parallel implementation method

Abstract

The invention relates to a multi-core DSP parallel method, and provides an accumulation cross-correlation envelope alignment 8-core DSP on-chip parallel implementation method. The method is characterized in that, to begin with, a DSP core 0 carries out accumulation cross-correlation envelope alignment processing on front Z PRF pulse echoes of a frame of radar echo data to obtain a stable global reference signal refg, which is shared in follow-up alignment algorithms of cores 0-7, wherein Z is a multiple of 8; then, the DSP cores 0-7 carry out follow-up 8 PRF accumulation cross-correlation envelope alignment parallel processing according to the global reference signal refg, and the cores obtain aligned echo signals and local reference signals refI, I=[0,1,2,3,4,5,6,7] respectively, which represents the current core sequence; and then, the DSP core 0 detects whether all pulse signals in the frame are aligned, and if so, outputs the aligned echo signals, or otherwise, the DSP core 0 updates the global reference signal according to the eight reference signals ref[0-7] output by the eight cores, and the steps above are repeated. The method is suitable for accumulation cross-correlation envelope alignment 8-core DSP on-chip parallel implementation.

Description

technical field [0001] The invention relates to a multi-core DSP parallel method, in particular to a multi-core DSP parallel method for realizing a large amount of calculation, pulse-by-pulse recursive operation, and an accumulated cross-correlation envelope alignment method. Background technique [0002] In Inverse Synthetic Aperture Radar (ISAR), the quality of envelope alignment will have a significant impact on the final imaging quality. For two adjacent echoes, the rotation angle of the target is generally less than 0.01o, so their real envelopes are very similar, and their correlation coefficients are generally above 0.95. Based on the above facts, many experts and scholars at home and abroad have conducted a lot of research on the envelope alignment algorithm. It can be seen from the published literature that there are mainly the following three ways: [0003] 1. The maximum criterion of the cross-correlation function; [0004] 2. Minimum entropy criterion; [0005...

AI Technical Summary

Problems solved by technology

The disadvantage is that this algorithm can only be operated sequentially according to the pulse echo, and cannot be calculated in parallel to improve efficiency.

Images