The Optimal Clustering Center As Target Identification Method Of One Dimensional Distance Of The Fpga Design And Implementation

Radar target recognition is one of the important research directions in modern radar field, and plays a key role in military and national defense areas. As the high-resolution range profile(HRRP) based on wideband radar reflects targets’specific structure and shape information, lots of studies have showed it can effectively identify targets. Recently, with the development of FPGA, a high density and high performance FPGA provides a good platform for target recognition system based on range profile. FPGA target recognition system can meet the high-speed and real-time requests of the identification of high-speed moving targets, such as aircraft and missiles. So, it have an important military significance.This dissertation focused on studying the range profile target recognition methods based on the optimal cluster centers algorithms, which were designed and implemented in FPGA device. The main contents were as follows:1.Analyze the range profile target recognition methods based on the optimal cluster centers algorithms; simulate the algorithms in MATLAB; do the structural decomposition about the implementation of these algorithms in FPGA device.2.The basic single-precision floating point operation module was designed and implemented in FPGA, which contains the basic four arithmetic module, the root operation module and the accumulate module. The design of plus and minus operations module adopted two-path design methods; the division module adopted Anderson algorithms, and the method of initialization in Anderson algorithm was improved; the root operation module adopted Taylor approximation algorithms; the accumulation module adopted pipeline architecture. The simulation showed that all the modules were in high accuracy.3. Study how the256-point single-precision floating point FFT calculation module was designed and implemented in FPGA, and simulate these algorithms in FPGA. This module adopted pipelined architecture, which can effectively improve the rate of throughput.4. Study how the matrix operations module that used in the optimal cluster centers algorithms was designed and implemented in FPGA. This module contains three operation parts:the vector2-norm module, the matrix multiplication module and the matrix inversion module. The matrix multiplication module adopted parallel structure, which can effectively reduce calculation delay; the matrix inversion adopted the block matrix method, which can effectively reduce the consumption of DSP module integrated in FPGA.