Characteristics Of Sub-one-dimensional Space Radar Target Like Fpga Identification Method To Achieve The Key Technology Research

Radar target recognition is one of the significant branches in modern radar application.It plays an important role in the modern weapon development programme and has produced plenty of high performance algorithms through years of develpment.Classical eigen-subspace and canonical subspace recognition algorithms based on target range profile had been developed. In this dissertation, we focused on the implementation of eigen-subspace extraction algorithm based on FPGA.Jacobi algorithm is a classical method used to solve SVD of martrix,it can be implemented more easily by parallel method than other algorithms.CORDIC algorithm takes an important role in Jacobi design and it has lots of excellence in hardware implementation and is implemented by FPGA.The dissertation includes:1.After analyzing the feasibility of algorithm's implementation based on FPGA, a corresponding design scheme is proposed.2.Research works were made on Jacobi method for SVD (singularity value decomposition) calculation, and CORDIC (Coordinate Rotational Digital Computer) algorithm for Jacobi method implementation. On the other hand, research is also referred to my design's working mode and FPGA implementation technologies, including convergence analysis, data precision, operation mode, implementation, etc.3.This dissertation introduces CORDIC processor's implementation. A 2-order matrix SVD module was designed. Functional and timing simulations based on Xilinx's Virtex4 series were provided for validation.4.Higher order matrix SVD method is discussed for future research work. The pipelining design of 2 range martrix's SVD has been implemented using Virtex4 xc4vlx200.