| High Range Resolution Profile(HRRP) has drawn great attention in radar automatic target recognition(RATR) community, and a lot of classical HRRP recognition methods have been proposed recently. HRRP is the summations of projection of the complex echoes returned from the target scattering centers along the radar line-of-sight based on target scattering center model theory,the scattering character for planes is determined mainly by strong scattering points which enjoys sparse distribution. It is beneficial for recognition if strong scattering points could be extracted, but there are few recognition algorithms based on scattering points. In this thesis we extract amplitude and location of scattering points under the assumption of target scattering center model theory and sparse representation, then propose a recognition algorithm based on point matching theory, and realize the real-time HRRP recognition system. The main content is summarized as follows.Firstly, we deduce the formula of HRRP echoes from the target scattering center model theory, and realize the extracting method of scattering points and denoising method of HRRP based on Orthogonal Matching Pursuit(OMP) of sparse representation.Secondly, we introduce basic definition of Hausdorff distance and one HRRP recognition method based on Hausdorff distance after extracting scattering points sets. Then one noise-robust recognition method is proposed by improvement. At last, good recognition results are obtained based on experiment of measured data.Finally, the design of target recognition system is introduced which contains host and signal processing board. This dissertation mainly studies two parts. 1)development of host software which includes introduction to development environment, design of process and framework, and software development which is completed in VC6.0 compiler. 2)development of DSP recognition program on signal processing board which includes introduction to hardware resources, realization of algorithms and optimization of program, and the development of DSP program with Visual DSP++ compiler. |
PDF Full Text Request