Systems Engineering and Electronics ›› 2018, Vol. 40 ›› Issue (5): 976-981.doi: 10.3969/j.issn.1001-506X.2018.05.03

Previous Articles     Next Articles

Adaptive sample selection algorithm based on sparse recovery spectral similarity

WANG Xiaoming1, LI Jun1, ZHANG Shengmiao1, LU Yan1, ZHANG Yongjie2   

  1. 1. School of Electronic Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China; 2. Science and Technology on MillimeterWave Laboratory, Beijing 100854, China
  • Online:2018-04-28 Published:2018-04-24

PDF (PC)

Abstract:

Training sample selection is inevitable for space-time adaptive processing in a heterogeneous environment. The existing training sample selection algorithm based on Fourier spectral similarity (FSPS) cannot meet the resolution requirement in both the target signal contaminated sample discarding step and the similar sample selecting step when the system’s degree of freedom (DoF) is small. To mitigate this problem, this paper proposes an adaptive training sample selection algorithm based on the sparse recovery spectrum. This study utilizes the cell under test (CUT) and initial training samples’ high-precision sparse recovery spectrums to select samples which have similar clutter characteristics with the CUT. The proposed method enhances the resolution for heterogeneous samples compared with FSPS while shows the robustness in a heterogeneous environment. Simulation results demonstrate that the proposed method has better selection results and system output performance than the FSPS method under a low system’s DoF situation.

[an error occurred while processing this directive]