科研之友
微信
新浪微博
Facebook
分享链接
摘要
The multiple measurement vector (MMV) problem aims to recover non-zero entries from multiple measurements. In order to discriminate and identify the target echo and repeater jamming in the mainlobe, multiple measurements are analysed and decomposed to form a two-dimensional reconstruction model by constructing two overcomplete dictionaries. It is an extension of the MMV problem. The two-dimensional reconstruction model can be converted to a single measurement vector problem by the Kronecker product and solved by basis pursuit. In order to reduce dimension, a random left matrix is adopted to propose the random Kronecker basis pursuit (RKBP) algorithm. Both the direction of arrival (DOA) and delay information of the target echo and jamming signals are obtained to identify them. Furthermore, RKBP outperforms sparse recovery for weighted subspace fitting which relies on second-order cone programming in the DOA resolution capability. Finally, simulation results verify the efficiency of the proposed algorithm.
- 出版日期2017-2
- 单位武汉理工大学; 中国人民解放军国防科学技术大学
