基于子空间追踪重构算法的改进
Improved Reconstruction Algorithm Based on Subspace Pursuit
-
摘要: 压缩感知的重构算法中,子空间追踪算法(Subspace Pursuit,SP)是近几年来研究的热点.针对SP算法,引进前瞻策略和自适应选择步长的方法,来改进其重构效果.实验结果表明:自适应前瞻SP算法(Adaptively Look Ahead Subspace Pursuit,ALASP)重构效果优于其他匹配追踪算法.Abstract: Subspace Pursuit is one of the hot research topic.To achieve a better performance,developed a look ahead strategy that adaptively chooses the best number of coefficients.Subspace Pursuit algorithm use a support-set expansion by a fixed number of coefficients,leading to restricted performance.experimental results show that the new algorithm which can be call adaptively look ahead subspace pursuit performs better than other greedy pursuits algorithms.