作者: David Wipf , Jeong-Min Yun , Qing Ling
DOI: 10.1109/DCC.2015.68
关键词:
摘要: The simultaneous sparse approximation problem is concerned with recovering a set of multichannel signals that share common support pattern using incomplete or compressive measurements. Multichannel modifications greedy algorithms like orthogonal matching pursuit (OMP), as well convex mixed-norm extensions the Lasso, have typically been deployed for efficient signal estimation. While accurate recovery possible under certain circumstances, it has established these methods may all fail in regimes where traditional subspace techniques from array processing, notably MUSIC algorithm, can provably succeed. Against this backdrop several recent hybrid developed merge estimation step OMP-like procedures to obtain superior results, sometimes theoretical guarantees. In contrast, paper considers completely different approach built upon Bayesian sensing. particular, we demonstrate minor standard naturally best both worlds backed and empirical support, surpassing performance existing algorithms, especially when poor RIP conditions render alternative approaches ineffectual.