본문 바로가기
HOME> 논문 > 논문 검색상세

학위논문 상세정보

(A) study of greedy sparse signal recovery algorithm 원문보기

  • 저자

    권수혁

  • 학위수여기관

    Graduate School, Korea University

  • 학위구분

    국내박사

  • 학과

    컴퓨터電波通信工學科

  • 지도교수

    沈秉孝

  • 발행년도

    2014

  • 총페이지

    xi, 123장

  • 키워드

    신호처리;

  • 언어

    eng

  • 원문 URL

    http://www.riss.kr/link?id=T13541888&outLink=K  

  • 초록

    As a paradigm to acquire sparse signals at a rate significantly below Nyquist rate, compressive sensing (CS) has attracted considerable attention in recent years. The major goal of CS is to recover a high dimensional sparse signal from its low dimensional linear measurements. It is now well-known that if the measurement matrix satisfies the condition so called restricted isometry property (RIP) or mutual incoherence property (MIP), the sparse signal can be accurately recovered through properly designed recovery algorithms. Overall, the sparse recovery algorithms can be classified into two distinct categories: convex optimization method and greedy method. The convex optimization method recovers sparse signals using the linear programming (LP) technique. While this method provides excellent recovery performance, computational cost is often burdensome for large scale applications. As a cost-effective alternative of the convex optimization approach, greedy method has been widely used. The greedy method iteratively estimates the support or coefficients of the sparse signal to be recovered. In this dissertation, we investigate algorithms based on greedy method to improve sparse signal recovery performance. In the first part of the dissertation, we introduce an extension of the OMP for pursuing efficiency in reconstructing sparse signals. Our approach, henceforth referred to as generalized OMP (gOMP), is literally a generalization of the OMP in the sense that multiple $N$ indices are identified per iteration. Owing to the selection of multiple ``correct" indices, the gOMP algorithm is finished with much smaller number of iterations compared to the OMP. We show that the gOMP can perfectly reconstruct any $K$-sparse signal ($K > 1$), provided that the sensing matrix satisfies the RIP with $\delta_{NK}


 활용도 분석

  • 상세보기

    amChart 영역
  • 원문보기

    amChart 영역