A fast algorithm for joint sparse signal recovery in 1-bit compressed sensing

Abstract

In this letter, a new algorithm is proposed for fast 1-bit compressed sensing (CS) recovery from multiple measurement vectors (MMV). The proposed algorithm is based on the adaptive outlier pursuit-flip (AOP-f), which adaptively locates and corrects potential sign flips from the measurement vectors. To improve the performance, we introduce an erasure method to the proposed algorithm by setting ambiguous sign flips as erasures. Additionally, we present a new stopping criterion, which significantly reduces the time complexity of the proposed algorithm. Numerical results show that the proposed algorithm performs better than existing ones in terms of reconstruction performance and time complexity.