Date & Time: 

2023 / 04 / 07 (Fri) 14:20 - 16:20

Location: 

Delta Building R216, NTHU

Speaker: 

 楊明勳 教授

國立中央大學通訊工程學系

Topic: 

Ambiguity-Free Sparsity-Aware Phase Retrieval: Algorithms and Performance Analysis

Abstract: 

Phase retrieval is concerned with the reconstruction of a signal from magnitude-only linear measurements. This problem arises in many engineering and scientific applications, such as optical imaging, X-ray crystallography, astronomy, etc., mainly because sensing devices in optical systems can only capture the intensity of light waves. It also arises in wireless communications for the purpose of removing unreliable phase information. To address this, many sparsity-promoting algorithms for efficient signal reconstruction have also been developed in the literature. However, all existing works ensure signal recovery only up to a global phase ambiguity.

In this talk, a novel CS-based approach that instead utilizes the magnitude of affine measurements to achieve ambiguity-free signal reconstruction will be presented. It will be shown that in the noise-free case, perfect signal recovery is guaranteed subject to a mild condition on the signal sparsity. The proposed approach is then extended to two noisy scenarios, namely, sparse noise (or outliers) and non-sparse bounded noise. We will show that for both cases, perfect support identification is still ensured under mild conditions on the noise model, and under perfect support identification, analytic performance guarantees are derived accordingly to justify the stable reconstruction of nonzero signal entries.

Biography: 

Ming-Hsun Yang received the Ph.D. degree in communications engineering from National Chiao Tung University, Taiwan, in 2018. From 2019 to 2021, he was a Post-Doctoral Researcher at the Institute of Communications Engineering, National Chiao Tung University. Since 2022, he has been with the Department of Communication Engineering, National Central University, where he is currently an Assistant Professor. His research interests include distributed learning, statistical signal processing and compressed sensing with applications in wireless sensor networks.