【115/10/22】16:20-17:10 梁惟㨗 助理教授 (國立屏東大學應用數學系)
發佈日期 :
2026-10-15
| Colloquium | |
|---|---|
|
|
|
| Time | 2026-10-22 16:20-17:10 |
| Venue | 數學館1樓 31106 教室 |
| Speaker | 梁惟㨗 助理教授 (國立屏東大學應用數學系) |
| Title | Distributed Compressive Sensing: Sparse Recovery and the Role of Signal Diversity |
| Abstract | Compressive Sensing (CS) provides a framework for recovering sparse signals from a limited number of linear measurements. Distributed Compressive Sensing (DCS) extends this framework by exploiting joint sparsity among multiple signals. However, existing theoretical recovery guarantees often fail to fully characterize the advantages of multiple measurement vectors (MMVs) over a single measurement vector (SMV). In this talk, I will introduce the fundamental concepts of compressive sensing and discuss how the characteristics of signal ensembles affect recovery performance. In particular, I will focus on the role of Euclidean distances between signals in the deterministic MMV framework. Using simultaneous orthogonal matching pursuit (SOMP) as a case study, I will present modified recovery strategies and sufficient conditions for exact recovery based on the restricted isometry property (RIP). Finally, numerical experiments will be presented to illustrate how different signal ensembles and the number of measurement vectors influence recovery performance. |
|
|
|
