Institute of Information Science, Academia Sinica

Recent advance in compressive sensing enables efficient identification and recovery of sparse components over a large number of unknowns. Similarly, contention resolution, an important building block of medium access control protocols, requires identifying a small set of active hosts that need radio channel access and hence benefits from compressive sensing. In this talk, I will present two MAC designs that exploit compressive sensing for multiuser MIMO networks and traditional single user wireless LANs, respectively. We show that our designs can reduce the contention overhead and boost up the network throughput. Furthermore, compressive sensing facilitates channel response estimation, and as a result, simplifies the MAC protocol design for multiuser MIMO networks.
 

Tsung-Han Lin is a Ph.D. candidate in Computer Science in Harvard University. He received his B.S. and M.S. degrees from National Taiwan University both in Electrical Engineering. His research interests include wireless networks, compressive sensing, wireless networking for unmanned aerial vehicles (UAVs), and wireless sensor networks.