TY - GEN
T1 - Sparsity aware dynamic distributed compressive spectrum sensing and scheduling
AU - Michelusi, Nicolo
AU - Mitra, Urbashi
N1 - Funding Information:
research was funded in part by one or more of the following grants: NSF CCF-1117896, NSF CNS-1213128, NSF CCF-1410009, AFOSR FA9550-12-1-0215, DOT CA-26-7084-00, NSF CPS-1446901 and ONR N00014-09-1-0700.
Publisher Copyright:
© 2015 IEEE.
PY - 2016/2/26
Y1 - 2016/2/26
N2 - A cross-layer framework for resource constrained dynamic distributed spectrum sensing and scheduling is presented. A network of secondary users (SUs) opportunistically communicate over portions of the spectrum estimated to be unused by other systems. A central controller (CC) schedules the traffic of the SUs, based on distributed compressed measurements collected by the SUs. Sensing and access are jointly controlled to maximize the SU throughput, with constraints on PU throughput degradation and SU cost. Sparsity in the network dynamics is exploited: leveraging a prior spectrum occupancy estimate, the CC needs to estimate only a residual uncertainty vector via sparse recovery techniques. The trade-off between achieving accurate spectrum estimates, high throughput, and low state information overhead, is optimized via dynamic programming.
AB - A cross-layer framework for resource constrained dynamic distributed spectrum sensing and scheduling is presented. A network of secondary users (SUs) opportunistically communicate over portions of the spectrum estimated to be unused by other systems. A central controller (CC) schedules the traffic of the SUs, based on distributed compressed measurements collected by the SUs. Sensing and access are jointly controlled to maximize the SU throughput, with constraints on PU throughput degradation and SU cost. Sparsity in the network dynamics is exploited: leveraging a prior spectrum occupancy estimate, the CC needs to estimate only a residual uncertainty vector via sparse recovery techniques. The trade-off between achieving accurate spectrum estimates, high throughput, and low state information overhead, is optimized via dynamic programming.
UR - http://www.scopus.com/inward/record.url?scp=84969850685&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84969850685&partnerID=8YFLogxK
U2 - 10.1109/ACSSC.2015.7421312
DO - 10.1109/ACSSC.2015.7421312
M3 - Conference contribution
AN - SCOPUS:84969850685
T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers
SP - 1109
EP - 1113
BT - Conference Record of the 49th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015
A2 - Matthews, Michael B.
PB - IEEE Computer Society
T2 - 49th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015
Y2 - 8 November 2015 through 11 November 2015
ER -