Enhancing the reliability of cognitive radio networks via channel assignment: risk analysis and redundancy allocation

Abstract

A key challenge in cognitive radio networks is the unreliability of cognitive radio links due to the interruptions from primary users. To enable reliable data transmission over cognitive radio networks, the large available bandwidth obtained from cognitive radio can be used to improve the system redundancy in order to guarantee the system reliability. Using the terminology in the theory of optimal reliability design in industrial engineering, a single path data flow can be considered as a parallel-series system, in which each cut set represents a secondary user and contains several assigned channels. A heuristic algorithm is proposed to assign the channels in the licensed spectrum band to different cut sets in the parallel-series system to minimize the overall risk. By adopting the concept of cut set hazard, the algorithm is applied in both single path and multiple path situations. It is also extended to the decentralized scenario. Channel reuse, which incurs the correlation of failures in the system, is also considered for the channel assignment. Simulation shows that the proposed algorithms can significantly improve the reliability of cognitive radio networks.