A node grouping algorithm for joint relay selection and resource allocation in cooperative cognitive radio networks

Compared with traditional wireless networks, cognitive radio network (CRN) has a time-variant and unbalanced spectrum distribution. Cooperative relaying technology, which enables resource-rich users to assist resource-short users, can effectively improve spectrum utilization as well as system fairness in CRN. This paper focuses on the R-C-P strategy, i.e. joint relay selection (R), channel assignment (C) and power allocation (P) strategy, for uplink OFDMA cooperative relaying CRN. The R- C-P scheme is formulated as a nonlinear optimization problem with the objective to improve both throughput and fairness, and a heuristic node grouping algorithm (NGA) is proposed to derive a polynomial-time suboptimal solution to the problem. NGA selects relay and allocates resource according to spectrum availability and traffic demands of cognitive users, effectively exploiting both space and frequency diversity of the system. Simulation results show that, NGA significantly improves system fairness level while not reducing the throughput, which has a better performance than other algorithms.