A Fair Multi-Channel Assignment Algorithm With Practical Implementation in Distributed Cognitive Radio Networks

Designing an efficient spectrum assignment (SA) mechanism is a key issue for realizing dynamic spectrum access in cognitive radio network. In multi-channel selection based SA schemes, secondary users (SUs) are able to utilize multiple channels simultaneously to enhance the network throughput. However, a fairness problem may happen if few SUs utilize too many idle data channels that other SUs are left with no idle channels, thus increasing the blocking probability and reducing the fairness. Aiming at improving the network throughput with multi-channel selection capability while maintaining fairness among the SUs, in this paper, we propose a fair multi-channel assignment scheme (FMCA) for distributed cognitive radio networks. For the FMCA scheme, we design a new MAC framework for sensing and access contention resolution, which is integrated into the FMCA scheme. Channel-aggregation (CA) technique is used in each SU to enable the multi-channel selection ability. Considering both of the idle data channel utilization efficiency and the transmit power budget constrained CA ability of each SU, we analytically formulate a channel assignment problem according to the well-known Jain's fairness criterion. Our objective is to find a channel assignment with maximal fairness index for all SUs. The optimization problem is turned out to be a quadratic integer programming (QIP). According to the definition of Jain's fairness criterion, we design an algorithm to get the optimal solution of the QIP. With the optimal channel assignment solution, the FMCA scheme is realized in the channel assignment phase of the proposed MAC protocol. Extensive simulation results show that the proposed FMCA scheme gets a good tradeoff between throughput and fairness compared with the existing SA schemes.