TAMSA: Two-Stage Auction Mechanism for Spectrum Allocation in Cooperative Cognitive Radio Networks

Cooperative cognitive radio networks have been proposed to address spectrum starvation problem and enhance the transmission rate of mobile devices. Most works assume one user could afford the whole spectrum and neglect the selfishness nature, which is not practical. Based on group-buying, a two-stage auction mechanism named TAMSA is proposed to guarantee the quality of service and improve the utilization ratio of spectrum resources. TAMSA is an incentive mechanism involving the primary users (PUs) and relay nodes. TAMSA can also reduce the cost of the secondary users (SUs) and increase utilities for both PUs and relay nodes. In the first stage, SUs submit their budgets, valuations and demands for spectrum resources to relay nodes in group-buying, relay nodes calculate revenues and determine the winning SUs. In the second stage, we execute VCG auction between the relay nodes and PUs, with a maximum-weighted-matching algorithm. TAMSA can effectively allocate spectrum resources to meet the demands of SUs. We show that TAMSA is truthful, individual rational and computational efficient. Extensive simulation results show that TAMSA outperforms random algorithm by 256% in terms of average utility of PUs. TAMSA is able to improve the average utility of SUs and relay nodes significantly up to 213% and 10 times respectively. TAMSA is further improved by 28.33% and 78.65% in terms of average utility of PUs over TASG and TACC, respectively.