A Double-Auction-Based Mechanism to Stimulate Secondary Users for Cooperative Sensing in Cognitive Radio Networks

For the purpose of enhancing the accuracy of primary user (PU) transmission detection, using multiple secondary users (SUs) to sense cooperatively, instead of sensing individually, has recently been suggested. However, there are some issues: The first is whether the SUs are willing to sense the PU. It is better to give SUs incentives, or they may not behave cooperatively. The second is the system energy cost. It will be unnecessary and would be a waste of energy to let all the SUs sense together if some of them can work just as well. The third is the fusion time. If there are too many sensing SUs, it will take too long to collect the results and fuse them. To solve these issues, we propose the Secondary User stiMulating (SUM) mechanism. SUM has two parts: sensing auctions and report fusion. In sensing auctions, we introduce double auction into cooperative sensing, which is a pioneer work. We design a truthful winner determination and pricing mechanism. In report fusion, we propose a greedy algorithm to first satisfy the PU throughput demand and then maximize the SU throughput; this algorithm further stimulates the SU to use cooperative sensing. We prove that our auction mechanism is truthful and that the fusion problem is NP-hard. The simulation results show that SUM is effective in terms of saving the system energy cost, reducing the report fusion time, and maintaining the system throughput.