Study On Spectrum Decision Technology In Cognitive Wireless Mesh Networks

Cognitive Wireless Mesh Networks (CogWMN) which can meet the requirements of ubiquitous wireless communications have been a crucial development trend of future wireless networks. The introduction of cognitive radio (CR) technology relieves the spectrum woes faced by global wireless communications. This is one reason; another is that the mesh structure provides an effective way for the integration of heterogeneous networks, offering users a secure and flexible access mechanism and better services through the complementary characteristics among different networks. Thanks to the cognitive ability, mesh users can find spectrum opportunities through spectrum sensing and access them intelligently. The preciousness and scarcity of wireless spectrum resources requests cognitive users to efficiently employ these spectrum opportunities. Spectrum decision is a key process to realize efficient spectrum management and usage. A good spectrum decision scheme requires not only the consideration about the traffic statistics of both licensed users and cognitive users, but also some analysis for spectrum characteristic parameters, spectrum handoff times and network-layer routing, to achieve optimal spectrum utilization.Based on the extensive research of spectrum decision technology in CogWMN, some problems are pointed out and the corresponding improved approaches are proposed in this paper. Firstly, considering the overall transmission delay of cognitive users and the utilization of available channels, a probability-based greedy spectrum decision algorithm is proposed, in which the probabilistic method ensures the proportionality of channel utilization through pre-assigning a probability vector for all the available channels and the greedy algorithm helps cognitive users select its operating channel that brings minimum transmission delay. The simulation results show that the algorithm performs well in decreasing the transmission delay of cognitive users, and simultaneously can alleviate channel congest caused by the competition among users, making a balance in channel utilization. Secondly, because of the preemptive priority of licensed users multiple spectrum handoffs may occur during the transmission process of cognitive users, incurring the increase in actual transmission time. Therefore a joint design between spectrum handoff based on interference temperature model and impartial spectrum decision is proposed. The simulation results show that the scheme can enhance the spectrum utilization, lower the outage probability of cognitive users and improve the network performance. Either of the aforementioned spectrum decision schemes is presented for the per-hop link. In light of the multi-hop transmission characteristic of CogWMN, finally, on the basis of the combination of spectrum overlay and spectrum underlay, a cross-layer design between spectrum decision and routing is proposed with the objective of the end-to-end transmission performance of cognitive users. The simulation results show that the cross-layer strategy can increase the spectrum utilization and decrease the end-to-end cumulative delay of cognitive users, revealing significant advantages on the end-to-end transmission performance of CogWMN.