Abstract
Cognitive Radio (CR) provides a promising means to the more efficient use of available spectrum. Routing in multi-hop wireless networks remains challenging and introduction of CR technology has created additional demands on routing within Cognitive Radio Mobile Ad-hoc Networks (CR-MANETs). To address these challenges, spectrum-aware routing protocols aiming at dynamic utilization of the so-called spectrum opportunities have been developed recently to improve end-to-end performance of the network for example in terms of Delay, Packet Loss and Throughput. One of the bottlenecks in the performance of ad hoc networks has been the lack of a load balancing mechanism. With the addition of potential routing opportunities introduced by CR technology, a load balanced routing protocol which can utilize SOPs into the load balancing mechanism is a missing puzzle in the problem of routing in CR-technology. Quantum game theory provides a framework to utilize entangled particles with the aim of affecting decision-making process of distant players. Hence, this theory has the potential to be used as a framework to target the load balancing problem in ad hoc networks. First, a novel spectrum-aware routing protocol based on OLSR as the basis of implementation is proposed in this research. The proposed algorithm utilizes ETX as the link quality estimation metric and provides the best weigh end-to-end paths based on generalization of Dijkstra’s algorithm to multigraphs. The results demonstrate that the proposed algorithm outperforms the existing baseline OLSR routing algorithm. Due to the instability in the end-to-end delay performance of the proposed algorithm, backpressure algorithm is identified as a potential solution to stabilize queues in the network and target the shortcoming of the proposed algorithm. Hence, a novel spectrum-aware routing algorithm based on backpressure load balancing mechanism is proposed and compared against the baseline OLSR and the proposed spectrum-aware OLSR algorithm. The OLSR backpressure spectrum-aware (OLSR-BSA) routing algorithm not only optimizes route computation based on the predefined cost metric but also incorporates the queue gradients of backpressure algorithm to perform load balancing. The results proof that the backpressure algorithm can efficiently utilize the SOPs in the load balancing optimization problem and results a performance and stability gain both in terms of end-to-end delay and packet delivery ratio. The instability resulted by inaccuracy of queue information in the proposed OLSR-BSA algorithm motivated our research to explore the problem of load balancing from a completely new perspective of quantum game theory. We have formulated the problem of load balancing in ad hoc networks using quantum game theory and proposed a novel routing algorithm so called Quantum Load Balanced OLSR (QLB-OLSR). The simulation results demonstrate a significant load balancing stability gain against the baseline OLSR routing algorithm. Key words: Spectrum, Routing, Cognitive Radio, Spectrum-aware routing, Backpressure Algorithm, Quantum Game Theory, Quantum Load balancing Email: s.shariat@surrey.ac.uk