Research On Cognitive Routing Protocol In Wireless Mesh Network

Wireless mesh network is a kind of access network with high bandwidth,self-orgnization and roubustness,applying to metropolitan area network,emergency communication and other fields.As a practical network,wireless mesh network has to provide routing protocols with high-throughput and high-reliability to meet various demands of users.Therefore,the study of efficient routing protocols is one of core problems of wireless mesh network.Because of a shortage of adaptive ability,static wireless mesh network perhaps causes a phenomenon that packet loss rate is increasing when it faces complex network environment.In order to solve the above problem,the cognitive process of cognitive network is introduced in wireless mesh network,which can adjust optimal transmission path according to change in the network environment,to provide the best end-to-end performance for users.This paper focuses on the research of routing protocol under the above network environment.It analyzes the classic network coding awareness routing protocol,and adopts routing optimization algorithm with cognitive function,which is used to optimize network coding awareness routing for improving the network performance of wireless mesh network.Due to the demand of low delay and high network throughput,the distributed coding-aware routing algorithm based on genetic algorithm(GA_DCAR)is proposed in wireless mesh network.The algorithm realizes cognitive function by genetic algorithm,designs the rule of detection and update of network coding opportunities according to the characteristics of coding-aware routing and genetic algorithm,and improves the ability searching routing by genetic operation,which reduces transmission delay and improves network throughput.For the problem of link interference in wireless mesh network,a routing algorithm based on ant colony optimization is proposed——ACAR(Coding-aware Routing Algorithm Based on Ant Colony Algorithm).The algorithm adopts cross-layer design and ant colony algorithm to optimize GA_DCAR.It selects routing through probability,and improves the ability searching routing by adjusting the cost of link.As routing request packets finding paths have the ability to be aware of local network status,it can predict link interference ratio and achieve the goal of reducing interference and improving network throughput.Finally,the advantages of GA_DCAR and ACAR are verified on the NS2 simulation platform by comparing with DCAR protocol.