Performance Analysis Of Network Coding Aware Opportunistic Routing In Cognitive Radio Networks

In recent years, more and more users start to use mobile phones to access theInternet. And this brings people many resources and much convenience. However,the spectrum resources in wireless networks are limited. Compared with so manyusers, spectrum resources are relatively scarce. According to the statistics, mostfrequency bands are used by few users, and this make some frequency bands stayidle for some time. Most of the spectrum resources are allocated to fixed users, andonly authorized users can access these spectrums. However, these authorized usersdo not fully make use of their spectrum resources. In contrast, non-authorized userscan only grab the general frequencies opportunistically. To solve this problem andguarantee the quality of user experiences, the cognitive radio networks (CRNs)appear.This paper focuses on the network coding aware opportunistic routing incognitive radio networks. We proposes a new network coding-aware opportunisticrouting protocol by applying opportunistic routing technology into CRNs.Opportunistic routing can improve the performance of wireless multi-hop networksby using the broadcasting nature of the wireless channel. Opportunity routing usesmultiple paths to reduce the system packet loss rate, but this is also a challenge. Itneeds to control the number of duplicate packets in the network. MORE is asuccessful opportunistic routing protocol which is applied into the traditionalwireless networks. However, since the dynamic environment and the uncertainty ofthe spectrum in cognitive radio networks, MORE cannot be directly used in thecognitive radio networks. Thus, this thesis presents an algorithm of coding factor.This thesis also proposes a new method for link delay estimation and a channelselection algorithm on the link layer based on the natures of cognitive radionetworks. Channel decision is an important part in the CRNs. In this thesis, weutilize linear programming to calculate the best decision of the network channelstatic time point through several iterations.Finally, this thesis uses NS2as a platform and writes the link layer module.Then we simulate the new routing protocol, NCOR. The simulation indicates that our new routing protocol could improve the performance of CRNs. With differentpacket loss rate, number of channels, probability of PU appearance, transmissionrate of source and the length of queue, NCOR can achieve a higher throughput andimprove the performance of the network.