ACO-based routing algorithms for wireless mesh networks

The popularity of Wireless Mesh Networks (WMNs) is growing exponentially in recent years, due to their flexible deployment and compatible communication features. As a key technology for next-generation wireless networking, WMNs promise an attractive future to both academic and industrial world. However, current WMNs are short in optimal routing protocols. Instead, many WMNs use the routing algorithms from ad hoc networks, which have different network features. Thus, routing becomes the most urgent issue that needs to be solved. In this thesis, routing problems in WMNs are discussed in different aspects, and then several proposed solutions in state-of-the-art are introduced with their advantages and disadvantages. Ant-In-Mesh routing protocol and the enhanced version are proposed for WMNs, inspired by traditional Ant Colony Optimization (ACO) algorithm, to deal with new challenging characters of WMNs. Periodical Mesh update is performed between neighbors, to keep the network alive. With these updated information at all the hosts, various Ants can collect the fresh routing data while they are launched for different purposes, also, the per-hop and end-to-end routing metrics can be calculated. Upon new connection requests, route discovery is carried out. After the routes are set up, proactive route maintenance is performed on each route. Several popular routing protocols and our algorithms are simulated. and compared using Qualnet. The simulation results show that our algorithms outperform the others, in terms of packet delivery ratio and end-to-end delay, as the mobility and network size increase.