QoS Routing Research Based On Improved Ant Colony Algorithm

The basic mission of QoS Routing is finding a route that meets the needs of user's QoS requirement. But most of the arithmetic is not equal to multi-constraints QoS routing problem. Ant algorithm is a new kind of meta-heuristic algorithms for solving combinatorial optimization problems or continuous function optimization problems. The characteristic of parallel, positive feedback and robust is strongly showed during ant colony's forage, Ant algorithm also has these characteristics, so the research of theory and utility of ant algorithm has great merit.The thesis analyzes the basic theory and model of ant algorithm, introduces the features of ant algorithm and its research state. The slow convergence and stagnation behavior are the main drawback of basic ant algorithm, so the thesis proposes two methods to overcome these shortcomings. First, improvement based on self-adapt antilogy: combines with local and global pheromone's adjustment to update route's pheromone dynamically, alter some correlative parameters of pheromone adaptively; Second, improvement base on aberrance antilogy: when the algorithm get into stagnancy, mutate the algorithm to make the optimization result break away from the local optimization area. The improved algorithm is effective and feasible for QoS routing problem.Finally, the thesis frames a network model in OPNET, do simulation and verify on the unimproved and improved algorithm, analyze and compare their different results. The results signify the improved algorithm has better convergence and is able to improve the quality of packet transmission effectively. Meanwhile, the improved algorithm still retains ant algorithm's expansibility.