Dissertation > Excellent graduate degree dissertation topics show

Research on QoS Multicast Routing Algorithms Based on ACO in Ad Hoc Networks

Author: ZhangYanHui
Tutor: YuanPing
School: Northeastern University
Course: Communication and Information System
Keywords: ACO QoS multicast Ad Hoc network
CLC: TN929.5
Type: Master's thesis
Year: 2009
Downloads: 11
Quote: 0
Read: Download Dissertation

Abstract

With the development of wireless technology and the popularization of portable devices, the need for new mobile communication service has increasingly expanding. Conform to this trend, as a multi-hop, infrastructure-less and self-organized network, Ad Hoc network received increasing attention and is becoming one of the hot directions for researching. Since the distribution, trend nature, randomness and asynchrony are similar to ACO algorithm in Ad Hoc network, enlighten people apply the ACO algorithm to resolve the Ad Hoc network route problem thereby. For this purpose, this paper mainly studied ACO algorithm and use ACO algorithm to solve QoS multicast problem in Ad Hoc network.Multicast route algorithm satisfying the QoS constrained is a hot spot and difficult problem in the Ad Hoc network studies. Compared with tradition determinate routing, the distributed intelligence algorithm can adapt to the Ad Hoc net environment better. In this article, we made a new ant algorithm (QMCOP-ACO for short) that includes QoS parameters such as delay, bandwidth, delay-jitter and cost to solve the QoS multicast routing problem in Ad Hoc network. In QMCOP-ACO algorithm, the ants are treated as packets. The source node is treated as the nest and the destination node is treated as the food. Each node maintains a probabilistic routing table. The routing tables are also called pheromone tables. Packets according to pheromone tables to find next hop with a higher probability those links associated with higher pheromone values. After pheromone iterative many times, the ant will ultimately found the least costly route which meet the multiple QoS constraints. As the Ad Hoc network topology changes frequently, the important thing is to improve the convergence rate of algorithm in order to adapt to dynamic network environments in Ad Hoc networks. So this algorithm is aimed at to improve convergence speed ,which add certainly fit in with tactics to improve renewal speed, and use the award and punishment probabilistic mechanism to change converge speed, we also add history award to accelerate the algorithm to convergence. Simulate result indicate, this algorithm has quicker convergence speed and adapts to the topology t changes in Ad Hoc network, algorithm was testified correctness and validity.When the network size become large in Ad Hoc, the initial pheromone distribution is very small.Ants search for feasible path may become randomness, it will lead to slow convergence of algorithm. So we want to make use of the ability that node can gain the position information in Ad Hoc network to guide the ant looking for road process, we suggest a new algorithm owing to visual ant to solve big size QoS multicast problem in Ad Hoc network. Visual Ant using location information of nodes providing by GPS to calculate the deflection angle of next path, then use deflection angle get the path deflection value, then use deterministic routing formula and path deflection value to find a viable path. The simulation results show that the new algorithm improve the slow convergence problem in ACO.

Related Dissertations

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Wireless communications > Mobile Communications
© 2012 www.DissertationTopic.Net Mobile