Research On QoS Guarantee Technologies For The MANET

MANET, namely Moblie Ad Hoc Network, is type of self-organized multi-hop wireless network which the node can move independently and freely. At present, in many active fields, such as business, military, emergency and so on, MANET is more and more important and its main features, such as no center, self-organizing and independent of the fixed network infrastructure, strong survivability make it more and more widely used in many fields.In recent years, the rapid popularization of mobile devices, wireless communication technology keeps on developing, these are the conditions for the outbreak of multimedia services, QoS performance of the network is under great pressure. But the nodes in MANET are independent and are not constrained to move freely, the topology of MANET is instable, the link duration of MANET is short and the frequent route discovery process of MANET, all these factors affect the reliable and stable operation of the business. So as to get over this difficulty, in this research, the performance of the existing routing algorithms are analyzed and researched, and two classic algorithms: Ant Colony algorithm and Particle Swarm Optimization algorithm are selected to study. Then integrate the two algorithms, the fusion algorithm is applied to the MANET routing improvement, the main research contents are as follows:First of all, combined with the current popular intelligent algorithm research trends, in this research, ant colony algorithm is studied and then do the research on particle swarm optimization algorithm. Now, the main mean to solve nonlinear combinatorial optimization problem such as the QoS routing problem is use the heuristic intelligent optimization algorithm. Ant colony algorithm is widely used and has better application effect on this problem, but it also has inherent defects: slow search speed and premature convergence. In the meantime, the particle swarm optimization algorithm has the characteristics of faster search speed and easier to fly over the local optimum Local optimal solution. Therefore, this paper considers the integration of these two algorithms. Step 1: obtain the alternative path set through operating the ant colony algorithm; Step 2: select the best path from the alternative path set through operating the particle swarm optimization algorithm. Through these 2 steps, an ant colony-particle swarm fusion optimization algorithm is proposed. Simulating through MATLAB, so as to prove the fusion algorithm is effective, the convergence process and performance of the fusion algorithm and the other three in solving typical path problem traveling salesman problem are compared.Secondly, based on the fusion algorithm which are presented, the optimization and improvement of the routing discovery and maintenance phase of the existing MANET routing protocols are presented, the APQ-AODV as Ant Colony-Particle Swarm QoS Routing Protocol based on AODV, is proposed. Relative to AODV, APQ-AODV are improved in three aspects: packet delivery fraction, route discovery frequency, routing packets overhead. Make a comparison with the ant colony algorithm-based routing protocol in existence, APQ-AODV shorten route discovery time and is better suited to the dynamic network topology, at the same time, the chosen path has better performance. In this research, use the NS2 to simulate the performance of APQ-AODV, and select AODV as well as ARA to compare, to prove the APQ-AODV is effective. By controlling the node pause time and contract rate, set the corresponding environment respectively, then the 3 routing protocols are simulated, then compared the PDR, RPO, RDF and AVGD. Through the results we can know that as the APQ-AODV is not based on the shortest path, its AVGD performance is worse than AODV, expect this, in the other three areas have improved, and as the node pause time is suspended, the contract rate is increased, the advantages of APQ-AODV are obviously; relative to the ARA protocol, the APQ-AODV comprehensive performance has been enhanced. The results of these experiments tell that the proposed routing protocol has improved in several parameters.