Mobile IP Routing Based On Genetic Algorithm And Performance Evaluation

Two technological advances in recent years have radically altered the nature of computing for most computer users. The first is mobility. Laptop computers now represent the fastest growing segment of the computer market. Most observers expect that laptop computers, palmtop computers, networked personal, digital assistants, and other such mobile computers will eventually represent the majority of stations connected to Internet. The advantage of mobile computing is that users may access all their applications for any location, whether they are in another building or a different state. The second advance is the widespread use of the Internet for communication, file transfer, and World Wide Web connectivity. This thesis describes how to resolve the routing problem in mobile IP using genetic algorithms including optimized routing and routing via a designated node such as triangle routing and reverse tunneling routing and how to evaluate the performance of mobile IP based on wireless LAN.This paper focuses on the following research work:1.proposes a genetic algorithm based on matrix encoding for next hop route with load and time delay;2.proposes a genetic algorithm for shortest route via designated node, solves the problem which Dijkastra algorithm need two times or more to compute;3.proposes a scheme of dividing populations and dividing time in order to improve the success ratio;4.proposes a network model for directional motion, performs evaluation for mobile IP based on WLAN, and concludes some useful results.Chapter 2 details the base mobile IP principles which are similar to specification. Some terms on mobile IP are defined in this chapter. Genetic algorithms are used to solve the optimized problems. We apply genetic algorithms in routing problems of mobile IP. Routing problem such as triangle routing and reverse tunneling routing are analyzed and genetic algorithms are connected with routing problems. Chapter 2 also deals with evaluating the performance of mobile IP based on wireless LAN in network simulator 2(NS-2).Chapter 3 deals with optimized routing problem with load balance and time delay. Genetic algorithm was a basic algorithm for optimum problem. The routing problemwith load balance was described and a mode was made for it. Genetic algorithm with matrix encoding, different mutate operator, crossover operator and selection operator based on order rank selection algorithm was applied to solve the problem. Genetic algorithm was approved to be convergence. Computation complexity was calculated. Simulation results show that convergence of genetic algorithm based on matrix is good. The relation between cost and population size or crossover probability is analyzed. The genetic algorithm with load balance can solve routing problem and can be applied in mobile IP and next hop route.Chapter 4 collaborates on routing problem via a designated node. Due to the excellent performance in Nondeterministic Polynomial problem, such as TSP, we choose the genetic algorithm to implement triangle routing in Mobile IP protocol. In order to get the best routing from the correspondent node to the mobile node, which passes the mobile node's home agent, the proposed method has been improved by a series of experiments. Thus, the scheme is to be proved useful steeply in this essay. It is in accord with the view of Chang Wook Ahn.Chapter 5 performs a simulative evaluation of mobile IP via ns-2 for a scenario comprising of one home agent and two foreign agents based on IEEE 802.11 wireless LAN standard. We describe a model which is adaptive to directional motion such as high speed vehicle. We represent the bandwidth and loss packet under constant bit rate(CBR) traffic. We also analyze the relation between loss packet and time delay/speed. We conclude that handoff needs good control of speed and time delay.