Research Of Multicast Routing Algorithm Based On Genetic Algorithm And Fuzzy Logic

Multicast, is a communicating process that the one sender transmit data to multiple receivers. By using multicast transmission, the data will occupy less bandwidth to meet the large-scale client receiving demand, thus the information can be transferred more effectively in the Internet. The key of multicast communication is to select the multicast route. That is how to build a multicast distributing tree which is used to meet the demand of users' QOS. Therefore, to find out the multicast route at less cost that meets the QOS requirements of applications, thus guarantee the normal efficient operation of multicast system, will be of very important significance.The existing and general multicast algorithms are such as flooding, DVMRP, MOSPF, CBT and so on. But flooding will consume too much resource; although DVMRP pruning has large efficiency, it needs a lot of place to save built-tree; the weakness of MOSPF is that it needs transmit link status information between routers periodically so MOSPF is thought to be not suitable for multicast; if CBT is used, for some source of the net, the result of this algorithm will be not ideal, especially when the nodes of the net are distributed not so symmetrically.Firstly, this paper elaborates the theoretical basis: genetic algorithms and fuzzy logic. The algorithm is designed for those nets which have high real-time demand. So it has adopted the heuristic idea based on genetic algorithm which will get fairly good result in finite time waste. Furthermore, the function of degree of adaptability built has used fuzzy logic control technical.In this paper, rationale of the algorithm has been expounded above all: genetic algorithm and fuzzy logic. Genetic algorithm is a type of random searching method which builds on evolutionary rule from biological kingdom. It is the key technical about Pls use full name. Fuzzy logic imitates human's thinking methods to denote and analyze dubious or imprecise information. Then, this paper presents an multicast algorithm which combines the advantages of these two theories, the algorithm presents a new method for solving multicast routing program. The algorithm has considered the measure of data transforming: delay, bandwidth and jitter and has chosen function of degree of adaptability by means of fuzzy logic. It has used simple fuzzy judging method to reduce computational complexity as to get excellent real-time efficiency. Finally, in this paper the simulation tool NS-2 has been used to simulate the algorithm. According to the simulation result, the performance of the algorithm has been evaluated. This paper also give some suggestion for the further improvement of the algorithm.