Research On Some Optimazation Issues In Routing Algoirthms

The Internet has become the most important modern information society infrastructure and an important part of people's life and work. But Internet transmission model " best-effort " service is unable to meet all kinds of multimedia applications and the users network transmission quality requirements. Therefore, to improve the efficiency of network resource utilization, to provide a high-quality service for users are a hot topic of Internet research area. The study in recent years shows that the network routing algorithm plays a very key role to achieve the quality of service network. This paper mainly studies the quality of services unicast and multicast algorithms, in which the quite significant results are summarized below:1. In Chapterâ…¢,tabu search is introducted into multi-constrained unicast QoSR calculation. First several QoS metrics are transformed into a single energy by using an energy function. Then based on Dijkstra algorithm, the candidate sets and evaluation function are constructed by tabu search. The approximate optimal solution is find by the iterative method in tabu search. Simulation results show that the algorithm performance is stable with high success rate and low price.2. Chapterâ…£defines the delay constraint minimum cost multicast routing problem, and then presents multicast routing algorithms,respectively based on mixed strategy of genetic and tabu,and the ant algorithm. Simulation results show that algorithm is stability with a fast convergence and good price performance. It provides favorable solutions to multicast routing problems.3. In Chaperâ…¤,deals with the minimum cost multicast routing problems with the nodes CPU, the buffer and bandwidth constraints. It gives a unified model and discusses three multicast routing algorithms, respectively based on the simulated annealing,the genetic algorithm,the tabu search. Simulation results show that the algorithms convergence is fast and low price, and can be met the QoS requirements.4. In Chapterâ…¥,CMCMRA and SMCMRA algorithms are presented for the multi-constrained minimum cost steiner problem, respectively based on the CBT idea and SPH idea. The performance analysis indicates that these two algorithms are easy to implement and relatively low complexity. Finally, the simulation result show that algorithms have a low price performance,in which additional constrained QoS requirements can be met. 5. In Chapterâ…¦, a core selection algorithm is proposed for the core seletion problem with QoS requirements in the core based multicast routing protocol. It selects a minimum number cores form core candidate set, and make all group members satisfy end-to-end quality of service constraints. The simulation results show that the algorithm is feasible and effective in selecting core nodes and satisfying end-to-end QoS constraints.