Font Size: a A A

QoS Anycast Routing Problem Based On Unicast

Posted on:2007-08-13Degree:MasterType:Thesis
Country:ChinaCandidate:L L FengFull Text:PDF
GTID:2178360185987116Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the fast growth of the Internet in recent years, more and more people pay attention to the problem of QoS in IPv6.If we want to solve the QoS problem,we need the skill to manage unicast,multicast and anycast. At the present ,more attentions have been paid on unicast and multicat,less research have been launched on anycast.So there have lots of problem on anycast QoS need us to research.In this paper,we make a study in the field of anycast service and according to various QoS parameters define anycast QoS routing model.In order to solve different requirement,that is delay-Constraint and multi-QoS routing ,proposed by client ,we designed different GA in this thesis.Now ,more and more real-time service appeared ,Delay-Constraint and less cost is a hot topic in routing research .Aiming at this problem ,we use SAGA() proposed by Paul L.Stoffa to solve this problem.This algorithm can make up the shorting of GA.Except of Delay-Constraint ,there are lots of service need to think abort other conditions,such as bandwidth,Delay-Jitter etc.In this thesis ,we consider three QoS parameters together ,that is Delay,Delay-Jitter,Pack-Loss etc.We called these problem as multi-QoS routing Problem .In this algorithm ,we add adaptive GA and fitness scaling to solve this problem.Adaptive GA can simple the process of designing crossover and mutation's probability, fitness scaling can keep the various of population.In this paper ,we also develop the operation of initiating population. We use unicast instead of depth-first search .Because unicast can keep network-layer and application-layer together.Unicast also can develop initial population.In order to vertify the validity of our algorithms,We build up a network anycast routing simulator.Simulator results show that the algorithm has a better performance.
Keywords/Search Tags:anycast QoS routing, genetic algorithm, multiple QoS parameters constrained, Delay-Constrained
PDF Full Text Request
Related items