Research On Optimization Strategies Of Traffic Transmission Based In Complex Load-networks

With the rapid development of society,various social networks such as communication networks have greatly facilitated people's lives and are increasingly burdened,resulting in network congestion becoming an inevitable phenomenon.The emergence of network congestion affects the normal operation of society and even causes serious economic losses.Therefore,research on network traffic transmission optimization strategy is an important issue that we need to solve.Usually,network traffic transmission performance is affected by three factors: Network resources,routing strategy and network topology This paper summarizes and summarizes the previous researches on the optimization strategy of complex load-networks traffic transmission,and proposes corresponding traffic transmission optimization strategies from these three perspectives.The main research work and achievements of this paper are as follows:(1)A method of capacity allocation based on nodes and edges is proposed.In view of the fact that only the processing capacity limitation of nodes or edges is considered in the existing research,which is inconsistent with the reality.This paper proposes a processing capacity allocation strategy based on node betweenness and edge betweenness,and verifies the rationality of the strategy by simulation experiments.The processing capacity allocation method is closer to the reality.(2)A routing method based on dynamic queue length is proposed.Similarly,in the existing research,only the impact of the real-time packet queuing length of nodes or edges on the complex load-networks traffic transmission which is inconsistent with the reality.This paper proposes a dynamic routing strategy based on the length of nodes and edges queuing,and combined with the nodes betweenness playing a leading role.Theoretical analysis and simulation experiments verify the rationality of the proposed strategy and analyze the dynamic routing strategy closer to the reality.(3)An edge-adding method based on domain similarity is proposed.By comparing and analyzing the existing network edge-adding methods,this paper proposes a new edge-increasing method based on the domain similarity index,and proves the advantage of this edge-increasing mode in improving the performance of scale-free network traffic transmission in simulation experiments.(4)An edge-adding method based on the uniformity of network structure is proposed.Considering the diversity and difference of topological structure in large-scale network topology,the traditional method of edge-adding according to a certain rule has limitations.This paper is inspired by the more uniform distribution of nodes and the higher the network transmission capacity.Entropy theory evaluates the structural uniformity of the network,which is used as the standard edge-adding,and the advantages of this strategy are verified in simulation experiments.