Research On SDN Architecture And Performance Of Shared Bicycle Networking Based On Topology Awareness

Intelligent transportation system has become particularly important with the introduction of the concept of "smart city".The shared bicycle system,as a part of the intelligent transportation system,is developing extremely rapidly.However,the topology of the current shared bicycle network is simple and the data transmitted is single,and there is no unified solution for sensing multiple sources of big data.For this reason,this thesis focuses on the network layer of bike sharing systems based on the following main work.First of all,a network structure suitable for shared bicycle is proposed and studied in this thesis.Then the bike sharing system is studied and analyzed in different application scenarios to explore its potential value.Therefore,this thesis puts forward the future application scenarios of shared bicycle under different circumstances.Secondly,this thesis proposes a multi-hop routing algorithm based on genetic algorithm for clustering.Aiming at the characteristics of shared bicycle mobility,the algorithm adds random speed factor in cluster head election and uses genetic algorithm for optimal path selection during data forwarding.Comparing the genetic algorithm-based split-cluster routing algorithm with the classical LEACH algorithm and CECA algorithm with under the same conditions,it is found that the algorithm proposed in this paper can slow down the node death process,thus improving the network performance.Finally,this thesis proposes a Software Defined Network(SDN)routing algorithm based on genetic algorithm,which routing algorithm based on genetic algorithm takes the real-time bandwidth and delay as the design factors of the fitness function.The routing algorithm can realize the load balancing of network resources and reduce the time of terminal communication.The optimal path algorithm based on hops,the optimal path algorithm based on delay and the optimal path algorithm based on bandwidth are simulated by Mininet software and Ryu controller.The experimental results show that the genetic algorithm based routing algorithm has lower latency compared to the other three algorithms when implementing end-to-end data transmission in large scale networks.