Research On Energy Harvesting WSN Node Deployment Strategy For Railway Environment Monitoring

With the rapid development of high-speed railway,China has the fastest and longest high-speed railway network in the world,which brings great convenience to passengers'travel.Meanwhile,the safety of train operation has become an increasingly concerned issue.The key to train safety operation is real-time monitoring of train operating environment.At present,Chinese railway environmental monitoring still uses high-complexity wired network for monitoring,because the railway environment is mostly in remote areas,the deployment is cumbersome and the maintenance cost is high.Therefore,wireless sensor network(WSN)is introduced in the railway environment monitoring.Compared with other monitoring systems,WSN network is characterized by dynamic self-organization,fast data processing speed,high stability,low deployment cost,strong compatibility and large-scale deployment.However,due to the limited energy of WSN nodes,the phenomenon of"energy hole"is especially serious in linear network,which reduces the network performance of WSN.In this dissertation,WSN node deployment in railway monitoring environment is studied.With the goal of solving the energy consumption imbalance at the cluster head,a banded WSN non-uniform optimal cluster deployment strategy suitable for railway is proposed to monitor the train operation along the railway.The dissertation mainly has the following research results:First,the existing linear WSN node deployment strategies are analyzed.To solve the problem of"energy hole"caused by unbalanced energy consumption at the cluster head of linear WSN network based on clustering.In this dissertation,a banded WSN non-uniform optimal clustering algorithm is proposed.On the basis of non-uniform clustering of banded network according to geometric sequence,the energy consumption of multi-hop cluster heads was analyzed,and the deployment density function of sensor nodes in each region was obtained,in order to balance the energy consumption of cluster head.At the same time,the concept of cluster head weight is introduced,which makes it more likely that the node with more residual energy and close to sink will become the cluster head.The corresponding routing protocol is designed,and the sleep mechanism and the cluster head rotation mechanism are introduced to effectively prolong the network lifetime.Firstly,the optimal cluster number of non-uniform optimal clustering algorithm proposed in this dissertation was obtained by MATLAB software simulation under different deployment lengths.Secondly,the performance of this optimization algorithm is compared with that of non-uniform clustering and uniform clustering algorithms in aspects of network lifetime,energy utilization,residual energy ratio and network energy consumption.The simulation results show that the proposed non-uniform optimal clustering algorithm can effectively balance the energy consumption of cluster head,increase the energy utilization rate,and improve the network life cycle by 172and 441 rounds respectively compared with non-uniform clustering and uniform clustering.Second,according to the railway application scenario,most of the rails are in remote places,and the maintenance work such as battery replacement is heavy.For this,using solar energy collection of sensor nodes,with energy harvesting wireless sensor network(EH-WSN).For solar energy's characteristics of randomness and instability,an improved weather conditioned moving average(WCMA)energy prediction algorithm is proposed,the node energy model with energy harvesting(EH)and prediction is constructed.In terms of the weight of cluster head,considering the energy collected by the node,it is more likely that the node with more residual energy,which is close to sink and which collects more energy,will become the cluster head.The performances of the average packet delivery ratio,throughput density,throughput fairness are simulated in EH-WSN and WSN,as well as the transmission power P_i,node deployment density and the signal-to-interference noise ratio(SINR)threshold?.The results show that the average packet delivery rate,throughput density and throughput fairness of EH-WSN are improved effectively,compared with WSN powered by battery.The proposed strategy has good scalability,suitable for the non-uniform deployment in banded EH-WSN,especially with a large number of nodes and a long distance.