Study On Performance Analysis Of Stochastic Network Calculus And Network Conding Based Wireless Sensor Network

Recently, as e-commerce and The Internet of things(IoT) develop rapidly, wireless sensor network (WSN) has become an important branch of the IoT and its application and performance optimization have also become a focus of scientific research. WSN applications get higher requirement on the performance of the network system. Finding effective methods to improve performances of wireless sensor network and using an effective evaluation tool for the analysis of network system are with important practical significance. In this paper, we focus on the performance analysises of hierarchical WSN and distributed WSN base on the stochastic network calculus theory. With the network coding theory, the supporting QoS Differented Services(DiffServ) wireless sensor network model and the distributed wireless sensor network model were improved respectively. Base on theory of stochastic network calculus, the delay, data backlog and throughput performances of these two improved models were analyzed and compared. Specific work is as follows:In this paper, in view of the stringent quality of service (QoS) guarantees for real-time network applications of a cluster tree wireless sensor network and base on a DiffServ WSN node model which supports QoS service, a new DiffServ WSN node model with network coding was proposed. In the new model, coding features were added to the node basing on the theory of Network Coding. We then analyzed the performance upper boundries of delay and data backlog of the new model by the tools of stochastic network calculus theory. The numerical analysis shows that with Network Coding we can effectively improve the delay, data backlog, etc. of an DiffServ wireless sensor network. Besides, the new DiffServ model with Network Coding can privide guidances for the evaluation of a Network Coding strategy and its application in an Real-time WSN.In this paper, base on the distributed wireless sensor network model, the characteristics of Network Coding were applied to the network node and a new distributed WSN model with encoding function was proposed. we analyzed the stochastic arrival curve and the stochastic service curve of the new model. We also used the theory of stochastic network calculus to analyze the upper bounds of the end-to-end performances of the coding model, such as the end-to-end delay and throughtput. Finally, by numerical analysis we obtained the upper bounds of the end-to-end delay and the throughput of the coding model and compared them to the performances of traditional distributed wireless sensor network. The comparison result shows that the Network Coding way can effectively improve the end-to-end performance of the distributed wireless sensor network model. The analysis results can provide theoretical guidance for the application and the evaluation of a new Network Coding strategy in a Distributed WSN, which is with a vital significance.