Research On Working Strategy Of Wireless Sensor Network Based On Information Freshness

Wireless Sensor Networks(WSN)is a network composed of large number of sensor nodes and a base station.It mainly collects relevant information about the scene where the nodes are located.It is one of the indispensable ways for modern communication networks to obtain perception information.In recent years,the application range of WSN has become wider,but the energy carried by sensor nodes is limited and cannot be replenished.The limited battery capacity of sensor node has become a constraint condition for the development of WSN.With the development of energy harvesting networks,more attention is paid to introducing renewable energy into WSN.However,due to the instability of renewable energy supply,energy management research in WSNs faces enormous challenges.With the development of 5G technology,more and more real-time application-oriented services such as the Internet of Vehicles are flourishing development.How to realize realtime perception of the physical environment and monitor its status to complete intelligent decision-making have become an urgent problem in WSN research.The Age of Information(Ao I)as an important indicator to measure the freshness of information is defined as the time interval between the generation time of the latest data packet received by the receiver and the current time.It becomes a metric for evaluating and judging the timeliness of data packets in the network.In this paper,we study the timely acquisition of sensing information in WSN.The research work is carried out on energy-collectable WSNs from the acquisition of sensing information and the transmission of data packets.The specific research work of this paper is as follows:First of all,how to achieve accurate data packet collection in WSN becomes the primary issue as the randomization of energy collection of sensor nodes and the time-varying characteristics of wireless communication channels.The average Ao I optimization equation is established on the basis of analyzing the relationship between the sampling strategy of data packets,the energy collection strategy of nodes,and the allocation of transmit power.In order to solve the optimization problem,Lyapunov optimization theory is introduced into this paper.This optimization theory transforms the optimization problem into a drift plus penalty minimization problem through virtual queue technology.And using the global search algorithm to get the optimal sampling strategy.Simulations analyze the effectiveness of this strategy.Secondly,in order to minimize the average Ao I in the network,the management strategy of data packets that collected by sensor nodes in the server is studied to ensure the timely delivery of perception information.Assuming that the arrival of data packets of sensor nodes conforms to Poisson process and the server service time obeys exponential distribution.The average Ao I expression of the network is deduced by using Stochastic Hybrid Systems(SHS)technology and queuing theory.According to the difference of server processing rate,the network is divided into homogeneous network and heterogeneous network.And different source-aware packet management strategies are proposed in different networks.Finally,through simulation research,the proposed strategy is compared with Last Come First Serve(LCFS)and LCFS with preemption-in-service(LCFS-S)strategies,and the simulation results verify the effectiveness of this strategy.