Research On MAC Protocol For Service Differentiation In Wireless Sensor Networks

Wireless sensor networks(WSNs)is listed as the first of ten emerging technologies to change the world in the 21 st century,and is widely used in all walks of life.With the continuous development of modern intelligent devices and the continuous change of application requirements,the number of nodes and sensors in WSNs is growing,while the types and number of services are also increasing.However,different services have different quality of service requirements.In order to meet the transmission needs of various services in resource constrained WSNs,providing quality of service and service differentiation will become an inevitable trend.In this paper,a statistical priority based multiple access(SPMA)protocol is introduced into the MAC layer of WSNs.The three key parameters of SPMA protocol are studied and optimized: the sending threshold of each priority service,backoff time and channel load.In this paper,the professional network simulation tool OPNET is used to build the simulation platform,and the improved protocol is simulated to verify its advantages.Firstly,in order to solve the problem of low flexibility of fixed threshold,which can not cope with network changes and waste channel resources,this paper proposes a dynamic threshold algorithm based on channel load statistics,which dynamically adjusts the transmission threshold of each priority service according to the service transmission and channel occupation.Simulation results show that the proposed algorithm increases the success probability of low priority services,enhances the utilization of channel resources and improves the system throughput on the basis of guaranteeing the quality of service of high priority services.Secondly,aiming at the problem that the traditional backoff algorithm does not support service differentiation and can not cope with network changes,this paper proposes an adaptive backoff algorithm based on service differentiation.Each priority service dynamically adjusts the backoff time according to the channel occupation status,so as to reduce unnecessary backoff under light load,and quickly reduce the channel access under heavy load,so as to provide higher priority services with better performance.Simulation results show that the proposed backoff algorithm provides obvious service differentiation,effectively reduces the packet transmission delay,and improves the quality of service of high priority services under heavy load.Finally,aiming at the problem that the existing channel load prediction algorithms can not accurately describe the channel occupancy,this paper combines the differential moving autoregressive average model and Kalman filter theory to predict the channel load and effectively control the channel access.The simulation results show that the multi priority MAC protocol based on channel load prediction achieves service differentiation,improves the success probability of high priority services,reduces the transmission delay,and obtains better services.