WSN Routing Algorithm Based On Link Quality Perception

Wireless sensor network(WSN)is a new technology to realize area monitoring by collecting,processing and aggregating data in specific area.Due to the limited energy and transmission power of sensor nodes,they are often affected by noise,multipath effect,signal interference and other factors in the process of wireless communication.Generally,The link quality in the network shows a strong uncertainty.Communication with low-quality links will lead to the failure of data transmission and frequent data retransmission,which will not only increase the network delay,but also affect the quality of data transmission,thus increasing the energy consumption of nodes and shortening the life cycle of network.Therefore,a link quality estimation model for the application scenario with uncertain link quality is put forward in this paper.Then,a link quality perception based WSN routing algorithm is put forward.The summary of what the paper mainly done is as below.(1)The basic theories of link quality estimation and WSN routing protocol are analysed.Position characteristics,Temporal characteristics and asymmetry of link quality are summarized.Some hardware parameters and software parameters based link quality estimation methods are introduced.The definition,problems,classification and performance index of WSN routing protocol are analyzed.Some classical routing protocols are introduced.(2)First of all,the link data set for storing link parameter samples is constructed,and the hardware devices and collection strategies for collecting samples are introduced.in the preprocessing stage of link data set,the data denoising model based on iForest algorithm is trained and the link data set is denoised,and then the data normalization is completed.Finally,The link quality estimation model based on GBDT regression algorithm is trained with preprocessed data set.The Jupyter Notebook of Anaconda platform is used to compare and analyze the algorithm performance.The experimental results show that the denoising effect of iForest is better than other models.The GBDT-iForest model trained with iForest denoised data has higher estimation accuracy than other models.(3)Based on the estimation model of link quality,the parameter named LIN(Link indicator between Nodes)is proposed to synthetically evaluate link quality and communication cost.By sensing the quality of all links in the whole network,an optimal clustering strategy based on PSO optimization algorithm and Communication indicators between nodes is proposed.Then,the power adaptive transmission method based on signal-to-noise ratio and the corresponding routing algorithm: LQP-PA are proposed for the data transmission stage.On this basis,LQP-PAMH,a routing algorithm combining multi-hop transmission and power adaptive transmission is proposed.Matlab 2014 is used as the simulation platform to simulate the WSN application scenario under the condition of complex link quality,the performance of routing algorithm of this paper ia compared with other similar routing algorithms.Experimental results show that LQP-PAMH is superior to other similar algorithms in terms of communication performance in clusters,communication performance of multi-hop path and overall network performance.And LQP-PAMH has stronger adaptability to the change of monitoring area.