Data Collection Optimization Method for Wireless Sensor Networks Based on Linear Regression

2019 2nd International Conference on Advanced Materials, Intelligent Manufacturing and Automation

Zhang, Meng^1 ; Zhang, Xiaomei^1 ; Huang, Yinghui^1

School of Information Science and Technology, Nantong University, Nantong, Jiangsu

226019, China^1

关键词: Communication overheads;    Linear regression models;    Low computational complexity;    Network energy consumption;    Optimization strategy;    Predict and estimate;    Spatial-temporal correlation;    Structural characteristics;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/569/3/032064/pdf
DOI  :  10.1088/1757-899X/569/3/032064

来源: IOP

【 摘 要 】

In the event monitoring applications of wireless sensor networks, the sensing data collected by sensor nodes near the same monitoring area has a great spatial-temporal correlation. In order to reduce the amount of data transmission in the network and the energy consumption of communication among nodes, an energy-efficient distributed data collection optimization strategy based on linear regression for wireless sensor networks is proposed. linear regression model of local sensing data is constructed to represent and predict the actual sensing data monitoring values of sensor nodes. Within the allowable range of errors, the node does not need to transmit the actual monitoring sensing data to the sink node, but only transmits the parameter information of the regression model basis function. Without losing the basic structural characteristics of data, the communication overhead caused by frequent data transmission between sensor nodes is effectively reduced, and the linear regression model of sensing data adopts the incremental update method with low computational complexity. he simulation results show that the data collection optimization strategy based on linear regression can effectively predict and estimate perceptual data with less network energy consumption, and achieve the goal of reducing network energy consumption.