Narrow Results

In wireless sensor networks, the loss of sensor data is inevitable due to some uncertainty factors such as limited node resources, network link instability and so on, which will affect the data quality to some extent. In this paper, the linear interpolation model based on temporal correlation was established to solve this problem, and in order to improve the accuracy of missing data estimation, the attribute correlation of sensor data was considered to establish the regression model. Experiment results based on real dataset show that the proposed algorithm gains lower error rate and improve the recovery accuracy of sensor data effectively.