Data Aggregation Algorithm based on Autoregressive Model in Wireless Sensor Networks
Hanxiao Zhi, He Xu, Peng Li · 2020
In wireless sensor networks, sensors are used to monitor environmental information. However, in the process of monitoring, sensors are affected by system noise and environmental noise. There has a deviation between the measured value and the true value. In order to reduce the deviation and obtain more accurate observation result, a data aggregation method based on autoregressive model is proposed in this paper. Firstly uses autoregressive model to fit the historical data in a period of time to get the predicted value at the next moment. According to the orthogonal principle, the measurement variance is calculated by the predicted value and the measured value, and the optimal weighting coefficient is obtained by Lagrange multiplier method in the sense of minimum variance, and then gets the result of aggregation. The algorithm does not need any prior information. Simulation result shows that the overall error and the error fluctuation amplitude of the algorithm are small, the aggregation results are stable and accurate.