Automatic ARIMA Time Series Modeling for Data Aggregation in Wireless Sensor Networks

One of the most natural and primary ways of data collection in wireless sensor networks is to periodically report sensed data values from sensor node to aggregator. However, this kind of data collection mechanism comes at the cost of power consumption and packet collision. In this paper, we developed an automatic ARIMA (Auto Regressive Integrated Moving Average) modeling based data aggregation scheme. The main idea behind this scheme is to decrease the number of transmitted data values between sensor nodes and aggregator by using time series prediction model. The proposed scheme can effectively save the precious battery energy of wireless sensor node while keeping the predicted data values of aggregator within application defined error threshold. We show through experiments with real data that the predicted values of our proposed scheme fit the real sensed values very well and fewer messages are transmitted between sensor node and aggregator.