Energy Conservation Approach for Precision-Insensitive Wireless Sensor Applications

Many wireless sensor applications operate on a data set that requires a relatively high degree of precision. Small changes in the data collected from the working field by sensor nodes result in different outcomes as perceived by the data-sensitive application. A large amount of energy is consumed to maintain such sensitivity when all data detected by the sensor nodes are always transmitted to the receiver. Conversely, some applications do not need such high degree of precision. In this paper, we propose two approaches for determining the working schedule of each sensor node for precision-insensitive wireless sensor applications. In the first approach, Moving Average Base Station (MABS) model, the base station computes a decision model to be used by each sensor node in deciding whether or not data should be transmitted back to the base station. Data received by the base station is used to compute a statistical model which determines the acceptance range of the sensed data. These bounds, which are specific to each node, are transmitted to the respective nodes as a model for their decision. An alternative approach, termed moving average sensor mode (MASN) model, works in a similar fashion, but with the sensors capable of establishing the decision model on their own. The outcome is the ability to reduce energy consumption and, hence, extend the overall system lifetime