An LMS Based Data Reduction Technique for Energy Conservation in Wireless Sensor Network ( WSN )

In preceding existence, Wireless Sensor Networks (WSNs) have gained an amplified attention from the research community and extended its boundaries in commercial, industrial and medical domains. The tailback in WSN based application development is the inadequate energy source. As sensor nodes are generally battery-powered devices, the critical aspect is to lessen the energy consumption of nodes, so that the network lifetime can be extended over a reasonable time span. The energy of a node is consumed by sensor, processor and radio interface, of which the radio interface is the principal consumer of nodal energy. Thus reduction of data to be transmitted can effectively lessen the energy consumption, bandwidth requirement and network congestions. Data reduction strategies aim at reducing the sum of data sent by each node by predicting the measured values both at the source and the sink node, thus only requiring nodes to send the readings that depart from the prediction. While effectively plunging power consumption, such techniques so far needed to rely on a prior knowledge to properly model the estimated values. In this paper, an adaptive filter based on Least Mean Square (LMS) algorithm is employed that requires no prior modeling, allowing nodes to work independently and without using global model parameters. In this method, the data loss and node failures are also taken in to deliberation and appropriate techniques are included to reduce the prediction error. This work also involves dynamic adaptation of step size during different modes of adaptive filter.