Polynomial regression for data gathering in environmental monitoring applications

Abstract

How to prolong the lifetime of wireless sensor networks is one of the most important design issues. In order to tackle this issue, we propose an energy-efficient polynomial regression-based data gathering algorithm in environmental monitoring applications. Each sensor node in the network fits a regression function with its sensed data in most recent rounds, and sends coefficients of the regression function and some related parameters to the sink node instead of sending the sensed data. Theoretical analysis and simulation studies show that the proposed algorithm can greatly reduce data transmissions among the sensor nodes, with significant energy savings on the sensor nodes and thus extending lifetime of the entire network.