A sensor positioning algorithm for the EMMON large scale Wireless Sensor Network architecture

Abstract

In this paper a sensor positioning scheme is proposed for the EMMON architecture of Wireless Sensor Networks (WSN). This is a recently developed new multi - tiered hierarchical structure imposing the need for a distributed, low - complexity positioning algorithm. In this work we developed an integrated positioning scheme which involves an on-line calibration phase and a subsequent assessment of the parameters of a fuzzy inference system used to generate distance approximations based on RSSI values. In the calibration phase a sensor node is placed arbitrarily in a number of known positions and collects RSSI measurements from all anchor nodes. A data set consisting of RSSI values vs. distance is created and is used as training data set of the fuzzy inference system. Finally experimental results are presented which show the improvement achieved in the accuracy of the proposed scheme against other range-based schemes.