Predictive channel access scheme for Wireless Sensor Networks using Received Signal Strength statistics

Abstract

Number of indoor Wireless Sensor Network (WSN) applications are increasing day by day. However, there are many interferers (like 802.11× and 802.15.4 devices) effecting 2.4 GHz band. Moreover, co-existence is a bigger challenge for WSNs because of their tiny structure and resource constraints. In this paper, a novel method is proposed to predict the near future channel quality using the statistical channel noise history. By using the prediction method, WSN node channel access can be scheduled when the channel is free in order to increase performance. In this paper, first, noisy channels are analyzed to find a prediction metric. Then, with the help of the prediction algorithm developed, channel access is performed. The results obtained from the test bed show that our approach incurs up to 60% to 90% reduction in noisy channel access which satisfies less packet loss and less interference with non-WSN communications.