M/G/1 QUEUE-BASED REDUCTION OF POWER CONSUMPTION AND LATENCY IN WIRELESS SENSOR NETWORKS

Abstract

Longevity of wireless sensor networks (WSN) is dependent on the optimal utilization of power supply. To make optimal utilization of power and increase the operational life of the network, we present two techniques with the goal of decreasing the consumption of power in a sensor node by incorporating queuing theory. We analyze the performance of wireless sensor networks that implement a M/G/1 queue with two different queuing policies. The analysis is done with respect to two important aspects: power consumption and latency delay. The results of the analysis illustrate the fact that the power consumed at a wireless sensor node can be reduced significantly by optimal selection of thresholds. We also compare the two policies in terms of power consumption and latency and find that the Min (N, T) policy is better equipped to not only reduce the power consumption but also reduces the latency delay caused due to the introduction of the queuing thresholds. The results indicate that the schemes studied can be implemented in practical scenarios as they are effective in reducing power consumption and increasing the operational life of a WSN.