Abstract
Conference Title: 2015 36th IEEE Sarnoff Symposium Conference Start Date: 2015, Sept. 20 Conference End Date: 2015, Sept. 22 Conference Location: Newark, NJ, USA Power consumption minimization in Wireless Sensor Networks (WSNs) has been discussed extensively in literature. Usually, a central node (a receiver) in WSNs consumes large amount of power due to the necessity to decode every received packet regardless of the fact that the transmission may suffer from packets collision. Current collision detection mechanisms in WSNs have largely been revolving around direct demodulation and full decoding of received packets. The obvious drawback of full decoding of a received packet is the need to expend a significant amount of energy and processing complexity in order to fully-decode a packet, only to discover the packet is illegible due to a collision. In this paper, we propose a suite of novel, yet simple and power-efficient technique to detect a collision at the receiver side of WSNs without the need for full-decoding of the received packet. Our novel approach aims at detecting collision through fast examination of the signal statistics of a short snippet of the received packet via a relatively small number of computations over a small number of received IQ samples. We also present a complexity and power-saving comparison between our novel approach and a conventional full-decoding approach (Log-MAP algorithm) to demonstrate the significant power and complexity saving advantage of our approach. In addition, we demonstrate how to tune various design parameters in order to allow a system designer multiple degrees of freedom for design trade-offs and optimization.