Data salmon: A greedy mobile basestation protocol for efficient data collection in wireless sensor networks

Our work addresses the spatiotemporally varying nature of data traffic in environmental monitoring and surveillance applications. By employing a network-controlled mobile basestation (MB), we present a simple energy-efficient data collection protocol for wireless sensor networks (WSNs). In contrast to the existing MB-based solutions where WSN nodes buffer data passively until visited by an MB, our protocol maintains an always-on multihop connectivity to the MB by means of an efficient distributed tracking mechanism. This allows the nodes to forward their data in a timely fashion, avoiding latencies due to long-term buffering. Our protocol progressively relocates the MB closer to the regions that produce higher data rates and reduces the average weighted multihop traffic, enabling energy savings. Using the convexity of the cost function, we prove that our local and greedy protocol is in fact optimal.