Energy Efficient Data Gathering for WSN-based Context-aware Applications

Abstract

To save power, many previous works propose to use wake-up scheduling to support energy efficient data gathering in wireless sensor networks (WSNs). However, these works can only work well for regular data patterns. In many WSN-based context-aware applications, the network may randomly have some events, which may be urgent alarms or instant signals. Assume that the time is divided into fixed-length slots, and n consecutive slots are grouped as a frame. We propose a new wake up and sleep model for nodes such that nodes only wake up in some slot periodically. In this work, we design a two-phase scheme to support both regular and event data reporting, and the goal is to minimise regular report, event detection, and event reporting latencies. In the designed scheme, the first phase assigns each node one regular slot and several event slots by considering how to reduce above three kinds of latencies. In the second phase, nodes operate in a distributed manner to sense and report regular or event data. Simulation results show that our designs can indeed achieve our goal.