"Self-Organized Scheduling of Node Activity in Large-Scale Wireless Sensor Networks"
Advances in MEMS (Micro-Electro-Mechanical Systems) technology have enabled the development of extremely small multi-functional sensing devices. Due to their miniature size, these micro-sensors can blend seamlessly with the environment and sense intricate information. Some key advantages of such pervasive sensor networks are: a) Ubiquitous, non-intrusive nature; b) Random deployment with limited or no pre-established infrastructure; c) Minimal supervision since nodes can self-organize into a functional network through local collaboration, d) Cost effectiveness; e) Flexibility; f) Scalability; g) Simple expandability; and e) Robustness. Large-scale sensor networks are being envisioned and applied in a wide range of scenarios like sensors embedded in a bridge to monitor for cracks, deployed in a field to track enemy movements or scattered in a forest to detect a fire breakout at an early stage. Field coverage is a critical issue for such event monitoring wireless networks. Given the spatiotemporal nature of the phenomena being observed, the sensor network must be able to detect and report its occurrence as quickly and accurately as possible. This leads to the problem of coverage: Ensuring that no part of the field remains unsensed for more than a specified duration. The sensor nodes used in random networks have limited resources (on-board battery, processing power and storage) and are vulnerable to failure. As individual nodes are lost, coverage (and communication connectivity) drops, rendering the network unusable. One way around this is to "over-deploy" nodes, i.e., start the network off with a significantly higher node density than is needed, and only turn them on when necessary. This turns the coverage problem into one of scheduling. Since the network is assumed to be unattended, such scheduling must be done by the nodes themselves in a self-organized fashion. To conserve energy, as few nodes as possible must be switched on at a time, but the load across the nodes must also be balanced to avoid creating "holes" in the network when highly loaded nodes fail. This ensures than coverage is maintained while maximizing the network's lifetime and producing graceful rather than catastrophic degradation. In this paper, we compare several decentralized, self-organized methods for scheduling nodes to obtain effective coverage. The focus in on assessing how well individual nodes in the network can locally estimate and optimize their schedules in order to achieve complete global coverage and a longer network lifetime at a lower energy cost. The primary constraints are energy spent on information exchange for setup, the communicating radius which limits the extent of local information available to the node, and the latency in determining the optimal schedule. We also look at whether some scheduling methods provide greater robustness to random loss of nodes (due to events in the environment) than others.