Novel Metric for Identifying Energy-Vulnerable Nodes and Corresponding Proactive Schemes in Wireless Sensor Network
Joan Cortes, Qi Wang, John A. Dunlop · 2009
The resource-constrained nodes in wireless sensor networks (WSNs) are prone to failures, especially due to battery depletion. Well-defined effective metrics to identify the potential vulnerability in advance are essential for failure prevention. We propose a novel and advantageous metric called Battery Index to predict energy-vulnerable nodes at the very early stage of the network lifetime. Enabled by this metric, two proactive schemes are designed to prolong the lifetime of randomly deployed WSNs through dynamic routing and load balancing, and mobile robot- based node rescue, respectively. The proposed metric and schemes are validated through simulations employing realistic models, and numerical results show significant improvement in network lifetime (or usability) and resilience.