Memento: A Health Monitoring System for Wireless Sensor Networks

Abstract

Wireless sensor networks are deployed today to monitor the environment, but their own health status is relatively opaque to network administrators, in most cases. Our system, Memento, provides failure detection and symptom alerts, while being frugal in the use of energy and bandwidth. Memento has two parts: an energy-efficient protocol to deliver state summaries, and a distributed failure detector module. The failure detector is robust to packet losses, and attempts to ensure that reports of failure will not exceed a specified false positive rate. We show that distributed monitoring of a subset of well-connected neighbors using a variance-bound based failure detector achieves the lowest rate of false positives, suitable for use in practice. We evaluate our findings using an implementation for the TinyOS platform on the Mica2 motes on a 55-node network, and find that Memento achieves a 80-90% reduction in bandwidth use compared to standard data collection methods