Stopping time improves detection rules in distributed sensor networks
On the Sequential Test and Distributed Detection
Information Theory
Summary
This paper explains how deciding when to stop collecting information can help make better decisions in systems that detect events using multiple sensors. The authors give a simple way to define this stopping point and show how to create the best decision rules in both single-center and spread-out detection systems. They find upper limits on how long one should wait before making a decision and verify these limits behave in expected ways. Their approach works even when many sensors are arranged in complex, one-way networks.
What this means in practice
- •For network operators: Cut down delays in sensor event detection by setting optimal stopping times adapted to network layouts.
- •For industrial automation teams: Design faster fault detection systems in factories using these stopping time rules in distributed sensor networks.
Authors
Earnest Akofor
Abstract
We present a simple definition of stopping time and its role in the formulation of sequential tests for both centralized and distributed detection, providing a straightforward procedure for obtaining optimal decision rules. Upper bounds for optimal stopping time are derived and numerically shown to possess certain qualitative features expected of the optimal stopping time. The results are extended to any distributed detection network in the form of an acyclic directed graph.