Data-Efficient Quickest Change Detection in Minimax Settings

TL;DR

This paper introduces a data-efficient quickest change detection method using the DE-CuSum algorithm, optimizing observation costs while maintaining detection performance under minimax constraints.

Contribution

It proposes the DE-CuSum algorithm for minimax quickest change detection with observation cost constraints, demonstrating asymptotic optimality and superior performance over fractional sampling.

Findings

DE-CuSum is asymptotically optimal as false alarm rate approaches zero.

The algorithm outperforms fractional sampling in trade-off curves.

Numerical results confirm the effectiveness of the proposed method.

Abstract

The classical problem of quickest change detection is studied with an additional constraint on the cost of observations used in the detection process. The change point is modeled as an unknown constant, and minimax formulations are proposed for the problem. The objective in these formulations is to find a stopping time and an on-off observation control policy for the observation sequence, to minimize a version of the worst possible average delay, subject to constraints on the false alarm rate and the fraction of time observations are taken before change. An algorithm called DE-CuSum is proposed and is shown to be asymptotically optimal for the proposed formulations, as the false alarm rate goes to zero. Numerical results are used to show that the DE-CuSum algorithm has good trade-off curves and performs significantly better than the approach of fractional sampling, in which the observations are skipped using the outcome of a sequence of coin tosses, independent of the observation process. This work is guided by the insights gained from an earlier study of a Bayesian version of this problem.