Managing limited sensors to improve remote data estimation accuracy
Resource-Constrained Semantic-Aware Remote Estimation with Overlapping Sensor Coverage
Information Theory
Summary
This paper looks at how to best use a limited number of sensors that share a communication line but have different qualities, like reliability and delay, to estimate information from several sources. The authors consider how to schedule sensor use in a way that balances accuracy and resource limits, such as how often each sensor can send data. They model the problem mathematically to find efficient strategies that minimize errors in the estimated data. Their results show that optimal solutions can be simplified and involve a mix of straightforward strategies, and they explain how the limits on sensor use affect the best approach.
Markov decision processfinite-state Markov sourcessensor schedulingtransmission constraintsLagrangian optimizationtime-division multiple-accesspolicy randomizationdual subgradient ascent
Authors
Bowen Sun, Nikolaos Pappas
Abstract
We study semantic-aware remote estimation of multiple finite-state Markov sources observed by K sensors with overlapping coverage. The sensors share a time-division multiple-access uplink and differ in transmission reliability, delivery delay, and transmission budget. In each slot, the scheduler jointly selects a source and one of its monitoring sensors, or remains idle, to minimize the long-run average cost of actuation error subject to global and per-sensor transmission-frequency constraints. We formulate this problem as a finite average-cost constrained Markov decision process. We show that the transmission resource functions have rank at most K, although the global constraint may still restrict the feasible region. Consequently, the Lagrangian depends only on K effective transmission costs, and an optimal constrained solution can be represented using at most K+1 deterministic policy-recurrent-class components. We further characterize the piecewise-affine concave Lagrangian value function and derive projected dual subgradient ascent over an explicitly bounded multiplier set. Numerical results illustrate the value-function structure, the need for policy randomization in a representative instance, and the interaction between global and per-sensor transmission budgets.