Efficient Sensor Scheduling Strategy Based on Spatio-temporal Scope Information Model

TL;DR

This paper introduces a spatio-temporal scope information model for IoT sensor networks, proposing optimal scheduling mechanisms and validating their performance through experiments and theoretical analysis.

Contribution

It presents a novel SSIM model for quantifying valuable sensor data and develops both single-step and long-term optimal scheduling strategies.

Findings

Theoretical bounds for scheduling results are derived.

Q-learning effectively optimizes long-term scheduling.

Experimental results validate the proposed mechanisms.

Abstract

In this paper, based on the spatio-temporal correlation of sensor nodes in the Internet of Things (IoT), a Spatio-temporal Scope information model (SSIM) is proposed to quantify the scope valuable information of sensor data, which decays with space and time, to guide the system for efficient decision making in the sensed region. A simple sensor monitoring system containing three sensor nodes is considered, and two optimal scheduling decision mechanisms, single-step optimal and long-term optimal decision mechanisms, are proposed for the optimization problem. For the single-step mechanism, the scheduling results are analyzed theoretically, and approximate numerical bounds on the node layout between some of the scheduling results are obtained, consistent with the simulation results. For the long-term mechanism, the scheduling results with different node layouts are obtained using the Q-learning algorithm. The performance of the two mechanisms is verified by conducting experiments using the relative humidity dataset, and the differences in performance of the two mechanisms are discussed; in addition, the limitations of the model are summarized.