Abstract
To provide an accurate environmental map (EM) while avoiding unnecessary transmissions of Internet of Things (IoT) devices, we propose a spatiotemporal correlation-based environmental monitoring system (ST-EMS). In ST-EMS, IoT devices decide whether to transmit the sensed data to an IoT gateway (GW) or not by considering the temporal correlation in the sensed data and energy level. Through a Markov decision process (MDP) formulation, the optimal policy is obtained and it is proved that the optimal policy of MDP has an implementation-friendly threshold structure by using the submodularity concept. Also, the IoT GW in ST-EMS restores EM and improves its accuracy by exploiting the spatial correlation among sensed data using probabilistic matrix factorization. Evaluation results demonstrate that ST-EMS can improve the expected total reward significantly compared with other schemes and achieve low mean square error of 1% in EM restoration.
Original language | English |
---|---|
Article number | 8588359 |
Pages (from-to) | 2958-2968 |
Number of pages | 11 |
Journal | IEEE Transactions on Industrial Informatics |
Volume | 15 |
Issue number | 5 |
DOIs | |
Publication status | Published - 2019 May |
Keywords
- Energy harvesting
- Internet of Things (IoT)
- Markov decision process (MDP)
- monitoring service
- probabilistic matrix factorization (PMF)
- spatiotemporal correlation
ASJC Scopus subject areas
- Control and Systems Engineering
- Information Systems
- Computer Science Applications
- Electrical and Electronic Engineering