Abstract
Even though energy harvesting is a promising technology for energy-efficient surveillance systems, energy harvesting levels are highly dynamic depending on the time and location. Thus, the deployment of non-energy-harvesting sensor nodes and sophisticated sleep scheduling of sensor nodes are necessary for performance guaranteed surveillance systems. In this paper, we present a software-defined surveillance system (SDSS) in which a centralized controller determines the sleep schedules of energy harvesting and non-energy harvesting sensor nodes on the basis of the collected information such as the spatial distribution of targets and the energy levels of sensor nodes. To derive the optimal sleep schedules minimizing the number of active sensor nodes while providing sufficient surveillance performance, a constraint Markov decision process problem is formulated and the optimal policy on sleep scheduling is obtained by linear programming. The evaluation results demonstrate that the SDSS with the optimal policy can reduce energy consumption by employing fewer active sensor nodes while providing the required level of target monitoring probability.
| Original language | English |
|---|---|
| Journal | IEEE Internet of Things Journal |
| DOIs | |
| Publication status | Accepted/In press - 2018 Jan 22 |
Fingerprint
Keywords
- constraint Markov decision process (CMDP).
- Energy consumption
- Energy harvesting
- energy harvesting
- Internet of Things (IoT)
- Optimization
- Schedules
- sleep scheduling
- Surveillance
- surveillance system
- target monitoring
- Wireless sensor networks
ASJC Scopus subject areas
- Signal Processing
- Information Systems
- Hardware and Architecture
- Computer Science Applications
- Computer Networks and Communications
Cite this
A Software-Defined Surveillance System with Energy Harvesting : Design and Performance Optimization. / Ko, Haneul; Pack, Sangheon.
In: IEEE Internet of Things Journal, 22.01.2018.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - A Software-Defined Surveillance System with Energy Harvesting
T2 - Design and Performance Optimization
AU - Ko, Haneul
AU - Pack, Sangheon
PY - 2018/1/22
Y1 - 2018/1/22
N2 - Even though energy harvesting is a promising technology for energy-efficient surveillance systems, energy harvesting levels are highly dynamic depending on the time and location. Thus, the deployment of non-energy-harvesting sensor nodes and sophisticated sleep scheduling of sensor nodes are necessary for performance guaranteed surveillance systems. In this paper, we present a software-defined surveillance system (SDSS) in which a centralized controller determines the sleep schedules of energy harvesting and non-energy harvesting sensor nodes on the basis of the collected information such as the spatial distribution of targets and the energy levels of sensor nodes. To derive the optimal sleep schedules minimizing the number of active sensor nodes while providing sufficient surveillance performance, a constraint Markov decision process problem is formulated and the optimal policy on sleep scheduling is obtained by linear programming. The evaluation results demonstrate that the SDSS with the optimal policy can reduce energy consumption by employing fewer active sensor nodes while providing the required level of target monitoring probability.
AB - Even though energy harvesting is a promising technology for energy-efficient surveillance systems, energy harvesting levels are highly dynamic depending on the time and location. Thus, the deployment of non-energy-harvesting sensor nodes and sophisticated sleep scheduling of sensor nodes are necessary for performance guaranteed surveillance systems. In this paper, we present a software-defined surveillance system (SDSS) in which a centralized controller determines the sleep schedules of energy harvesting and non-energy harvesting sensor nodes on the basis of the collected information such as the spatial distribution of targets and the energy levels of sensor nodes. To derive the optimal sleep schedules minimizing the number of active sensor nodes while providing sufficient surveillance performance, a constraint Markov decision process problem is formulated and the optimal policy on sleep scheduling is obtained by linear programming. The evaluation results demonstrate that the SDSS with the optimal policy can reduce energy consumption by employing fewer active sensor nodes while providing the required level of target monitoring probability.
KW - constraint Markov decision process (CMDP).
KW - Energy consumption
KW - Energy harvesting
KW - energy harvesting
KW - Internet of Things (IoT)
KW - Optimization
KW - Schedules
KW - sleep scheduling
KW - Surveillance
KW - surveillance system
KW - target monitoring
KW - Wireless sensor networks
UR - http://www.scopus.com/inward/record.url?scp=85040966911&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85040966911&partnerID=8YFLogxK
U2 - 10.1109/JIOT.2018.2797174
DO - 10.1109/JIOT.2018.2797174
M3 - Article
AN - SCOPUS:85040966911
JO - IEEE Internet of Things Journal
JF - IEEE Internet of Things Journal
SN - 2327-4662
ER -