TY - JOUR

T1 - Fog Assisted 6TiSCH Tri-Layer Network Architecture for Adaptive Scheduling and Energy-Efficient Offloading Using Rank-Based Q-learning in Smart Industries

AU - Rafiq, Ahsan

AU - Ping, Wang

AU - Min, Wei

AU - Muthanna, Mohammad Saleh Ali

N1 - Publisher Copyright: © 2001-2012 IEEE.

PY - 2021/11/15

Y1 - 2021/11/15

N2 - The smart industry is a significant realization of the internet of things. IPv6 over time-slotted channel hopping (6TiSCH) is the new standard that has emerged as a promising technology for industrial communications. The demanding issues of smart industries include scalability, latency, and energy efficiency. Existing studies have presented scheduling schemes for 6TiSCH-based industries. However, these works are not suitable for practical scenarios where emergencies frequently occur. Hence, this study aims to improve scalability, latency, and energy efficiency in 6TiSCH-based smart industries for practical scenarios. A new fog assisted 6TiSCH Tri-Layer architecture is designed with a network layer, smart communication layer, and cloud layer. Scalability and latency are realized by fog computing in the smart communication layer. The rough set-based root selection (R2S) algorithm improves the network layer’s energy efficiency by constructing an R2S and destination-oriented direct acyclic graph. Emergency cognizant distributed adaptive scheduling minimizes the latency for data transmission through fuzzy Bayesian learning-based parent selection and multiobjective gravitational search algorithm-based channel selection. In the fog layer, the rank-based Q-learning algorithm performs offloading to manage the energy consumption among fog nodes. Industrial data are stored in the cloud layer where they can be accessed by end users. The performance is evaluated by modeling the tri-layer 6TiSCH network in network simulator-3.26. Experiments show that the proposed work achieves satisfactory results in terms of energy consumption, latency, end-to-end delay, response time, and transmission efficiency.

AB - The smart industry is a significant realization of the internet of things. IPv6 over time-slotted channel hopping (6TiSCH) is the new standard that has emerged as a promising technology for industrial communications. The demanding issues of smart industries include scalability, latency, and energy efficiency. Existing studies have presented scheduling schemes for 6TiSCH-based industries. However, these works are not suitable for practical scenarios where emergencies frequently occur. Hence, this study aims to improve scalability, latency, and energy efficiency in 6TiSCH-based smart industries for practical scenarios. A new fog assisted 6TiSCH Tri-Layer architecture is designed with a network layer, smart communication layer, and cloud layer. Scalability and latency are realized by fog computing in the smart communication layer. The rough set-based root selection (R2S) algorithm improves the network layer’s energy efficiency by constructing an R2S and destination-oriented direct acyclic graph. Emergency cognizant distributed adaptive scheduling minimizes the latency for data transmission through fuzzy Bayesian learning-based parent selection and multiobjective gravitational search algorithm-based channel selection. In the fog layer, the rank-based Q-learning algorithm performs offloading to manage the energy consumption among fog nodes. Industrial data are stored in the cloud layer where they can be accessed by end users. The performance is evaluated by modeling the tri-layer 6TiSCH network in network simulator-3.26. Experiments show that the proposed work achieves satisfactory results in terms of energy consumption, latency, end-to-end delay, response time, and transmission efficiency.

KW - 6TiSCH

KW - Cloud computing

KW - Edge computing

KW - Energy efficiency

KW - Fog Computing

KW - Industries

KW - Job shop scheduling

KW - Offloading

KW - Processor scheduling

KW - R2S-DODAG

KW - Sensors

KW - Smart Industry

KW - fog computing

KW - offloading

KW - smart industry

KW - DEPLOYMENT

KW - CLOUD

KW - EDGE

KW - MANAGEMENT

KW - INTERNET

KW - ATTRIBUTE VALUES

KW - IOT

KW - THINGS

KW - SYSTEMS

KW - OPTIMIZATION

UR - http://www.scopus.com/inward/record.url?scp=85101487218&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/5af3694e-6abc-3236-a634-ee8120987cf5/

U2 - 10.1109/jsen.2021.3058976

DO - 10.1109/jsen.2021.3058976

M3 - Article

AN - SCOPUS:85101487218

VL - 21

SP - 25489

EP - 25507

JO - IEEE Sensors Journal

JF - IEEE Sensors Journal

SN - 1530-437X

IS - 22

ER -