Parallel Offloading in Green and Sustainable Mobile Edge Computing for Delay-Constrained IoT System

• Published in: IEEE transactions on vehicular technology Vol. 68; no. 12; pp. 12202 - 12214
• Main Authors: Deng, Yiqin, Chen, Zhigang, Yao, Xin, Hassan, Shahzad, Ibrahim, Ali. M. A.
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.12.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Clean energy; Cloud computing; Computer simulation; Constraints; Delay; Delays; Edge computing; Energy; Energy consumption; Energy harvesting; Energy management; Internet of Things; internet of things (IoT); lyapunov optimization; Maintenance costs; Mobile computing; Mobile edge computing; Optimization; partial computation offloading; Queues; Renewable energy; resource allocation; Resource management; Response time; Smart buildings; Sustainability; Task analysis; Technology assessment
• ISSN: 0018-9545; 1939-9359
• DOI: 10.1109/TVT.2019.2944926

Currently, the Internet of Things (IoT) solutions are playing an important role in numerous areas, especially in smart homes and buildings, health-care, vehicles, and energy. It will continue to expand in various fields in the future. However, some issues limit the further development of IoT technologies. First, the battery-powered feature increases the maintenance cost of replacing batteries for IoT devices. Second, existing Cloud-IoT frameworks are not able to cope with emerging delay-constrained applications in the IoT system due to its centralized mode of operation and the considerable communication delay. Existing studies neither satisfy the demand for the quick response in time-constraint IoT applications nor fundamentally solving the problem of energy sustainability. Therefore, this paper studies the problem of energy sustainability and timeliness in IoT system. Based on Energy Harvesting Technologies (EHT), the Green and Sustainable Mobile Edge Computing (GS-MEC) framework is proposed to make IoT devices self-powered by utilizing the green energy in the IoT environment. In this framework, we formulate the problem of minimizing response time and packet losses of tasks under the limitation of energy queue stability to improve the timeliness and reliability of task processing. Additionally, the dynamic parallel computing offloading and energy management (DPCOEM) algorithm is designed to solve the problem based on the Lyapunov optimization technology. Finally, theoretical analysis demonstrates the effectiveness of the proposed algorithm, and the numerical result of simulation shows that the average performance of the proposed algorithm is an order of magnitude better than state-of-the-art algorithms.