Online learning offloading framework for heterogeneous mobile edge computing system

Abstract

Cloud of Things (CoT) is a significant paradigm for bridging cloud resource and mobile terminals. Mobile edge computing (MEC) is a supporting architecture for CoT. The objectives of this paper are to describe and evaluate a method to handle the computation offloading problem during user mobility which minimizes the offloading failure rate in heterogeneous network. Furthermore, users' mobility and their choices for offloading lead to the everchanging condition of wireless network and opportunistic resource available. By modeling such dynamic mobile edge environment, quantizing the user cost, failure penalty and diversified QoS requirements, computation offloading problem is converted into an online decision-making problem in a stochastic process. We divide the decisionmaking into two phases: offloading planning phase and offloading running phase. In both phases the learning agent can continuously improve the control policy. We also conduct a failure recovery policy to tackle different types of failure and is included in the decision-making process. The numerical results show that the proposed online learning offloading method for mobile users can derive the optimal offloading scheme compared with the baseline algorithms.

Computing Unloading Strategy of Massive Internet of Things Devices Based on Game Theory in Mobile Edge Computing

“…According to the concept of better response in the potential game in formula (14), mobile users can choose a relatively superior offload strategy in each time slot. According to the concept of the optimal response of the potential game in formula (15), each mobile user is the optimal participant in the strategy set of other mobile users [37,38].…”

An Intelligent Opportunistic Routing Algorithm for Wireless Sensor Networks and Its Application Towards e-Healthcare

“…In recent times, the cloud of things (CoT) has gained immense popularity, due to its ability to offer an enormous amount of resources to wireless networks and heterogeneous mobile edge computing systems. The CoT makes the opportunistic decision-making during the online processing of tasks for load sharing, and makes the overall network reliable and efficient [ 22 ]. The cloud of things framework can significantly improve communication gaps between cloud resources and other mobile devices.…”

Real-time task processing method based on edge computing for spinning CPS

“…Given that technologies related to cloud computing cannot efficiently handle massive amounts of data from devices located at the edge of networks, computing models continue to be developed; these include micro-data centers [22], cloud-sea computing for humans and subsystems facing the physical world [23], cloud integration sensor networking, peer-to-peer networking, network virtualization, configuration management technology for fog computing [24][25][26], and cloudlet. Cloudlet focuses on business logic to connect cloud mobile devices and cloud servers in mobile cloud computing [27] and is considered to be the key to reduce the mobile core utilization and latency of mobile terminal users at the edge of mobile computing [28][29][30][31]. These technologies enable computing on the edge of networks and represent the development of executing edge computing of downstream data-represented cloud services and upstream data-represented IoT services [32].…”