A Deep Learning Approach for Energy Efficient Computational Offloading in Mobile Edge Computing

Zaiwar Ali,Sadia Khaf,Lei Jiao,Ghulam Abbas,Ziaul Haq Abbas,Thar Baker

doi:10.1109/access.2019.2947053

Abstract

Mobile edge computing (MEC) has shown tremendous potential as a means for computationally intensive mobile applications by partially or entirely offloading computations to a nearby server to minimize the energy consumption of user equipment (UE). However, the task of selecting an optimal set of components to offload considering the amount of data transfer as well as the latency in communication is a complex problem. In this paper, we propose a novel energy-efficient deep learning based offloading scheme (EEDOS) to train a deep learning based smart decision-making algorithm that selects an optimal set of application components based on remaining energy of UEs, energy consumption by application components, network conditions, computational load, amount of data transfer, and delays in communication. We formulate the cost function involving all aforementioned factors, obtain the cost for all possible combinations of component offloading policies, select the optimal policies over an exhaustive dataset, and train a deep learning network as an alternative for the extensive computations involved. Simulation results show that our proposed model is promising in terms of accuracy and energy consumption of UEs.

Highlights

Mobile and wearable devices, after referred to as user equipment (UE), have experienced a tremendous increase in computational power over the years but, the applications running on these devices are becoming increasingly complex at the same time [1], [2]
WORK In this paper, we demonstrated a novel approach to intelligently offload application components to cloudlets using comprehensive mathematical modeling and deep learning approach named as efficient deep learning based offloading scheme (EEDOS)
We modeled a cost function for the application execution on UEs as well as on the cloud server under the constraints of energy consumption, network conditions, delays, and available computation resources

Summary

Introduction

After referred to as user equipment (UE), have experienced a tremendous increase in computational power over the years but, the applications running on these devices are becoming increasingly complex at the same time [1], [2]. The task of executing computationally intensive applications on devices is not fully prepared to handle the computational workload, which demands an alternative solution. Cloud computing gained much popularity as a promising alternative [3]. The delays involved in communication between the UEs and the cloud servers pose serious challenges on the viability of such solutions [4]. Mobile Cloud Computing (MCC) is not an effective solution to manage the computational needs of mobile devices [5], [6].

Objectives

Results

Conclusion