Joint Connection Modes, Uplink Paths and Computational Tasks Assignment for Unmanned Mining Vehicles’ Energy Saving in Mobile Edge Computing Networks

Rui Xiong,Lijing Li,Xiaosu Yi,Huasong Zeng,Chunxi Zhang

doi:10.1109/access.2020.3013714

Rui Xiong, Lijing Li + Show 3 more

Open Access

https://doi.org/10.1109/access.2020.3013714

Copy DOI

Abstract

At present, most unmanned mining vehicles (UMVs) adopt batteries to meet the requirements of low power consumption in driving control systems, and saving energy is the key to increase the working time and production efficiency. Mobile edge computing (MEC) is an effective technology that can improve the driving performance, whereas reduces the power consumption caused by the UMV’s CPU. However, sending more offloading tasks to MEC servers means higher wireless channel transmission power, and especially in mining areas, where the communication quality of wireless channels are easily deteriorated by dust, rocks and ravines. To solve this contradiction, this article firstly analyzes the UMVs’ consumption of computational power and communicational power based on the proposed MEC architecture. Then, considering that flexible connection methods can reduce the end-to-end delay of offloading tasks and improve the use efficiency of link resources, a joint connection modes, uplink paths and computational tasks assignment method is proposed to reduce the power consumption under a strict delay constraint. Furthermore, a novel algorithm is presented to obtain the optimal parameters. Finally, through a simulation experiment, the effectiveness of this method in reducing the power consumption compared with the shortest path method is proved.

Highlights

With the rapid development of intelligent driving technology, unmanned mining vehicle (UMV) has become an effective tool that can improve production efficiency and reduce labor costs in mining areas
All UMVs are traveling at a low speed and randomly distributed in the mining area
Based on the above evaluations, the proposed method based on joint connection modes, uplink paths and computational tasks assignment can reduce the power consumption by approximately 20% compared with optimal tasks assignment method with the shortest path and 40% compared with optimal tasks assignment method with the fixed path, which demonstrates the effectiveness of the proposed method

Summary

INTRODUCTION

With the rapid development of intelligent driving technology, unmanned mining vehicle (UMV) has become an effective tool that can improve production efficiency and reduce labor costs in mining areas. A novel method that comprehensively considers the factors including connection modes, uplink paths, communicational loads, and computational tasks is proposed to VOLUME 8, 2020 achieve the minimum power consumption for UMV in the mining area, and a traversal algorithm is presented to calculate the optimal parameters of this method. Due to the high computational complexity of the formulated problem, a heuristic method is designed from the aspects of offloading decision, channel assignment, and power control He et al [15] presented an integrated framework that can enable dynamic orchestration of networking, caching, and computing resources to improve the performance of generation vehicular networks, and proposed a resource allocation strategy using a novel deep reinforcement learning approach to resolve the joint optimization problem, where the gains of networking, caching and computing are taken into consideration. Based on this network architecture, the following analyzes the power consumption in two aspects

COMPUTATIONAL MODEL

COMMUNICATIONAL MODEL

SIMULATION RESULTS

CONCLUSION