Device-centric Energy Optimization for Edge Cloud Offloading

TL;DR

This paper develops an analytical framework to optimize energy consumption in edge cloud offloading, balancing local processing and data transmission for multiple devices, resulting in reduced device energy use.

Contribution

It introduces a closed-form solution for optimal offloading decisions that minimizes device energy consumption in edge cloud systems.

Findings

Significant reduction in device energy consumption achieved.

Optimal offloading data amounts derived for given resources.

Analytical framework applicable to multi-device scenarios.

Abstract

A wireless system is considered, where, computationally complex algorithms are offloaded from user devices to an edge cloud server, for the purpose of efficient battery usage. The main focus of this paper is to characterize and analyze, the trade-off between the energy consumed for processing the data locally, and for offloading. An analytical framework is presented, that minimizes the in-device energy consumption, by providing an optimal offloading decision for multiple user devices. A closed form solution is obtained for the offloading decision. The solution also provides the amount of computational data that should be offloaded, for the given computational and communication resources. Consequently, reduction in the energy consumption is observed.