Collaborative Computation Offloading in Wireless Powered Mobile-Edge Computing Systems

TL;DR

This paper introduces a cooperative offloading model in wireless powered MEC systems where users assist each other and locally compute tasks to optimize data processing within time constraints.

Contribution

It presents a novel user cooperation framework with joint optimization of offloading, local computing, and resource allocation in wireless powered MEC systems.

Findings

Cooperation improves computation efficiency over benchmarks.

Optimized resource allocation enhances data processing capacity.

Local computing reduces energy and time costs.

Abstract

This paper studies a novel user cooperation model in a wireless powered mobile edge computing system where two wireless users harvest wireless power transferred by one energy node and can offload part of their computation tasks to an edge server (ES) for remote execution. In particular, we consider that the direct communication link between one user to the ES is blocked, such that the other user acts as a relay to forward its offloading data to the server. Meanwhile, instead of forwarding all the received task data, we also allow the helping user to compute part of the received task locally to reduce the potentially high energy and time cost on task offloading to the ES. Our aim is to maximize the amount of data that can be processed within a given time frame of the two users by jointly optimizing the amount of task data computed at each device (users and ES), the system time allocation, the transmit power and CPU frequency of the users. We propose an efficient method to find the optimal solution and show that the proposed user cooperation can effectively enhance the computation performance of the system compared to other representative benchmark methods under different scenarios.