BS-assisted Task Offloading for D2D Networks with Presence of User Mobility

TL;DR

This paper proposes a BS-assisted D2D task offloading scheme in mobile edge computing, addressing mobility challenges by balancing resource consumption and completion time through an optimized algorithm.

Contribution

It introduces a novel BS-assisted D2D offloading framework with a proven complex optimization problem and a Lagrangian duality-based algorithm for efficient task offloading.

Findings

Algorithm achieves close-to-optimal performance.

Trade-off analysis between resource consumption and completion time.

Structural insights into optimal offloading strategies.

Abstract

Task offloading is a key component in mobile edge computing. Offloading a task to a remote server takes communication and networking resources. An alternative is device-todevice (D2D) offloading, where a task of a device is offloaded to some device having computational resource available. The latter requires that the devices are within the range of each other, first for task collection, and later for result gathering. Hence, in mobility scenarios, the performance of D2D offloading will suffer if the contact rates between the devices are low. We enhance the setup to base station (BS) assisted D2D offloading, namely, a BS can act as a relay for task distribution or result collection. However, this would imply additional consumption of wireless resource. The associated cost and the improvement in completion time of task offloading compose a fundamental trade-off. For the resulting optimization problem, we mathematically prove the complexity, and propose an algorithm using Lagrangian duality. The simulation results demonstrate not only that the algorithm has close-to-optimal performance, but also provide structural insights of the optimal trade-off.