A Novel Matching Theory-Based Framework for Computation Offloading in Device-to-Device Communication

Abstract

The increasing demand for multimedia mobile applications creates challenges for smartphones that have low memory or slow processor. In this paper, we explore how local cooperation among the devices can help alleviate this problem to some extent. Local cooperation can be achieved through device-to-device communication which is expected to be a significant feature of next generation cellular networks. The cooperation can be achieved by having smartphones with insufficient resources offload their jobs to neighboring smartphones with higher computational capability. We model it as a stable matching problem. We also propose a novel flexible framework and simple distributed algorithms to pair mobile devices of heterogeneous capabilities.