Joint Optimization of Relaying Rate and Energy Consumption for Cooperative Mobile Edge Computing

Mobile edge computing networks with energy constrained users, which do not have access to licensed spectrum, might face difficulty in completing delay sensitive computational tasks due to lack of proper offloading spectrum. This paper considers a mobile edge computing scenario in the context of cooperative communication, where an user (with no access to licensed spectrum) relays a licensed spectrum holder's data to get access to the licensed spectrum for computation task offloading. We consider the relay to have a heavy computational task to complete by a given time duration. However, due to less computation power, the relay might not be able to meet the timeline. In this regard, the relay might offload partial computational task over the licensed spectrum to a more computationally powerful node, e.g., mobile edge computing server. The licensed spectrum holder might be interested in maximizing the relaying rate; whereas, the relay might want to minimize the total energy consumption for the task computation. We consider the relay partially offloading it's computational task and formulate an optimization problem capturing two different interests for the licensed spectrum holder and the relay. In the optimization problem, we consider relevant constraints to make it more general. The optimization problem is observed to be non-convex. However, after analysis, we propose a less complex iterative algorithm. We observe that joint optimization provides a better trade-off between the relaying rate and the energy consumption at the WU, which is not observed for individual relaying rate maximization and energy consumption minimization.