Distributed Resource Allocation for NOMA-Based Mobile Edge Computing with Content Caching

Abstract

Nonorthogonal multiple access (NOMA) and content caching have been introduced to effectively utilize and save bandwidth resources, thus relieving network transmission pressure. Therefore, we develop a computation offloading framework blending NOMA and content caching to further improve the computational performance. We formulate the joint computation offloading decision, resources allocation, and caching decision as an utility maximization problem under the constrains of bandwidth, computing and storage resources. We convert the original non-convex problem to a convex problem via variable relaxation and substitution. We then decompose it and introduce an alternative direction method of multipliers based distributed algorithm to solve it. Finally, simulation results show that our algorithm can quickly converge and reach up to 95% of the optimal value. It also performs well over various parameter settings.