Hybrid Computation Offloading in Fog and Cloud Networks with Non-Orthogonal Multiple Access

Fog Computing is proposed as a promising technique that extends the computing infrastructure from the cloud center to the network edge. To enable applications to offload efficiently, the multiple resource allocation, including computation, spectrum, and energy resource as well as user scheduling, plays a quite important role in future networks. However, due to the constrained computing, storage and radio resource of fog nodes (FNs), the multiple resource allocation would be a fundamental problem and it should be well investigated. Thus, in this paper, we focus on this problem in a general multi-user multi-FN system with a remote cloud in small cell networks, where each user has multiple independent tasks that can be executed at FNs or at the remote cloud. The multiple FNs are response for performing the task distribution, user scheduling, and resource allocation dynamically and independently. Furthermore, to reduce offloading transmission latency and release the constraint of the limited radio resource, non-orthogonal multiple access ( NOMA), which enables the multiple users to transmit data to the same FN on the same radio resource, is introduced into fog and cloud networks. To this end, we formulate the offloading decision, user scheduling, and resource allocation problem as an optimization problem. Our objective is to minimize the total system cost of energy as well as the delay of users. Furthermore, we transform the original problem into a convex problem and then decompose it in a distributed and efficient way. Simulation results show that the proposed approach achieves a better performance compared with existing schemes.