Summary

Mobile edge computing (MEC) has been considered as a promising solution for enabling computation-intensive yet latency-sensitive applications at resource-constrained wireless devices. Due to the advanced non-orthogonal multiple access (NOMA) for next-generation wireless access networks, NOMA-empowered MEC enables a flexible and spectrum efficient multi-access task offloading approach in future heterogeneous small-cell networks. In this chapter, we first review the recent advances in NOMA-empowered MEC and edge intelligence. Then, as a concrete design example, we leverage the small-cell dual connectivity in heterogeneous small-cell networks and study a paradigm of dual computation offloading in which an edge-computing user can simultaneously offload partial workloads to a cloudlet server (CS) co-located at the macro base station and an edge server (ES) co-located at a small-cell based-station. To facilitate the multi-user dual computation offloading, we exploit a hybrid NOMA and frequency division multiple access (FDMA) transmission in which the edge users from the NOMA groups for offloading their respective workloads to different ESs at different small-cell base stations. Meanwhile, all users use FDMA for offloading their workloads to the CS at the macro base station. A joint optimization of the users' partial offloading decisions, the hybrid NOMA-FDMA transmission, as well as the processing rate allocations at the ESs and the CS, is formulated to minimize the overall task completion latency. An efficient algorithm is proposed to solve the joint optimization problem. Numerical results are provided to validate the effectiveness and efficiency of our proposed algorithms.