NOMA-D2D Technology-aided Low delay resource allocation for MEC networks

In this paper, we propose a joint communication and computing resource allocation scheme in mobile edge computing (MEC) networks aided by non-orthogonal multiple access (NOMA) device-to-device (D2D) technology, in which users are allowed to offload computing tasks to edge servers or D2D receivers. In this case, our optimization goal is to minimize the system delay under consideration of imperfect channel state information (CSI), which results in a multivariate coupled mixed-fraction non-linear programming problem being formulated. To solve this problem, we firstly decouple it into two sub-phases, namely the communication stage and the computing stage, and derive the closed-form expression of the power allocation coefficients without perfect CSI simultaneously. Then a spectrum resource allocation algorithm based on KM (KM-SRA) and a computing resource allocation algorithm based on maximum gain (MG-CRA) are proposed. Finally, by balancing the transmission delay and the computing delay, we propose an iterative algorithm to optimize the system delay according to KM-SRA and MG-CRA. The simulations illustrate that the proposed scheme can reduce the system delay and degrade communication interruption.