Delay Aware Secure Computation Offloading in NOMA aided MEC for IoV Networks

In this paper, we investigate a multi-vehicle multitask non-orthogonal multiple access (NOMA) based mobile edge computing (MEC) system with passive eavesdropping vehicles. To enhance the performance of edge vehicles, we propose a vehicle grouping and pairing method (GPM) to employ the vehicle near to MEC acting as a full-duplex (FD) relay to assist edge vehicles. In order to improve the transmission security, artificial noise (AN) is used to interfere with eavesdropping vehicles. The approximate expression of secrecy outage probability of the system (SOPS) is derived in closed form. This paper aims to minimize the total delay of task completion of edge vehicles by jointly optimizing vehicle task division, power allocation and transmit beamforming. We design a power allocation and task scheduling algorithm relying on genetic algorithm (GA-PATS) to solve our formulated mixed-integer non-linear programming (MINP) problem. Numerical results demonstrate the superiority of our proposed scheme in the aspect of system security and transmission delay.