Delay Aware Secure Offloading for NOMA-Assisted Mobile Edge Computing in Internet of Vehicles

TL;DR

This paper proposes a delay-aware secure offloading scheme for NOMA-assisted MEC in Internet of Vehicles, utilizing vehicle relays and artificial noise to enhance security and reduce delay.

Contribution

It introduces a novel vehicle grouping and pairing method with artificial noise for security, and a genetic algorithm-based optimization for delay minimization.

Findings

The scheme improves system security against eavesdropping.

It reduces total task completion delay for edge vehicles.

Numerical results confirm the scheme's effectiveness in security and delay performance.

Abstract

In this paper, a multi-vehicle multi-task nonorthogonal multiple access (NOMA) assisted mobile edge computing (MEC) system with passive eavesdropping vehicles is investigated. To heighten the performance of edge vehicles, we propose a vehicle grouping pairing method, which utilizes vehicles near the MEC as full-duplex relays to assist edge vehicles. For promoting transmission security, we employ artificial noise to interrupt eavesdropping vehicles. Furthermore, we derive the approximate expression of secrecy outage probability of the system. The combined optimization of vehicle task division, power allocation, and transmit beamforming is formulated to minimize the total delay of task completion of edge vehicles. Then, we design a power allocation and task scheduling algorithm based on genetic algorithm to solve the mixed-integer nonlinear programming problem. Numerical results demonstrate the superiority of our proposed scheme in terms of system security and transmission delay.