Learning Based and Physical-layer Assisted Secure Computation Offloading in Vehicular Spectrum Sharing Networks

Massive computing tasks have been generated with the widespread applications of big data analysis in vehicular edge computing (VEC) networks. However, the offloading process of the VEC networks suffers a threat of information leakage. The physical layer security (PLS) technology is an effective security solution to protect confidential information. Furthermore, the contradiction between massive data transmission and limited communication resources promotes an urgent need for a proper scheme to improve resource utilization. In this paper, we design a joint secure offloading and resource allocation (SoRA) scheme based on PLS technology and spectrum sharing architecture. We aim at minimizing the system processing delay of all vehicular users (VUs) while ensuring the security of information, by jointly optimizing the spectrum access, transmit power and computing resource allocation. Then we adopt a multi-agent deep reinforcement learning algorithm to solve the optimization problem. With proper training, we demonstrate that the VU agents can successfully cooperate to improve the system processing delay and ensure the security of the offloading process.

I. Introduction

The application of Big Data on the Internet of Vehicle (IoV) promotes vehicles to generate increasingly more delay-sensitive tasks for supporting various novel services. The emerging mobile edge computing (MEC) has been regarded as a promising technology that can provide fast and energy-efficient computing services for multiple vehicular users (VUs) [1], [2]. However, the offloading process suffers a high risk of confidential information leakage. Physical-layer security (PLS) techniques are effective security mechanisms, which can protect the confidential computing task information (CTI) from eavesdropping by exploiting the characteristics of the wireless channel [3], [4]. The authors in [3] propose a secure computation offloading model, where joint AN-assisted beamforming and wiretap coding schemes are used to prevent a multi-antenna eavesdropper from wiretapping. The authors in [4] employ the physical-layer security approach to protect the wireless computation offloading in MEC, aiming at minimizing the weighted sum-energy consumption. However, the above researches only consider a static eavesdropping scene with one eavesdropper, which are not suitable for dynamic IoV scenarios with multiple eavesdroppers.