MEC-NOMA系统的物理层安全性能评估-《集美大学学报（自然科学版）》

文章信息/Info

Author(s):: ZHANG Min; College of Computer Engineering,Jimei University,Xiamen 361021,China

Keywords:: mobile edge computing; nonorthogonal multile access; PHY-security; security outage probability; convolutional neural network

摘要:: 面向移动边缘计算（mobile edge computing,MEC）场景，建立了一个存在主动攻击者的非正交多址协议（nonorthogonal multile access,NOMA）网络传输模型,并设计一种卷积神经网络（convolutional neural network,CNN）模型来评估该传输模型的安全中断概率（security outage probability，SOP）。研究结果表明：所提出的MECNOMA系统不仅提高了SOP，而且能够对抗主动窃听者的攻击；此外，通过CNN模型评估的SOP与其他方法（蒙特卡洛方法和数学解析表达式）非常接近，且执行时间更短。

Abstract:: In this paper,a NOMA model with active eavesdropper is established for mobile edge computing(MEC) scenarios,and evaluated the security outage probability (SOP) of the transport model by convolutional neural network (CNN) model.The results show that the proposed MEC-NOMA system not only improves the SOP,but also overcomes the active eavesdropper attacks.In addition,the SOP estimated by CNN model is very close to other methods (Monte Carlo method and analytical expression),but the execution time is shorter.