面向频谱接入深度强化学习模型的后门攻击方法  

Backdoor Attack Against Deep Reinforcement Learning-based Spectrum Access Model

作　　者：魏楠 魏祥麟 范建华 薛羽[1] 胡永扬 WEI Nan;WEI Xianglin;FAN Jianhua;XUE Yu;HU Yongyang(School of Computer and Software,Nanjing University of Information Science and Technology,Nanjing 210044,China;The 63rd Research Institute,National University of Defense Technology,Nanjing 210007,China)

机构地区：[1]南京信息工程大学计算机与软件学院,南京210044 [2]国防科技大学第六十三研究所,南京210007

出　　处：《计算机科学》2023年第1期351-361,共11页Computer Science

摘　　要：深度强化学习(Deep Reinforcement Learning,DRL)方法以其在智能体感知和决策方面的优势,在多用户智能动态频谱接入问题上得到广泛关注。然而,深度神经网络的弱可解释性使得DRL模型容易受到后门攻击威胁。针对认知无线网络下基于深度强化学习模型的动态频谱接入(Dynamic Spectrum Access,DSA)场景,提出了一种非侵入、开销低的后门攻击方法。攻击者通过监听信道使用情况来选择非侵入的后门触发器,随后将后门样本添加到次用户的DRL模型训练池,并在训练阶段将后门植入DRL模型中;在推理阶段,攻击者主动发送信号激活模型中的触发器,使次用户做出目标动作,降低次用户的信道接入成功率。仿真结果表明,所提后门攻击方法能够在不同规模的DSA场景下达到90%以上的攻击成功率,相比持续攻击可以减少20%~30%的攻击开销,并适用于3种不同类型的DRL模型。Deep reinforcement learning(DRL)has attracted much attention in multi-user intelligent dynamic spectrum access due to its advantages in sensing and decision making.However,the weak interpretability of deep neural networks(DNNs)makes DRL models vulnerable to backdoor attacks.In this paper,a non-invasive backdoor attack method with low-cost is proposed against DSA-oriented DRL models in cognitive wireless networks.The attacker monitors the wireless channels to select backdoor triggers,and generates backdoor samples into the experience pool of a secondary user’s DRL model.Then,the trigger can be implanted into the DRL model during the training phase.The attacker actively sends signals to activate the triggers in the DRL model during the inference phase,inducing secondary users to take the actions set by the attacker,thereby reducing their success rate of channel access.A series of simulation show that the proposed backdoor attack method can reduce the attack cost by 20%~30%while achieving an attack success rate over 90%,and is suitable for three different DRL models.

关 键 词：动态频谱接入 深度强化学习 后门攻击 触发器 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TN915.08[自动化与计算机技术—控制科学与工程]