Downlink Power Allocation in Massive MIMO via Deep Learning: Adversarial Attacks and Training

被引:6
|
作者
Manoj, B. R. [1 ,2 ]
Sadeghi, Meysam [3 ]
Larsson, Erik G. [4 ]
机构
[1] Linkoping Univ, Dept Elect Engn, Div Commun Syst, S-58183 Linkoping, Sweden
[2] Indian Inst Technol Guwahati, Dept Elect & Elect Engn, Gauhati 781039, India
[3] Linkoping Univ, Div Commun Syst, S-58183 Linkoping, Sweden
[4] Linkoping Univ, Dept Elect Engn ISY, S-58183 Linkoping, Sweden
关键词
Adversarial attacks; adversarial training; black-box attack; deep neural networks; massive MIMO; regression; white-box attack; CLASSIFICATION; NETWORKS;
D O I
10.1109/TCCN.2022.3147203
中图分类号
TN [电子技术、通信技术];
学科分类号
0809 ;
摘要
The successful emergence of deep learning (DL) in wireless system applications has raised concerns about new security-related challenges. One such security challenge is adversarial attacks. Although there has been much work demonstrating the susceptibility of DL-based classification tasks to adversarial attacks, regression-based problems in the context of a wireless system have not been studied so far from an attack perspective. The aim of this paper is twofold: (i) we consider a regression problem in a wireless setting and show that adversarial attacks can break the DL-based approach and (ii) we analyze the effectiveness of adversarial training as a defensive technique in adversarial settings and show that the robustness of DL-based wireless system against attacks improves significantly. Specifically, the wireless application considered in this paper is the DL-based power allocation in the downlink of a multicell massive multi-input-multi-output system, where the goal of the attack is to yield an infeasible solution by the DL model. We extend the gradient-based adversarial attacks: fast gradient sign method (FGSM), momentum iterative FGSM, and projected gradient descent method to analyze the susceptibility of the considered wireless application with and without adversarial training. We analyze the deep neural network (DNN) models performance against these attacks, where the adversarial perturbations are crafted using both the white-box and black-box attacks.
引用
收藏
页码:707 / 719
页数:13
相关论文
共 50 条
  • [1] Adversarial Attacks on Deep Learning Based Power Allocation in a Massive MIMO Network
    Manoj, B. R.
    Sadeghi, Meysam
    Larsson, Erik G.
    IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC 2021), 2021,
  • [2] Defending Adversarial Attacks on Deep Learning-Based Power Allocation in Massive MIMO Using Denoising Autoencoders
    Sahay, Rajeev
    Zhang, Minjun
    Love, David J. J.
    Brinton, Christopher G. G.
    IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING, 2023, 9 (04) : 913 - 926
  • [3] Deep Learning Power Allocation in Massive MIMO
    Sanguinetti, Luca
    Zappone, Alessio
    Debbah, Merouane
    2018 CONFERENCE RECORD OF 52ND ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS, AND COMPUTERS, 2018, : 1257 - 1261
  • [4] Universal Adversarial Attacks on Neural Networks for Power Allocation in a Massive MIMO System
    Santos, Pablo Millan
    Manoj, B. R.
    Sadeghi, Meysam
    Larsson, Erik G.
    IEEE WIRELESS COMMUNICATIONS LETTERS, 2022, 11 (01) : 67 - 71
  • [5] Power Allocation for Downlink Training in Cell-Free Massive MIMO Networks
    Izadi, Arezoo
    Razavizadeh, S. Mohammad
    Saatlou, O.
    2020 10TH INTERNATIONAL SYMPOSIUM ON TELECOMMUNICATIONS (IST), 2020, : 111 - 115
  • [6] Massive MU-MIMO Downlink Training with Orthogonal Pilots Allocation
    Qian, Cheng
    Yang, Haifen
    Li, Guangjun
    2018 41ST INTERNATIONAL CONFERENCE ON TELECOMMUNICATIONS AND SIGNAL PROCESSING (TSP), 2018, : 559 - 564
  • [7] Downlink Power Control for Cell-Free Massive MIMO With Deep Reinforcement Learning
    Luo, Lirui
    Zhang, Jiayi
    Chen, Shuaifei
    Zhang, Xiaodan
    Ai, Bo
    Ng, Derrick Wing Kwan
    IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, 2022, 71 (06) : 6772 - 6777
  • [8] Learning-Based Downlink Power Allocation in Cell-Free Massive MIMO Systems
    Zaher, Mahmoud
    Demir, Ozlem Tugfe
    Bjornson, Emil
    Petrova, Marina
    IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2023, 22 (01) : 174 - 188
  • [9] Outage Analysis of Downlink URLLC in Massive MIMO systems with Power Allocation
    Bana, Alexandru-Sabin
    Sanguinetti, Luca
    De Carvalho, Elisabeth
    Popovski, Petar
    CONFERENCE RECORD OF THE 2019 FIFTY-THIRD ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS & COMPUTERS, 2019, : 1394 - 1398
  • [10] Power Allocation in Cell-Free Massive MIMO: A Deep Learning Method
    Zhao, Yu
    Niemegeers, Ignas G.
    De Groot, Sonia Heemstra
    IEEE ACCESS, 2020, 8 : 87185 - 87200