Adversarial Attack on DL-based Massive MIMO CSI Feedback

TL;DR

The authors' simulation results show the destructive effect adversarial attack causes on DL-based CSI feedback by analyzing the performance of normalized mean square error and find that the jamming attack could be prevented with certain precautions.

Abstract

With the increasing application of deep learning (DL) algorithms in wireless communications, the physical layer faces new challenges caused by adversarial attack. Such attack has significantly affected the neural network in computer vision. We chose DL-based analog channel state information (CSI) to show the effect of adversarial attack on DL-based communication system. We present a practical method to craft white-box adversarial attack on DL-based CSI feedback process. Our simulation results showed the destructive effect adversarial attack caused on DL-based CSI feedback by analyzing the performance of normalized mean square error. We also launched a jamming attack for comparison and found that the jamming attack could be prevented with certain precautions. As DL algorithm becomes the trend in developing wireless communication, this work raises concerns regarding the security in the use of DL-based algorithms.

Figures (5)

• Figure 1: Architecture of CsiNet: An encoder constructed with convolutional, reshape, and fully connected layers; Decoder with fully connected and reshape layers and two RefineNet units connected in series. RefineNet unit is blocked specifically.
• Figure 2: NMSE of CsiNet versus SNR with adversarial and jamming attacks for the indoor scenario with $\gamma$ set to be 1/4.
• Figure 3: NMSE of CsiNet trained in AWGN channel versus SNR with adversarial attack for the indoor scenario with $\gamma$ set to be 1/4.
• Figure 4: NMSE of CsiNet trained in different compression rates versus SNR with adversarial attack for the indoor scenario.
• Figure 5: NMSE of CsiNet trained for outdoor scenario versus SNR with adversarial attack with $\gamma$ set to be 1/4.