Jamming-Assisted Proactive Eavesdropping Over Two Suspicious Communication Links

Haiyang Zhang, Lingjie Duan, Rui ZhangIEEE Trans. Wireless Communications 2020 · DOI

TL;DR

This paper introduces a novel proactive eavesdropping strategy where a legitimate monitor uses jamming to disrupt two suspicious communication links simultaneously. The authors characterize the achievable eavesdropping rate region by optimizing the jamming strategy, demonstrating significant performance gains over passive eavesdropping, especially in challenging interference environments.

What Problem Does This Paper Address?

In today's increasingly complex wireless landscape, infrastructure-free networks like D2D and ad hoc networks are vulnerable to misuse by malicious actors for surveillance and data theft. While wireless surveillance techniques exist, they often struggle with scenarios involving multiple suspicious links or unfavorable channel conditions. Existing proactive eavesdropping methods, which involve simultaneous jamming and listening, typically focus on a single suspicious link. This paper addresses the challenge of extending proactive eavesdropping to a scenario with two suspicious communication links, where the legitimate monitor aims to eavesdrop both simultaneously while using jamming to reduce their transmission rates and facilitate this eavesdropping. The core problem is to determine the optimal jamming strategy to maximize the achievable eavesdropping rate region for the monitor.

Methods

The paper proposes a "jamming-assisted proactive eavesdropping" approach. The legitimate monitor, equipped with multiple antennas and operating in full-duplex mode, strategically jams the suspicious receivers to lower their data rates. The achievable eavesdropping rate region is defined as the intersection of the achievable rate regions for the suspicious links (which the monitor can influence via jamming) and the eavesdropping links (which are fixed for a given receiver, like MMSE).

Key methodological contributions include:

  • Characterizing the Achievable Rate Region: The authors analytically derive the achievable rate region for the suspicious links by optimizing the monitor's jamming transmit covariance matrix under a power budget constraint.
  • MMSE Receiver Analysis: They first focus on the Minimum Mean Squared Error (MMSE) receiver case, deriving closed-form expressions for the upper and lower bounds of the suspicious links' rate region.
  • Time-Sharing Strategy: Recognizing that the suspicious links' rate region can be non-convex, they introduce a time-sharing based jamming strategy to enlarge this region.
  • MMSE-SIC Extension: The study is extended to the more advanced MMSE with Successive Interference Cancellation (MMSE-SIC) receiver, which offers a larger achievable rate region. This involves jointly optimizing time-sharing factors for different decoding orders.
  • Analytical Optimization: The core of the method involves solving complex optimization problems to find the optimal jamming beamforming vectors and covariance matrices. This includes leveraging techniques like semidefinite programming (SDP) and deriving closed-form solutions.

Datasets

Not discussed in the paper. The paper focuses on theoretical analysis and simulation-based results derived from a defined system model.

Key Insights and Ideas

  • Jamming as a Tool for Eavesdropping: The paper demonstrates that strategic jamming, rather than just being a disruptive force, can be a powerful tool to enable eavesdropping by degrading the target links' performance to a level where the monitor can successfully decode them.
  • Rate Region Intersection: The achievable eavesdropping rate region is fundamentally limited by the intersection of what the suspicious links can transmit and what the eavesdropper can decode. Jamming directly influences the former.
  • Time-Sharing for Convexity: The non-convexity of the achievable rate region for suspicious links is a significant challenge. Time-sharing between different jamming strategies effectively creates a convex hull, expanding the achievable region.
  • MMSE-SIC Advantage: The MMSE-SIC receiver offers a substantial improvement in the achievable eavesdropping rate region compared to the standard MMSE receiver, particularly in scenarios with high co-channel interference.
  • Spatial Degrees of Freedom: Increasing the number of jamming antennas (Nt) provides the legitimate monitor with more spatial degrees of freedom, allowing for more sophisticated jamming strategies and a larger achievable eavesdropping rate region.

Key Results

  • Characterization of Rate Regions: The paper provides theoretical characterizations of the achievable eavesdropping rate regions for both MMSE and MMSE-SIC receivers, with and without time-sharing jamming strategies.
  • Performance Gains: Simulation results show significant performance gains from proactive eavesdropping over passive eavesdropping.
  • Impact of Jamming Power: Increasing the jamming transmit power budget (Pmax) allows the monitor to further reduce suspicious links' rates, leading to larger achievable eavesdropping rate regions.
  • Location Matters: The legitimate monitor's location significantly impacts the achievable rate region, especially for the MMSE-SIC receiver, due to varying signal attenuation and interference levels.
  • MMSE-SIC Superiority: The achievable eavesdropping rate region for the MMSE-SIC receiver is notably larger than that for the MMSE receiver, especially in strong co-channel interference environments.
  • Effect of Jamming Antennas: Increasing the number of jamming antennas (Nt) from 2 to 5 significantly enlarges the achievable eavesdropping rate region.

Limitations and Future Work

  • Perfect CSI Assumption: The analysis assumes perfect Channel State Information (CSI) for all links, which may not always be realistic in practice.
  • Suspicious Transmitter Ignorance: The model assumes suspicious transmitters are unaware of the monitor and do not employ anti-eavesdropping or anti-jamming measures.
  • Perfect Self-Interference Cancellation: For the MMSE receiver analysis, perfect self-interference cancellation is assumed. While the paper later discusses imperfect cancellation, the primary analysis relies on this ideal scenario.

Future work directions suggested by the authors include:

  • Extending to Multiple Links: Generalizing the model to handle more than two suspicious communication links.
  • Advanced Transmitters/Receivers: Considering more complex scenarios with multiple antennas at the suspicious links' transmitters and receivers.
  • Adaptive Anti-Jamming: Investigating scenarios where suspicious users can detect jamming and employ adaptive anti-jamming techniques (e.g., random frequency hopping), leading to a dynamic interplay between the monitor and the suspicious users.

Summary updated

BibTeX
@article{zhang2020jamming,
  author = {Haiyang Zhang and Lingjie Duan and Rui Zhang},
  title = {{Jamming-Assisted Proactive Eavesdropping Over Two Suspicious Communication Links}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2020},
  doi = {10.1109/twc.2020.2987566},
}
Figure