[Paper Review] Safeguarding Wireless Network with UAVs: A Physical Layer Security Perspective

Qingqing Wu, Weidong Mei+1 more|arXiv (Cornell University)|Feb 7, 2019

UAV Applications and OptimizationEngineering15 references20 citations

TL;DR

This paper proposes physical-layer security techniques to protect both UAV-to-ground and ground-to-UAV communications from eavesdropping and jamming attacks in 5G networks. By leveraging cooperative artificial noise from multiple ground base stations, it achieves significant secrecy rate gains—especially with increasing numbers of cooperating jammers—effectively countering strong line-of-sight eavesdropping links from UAVs and terrestrial jammers.

ABSTRACT

Integrating unmanned aerial vehicles (UAVs) into future wireless systems such as the fifth-generation (5G) cellular network is anticipated to bring significant benefits for both UAV and telecommunication industries. Generally speaking, UAVs can be used as new aerial platforms in the cellular network to provide communication services for terrestrial users, or become new aerial users of the cellular network served by the terrestrial base stations. Due to their high altitude, UAVs usually have dominant line-of-sight (LoS) channels with the ground nodes, which, however, pose new security challenges to future wireless networks with widely deployed UAVs. On one hand, UAV-ground communications are more prone than terrestrial communications to eavesdropping and jamming attacks by malicious nodes on the ground. On the other hand, compared to malicious ground nodes, malicious UAVs can launch more effective eavesdropping and jamming attacks to terrestrial communications. Motivated by the above, in this article, we aim to identify such new issues from a physical-layer security viewpoint and propose novel solutions to tackle them efficiently. Numerical results are provided to validate their effectiveness and promising directions for future research are also discussed.

Motivation & Objective

• Address the heightened vulnerability of UAV-ground communications to eavesdropping and jamming due to dominant line-of-sight (LoS) channels.
• Tackle the emerging threat of malicious UAVs acting as high-capacity eavesdroppers or jammers in terrestrial cellular networks.
• Develop physical-layer security mechanisms that are robust to the unique propagation characteristics of LoS-dominant air-to-ground and ground-to-air links.
• Enable secure communication for both legitimate UAVs and terrestrial users in future 5G networks with high UAV density.
• Explore practical defense strategies against advanced attacks such as GPS spoofing and malicious UAV detection.

Proposed method

• Propose cooperative artificial noise (AN) transmission from multiple ground base stations (BSs) to degrade the eavesdropping quality at a malicious UAV.
• Design a jamming beamforming strategy that maximizes interference at the UAV while minimizing impact on the legitimate ground receiver.
• Utilize directional antenna patterns with downtilt (10°) at BSs to focus jamming energy toward the UAV and reduce interference to terrestrial users.
• Model the G2A (ground-to-air) eavesdropping channel as LoS-dominant and the A2G (air-to-ground) legitimate link as fading-limited, reflecting real-world propagation differences.
• Apply stochastic geometry and signal-to-interference-plus-noise ratio (SINR) analysis to evaluate secrecy rate performance under varying numbers of cooperating BSs.
• Simulate secrecy rate performance as a function of the number of cooperative BSs and transmit power, with fixed UAV altitude and distance from the serving BS.

Experimental results

Research questions

• RQ1 How does the secrecy rate of UAV-ground communications degrade under strong LoS-based eavesdropping by a malicious UAV?
• RQ2 To what extent can cooperative artificial noise from multiple ground base stations improve the secrecy rate in the presence of a high-capacity UAV eavesdropper?
• RQ3 What is the impact of increasing the number of cooperating jamming BSs on the secrecy rate, and how does it scale with transmit power?
• RQ4 How does the secrecy rate performance compare when the legitimate link experiences fading while the UAV eavesdropper enjoys a LoS channel?
• RQ5 What are the key design trade-offs in deploying cooperative jamming for UAV security, especially regarding interference leakage to legitimate users?

Key findings

• The secrecy rate is nearly zero when no cooperative jamming BSs are active, due to the superior LoS channel quality at the UAV eavesdropper compared to the fading terrestrial link.
• The secrecy rate increases rapidly with the number of cooperating jamming BSs, as the UAV experiences strong interference while the legitimate receiver sees minimal degradation due to path loss and fading.
• Increasing the transmit power $P_T$ of the BSs further improves the secrecy rate, despite both legitimate and eavesdropping links being affected, because the legitimate link benefits more due to its lower path loss and better spatial diversity.
• The proposed cooperative jamming scheme effectively mitigates the threat of high-capacity UAV eavesdroppers by exploiting the directional and spatial separation between the UAV and ground users.
• The secrecy rate performance is highly sensitive to the number of cooperating BSs, indicating that even a small number of additional jammers can yield substantial security gains.
• The results demonstrate that cooperative jamming is a viable and effective physical-layer defense mechanism against LoS-dominant UAV-based eavesdropping in 5G and beyond networks.