PaperlayerView exact revision on arXiv

Source-linked AI summary

Physical Layer Security in UAV Systems: Challenges and Opportunities

Xiaofang Sun, Derrick Wing Kwan Ng, Zhiguo Ding, Yanqing Xu, Zhangdui Zhong

arXiv:1909.06785v1cs.ITeess.SP

TL;DR

The paper examines UAV physical-layer security issues and attacks, then surveys approaches including trajectory design, resource allocation, and cooperative UAVs. It also discusses NOMA, MIMO, and millimeter-wave techniques for jointly improving spectral efficiency and security, with an example highlighting UAV flexibility.

  • Problem

    UAV physical-layer security issues remain important because unresolved security issues constrain communication and computation capabilities.

  • Method

    The paper surveys passive and active eavesdropping attacks and discusses trajectory design, resource allocation, cooperative UAVs, NOMA, MIMO, and millimeter-wave security approaches.

  • Results

    NOMA always outperforms OMA in the considered scenarios, while jointly designing UAV trajectory and resource allocation demonstrates advantages from UAV flexibility.

  • Takeaways & Limitations

    NOMA, MIMO, and millimeter-wave techniques can support security provisioning while improving spectral efficiency in UAV systems.

Abstract

from arXiv · show

Unmanned aerial vehicle (UAV) wireless communications have experienced an upsurge of interest in both military and civilian applications, due to its high mobility, low cost, on-demand deployment, and inherent line-of-sight (LoS) air-to-ground channels. However, these benefits also make UAV wireless communication systems vulnerable to malicious eavesdropping attacks. In this article, we aim to examine the physical layer security issues in UAV systems. In particular, passive and active eavesdroppings are two primary attacks in UAV systems. We provide an overview on emerging techniques, such as trajectory design, resource allocation, and cooperative UAVs, to fight against both types of eavesdroppings in UAV wireless communication systems. Moreover, the applications of non-orthogonal multiple access, multiple-input and multiple-output, and millimeter wave in UAV systems are also proposed to improve the system spectral efficiency and to guarantee security simultaneously. Finally, we discuss some potential research directions and challenges in terms of physical layer security in UAV systems.

I. INTRODUCTION

UAV communications offer flexible, on-demand connectivity through mobility and LoS channels, but their openness and constrained resources create significant security challenges. The paper surveys physical-layer-security approaches and advanced technologies for addressing these challenges.

  • UAVs support swift, cost-efficient deployment and flexible three-dimensional mobility for emergency and temporary communication scenarios.Applications include disaster rescue, remote sensing, and firefighting.
  • LoS air-to-ground channels can make channel state information depend directly on node positions, supporting high-speed communication design.The LoS component dominates many practical scenarios, particularly rural areas and moderate UAV altitudes.
  • Limited battery storage, small size, and a typical communication-UAV weight below 25 kg restrict communication, computation, and cruising duration.These constraints motivate security methods with lower computational and energy demands.
  • Open LoS channels expose transmitted signals to interception, while UAV transceivers remain vulnerable to malicious jamming attacks.Consequently, secure information transfer is a central concern in UAV wireless communications.
  • Traditional encryption can impose high computational complexity and energy consumption, whereas physical layer security exploits wireless-channel randomness with greater computational efficiency.The paper presents physical layer security as an alternative suited to UAV constraints.
  • The paper surveys passive and active security attacks, trajectory and resource-allocation solutions, cooperative UAVs, and applications of NOMA, 3D beamforming, and mmWave.It also identifies unresolved research challenges and future directions for secure UAV communications.

II. SECURITY ATTACKS IN UAV SYSTEMS

UAV physical-layer security is evaluated through wiretap-channel concepts and secrecy metrics, with LoS channels creating vulnerability when the UAV transmits. The same LoS property can improve security when the UAV receives.

  • A typical physical-layer-security model contains a legitimate transmitter, legitimate receiver, and potential eavesdropper connected by a wiretap channel.Secrecy capacity and secrecy outage probability are the two fundamental evaluation metrics.
  • Secrecy capacity is defined from the Shannon capacities of the main and eavesdropping channels as CS = [CM − CE]+.Here, [x]+ equals max{x, 0}.
  • When the UAV is a legitimate transmitter, LoS air-to-ground channels facilitate reception by both the intended receiver and eavesdropper, increasing eavesdropping vulnerability.The paper therefore focuses mainly on this transmitter scenario when investigating eavesdroppers.
  • When the UAV acts as a legitimate receiver, LoS channels enhance physical-layer security, particularly against passive eavesdropping.Thus, the security effect of LoS depends on the UAV's communication role.

A. Passive Eavesdropping Scenarios

Passive eavesdroppers silently intercept confidential data, making their positions difficult to obtain and leaving UAV systems vulnerable. Security responses depend on whether complete, partial, or no position information is available.

  • Passive eavesdroppers intercept confidential data without degrading the legitimate receiver’s signal and usually remain silent.
  • 1) Full Position Information of Eavesdroppers: With stationary eavesdroppers, cameras or synthetic aperture radar can support position tracking and resource allocation for security.Obtaining precise positions requires costly hardware and adds equipment load to the UAV.
  • 2) Partial Position Information of Eavesdroppers: For quasi-stationary or regionally moving eavesdroppers, partial position information can be obtained through sensing and tracking, but hardware and energy costs remain high.Worst-case secrecy capacity and secrecy outage probability are common evaluation metrics in this setting.
  • 3) No Position Information of Eavesdroppers: When eavesdroppers hide physically, the UAV may lack position information entirely, making perfectly secure communications impossible to guarantee always.

B. Active Eavesdropping Scenarios

Active eavesdroppers are more harmful because they can degrade the legitimate channel, improve the eavesdropping channel, or perform both through full-duplex or cooperative half-duplex operation.

  • Active eavesdroppers attack security by transmitting jamming signals that degrade the legitimate channel capacity or improve the eavesdropping channel capacity.
  • Full-duplex eavesdropper: A full-duplex eavesdropper simultaneously transmits jamming noise and intercepts confidential signals.The UAV may increase transmit power, which can facilitate eavesdropping over LoS channels and consume more energy.
  • Cooperative half-duplex eavesdroppers: Multiple half-duplex eavesdroppers can cooperate by having some transmit jamming toward the legitimate receiver while others intercept the confidential signal.
  • When active eavesdroppers possess the legitimate receiver’s CSI, UAV systems become severely vulnerable because the attacks can efficiently interfere with the legitimate receiver.
  • Possible solutions: UAV mobility, flexible positioning, and advanced resource allocation are presented as properties and techniques for improving secrecy against active eavesdroppers.

SPECIFIC SECURITY ATTACKS AND POSSIBLE SOLUTIONS IN UAV WIRELESS COMMUNICATION SYSTEMS

The paper surveys trajectory design and joint trajectory–resource allocation as physical-layer security strategies for UAV communications. Joint optimization can coordinate movement, speed, and transmit power around legitimate users and eavesdroppers.

  • Joint Trajectory and Resource Allocation Design: Joint trajectory and resource allocation design aims to improve security and spectral efficiency under practical UAV constraints.The considered relay scenario maximizes system spectral efficiency while guaranteeing secure communications.
  • Joint Trajectory and Resource Allocation Design: The UAV flies at full speed away from an eavesdropper, then slows and increases transmit power near the legitimate receiver for confidential transmission.
  • Joint Trajectory and Resource Allocation Design: In the illustrated scenarios, the UAV first approaches the transmitter to cache data, then approaches the legitimate receiver while avoiding trajectories near the eavesdropper.
  • Joint Trajectory and Resource Allocation Design: When closer to the eavesdropper, the UAV caches data or remains silent and moves away; when closer to the legitimate receiver, it transmits confidential signals with positive secrecy rate and hovers above the receiver.
  • Joint Trajectory and Resource Allocation Design: The paper concludes that joint trajectory and resource allocation can enhance physical-layer security and improve spectral efficiency in UAV systems.

2) Robust Joint Design:

Robust joint design addresses uncertain eavesdropper locations by optimizing trajectory and resource allocation for worst-case security while adjusting transmission around legitimate users and uncertain regions.

  • Robust Joint Design: An uncertain eavesdropper location can be modeled as a region centered on the exact location, with radius determined by the amount of uncertainty.
  • Robust Joint Design: Worst-case robust design guarantees system QoS whenever the eavesdropper lies within the modeled uncertainty region.
  • Robust Joint Design: The optimized UAV flies close to the legitimate receiver to enhance the legitimate channel while cruising away from the eavesdropper’s uncertain region.
  • Robust Joint Design: When approaching an uncertain eavesdropper area, the UAV decreases or shuts down transmit power and accelerates away; near legitimate users, it uses higher transmit power.
  • Artificial Noise: Artificial noise transmitted through cooperating UAVs can increase confidential-channel capacity while reducing eavesdropping-channel capacity.Because artificial noise consumes transmit power, less power remains for the confidential signal.
  • Resource Allocation: Power allocation is important for secrecy capacity, but optimal allocation commonly requires solving non-trivial NP-hard problems, motivating computationally efficient suboptimal solutions.

B. Anti-Jamming Techniques

The paper presents cooperative UAV and multi-antenna strategies that adapt trajectory, resource allocation, and transmission design to eavesdropper-position knowledge. These methods strengthen legitimate reception while reducing information leakage or secrecy outage.

  • Cooperative transmission: Cooperative multi-point transmission lets UAVs strengthen legitimate signals, weaken eavesdropper reception, and assign different UAVs to confidential transmission or jamming.The UAVs jointly optimize trajectories and resource allocations for these roles.
  • Known eavesdropper positions: Multiple UAVs can form a virtual antenna-array through joint trajectory and resource allocation design when eavesdropper positions are known.This focuses information energy beams on legitimate receivers while reducing leakage.
  • Partial eavesdropper information: With partial or statistical eavesdropper-position information, robust joint trajectory and resource allocation can improve secrecy rate or reduce secrecy outage probability.The outage probability can be reduced below a certain level.
  • Absent eavesdropper information: Without eavesdropper-position information, cooperative jamming in the legitimate receiver’s orthogonal channel space and position adjustment degrade eavesdropper signals.Multi-antenna methods further use spatial degrees of freedom to improve legitimate reception and reduce leakage.
  • Millimeter-wave security: Millimeter-wave systems combine large bandwidth, line-of-sight channels, and directional transmission to help avoid eavesdropping and resist active attacks.Frequency hopping hides confidential data without requiring CSI and uses frequency diversity against jamming.

IV. ADVANCED APPROACHES FOR UAV SYSTEMS

The paper surveys NOMA as an advanced UAV technique for jointly improving spectral efficiency and physical-layer security. Its optimization balances public and confidential rates, while performance depends on eavesdropper proximity and non-convex design constraints.

  • Overview: NOMA is proposed to improve spectral efficiency while supporting secure UAV communications alongside beamforming and millimeter-wave techniques.The paper discusses these techniques as advanced approaches for physical-layer security.
  • Relay scenario: In relay scenarios with different security-clearance levels, NOMA forwards confidential and public information simultaneously when the potential eavesdropper has a bad channel.The lower-clearance receiver is treated as a potential eavesdropper.
  • Optimization: Joint optimization of transmission scheme, resource allocation, and UAV trajectory is generally non-convex, so an iterative successive-convex-approximation algorithm provides a computationally efficient suboptimal design.The mode selection between NOMA and unicast is chosen from the resource-allocation optimization.
  • NOMA versus OMA: NOMA always outperforms OMA in the considered scenarios, demonstrating a spectral-efficiency advantage in balancing public and confidential data rates.The comparison uses rate regions under different cruising durations.
  • Eavesdropper proximity: The accumulated public data rate drops rapidly when the eavesdropper is sufficiently close to the legitimate receiver.When secrecy-rate requirements cannot be satisfied, the accumulated public data rate is set to zero as a security-failure penalty.

B. Enhancing Physical Layer Security by Multi-antenna Technology

Multi-antenna beamforming exploits spatial degrees of freedom to strengthen legitimate reception and suppress eavesdroppers. Three-dimensional beamforming and millimeter-wave techniques offer additional directional and spectral-efficiency benefits, subject to channel and CSI conditions.

  • Multi-antenna security: Multi-antenna technology enhances secrecy by concurrently improving legitimate signal power and degrading eavesdropper signal strength.Its UAV applications are illustrated through beamforming approaches.
  • Traditional 2D beamforming: Traditional 2D beamforming can point toward the orthogonal space of an eavesdropper’s channel when complete CSI is available.This approach may underuse spatial degrees of freedom, creating a trade-off between legitimate-user strength and eavesdropper degradation.
  • UAV channel geometry: UAV altitude and line-of-sight channels enable beam separation across azimuth and elevation domains for legitimate receivers and eavesdroppers.Different altitudes and elevation angles help distinguish the receivers spatially.
  • Millimeter-wave limitation: Millimeter-wave performance depends on line-of-sight channel availability, although UAV channels facilitate high-data-rate communication through inherent line-of-sight conditions.The channel is described as sparse in the angular domain, supporting directional transmission.
  • Millimeter-wave techniques: Millimeter-wave frequency hopping hides confidential data across carrier frequencies without CSI and uses frequency diversity to improve robustness against active jamming.Its large bandwidth also supports high-data-rate and directional transmission.

V. OPEN ISSUES AND CHALLENGES

The paper identifies practical challenges involving channel realism, mobility-dependent CSI acquisition, and pilot contamination. These issues can cause inappropriate transmission strategies and motivate field validation and robust position-tracking mechanisms.

  • Practical UAV channel modeling: LoS air-to-ground models may be inaccurate in urban areas, motivating realistic channel modeling and field-test verification.The paper presents this as an open challenge for UAV security research.
  • Position and CSI acquisition: UAV mobility affects CSI acquisition for legitimate users and eavesdroppers, requiring pragmatic mechanisms to detect and track their positions.Efficient CSI algorithms must exploit physical wireless-channel properties.
  • Pilot contamination: Active eavesdroppers can transmit deterministic pilots identical to legitimate pilots, causing pilot contamination that misleads UAV transmission design.The resulting strategy may favor eavesdropper reception or bring the UAV close enough to increase leakage.

C. Physical Layer Security Against Malicious UAV Attacks

UAV mobility and flexibility create both new security opportunities and attack surfaces, including interception and jamming. The article surveys physical-layer approaches spanning trajectory and resource design, cooperation, advanced access and transmission techniques, and future challenges.

  • Security challenges: UAV mobility can enhance physical-layer security but also enable confidential-data interception and jamming against legitimate links.These threats create challenges beyond those posed by conventional terrestrial eavesdroppers.
  • Security challenges: Limited research has addressed UAV physical-layer security from a communication-theory perspective, motivating advanced protective techniques.The article frames this as an important aspect requiring further investigation.
  • Trajectory and resource design: Trajectory design offers a new security dimension, but its coupling with resource allocation makes optimization intractable and existing designs generally suboptimal.The performance gap between optimal and existing suboptimal solutions remains unclear, motivating algorithms that balance complexity and performance.
  • Resource constraints and cooperation: Restricted flight duration, onboard energy, and computational capability constrain sustainable secure UAV communication.Energy harvesting from solar and laser is identified as a possible on-the-fly energy supply, while UAV cooperation can share onboard resources for secure communication.
  • Proposed approaches: The article proposes approaches against both passive and active eavesdropping, including joint trajectory-resource allocation and cooperative UAV strategies.Its illustrative joint-design example demonstrates advantages from UAV flexibility; NOMA, MIMO, and mmWave are proposed to enhance security and spectral efficiency.
  • Proposed approaches: NOMA, MIMO, and mmWave are proposed for UAV systems to enhance physical-layer security and spectral efficiency simultaneously.The article also identifies potential research directions and challenges in UAV physical-layer security.
Loading 1909.06785v1…