PaperlayerView exact revision on arXiv

Source-linked AI summary

Counter-Unmanned Aircraft System(s) (C-UAS): State of the Art, Challenges and Future Trends

Jian Wang, Yongxin Liu, Houbing Song

arXiv:2008.12461v1eess.SP

TL;DR

UAS growth and reliance on computer and communication technologies create public-safety, national-security, and privacy risks. The paper surveys C-UAS detection and mitigation literature, identifies counter-UAS challenges, and evaluates technology trends. It concludes that systematic C-UAS development is important for safe airspace integration, while surveyed approaches face deployment and operational constraints.

  • Problem

    UAS-based threats place public safety, national security, and individual privacy at risk, creating an urgent need for technologies that detect, track, identify, and mitigate unsafe or unauthorized UAS.

  • Method

    The paper provides a comprehensive survey of C-UAS literature, identifies counter-UAS challenges, and evaluates detection and mitigation trends.

  • Results

    The survey covers acoustic, vision, passive RF, radar, and data-fusion detection, alongside physical capture and jamming mitigation, while identifying deployment constraints and technology vulnerabilities.

  • Takeaways & Limitations

    Systematic C-UAS technology development is presented as a foundation for safe, secure, and privacy-respecting integration of UAS into the airspace system.

Abstract

from arXiv · show

Unmanned aircraft systems (UAS), or unmanned aerial vehicles (UAVs), often referred to as drones, have been experiencing healthy growth in the United States and around the world. The positive uses of UAS have the potential to save lives, increase safety and efficiency, and enable more effective science and engineering research. However, UAS are subject to threats stemming from increasing reliance on computer and communication technologies, which place public safety, national security, and individual privacy at risk. To promote safe, secure and privacy-respecting UAS operations, there is an urgent need for innovative technologies for detecting, tracking, identifying and mitigating UAS. A Counter-UAS (C-UAS) system is defined as a system or device capable of lawfully and safely disabling, disrupting, or seizing control of an unmanned aircraft or unmanned aircraft system. Over the past 5 years, significant research efforts have been made to detect, and mitigate UAS: detection technologies are based on acoustic, vision, passive radio frequency, radar, and data fusion; and mitigation technologies include physical capture or jamming. In this paper, we provide a comprehensive survey of existing literature in the area of C-UAS, identify the challenges in countering unauthorized or unsafe UAS, and evaluate the trends of detection and mitigation for protecting against UAS-based threats. The objective of this survey paper is to present a systematic introduction of C-UAS technologies, thus fostering a research community committed to the safe integration of UAS into the airspace system.

I. INTRODUCTION

UAS growth and increasing reports of unsafe or unauthorized operations create risks to public safety, national security, and privacy. The paper surveys C-UAS detection and mitigation technologies, challenges, and trends to support safer airspace integration.

  • UAS provide potential safety, efficiency, research, recreational, commercial, and emergency-response benefits.Examples include aerial photography, package and medical-supply delivery, and search-and-rescue support.
  • UAS threats arise from reliance on computer and communication technologies and affect public safety, national security, and individual privacy.Threats include unsafe operations near aircraft or crowds, operations over sensitive facilities, and cameras directed into private residences.
  • A C-UAS system lawfully and safely disables, disrupts, or seizes control of an unmanned aircraft or system, typically through detection and mitigation subsystems.The ideal detection subsystem detects, tracks, and identifies UAS; the ideal mitigation subsystem minimizes collateral damage and engagement cost.
  • Research has explored acoustic, vision, passive RF, radar, data-fusion, physical-capture, jamming, and destruction approaches, but existing efforts remain limited in scalability, modularity, or affordability.These technologies address detection, tracking, identification, and mitigation of unauthorized or unsafe UAS.
  • The survey reviews C-UAS literature, identifies counter-UAS challenges, and evaluates detection and mitigation trends for protecting against UAS-based threats.Its stated objective is a systematic introduction of C-UAS technologies that supports safe integration into the airspace system.
  • A. Reported UAS Sightings: UAS sightings increased dramatically over the past five years, with monthly sightings shown from January 2014 through December 2019 and many occurring during summer months.The paper also reports sightings in most populated U.S. states.

B. UAS-based Threats

The paper classifies UAS-based threats into public safety, national security, and individual privacy categories. These threats include collision and control risks, malicious payloads and surveillance, and privacy invasion or harassment.

  • UAS-based threats are classified into public safety, national security, and individual privacy categories.The classification organizes the threat types discussed in the paper.
  • Public Safety: Public safety threats include operating UAS near aircraft, airports, crowds, public events, emergencies, or while impaired.The unmanned nature of operations also creates risks from limited pilot visibility and communications-link failure.
  • National Security: National security threats include transporting contraband or weaponized payloads, prohibited surveillance and reconnaissance, and intellectual-property theft.These threats target sensitive facilities, information, or people through airborne access and observation.
  • Individual Privacy: Privacy threats involve intentional disruption, harassment, or invading individuals’ privacy with UAS.The paper specifically identifies camera use directed inside private residences as a privacy-invading operation.
  • Recent threat examples are documented with their occurrence time and location, threat category, and corresponding consequences.The paper presents these dimensions in a table of recent UAS-based threats.

C. Airspace Restrictions Applicable to UAS Flights

UAS restrictions and rising sightings motivate integrated detection and mitigation systems. The paper surveys detection technologies and emphasizes deployment constraints, including environmental sensitivity, antenna spacing, computation, and vision-device design.

  • Airspace Restrictions: Security-sensitive airspace restrictions prohibit UAS operations from the ground up to 400 feet above ground level and apply across UAS types and purposes.Stadium restrictions apply during crowd gatherings, while Washington, DC has a 30-mile spatial restriction circle.
  • Airspace Restrictions: FAA rules require UAS weighing more than 0.55 pounds to be registered, while work or business flights under 55 pounds require remote-pilot certificates.The rules vary by user category, including recreational, commercial, public-safety, government, and educational operators.
  • Need for Integrated C-UAS: Rising sightings despite FAA guidelines create an urgent need for integrated systems that detect and negate UAS for safe, secure, privacy-respecting airspace operations.The paper identifies detection and mitigation as the two main technology areas needed for this integration.
  • Detection and Mitigation Requirements: Ideal detection should detect, track, and identify UAS with a small footprint, highly automated operations, and location functions.Ideal mitigation should lawfully and safely disable, disrupt, or seize control while limiting collateral damage and cost per engagement.
  • Detection Technologies: Since 2014, UAS detection research has used acoustic, vision, passive RF, radar, and data-fusion technologies, with data fusion described as most popular and acoustic approaches least popular.The paper presents this evolution in Fig. 3 and discusses each technology’s advantages and disadvantages.
  • Acoustic Detection: Acoustic detection captures UAS sound, extracts signal features, and uses them to determine whether UAS are approaching.Reported approaches analyze power or frequency spectra and may use machine learning, but performance can depend on weather conditions.
  • Acoustic Detection: Acoustic approaches can recognize and locate UAS precisely, but their nature limits deployment and detection at large scale.The paper notes that machine learning may improve acoustic detection performance.

B. Passive RF based Detection

Passive RF detection uses transmissions between UAS and remote controllers as evidence for detecting and localizing UAS. Approaches include spectrum analysis, traffic-pattern analysis, motion-pattern analysis, and RF localization, but their applicability depends on communication characteristics and operating conditions.

  • RF sensing: UAS communication links provide spectral patterns that passive RF receivers use to detect and localize aircraft.Software Defined Radio receivers are commonly used to intercept RF channels.
  • Spectrum analysis: ANN-based RF detection uses improved slope, skewness, and kurtosis features to recognize UAS signal spectra.The reported approach outperformed recognition technologies based on the individual spectrum features.
  • Traffic analysis: Traffic-based methods identify UAS from packet-length distributions, WiFi fingerprints, and other communication characteristics.These methods passively monitor wireless signals between UAS and controllers.
  • Protocol robustness: Spectrum- and traffic-based methods may fail to identify UAS using unknown telemetry protocols, motivating motion-pattern analysis from radio signals.Motion-pattern approaches seek protocol-independent evidence of UAS presence.
  • Localization: Passive RF systems also estimate direction of arrival and demonstrate the feasibility of detecting and locating small UAS with FPGA-based SDR hardware.Localization is treated as a distinct component of the detection procedure.
  • Vision-based comparison: Vision-based detection captures UAS images or video and applies segmentation, neural networks, thermal sensing, or dynamic vision to distinguish aircraft from backgrounds and birds.Deployment remains challenging across variable environments, weather conditions, mobility levels, and bird-confusion scenarios.

D. Radar based UAS Detection

Radar-based UAS detection exploits range, velocity, Doppler, and reflection signatures, using active or passive architectures with specialized signal processing. Higher-frequency carriers, MIMO arrays, noise radar, SDR platforms, and existing illuminators address small, slow targets, but deployment and processing remain limiting factors.

  • Active radar: Conventional radar struggles with small, slow UAS because it is designed primarily for larger aircraft with radar cross-sections above 1 m^2.Radar nevertheless offers day-and-night operation, weather independence, and simultaneous range and velocity measurement.
  • Active radar: Higher-frequency carriers and MIMO beamforming are the two principal strategies for increasing conventional radar resolution in UAS surveillance.Reported systems include X-band, W-band, and 24 GHz FMCW or UWB designs.
  • Active radar: MIMO radar can detect micro-UAS with lower carrier frequencies, including a small hexacopter detected using a 32 by 8 element L-band receiver array.An 8.75 GHz X-band FMCW radar is another reported implementation.
  • Specialized radar: Noise radar and SDR-based multimode radar target slow UAS with cost-efficient or configurable architectures, but SDR performance depends strongly on backend processing.Analog implementations showed higher updating rate and signal-to-noise ratio than the tested digital implementations.
  • Active radar limitations: Active radars require specially designed transmitters, making them harder to deploy and vulnerable to anti-radiation attacks.This is identified as an apparent drawback of active radar architectures.
  • Passive radar: Passive radar reuses sources such as cellular, WiFi, and digital television signals, with single-station and distributed synthetic architectures.Distributed systems use cellular or DVB infrastructures as illumination sources.
  • Passive radar: Passive radar studies use micro-Doppler signatures for detection and classification, including evidence that plastic propellers produce less visible signatures than carbon-fiber propellers.Passive radar can detect and track UAS, but acceptable accuracy may require substantial post-processing or multiple receivers.
  • Signal processing: Radar signal processing is organized around conventional signal features and learning-based pattern recognition, including micro-Doppler analysis, CNNs, and deep belief networks.These methods derive UAS features from noisy RF-front-end outputs and Doppler or spectral-correlation representations.

3) Posterior Signal Processing:

Posterior signal processing and data fusion address the limitations of single-sensor UAS detection by extracting target features and combining complementary sensing modalities. The survey organizes fusion into multiple-sensor, multiple-type-sensor, and multiple-algorithm approaches, while noting deployment-consistency challenges.

  • Signal processing: Radar posterior processing separates conventional signal-feature detection from learning-based pattern recognition for extracting weak UAS reflections from noisy RF outputs.Micro-Doppler signatures and neural networks are representative techniques.
  • Data fusion: Data fusion integrates multiple sources to produce information more consistent, accurate, and useful than any individual input.The stated motivation is to combine the strengths of different detection approaches.
  • Fusion categories: Fusion research is classified into multiple-sensor, multiple-type-sensor, and multiple-sensing-algorithm data fusion.The classification responds to limitations in detection range, accuracy, scenario robustness, and computational efficiency.
  • Multiple-sensor fusion: Multiple-sensor fusion combines distributed measurements, such as acoustic phase differences or RF signals from multiple antennas, to improve localization and detection functions.Sensor weighting and receiving-phase adjustment are used to improve accuracy.
  • Multiple-type-sensor fusion: Multiple-type-sensor fusion combines complementary sensors because single sensors cannot consistently cover variable environments, detection ranges, and requirements at acceptable cost.The survey describes combining long- and short-range technologies to improve accuracy and distance coverage.
  • Algorithm fusion: Sensing-algorithm fusion adapts feature extraction and recognition algorithms to sensor types and environmental scenarios.One example uses unsupervised learning to extract features from acoustic sensors under bird, aircraft, weather, and UAS conditions.
  • Deployment challenge: Data-fusion approaches show advantages over single methods, but achieving consistent detection across differently deployed schemes remains a challenge.The survey links deployment design to varying distances from restricted areas and heterogeneous detection requirements.
  • Mitigation outcomes: Mitigation distinguishes capturing and retrieving a UAS from disabling and dropping it, with RF jamming and hacking identified as technologies supporting retrieval.EMP, RF jamming, and hacking can disable sensors, circuits, control systems, or communications, and may also cause a crash.

IV. STATE OF THE ART MITIGATION

The survey characterizes UAS mitigation as immature and organizes it into physical capture, jamming, and exploitation of system vulnerabilities. Physical capture can disable mobility, while jamming disrupts sensors or communications but remains broad, energy-intensive, and slow to control precisely.

  • Mitigation categories: UAS mitigation is organized into physical capture, noise-based jamming, and exploitation of system or sensor vulnerabilities to acquire control.The architecture uses initial target, detection, and neutralization ranges to characterize engagement response.
  • Physical capture: Net capture disables drone mobility by deploying a net from a gun or specialized weapon when an unauthorized or unsafe UAS is located.Deployable net systems can be installed on aircraft or authenticated UAS.
  • Physical capture: λ = 2.5cm spatio-temporal electromagnetic pulses reportedly neutralized onboard radio-electronic systems over a 0.5 to 1 km range without harming biological protection.Physical capture methods are described as lightweight, quickly assembled, efficient, and low cost, but potentially harmful to pilots.
  • Jamming: Jamming interferes with UAS sensors or systems using noise signals, and the survey classifies jamming into three main categories.The reported categories include tone, sweep, and protocol-aware signals.
  • Network attacks: Network-oriented mitigation includes jamming, black-hole, replay, and LTE-focused attacks against UAS communications.One reported experiment approximated the efficient LTE-UAS jamming range at 60 m.
  • Jamming limitations: Jamming can neutralize drones without physical damage and operate at scale, but current systems are omnidirectional, energy-consuming, and slow to take effect.The survey identifies directional, controlled, and faster response as future requirements.

C. Vulnerabilities

The paper categorizes UAS vulnerabilities around GPS, sensors, communication protocols, and operating systems, while reviewing detection and mitigation technologies and their deployment constraints.

  • UAS vulnerabilities are exploited through GPS control, sensor manipulation, and communication-protocol attacks, including spoofing, modification, and intrusion.
  • Man-in-the-middle attacks can inject control commands, while cracking SDKs, reverse engineering, and GPS spoofing can hijack UAS.
  • UAS operating systems and embedded sensors expand exploitable attack surfaces, potentially increasing unauthorized-control risks.
  • Detection deployment depends strongly on sensor characteristics: acoustic and passive-RF systems face environmental, antenna, computation, or platform constraints, while vision systems support broader platforms.
  • The survey identifies millimeter-wave radar with data fusion as a promising detection direction, while physical capture is considered practical and reliable for neutralization.

A. UAS Detection

Detection technologies face environmental, hardware, computational, identification, and deployment limitations that constrain their robustness and portability.

  • Acoustic detection is affected by environmental variation, wind, and limited treatment of Doppler effects from moving UAS.
  • Passive-RF detection depends on multiple antennas, telemetry protocols, and RF front-ends, while current SDR devices are heavy, power-hungry, and poorly portable.
  • Vision and other detection approaches may have limited range or identification capacity and can be affected by illumination, interference, or high computation requirements.
  • Detection methods also exhibit constraints including limited range, high computational consumption, limited identification, encrypted-data inapplicability, and limited support for FHSS or DSSS schemes.
  • Deep learning and signal-data processing are presented as potential ways to improve RF detection accuracy, efficiency, and robustness.

3) Vision based UAS Detection:

Vision-based UAS detection remains an active research area because practical systems must address device constraints, visual ambiguity, and low-resource deployment requirements.

  • Vision detection still needs lightweight, small, and low-cost devices, along with camera-aperture adjustment to reduce sunlight-induced fading.
  • Birds and fixed-wing UAS can appear similar, motivating additional biological signals to distinguish these scenarios.
  • Low-size, weight, and power-consumption deep-learning algorithms and transfer learning remain important for portable vision systems.
  • Ground radar can support military detection, but civilian deployment in crowded stadiums and residential areas is difficult because radar systems are large and inflexible.

5) Data Fusion based UAS Detection:

The paper surveys data fusion and mitigation approaches while emphasizing practical constraints involving portability, collateral effects, targeting, power, and operating conditions.

  • Data fusion must combine video, radio, audio, and other heterogeneous formats for detecting unauthorized or unsafe UAS.
  • Physical capture is direct and deployable, but effective systems must be lightweight, scalable across UAS sizes, and easy to operate.
  • Directional EMP systems must limit harm to people, reduce beam divergence, and balance frequency-dependent effectiveness against transmission distance.
  • RF jamming consumes substantial power, affects areas rather than precise points, and fails against UAS navigating autonomously with internal GPS.
  • Current RF jamming devices are heavy, constrained in movement, and difficult to carry on surveillance UAS, motivating lightweight and efficient alternatives.
  • Hacking and spoofing approaches target outer interference, networks, and sensors, but remote operation, communication-channel recognition, and sensor-specific attacks remain open needs.

4) Hacking:

The paper identifies protocol recognition as part of hacking-related counter-UAS operations, enabling defenders to determine communication parameters and target the pilots’ command link. Figure 13 presents a unified framework for drone safety management.

  • Hacking:: Protocol recognition can identify the autopilot and communication protocols used by radio-controlled UAS.The stated purpose is to determine communication parameters and support attacks on the pilots’ command link.
  • Hacking:: Recognizing protocols in a software-defined radio could support decrypting communication packets and modifying them.
  • Hacking:: Figure 13 presents a unified framework for drone safety management.

VI. FUTURE TRENDS

The paper argues that future C-UAS should combine multiple detection and mitigation approaches with operational policies and coordinated responsibilities. It therefore proposes collaborative safety management involving authorities, manufacturers, and other entities.

  • A. Technical advancements: Simple detection approaches cannot reliably detect malicious UAS across variable environments and diverse aircraft materials and configurations.The paper recommends fusing multiple sensors with ground and aerial data.
  • A. Technical advancements: Future mitigation should combine countermeasures and prioritize navigating intruding UAS away from sensitive areas without harming pilots’ property.
  • B. Policies and standards: Operational policies can reduce mistaken operations by guiding pilots and providing standards for industry.
  • B. Policies and standards: Market-entry standards should require UAS to be physically controllable and identifiable, enabling remote intervention in restricted areas.
  • B. Policies and standards: Pilot training and certification should include safety operations and security knowledge to help avoid basic mistakes.
  • C. Unified and Secured Coordination Strategies: Because one approach cannot adequately protect the public, the paper proposes a unified framework for collaborative UAS safety management.
  • C. Unified and Secured Coordination Strategies: The framework assigns coordination roles to local UAS authorities, airspace authorities, and manufacturers.Manufacturers are expected to specify environmental requirements and provide privileged control interfaces for intervention.
  • VII. CONCLUDING REMARKS: The survey synthesizes C-UAS detection and mitigation literature, identifies countering challenges, and evaluates protection trends.
Loading 2008.12461v1…