Over the last ten years, the numbers of unmanned air vehicles (UAVs) or "drones" have changed from being just a few specialist systems, used for scientific data gathering and military purposes, to them proliferating in huge numbers. They are used across a broad range of different leisure, commercial and military activities. UAVs can be used for: movement of items in factories for manufacturing, passenger and freight transportation, can take various roles in the agriculture and forestry industries (dispensing seeds, watering and monitoring crops), remote sensing for the oil and gas industries, traffic flow monitoring, support of emergency services, hobbies, security, military and many other applications. The expansion in the use of unmanned air vehicles has come about due to the development of low cost, high performance stable platforms, employing equally low-cost communication and navigation systems supplemented by simple to use software and interfaces. Therefore, there is a need to be able to monitor the rapidly changing use of airspace, especially at low and normally neglected altitudes to ensure UAVs do not compromise safety or are used for malicious purposes. Radar is the only sensor able to perform this function on a 24-hour, all weather, wide-area basis. This book, concerned with radar surveillance of UAVs, has been compiled using contributions from the leading experts around the world to create a single body of knowledge on this important, yet still emerging, topic. It is aimed at advanced students and researchers with an interest in radar systems.
Inspec keywords: electronic countermeasures; autonomous aerial vehicles
Other keywords: counter UAS systems overview; radar countermeasures; unmanned aerial vehicles
Subjects: General electrical engineering topics; Mobile robots; General and management topics; Radar equipment, systems and applications; Education and training; Aerospace control; Telerobotics
This book concerned with radar surveillance of UAVs, has been compiled using contributions from the leading experts around the world to create a single body of knowledge on this important, yet still emerging, topic.
As many recent examples show, small unmanned aerial systems (UASs or drones,in the following) have become a real threat to both civil and military targets. While such advanced technology means a huge opportunity for the military and industry,its alternative, sinister use for criminal and terrorist purposes is no longer a fictitious risk, requiring a huge effort in terms of counteractions.
To design a system which can counter a particular threat you must, first understand what the threat is, what measures can be taken to counter it and who will operate the system to do this. In this chapter, we will consider these points first of all and then discuss how the various possibilities can be translated into requirements for a radar system.
This chapter presents a comprehensive survey of published work on millimetrewave radar applied to UAV detection and classification, covering the reported radar systems, UAV RCS characteristics and micro-Doppler signatures. Recent work onthe specific signatures of UAVs equipped with a variety of threat payloads is thendescribed. The chapter concludes with a review of classification methods that areapplied to millimetre wave radar data in order to discriminate UAVs.
As mentioned in the first chapter of this book, *it is urgent to develop effective countermeasures, such as unmanned aerial vehicle (UAV) detection, localization,tracking, recognition and interception systems, to eliminate the potential risks causedby UAVs. A radar exhibits unique advantages for the detection and tracking ofUAVs and lends itself well to a potentially effective countermeasure for UAVs.
Passive Bistatic Radars (PBRs) use non-cooperative illuminators of opportunity to detect, localise and track targets. They have attracted considerable research interest in recent years because they can be operated and deployed at a relatively low cost, are difficult to detect and hence allow covert operations in hostile environments and do not require the allocation of an increasingly more congested frequency spectrum. Various analogue and digital communication systems (such as television (TV) and radio broadcast systems) have been studied and exploited as illuminators of opportunity. However, despite the extensive work carried out on PBRs that exploit random communication signals, there has been limited research investigating the use of existing non-cooperative radar systems as illuminators of opportunity. The exploitation of radar signals to achieve passive bistatic detection is attracting attention as it may offer significant advantages. Because common radar waveforms are deterministic, a reference channel is essentially not required to detect a target. Prior intelligence or live estimations of the deterministic waveform design parameters allow the passive receiver to be matched with the illuminator of opportunity and thus generate a range-Doppler map. Radar signals are also designed for detection and provide Doppler tolerance, large bandwidths (which provide good range resolutions) and good compression ratios. This chapter presents a PBR solution that exploits non-random signals transmitted by a non-cooperative staring radar system to detect drones. Staring radars offer a constant illumination of the volume under surveillance and, unlike radar systems that deploy a rotating antenna, offers a continuous signal of opportunity. They are very attractive illuminators in particular for short-range applications to detect low Radar Cross Section (RCS) and slow moving targets.
This chapter reports the latest results of DVB-T-based PR for counter-drone operations obtained by the research groups of the University of Alcala ́ and Sapienza University of Rome. First, Section 6.2 reports a power budget analysis to provide a preliminary evaluation of the expected coverage of PR against drones. Subsequently, Section 6.3 describes the adopted processing scheme, properly tailored to be effective in the considered application. In detail, a significant effort has been devoted to the disturbance cancellation stage that represents one of the key stages within a conventional PR processing scheme. The Neyman-Pearson (NP) detector approximation and clutter modelling are investigated in Section 6.4. Multi-channel strategies, based on frequency and spatial diversity, are presented inSection 6.5 to improve the detection and localization performance of PRs. Finally,Section 6.6 draws our conclusion.
In this chapter, the concept of a multiband PR system has been investigated. The multisensory PR architecture has been introduced and appropriate waveforms of opportunity have been selected to provide the system with multitasking capabilities. Specifically, the proposed system exploits DVB-T signals to guarantee the required coverage against drones and UAV, whereas DVB-S and WiFi-based systems offer improved accuracy in terms of target localization and characterization at shorter range. Finally, the WiFi band operation allows to include in the fusion logic also device-based localization techniques based on the radiation spontaneously emitted by commercial drones.
Among the potential illuminators of opportunity (IO) that can be used in a passive radar (PR) aimed at unmanned aerial vehicle (UAV) monitoring, Global Navigation Satellite Systems (GNSS) represent an interesting alternative providing a global coverage opportunity, known signal structure and scalability. This chapter investigates the capabilities of a GNSS-based PR for UAV detection by looking at relevant aspects such as coverage, power budgets and processing schemes. Finally, the chapter experimentally assesses the capability of this type of system to detect UAVs.
This chapter reviews the similarities and differences between microUnmanned Aerial Vehicles (UAVs), also referred to as drones, and bird targets from the signals they present to radar sensors. The proliferation of small UAV platforms for commercial and personal use has increased significantly in recent years. All projections show a rapid increase in the utilisation of commercial civilian microUAV platforms across a number of use cases. In the United States alone, the consumer drone market was valued at $355M in 2015 and projected to be worth approximately $4BN by the year 2024. This broad range of applications includes capture of high-quality video imagery, remote surveying capabilities, delivery, agriculture and racing. The trend of increasing use has been highlighted in many journal articles and news reports, it seems that drones are here to stay in society andwill only increase in their integration in everyday life.
In this chapter, we present two algorithms for radar recognition of multiple UAVs. In the first algorithm, the cadence frequency spectrum (CFS) is presented as a kind of low-dimensional feature of small UAVs. The CFS of each class of small UAVs is obtained by accumulating cadence-velocity diagram (CVD) over the Doppler frequency axis. Experimental results on measured data have demonstrated that the combination of CFS and the K-means classifier is capable of providing asatisfactory accuracy of recognition of multiple UAVs. The second algorithm isbased on dictionary learning. The dictionary is learnt by performing the K-SVD algorithm on the CVDs of training samples from each class of small UAVs. Then, the learnt dictionaries of all the classes are combined together to generate a recognition dictionary. In this way, the problem of recognition of multiple UAVs isconverted to a problem of signal decomposition. Subsequently, the sparse representation of the CVD from multiple UAVs is solved by the orthogonal matchingpursuit (OMP) with the learnt recognition dictionary and recognition result isobtained by evaluating the magnitudes of the sparse representation coefficients. Experimental results on measured radar data have demonstrated the effectiveness of the dictionary learning-based algorithm for the recognition of multiple UAVs and the superiority of the dictionary learnt from CVD over that learnt from a time-frequency spectrogram. It is worth emphasizing that the training data sets used inthese two algorithms only consist of radar echoes reflected from every single class of small UAVs.
This chapter presents a summary of radar-based classification approaches developed for small drones carrying payloads. Specific focus is given to three types oftechniques that were validated on the same multistatic radar data set collected usingthe University College London (UCL)-netted radar NetRAD. These techniquesused, respectively, features extracted from the centre of mass and bandwidth of themicro-Doppler signatures; different radar data domains generated from the micro-Doppler data to be processed by pretrained Convolutional Neural Networks(CNNs) and spectral kurtosis analysis on the micro-Doppler.
In this chapter, we briefly describe the staring radar concept and show how itcan be used to detect and discriminate drones against a severe background of other targets such as birds and ground clutter. The methods developed for collecting ground-truth data from both control and opportune targets are described.Throughout, examples from real-world measurements will be used to illustrate how to generate labelled training data as well as demonstrating the target recognition performance with a supervised learning-based approach.
The aim of this book was to provide an overview of existing challenges and solutions in the field of counter unmanned aerial vehicles (UAVs) from both an industrial and an academic perspective, with a particular focus on radar techniques.