The phrase 'waveform design and diversity' refers to an area of radar research that focuses on novel transmission strategies as a way to improve performance in a variety of civil, defense and homeland security applications. Three basic principles are at the core of waveform diversity. First is the principle that any and all knowledge of the operational environment should be exploited in system design and operation. Second is the principle of the fully adaptive system, that is, that the system should respond to dynamic environmental conditions. Third is the principle of measurement diversity as a way to increase system robustness and expand the design trade space. Waveform design and diversity concepts can be found dating back to the mid-twentieth century. However, it has only been in the past decade or so, as academics and practitioners have rushed to exploit recent advances in radar hardware component technology, such as arbitrary waveform generation and linear power amplification, that waveform diversity has become a distinct area of research. The purpose of this book is to survey this burgeoning field in a way that brings together the diverse yet complementary topics that comprise it. The topics covered range from the purely theoretical to the applied, and the treatment of these topics ranges from tutorial explanation to forward-looking research discussions. The topics treated in this book include: classical waveform design and its extensions through information theory, multiple-input multiple-output systems, and the bio-inspired sensing perspective; the exploration of measurement diversity through distributed radar systems, in both cooperative and non-cooperative configurations; the optimal adaptation of the transmit waveform for target detection, tracking, and identification; and more. This representative cross-section of topics provides the reader with a chance to see the three principles of waveform diversity at work, and will hopefully point the way to further advances in this exciting area of research.
Inspec keywords: filtering theory; radar antennas; target tracking; MIMO radar; passive radar; radar tracking; road vehicle radar; sensor placement
Other keywords: space-time diversity; waveform-diverse radar pulse compression; passive bistatic radar waveform design; filter design; radar target classification; radar tracking; active antenna system; automotive radar system; noncooperative radar network; sensor placement; radar target detection; MIMO radar
Subjects: Radar equipment, systems and applications; Wireless sensor networks; Filtering methods in signal processing; Single antennas; General electrical engineering topics
IEEE Radar Standard P686/D2 (January 2008) defines waveform diversity as 'Adaptivity of the radar waveform to dynamically optimize the radar performance for the particular scenario and tasks. May also exploit adaptivity in other domains, including the antenna radiation pattern (both on transmit and receive), time domain, frequency domain, coding domain, and polarization domain'. As this definition indi cates, the term `waveform diversity' does not refer to a tangible object but to a remote sensing paradigm. The basic elements of the paradigm are measurement diversity, knowledge-aided processing and design, and transmitter adaptivity. The waveform diversity paradigm arose from the insatiable demands for remote sensing performance that are always present in military applications, and the application of wave form diversity has led to many interesting and promising remote sensing concepts. A host of recent activity testifies to the interest and promise of waveform diversity within the radar community.
The waveform determines the delay-Doppler response of a radar system. From that response, one can derive the radar's range and velocity resolution and their ambiguities. This chapter explains the concept and motivation for pulse compression. It then describes narrow-band signals and their major signal processing and analysis tools - the matched filter and the ambiguity function. These tools are then used to study classical pulse signals such as unmodulated rectangular pulse, linear-FM pulse and binary and polyphase-coded pulses. The key for Doppler resolution - the coherent pulse train - is then analysed. Additional topics are reduction of sidelobes (delay and spectrum); inter-pulse diversity, multicarrier waveforms; and periodic continuous waveforms (CW).
In the last decade, a number of researchers have used information-theoretic ideas and maximization of mutual information in the design of radar waveforms for adaptive waveform radar and multiple-input multiple-output (MIMO) radar. However, it is not clear under what circumstances these approaches lead to optimal or near-optimal results. In this chapter, we review the fundamental ideas behind the use of information theory and information measures in radar waveform design. We also briefly review some of the more recent results in this area.
In this chapter, we present the concept of the multistatic ambiguity function and how it can be used to develop sensor placement strategies in multistatic radar systems. The multistatic ambiguity function provides a complete description of a given multistatic system and serves as a perfect link between the system parameters and performance measures. It has been successfully used in the literature to assess and design receiver weighting and waveform selection rules. We put emphasis on adequate sensor placement as a way of improving multistatic radar system performances. We present several simulation results that illustrate the significance of proper sensor placement in system configurations with a single transmitter and multiple receivers. We analyse both the case when it is possible to reposition the transmitter (for fixed receivers) and the case when it is possible to move some receivers (for a fixed transmitter) in order to achieve the best system resolution. We also provide several examples that demonstrate potential benefits of combining waveform selection, receiver weighting and sensor placement strategies in multistatic radar systems.
In this chapter, we review various probing waveform transmission schemes for multiple-input multiple-output (MIMO) radar with co-located antennas. An orthogonal probing waveform set is required to separate the transmitted waveforms at the receiver side to achieve a large virtual array size afforded by the MIMO radar. This increased virtual aperture size provides MIMO radar systems with many advantages, including better spatial resolution, improved parameter identifiability and enhanced performance for ground moving target indication (GMTI) and radar imaging. We discuss several MIMO radar transmission schemes herein, including fast-time code division multiple access (FT-CDMA), frequency-division multiple access (FDMA), time division multiple access (TDMA), randomized TDMA (R-TDMA), Doppler division multiple access (DDMA) and slow-time CDMA (ST-CDMA). The merits and limitations of these transmission schemes will be discussed, and brief examples will be presented for illustration purposes.
Passive bistatic radar (PBR), exploiting broadcast, communications or radionavigation signals, has received a great deal of attention in recent years, and has a number of attractions compared with conventional radar systems. However, the waveforms are not fundamentally designed for radar operation, so their performance will in general be suboptimal. It is therefore important to understand the effect of the waveform on the performance of the PBR, so as to be able to choose the most appropriate illuminator and to use the waveform in the optimal way, and it is in this sense that PBR forms a subject of waveform diversity. This chapter reviews the properties of a range of different PBR waveforms and the processing methods used with them.
Echolocating mammals such as bats, whales and dolphins are able to detect, select and attack prey even in dense cluttered and often hostile environments. They have developed echolocation for over 50 million years and rely on exceptional performance for their survival. Although the frequencies and waveform parameters used by radar sensors and by echolocating mammals are not the same, there remain close parallels that suggest lessons can be learnt from nature. In this chapter, we investigate the behaviour and performance of echolocating bats in terms of detecting, locating, tracking and capturing prey. We show how echolocation calls are diversified in a dynamic and intelligent manner according to the task performed and relate the results to typical flight trajectories. We discuss how echoes may be transformed into a meaningful perception of the target and finally we explore how this information can be used to develop a new architecture for radar automatic target recognition (ATR).
Driving a car seems to be a safe action. However, there are about 5,000 fatalities on German streets every year, which is absolutely too much. All drivers have strong limitations when measuring the distance and the relative velocity of other cars, especially under bad weather conditions which are the reason for several accidents. Therefore, some technical assistance is highly welcome to every driver. The European Union has called all car manufacturers to intensify their research activities in protecting vulnerable road users and increasing the traffic safety. An automotive radar sensor in the 24 GHz frequency domain measures target range, radial velocity and azimuth angle simultaneously with high accuracy and resolution even in multitarget situations. The all-weather capability is an important additional feature of all radar systems. Therefore, this chapter considers automotive radar sensors as a basis for reliable driver assistant systems. The technical challenge is the simultaneous measurement of target range, radial velocity and azimuth angle. This task has a direct and strong relevance to the waveform design process. Therefore, the objective of this chapter is the waveform design for continuous wave radar systems. Several transmit signals are considered in detail and different proposals are discussed and compared, especially for automotive applications. Usually, the lateral velocity component cannot be measured by a radar sensor and a single measurement. This task can be of high importance in typical city traffic situations where the track direction of a car needs to be known in a very short time frame. A signal processing scheme will be shown in this chapter which indeed allows the additional measurement of the lateral velocity component based on a single observation.
Radar pulse compression is widely used to achieve high range resolution without the attendant high-peak power that would otherwise be required through the use of an unmodulated short pulse. The wide variety of different waveform structures also provides a means to distinguish the echoes from different radars that may coincide in space, frequency and time. Furthermore, the development of new waveform-diverse emission schemes relies on the ability to adequately separate the various emission components within the receiver. Viewing these scenarios as belonging to the general framework of signal separation problems, in which the different signals mutually interfere, this chapter describes an adaptive receive scheme through which these different signal components can be extricated so as to realize the required radar sensitivity. Specific applications addressed include multistatic radar, imaging for high Doppler scenarios and stepped-frequency transmission.
The purpose of this chapter is to introduce geometric aspects of waveform diversity in terms of bistatic ambiguity function (BAF) and Cramér-Rao lower bounds (CRLBs). These aspects are fundamental in the choice of bistatic channel or set of channels in multistatic systems for target kinematic parameters estimation. Exploiting the relation between the ambiguity function (AF) and the CRLB, it is possible to calculate the bistatic CRLBs of target range and velocity of each transmitter-receiver (TX-RX) pair as a function of the target kinematic parameters and to provide a local measure of the estimation accuracy of these parameters. The information gained through the calculation of the bistatic CRLBs can also be used to evaluate the performance of each channel of the multistatic system and therefore for the choice, along the trajectory of the target, of the optimum TX-RX pair (or a set of bistatic channels) for data fusion and target tracking. This chapter proposes an algorithm that specifies what channels should be discarded and what channels should be considered during the fusion process. The results shown here can also be used [1] to calculate the weighting coefficients for combining the signals arising from the sensors of the network, highlighting those sensors exhibiting the best performance and discarding those channels with the worst one.
In this chapter, we consider the problem of waveform design for radar sensors that operate in a non-cooperative network. This is a system in which multiple radars share some common features (e.g. the same carrier frequency), but they do not cooperate in the detection stage of processing (namely each sensor performs detection processing independently). Our goal is to increase the performance of a sensor of the network, and, at the same time, to limit the interference induced by this element of interest on the remaining sensors. The resulting problem is in general non-deterministic polynomial-hard, namely an optimal solution cannot be calculated in polynomial time. However, it is possible to relax the original problem into a semidefinite programming problem, which is convex. This last problem can easily be solved in polynomial time. Starting from an optimal solution to the relaxed problem, we construct a good solution of the original non-convex problem and evaluate its quality via the approximation bound. The proposed technique, referred to as `waveform design in non-cooperative environment' (WDNE), enjoys the benefits of polynomial time complexity.
The recent development of the full digital array technology paves the way to the design of multistatic RADAR systems relying on agile waveforms at emission. This new paradigm fits perfectly with the concept of phase conjugation (or time reversal if applied to large-band data). This technique allows indeed to adaptively build a wave focusing onto a target, leading to an improvement in detection range or in search time as compared to classical approaches. The DORT method, issued from phase conjugation, even permits to detect multiple targets from the knowledge of the multistatic matrix of the antenna array. Based on both a theoretical analysis and an experimental proof provided here, it appears that these methods appear as very promising, especially for cueing and ultra-fast reacquisition modes where the phase conjugation/DORT SNR requirements are more easily fulfilled.
In this chapter, different techniques for improving radar performance in detection, location and classification of targets, through simultaneous transmission of radar signals in different directions, are analysed and discussed. After a brief introduction to the benefits and limitations of wide instantaneous angular coverage, the principles of space-time coding - i.e. sending simultaneously different signals through the different subarrays of a phased-array antenna - are described and illustrated by a few examples, e.g. intra-pulse scanning, circulating pulse or interleaved scanning. Generic techniques for codes optimization are then examined and shown to open the way to space-time adaptivity. A quantitative analysis of diversity gain demonstrates the benefits of high resolution inherent to wideband space-time radar systems, and a tentative classification of space-time codings is proposed. The conclusion emphasizes the benefits to be obtained from those space-time techniques, which will become accessible in near future, taking advantage of the current evolution towards agile front-ends and smart mode management systems for future radars.
In this chapter, we consider the design of radar transmit waveforms that are optimal for the purpose of detecting multiple unknown targets in a known interference environment. Specifically, we introduce constraints on the waveform's autocorrelation function, which is of fundamental importance in practice. The incorporation of autocorrelation constraints leaves the problem analytically intractable, and numeric techniques must be employed. We introduce the concept of `waveform-optimized performance' in order to provide a framework for suggesting and analysing various waveform designs. In this context, the eigenfunction waveforms and Neyman-Pearson detectors found elsewhere might no longer be optimal.
The typical approach to radar target classification is to image the target with waveforms that provide high resolution and low sidelobes, and then to compare the target images to a template library. In this chapter, we reconsider whether imaging-based metrics for waveform design are best for target classification, and develop alternative design strategies that result in waveforms with improved classification ability, but not necessarily a good ambiguity function by traditional notions. After presenting waveform design strategies based on optimizing signal-to-noise ratio or mutual information from a wide-sense stationary (WSS) ensemble of target impulse responses, we apply the design methods to the problem of radar target classification through a two-step process. The first step is to modify the design methodology that was based on WSS targets to account for the finite duration of practical target responses. The second step is to use the target class probabilities and impulse response library to calculate a weighted power spectral variance over target classes, which is then substituted into the design equations. The use of target class probabilities enables the waveform to be adapted in response to previous transmissions. Waveform behaviour and performance are studied over several different clutter and noise scenarios. The target impulse response library for these studies is based on finite-difference time-domain (FDTD) simulation of a publically available CAD model of an F-16 aircraft.
Waveform-agile design approaches for target tracking involve the adaptive configuration of the next transmit waveform by optimizing some cost function such as the predicted mean-squared tracking estimation error. These approaches are shown to be advantageous for complex radar tracking, such as when the received waveforms have originated from multipath returns or the targets are embedded in dense clutter or are under obscuration. As a result, this chapter addresses waveform-agile sensing methodologies for new challenging tracking applications, including multiple-input multiple-output (MIMO) radar, urban terrain multipath exploitation radar in high clutter and integrated urban terrain MIMO radar.
Transmitting waveforms with different polarizations in radar systems provide more complete information about the target and its environment, ensuring a significant enhancement of the radar's performance. Conventional polarimetric radars transmit waveforms with a fixed polarization pattern, independent of the target and clutter characteristics. In this chapter, we explore the adaptive design of radar polarization waveforms. We focus on a closed-loop system that sequentially estimates the target and clutter scattering parameters and then uses these estimates to select the polarization of the subsequent waveforms. We demonstrate that the radar system performance is significantly improved when the polarization of the transmitted signal is optimally and adaptively selected to match the polarimetric aspects of the target and the environment. In particular, we include an overview of our recent results in polarimetric design for radar detection and tracking.
In this chapter, we consider the problem of knowledge-aided transmit signal and receive filter design for point-like target in signal-dependent clutter. We suppose that the radar system has access to a (potentially dynamic) database containing a geographical information system, characterizing the terrain to be illuminated, and some a priori electromagnetic reflectivity and spectral clutter models, allowing the raw prediction of the actual scattering environment. Hence, we devise an optimization procedure for the transmit signal and the receive filter that sequentially improves the signal-to-interference-plus-noise ratio (SINR). Each iteration of the algorithm, whose convergence is analytically proved, requires the solution of both a convex and a hidden convex optimization problem. The resulting computational complexity is linear with the number of iterations and polynomial with the receive filter length. At the analysis stage, we assess the performance of the proposed technique in the presence of either a homogeneous ground clutter scenario or a heterogeneous mixed land and sea clutter environment.