1.1.1 Concept of Ultrawideband Radar

The term ultrawide band (UWB) describes a radar s ystem with an effective signal bandwidth that is a large percentage of the center frequency (20%–25%). This wi de bandwidth gives the radar systems a small spatial resolution (centimeter to millimeter) measurements and good imaging capabilities. UWB radars use a varie ty of signal formats, including short-duration pulses (impulses), random noise signals, step frequency coding, and pulse sequence coding, to achieve wide bandwidths. The fine spatial resolution of these radars can provide precision-range measurements for locating and imaging applications. In the following chapters, you will find examples of UWB radar systems in specia l applications requiring fine spatial resolution measurement and in penetration of materials for construction, geophysical surveying, security, and medical purposes [1,2].

1.1.2 Editorial Objectives

The writers prepared this book to give engineers and technical managers practical information about the latest UWB radar theory through real-world examples. You will find descriptions of systems that demonstrate new techniques and possible development directions for advanced UWB radars. Explanatory material and basic useful information have been added to make the book readable for newcomers. You will find chapters devoted to history, time-domain propagation theory, governmental regulations, and propagation in solid media. Some chapters describe UWB radar applications and implementations with example s, showing solutions to practical problems and demonstrating new radar sensing capabilities. You will also find practical examples of UWB radar, such as through-wall imaging, large-target backscattering effects, ground permittivity estimation, medical applications, and measurement of small movements of large buildings. These special applications and the theory supporting their design demonstrate the theoretical principles of UWB radar. We hope these descriptions of past accomplishments of the radar systems will pave the way for future concepts and practical applications.

1.2.1 My Association with UWB Radar

During the late 1980s, I served as the director of Long Range Technology Planning at the U.S. Air Force Systems Command’s Electron ic Systems Division, Hanscom Air Force Base, Massachusetts, United States. My commanders directed me to use my judgment to identif y new technology trends that could affect future defense capabilities and requirements. This put me in touch with experts and new ideas from industry and academia.

Everybody recognized the threat from ballistic missiles, and the “Star Wars” S trategic Defense Initiative (SDI) program received all the h igh-level attention and funding. At the time SDI started, the first cruise missiles en tered the American and Soviet Union arsenals and presented a new threat to national security. Drawing on my experiences as an air defense artillery officer, I considered the cruise missile threat a major technical challenge. I approached my commander with the implications of cruise missile defense for the future mission of the Electronic Systems Division and suggested investigating the problems in detecting cruise missiles flying at low altitudes.

My analysis of the problem indicated that detecting low-flying and small radar cross-section (RCS) targets at long ranges would require (1) airborne radar systems with a look-down capability because of the radar horizon and (2) radar spatial resolution in the order of the physical size of the target (0.1–10 m). Fine spatial resolution implied a large bandwidth and brought the additional adv antage of target identification by analysis of the return signal and possible recognition by imaging. More bandwidth meant more information and greater possibilities of target detection through signal spectrum analysis. The resulting bandwidths ran ged from 1 00 MHz to 1 GHz depending on the spatial res olution. This rais ed potential issues about frequency spectrum allocation, interference, and technological capabilities.

1.2.2 Bandwidth and Radar Range Resolution

We can examine the basic concepts of pulse radar to understand the relation between signal bandwidt h and range resolution capability. From the basic pulse radar theory, we know that the range resolution Δr of a pulse radar signal of velocity c = 3 × 10⁸ m/s depends on the pulse duration τ so that Δr = cτ/2, where the factor 2 considers the round trip time of the return signal. Basic Fourier analysis and communications theory postulates that shorter pulses require higher modulation frequencies. For a sinusoidal pulse signal of duration τ, we can approximate the –3dB band width as b = 1/τ. Combining these, we get the range resolution relation Δr = cτ/2 = c/2b [1]. Therefore, a 30-cm (~1 ft.) spatial resolution requires a 500- MHz band width. For the low microwave frequency range (0.5–10 GHz), the 500- MHz band width becomes a large fraction of the center frequency f_c. This gives us the concept of fractional bandwidth b_f = b/f_c, usually ex pressed as a percentage. A large radar bandwidth also raises problems about interference and compatibility with other radar and communications systems. Figure 1.1 shows the bandwidth and fractional bandwidth re quired for different levels of range resolution. F igure 1.2 shows the power spectral density (PSD) for short sinusoidal and Gaussian pulses [1,3].

At this time, developers knew that large fractional band width signals would require radical new concepts in radio design and only had practical experience to guide the design process. Figure 1.2 shows two typical short-duration UWB signa ls and their power spectral density (PSD) plots. This illustrates the wide bandwidth resulting from short-pulse signals. Other signal formats, for example random and pseudorandom noise modulations, can provide small spatial resolution and fall into the UWB category.

1.2.3 Early Demonstrations of UWB Radar

During the 1980s and early 1990s, the Defen se Advanced Research Projects Agency (DARPA) sponsored technology development programs, whic h demonstrated the capabilities of UWB systems for foli age penetration and detection of concealed vehicles and objects in cluttered and visually obscured environments. Other programs demonstrated the capabilities of these systems for high-resolution mapping, radar imaging, and detection of buried or concealed objects [1].

I became part of a select group of military officers and defense officials who wanted to develop large bandwidth technology for defense applications. For management and promotional purposes, we invented the term “ultra-wideband” to characterize existing high-resolution radar concepts with names such as impulse, baseband, nonsinusoidal, video pulse, etc. [1,2]. Our original definition described UWB radar as a system with a fractional bandwidth exceeding 25% of the cente r frequency, where the fractional bandwidth b_f = 2(f_h − f_l)/(f_h + f_l), with f_h and f_l indicating the highest and the lowest frequencies of interest. We did not define the power level at f_h and f_l and assumed that it meant the usual −3 dB level.

At this time, most people thought only in terms of bandwidth an d did not consider the concept of relative bandwidth. We selected 25% because it described the typical spectra of the 1-ns pulse length signal, as shown in Figure 1.2. Since then I have seen nothing indicating any special physical significance to the 25% fractional bandwidth. Over time, “ultra-wideband” has changed t o the current spelling “ultrawideband.” Government organizations regulating radio frequency (RF) spectrum allocation have refined the definition to include different relative bandwidth percentages and power levels, with highly specific terms covering fre quencies, power spectrum levels, and measurement methods. Section 1.4 presents a summary of UWB definitions and regulation. Chapter 2 recounts the early history of UWB technology, and Chapter 4 includes extracts of those regulations for your easy reference. In this chapter, I present the summary of the issues and history of UWB radar to give you some background and context for the remaining chapters.

1.4.1 Earl y History of UWB Radar

As Terence Barrett will show in Chapter 2, UWB ground- and material-penetrating radar has a long history of use in geophysical surveying and civil engineering, which starts in the 1970s. During the early days, users found that their equipment did not interfere with other electronic systems and that they could operate under the radar horizon of official scrutiny. In the late 1980s and early 1990s, American defense research agencies sponsored demonstration projects using UWB radar for long-range applications such as locating buried objects and m ines, high-resolution foliage penetration, target detection, imaging, and remote mapping. These demonstrations used high-power signals for special purposes in restricted areas, which showed the effectiveness of the technology.

TABLE 1.1
A Partial List of UWB Radar-Related Books

The increased interest of developers and evolution of electronics technology prompted more organizations to enter the UWB field with practical commercial applications. This raised questions about licensing and interference with communications and locating systems. In 2000, the American Federal Commun ications Commission (FCC) requested comments about proposed regulations for unlicensed systems and received hundreds of comments from people in the UWB radar business. The resulting 2003 FCC regulations set emission limits for UWB devices and opened the way to develop unl icensed UWB radar and radio applications for secu rity surveillance, penetratio n of materials, through-wall vision, and medical imaging, which are described in the later chapters [4–6].

The American FCC and European Union (EU) rules effectively restrict unlicensed UWB radar devices to the short-range ap plications described in this book. Looking at the broader implications of the term and other ways to expand signal bandwidths still leaves opportunities for developing high-power, long-range syst ems under special licenses. However, the same rules have inspired new concepts in signal processing, integration, and detection to maximize performance within those regulatory constraints.

1.4.2 Standard Definitions of UWB Radar

In 2005, Stephen Johnston and Arnie Greenspan formed the IEEE Ultrawideband Radar Co mmittee an d Working Group to collect suggestions for establishing standard definitions for new UWB radar-related terms. I served as the lexicographer and editor for the project, which re sulted in the IEEE Stand ard 1672, Ultrawideband Radar Definitions.

UWB radar development has changed or added new meaning to many terms that you learned during your education and work. Bandwidth provides a typical example of changed meaning. In classical electronics classes, bandwidth meant the difference between the –3 dB points in the frequency response of a circuit, component, or power spectrum of a signal. For UWB systems, the regulatory definition of bandwidth can mean the difference between the highest and the lowest frequencies of interest or the frequency spread between the −3 and −10 dB points for the emitted signal, or it can mean any parameter that a particular agency wishes to define as the bandwidth for a specific purpose.

Whenever you have a question about the meaning of ultrawideband, read the article, specifications, description, etc. to find the intended or specialized meaning. Once you determine the specific meaning, you can act accordingly.

We now have three sources of definitions for the term ultrawideband. These include the following:

• IEEE STD 1 672 Ultrawideband Radar Definitions [4]

• United States Federal Commun ications Commission (FCC) 47 U.S.C. Sections 154, 302a, 303, 304, 307, 336, and 544A [5]

• Electronic Communications Committee within the European Conference of Postal and Telecommunications Administrations (ECC-CEPT), The Protection Requirements of Radiocommunications Systems Below 10.6 GHz from Generic UWB Applications, Helsinki, February 2005, ECC Report 64 [6]

All the sources above agree on the following definitions:

• The fractional bandwidth equals b_f = 2(f_h − f_l)/(f_h + f₁), where f_h indicates the highest and f_l the lowest frequency of interest depending on the definition of bandwidth

• The center frequency f_c = (f_h + f₁)/2

Ultrawideband has several different definitions, which depend on the particular source:

• From IEEE STD 1672: “Ultrawideband: (a) a term for a radar and communications technology utilizing simultaneously a band of frequencies that can span from hundreds of megahertz to the end of the microwave region. (b) Per FCC rules, a fractional bandwidth ≥ 0.2 or a bandwidth ≥ 500 MHz. (c) A signal having a fractional bandwidth of 25% (DARPA) or 20% (FCC) or greater below a 1 GHz operational center frequency. Above 1 GHz, a signal with 25% (or perhaps 20%) or 500 MHz (whichever is smaller). See ultrawideband device” [4].

• From the American FCC 47 U.S.C. Section 15.503 Definitions: “Ultra-wideband (UWB) transmitter: An intentional radiator that, at any point in time, has a fractional bandwidth equal to or greater than 0.20 or has a UWB bandwidth equal to or greater than 500 MHz, regardless of the fractional bandwidth.” “UWB Bandwidth: For the purpose of this subpart, the UWB bandwidth is the frequency band bounded by the points that are 10 dB below the highest radiated emission, as based on the complete transmission system including the antenna. The upper boundary is designated f_H and the lower boundary is designated f_L. The frequency at which the highest radiated emission occurs is designated f_M” [5].

• From Article 5(1) of Directive 2002/20/EC of the European Parliament and of the Council of 7 March 2002 on the authorization of electronic communications networks and services: “equipment using ultra-wideband technology” means equipment incorporating, as an integral part or as an accessory, technology for short-range radiocommunication, involving the intentional generation and transmission of radio-frequency energy that spreads over a frequency range wider than 50 MHz, which may overlap several frequency bands allocated to radiocommunication services” [6].

Table 1.2 summarizes the different bandwidth definitions. Chapter 4 presents the exact wording of the American FCC and EU rules, which limit the radiated power and spectrum of different classes of UWB devices.

Caution: UWB may mean whatever the user wants it to mean within special constraints. Designers need to know the government regulations, which set specific limits on the radiated power of UWB devices depending on their particular applications, for example, ground-penetrating radar, surveillance systems, and medical devices. The spectral descriptions and limits have meanings based on specific measurement techniques, which vary between American and European agencies. Anyone planning for commercial development of unlicensed UWB devices should refer to the latest edition of any applicable regulations, laws, and rules. Chapter 4 presents extracts of the American FCC and EU rules and definitions for unlicensed UWB systems.

1.4.3 Bandwidth Designations and Definitions

Radar signal bandwidth designations recognize several types of signals based on their fractional bandwidth. Depending upon the applications or particular definitions, the literature agrees on the bandwidth definitions shown in Table 1.2, subject to definition of the power level at the highest and lowest frequency signals. (Readers and buyers beware: Note that the −10 and −20 dB power bandwidth definitions permit manufacturers to advertise antennas and other components as ultrawideband, which would not have qualified as such earlier.)

TABLE 1.2
UWB Radar Definitions

Other bandwidth definitions include −3, −10, and −20 dB power; root-mean-square; and 90% and 99% energy [7].

Mokole et al. [7] suggested a bandwidth categorization method based on either the fractional bandwidth b_f or the band ratio b_r, as shown in the Table 1.3, from the original DARPA definition [2]. As an example of further specialization, the electromagnetic interference community has another set of bandwidth designations based on the same fractional and band ratio terms as shown in Table 1.4 [7]. Principles of Waveform Diversity and Design by Wicks et al. [8] provides a valuable reference regarding signal formats, bandwidths, spectra, ambiguity functions, applications, etc. [8].

1.6.1 Background

UWB propagation differs from the normal concepts of “continuous wave” propagation of sinusoidal waveforms. A quick review of electromagnetic induction shows that the induced electrical field varies as the time derivative of the incident magnetic (electrical) field, which is expressed by the Faraday’s law ε = −dΦ_B/dt, where ε indicates the electromotive force and Φ_B the magnetic flux [10].

If a signal has a sinusoidal waveform, then the derivative will appear as a cosine waveform (or the original sine wave with phase shift). For NB sinusoidal signals, we can safely design systems and other devices based on a linear transfer function with a sine wave input and output and the addition of a phase-shift term if needed. We can generally ignore waveform changes, assuming linear transfer functions and sufficient bandwidth for the purpose.

Because UWB waveforms generally have a nonsinusoidal character, we must consider the time-derivative effects on waveforms. The following section uses the Gaussian pulse to show the effects of time differentiation on nonsinusoidal waveforms and frequency spectrum characteristics. This method can be applied to other waveforms.

1.6.2 Example of Gaussian Pulse Transformation during Transmission and Reception

The Gaussian pulse gives a good approximation of the direct current (DC) pulse waveform used to excite transmitters in UWB radar systems. Equation 1.1 defines the basic Gaussian exciting signal with a time constant τ_c for the basic pulse waveform as

Equation 1.2 shows the normalized general family of transmitted time derivative Gaussian pulses as

where n is the integer number of the derivative. Do not confuse the time constant τ_c with the pulse width τ used earlier for the pulse duration. A Gaussian signal will almost always occur within ±3τ_c of τ = 0 as shown in Figure 1.3a. For design purposes with Gaussian pulses, you can assume the pulse width time constant τ ≈ 6τ_c. Note that the normalization of this function ensures that for all n, the Gaussian function will have unit maximum amplitude.

In the case of nonsinusoidal UWB waveforms such as the Gaussian pulse, each inductive action during transmission or reception changes the waveform through a time differentiation as shown in Figure 1.4 [11]. This leads to potential problems in designing receivers, correlation detection schemes, and signal integration, which may depend on the waveform. Chapter 3 discusses how and why UWB waveforms change and presents an analytical approach to predicting them. Chapters 3 and 11 discuss the approaches to UWB signal detection with unknown waveforms.

The change of waveform will affect the frequency content, and the nth derivative Gaussian pulse will have a frequency domain form as follows:

This gives a center frequency by solving for the critical point indicated by the derivative of Equation 1.3, which gives

which yields a center frequency for a Gaussian pulse of time constant τ_c and derivative n as ωc=2n/τc or

Figure 1.5 shows how the center frequency varies over a time constant τ_c value of two orders of magnitude from 1 to 10 ns for the derivatives n = 1, 2, 3, 4 [9].

We can determine the time constant τ_c for a desired center frequency by rearranging Equation 1.5:

Figure 1.6 shows the results of plotting Equation 1.6 for the cases n = 1, 2, 3, 4 [9].

Figures 1.3b and 1.4 show how the Gaussian waveform changes with each reception/transmission and changes the pulse PSD. Normalizing the equation to unit peak amplitude, the Gaussian pulse PSD becomes

Figure 1.7 shows how the PSD shifts for the different orders of Gaussian pulse derivatives [9].

1.6.3 Conclusions on Gaussian Pulse Signal Propagation

The ideal Gaussian pulse approximates a common type of electrical current pulse used to excite the antenna in some UWB radar systems. This makes an ideal example for modeling UWB impulse (video pulse) signal propagation and describing the waveform distortions that result from time differentiation in the inductive processes of antennas. Each reception/transmission of a Gaussian-derived signal produces a different waveform with changes in the center frequency and spectrum. Other nonsinusoidal waveforms will change with each inductive process in their own special way. Many UWB radars use nonsinusoidal signal waveforms to achieve large instantaneous bandwidths. Maximizing system performance requires the designer to recognize these changes and use them in designing receivers. You can apply the methods shown to any nonsinusoidal or short-duration waveform.

1.7.1 Open-Space Measurement and Surveillance

UWB radar can provide high-resolution ranging and sensing for applications such as security surveillance, airborne mapping, and control systems [1,2]. For example, UWB radars can provide high-resolution imaging and detect the small movements of structures caused by daily heating and cooling, as shown in Chapter 6 on random signal radars. Small spatial resolution becomes an important quality for medical applications such as heart and breath measurements and diagnosis of internal injuries, as shown in Chapter 9.

1.7.2 Material-Penetrating Remote-Sensing Applications

Electromagnetic waves can penetrate nonconducting materials. This lets UWB radars measure and image inside solid media, which enables a wide range of applications such as geophysical surveying, underground-utilities mapping, infrastructure inspection, archeological inspection, building evaluation, mine detection, and security, medical, and forensic applications. Because the detection of objects depends on the reflection of signals, which occurs when the index of refraction of media changes, the ability of the radar to discriminate between smaller returned signal changes will limit the performance of material-penetrating radars. Chapter 5 presents the basic theory of propagation in solid media. Chapters 12 through 14 describe the advances in material-penetrating and through-wall radar systems.

1.7.3 Medical Measurements and Imaging

The material-penetrating quality of radio signals can help to measure the position and motion of internal organs such as the heart and lungs. UWB radar signals can provide precise measurements for vital-signs monitoring, imaging, and tomography. Chapter 9 describes a UWB radar system developed by The Sensiotec Corporation for remote monitoring of vital signs in hospitals. Other developers have started commercial development of some Lawrence Livermore National Laboratory (LLNL) Micropower Impulse Radar applications including sensing of pneumothorax injury and detection of intracranial hematoma or stroke. Experimental work and patents for breast cancer detection show a great potential for providing radar imaging for diagnosis and treatment in small clinics or remote areas [12–29].

1.7.4 Security

The small spatial resolution UWB radar signal can penetrate clothing to provide images of persons and the contents of luggage and packages. This can help authorities to quickly and remotely detect concealed weapons and other contraband. UWB radar-based systems can eliminate the health hazards of ionizing radiation and the limitations of passive imaging systems. Chapter 17 describes the results of advanced work in this area by Camaro Inc.

1.7.5 Military Remote-Sensing Applications

UWB radar demonstrations during the 1980s and 1990s showed how high-power systems could provide accurate maps of minefields, locate buried cables, present high-resolution images, detect objects obscured by foliage, discriminate between buried mines and clutter, and provide accurate topographical maps of inaccessible terrain [1,2]. Several companies now market handheld through-wall radars for military and security forces (see Chapter 14). Future defense applications may include augmenting remote surveillance systems with a UWB radar-based tomographic imaging system for below-ground imaging [30].

1.8.1 Optimization of UWB Antenna Array Geometry

The intuitive approach to multiple antenna array design suggests a symmetrical and regularly spaced pattern. Engineers of Xaver Inc. applied communications theory to antenna array design and discovered that they could get more information content from an irregular placement of transmitting and receiving elements. Chapter 14 describes the theory of array optimization.

1.8.2 Synchronization of Antenna Array Signals

UWB antenna arrays present special problems in synchronization and timing of received signals for high-resolution imaging applications. Different delays in each transmitter–receiver channel mean that spatial data will require corrections to ensure that returns from the same point in space appear in proper sequences in data arrays for high-resolution image reconstruction. Digital technology now provides a way to continuously determine the delays in transmitter–receiver channels and apply these to the range position data for precise imaging. Chapter 16 describes the method used by engineers of Xaver Inc. to build an advanced high-resolution imaging radar.

1.8.3 Improvements in Receiver Signal-to-Noise Ratios

Simple physics indicates that the signal strength of similar cross-section target returns will decrease as the inverse fourth power of range. This increases the signal-to-noise ratio (SNR) in the receiver and limits reliable detection of distant targets. Some UWB radar designers have developed techniques to integrate multiple-range cell returns from long-range returns to improve the SNR and imaging performance. The Novelda Nano Scale Impulse Radar described in Chapter 11 shows an example of range cell signal integration and enhanced detection. Chapter 15 describes a method developed by Xaver Inc. to overcome this problem.

1.8.4 Multistatic Radar

In many chapters, you will find systems using antenna arrays for imaging. Increasing the information content by collecting both directly reflected and side-scattered radiation looks like a promising direction for future development. Figure 1.8 illustrates a concept for enhanced medical imaging or small-object detection with low, direct backscatter characteristics.

1.8.5 Higher Order Signal Processing to Augment Imaging and Target Identification

The wide spectrum of UWB signals presents possibilities for using the shifted frequency spectrum (waveform) of radar returns. In Introduction to Ultra-Wideband Radar Systems, Chapter 9, “Radar Cross Section and Target Scattering,” Van Blairicum explains this type of target identification using the singularity expansion method (SEM). This opens the possibility of both imaging the objects and identifying their material composition or shape. In Chapter 11, “High-Order Signal Processing for Ultrawideband Signals,” Sheby and Marmarelis presented the concept of higher order signal processing (HOSP) as a way to differentiate between reflecting materials [2]. Combining SEM and HOSP with imaging might give UWB radar systems further capabilities for medical imaging and diagnostics when combined with microwave tomography.