Measurement Science and Technology

TOPICAL REVIEW

Performance and non-destructive evaluationmethods of airborne radome and stealth structuresTo cite this article: Ravi Panwar and Jung Ryul Lee 2018 Meas. Sci. Technol. 29 062001

View the article online for updates and enhancements.

Related contentDevelopment of dual port scanning freespace measurement systemJong-Min Hyun, Hasan Ahmed and Jung-Ryul Lee

-

Imaging with terahertz radiationWai Lam Chan, Jason Deibel and DanielM Mittleman

-

The 2017 terahertz science andtechnology roadmapS S Dhillon, M S Vitiello, E H Linfield et al.

-

Recent citationsAnisotropy of W-band complex permittivityin Al2O3Maxwell R Telmer et al

-

Microwave imaging of thick compositestructures using circular aperture probeMohammed Saif ur Rahman et al

-

This content was downloaded from IP address 143.248.247.199 on 07/05/2019 at 11:32

1 © 2018 IOP Publishing Ltd Printed in the UK

Measurement Science and Technology

R Panwar and J R Lee

Printed in the UK

062001

MSTCEP

© 2018 IOP Publishing Ltd

29

Meas. Sci. Technol.

MST

10.1088/1361-6501/aaa8aa

6

Measurement Science and Technology

Ravi Panwar1 and Jung Ryul Lee2

1 Discipline of Electronics and Communication Engineering, Indian Institute of Information Technology Design & Manufacturing Jabalpur (IIITDM Jabalpur), Madhya Pradesh, India2 Department of Aerospace Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, People’s Republic of Korea

E-mail: leejrr@kaist.ac.kr

Received 16 November 2016, revised 21 December 2017Accepted for publication 18 January 2018Published 18 April 2018

AbstractIn the past few years, great effort has been devoted to the fabrication of highly efficient, broadband radome and stealth (R&S) structures for distinct control, guidance, surveillance and communication applications for airborne platforms. The evaluation of non-planar aircraft R&S structures in terms of their electromagnetic performance and structural damage is still a very challenging task. In this article, distinct measurement techniques are discussed for the electromagnetic performance and non-destructive evaluation (NDE) of R&S structures. This paper deals with an overview of the transmission line method and free space measurement based microwave measurement techniques for the electromagnetic performance evaluation of R&S structures. In addition, various conventional as well as advanced methods, such as millimetre and terahertz wave based imaging techniques with great potential for NDE of load bearing R&S structures, are also discussed in detail. A glimpse of in situ NDE techniques with corresponding experimental setup for R&S structures is also presented. The basic concepts, measurement ranges and their instrumentation, measurement method of different R&S structures and some miscellaneous topics are discussed in detail. Some of the challenges and issues pertaining to the measurement of curved R&S structures are also presented. This study also lists various mathematical models and analytical techniques for the electromagnetic performance evaluation and NDE of R&S structures. The research directions described in this study may be of interest to the scientific community in the aerospace sectors.

Keywords: non-destructive evaluation (NDE), radome and stealth structures, imaging techniques, electromagnetics, airborne

(Some figures may appear in colour only in the online journal)

Glossary listing the acronyms used in the manuscript

Abbreviations DefinitionBRCS Bistatic radar cross section

CATR Compact antenna test rangeCNT Carbon nanotubeCSR Coplanar strip-line resonatorCW-THz-WI Continuous wave terahertz wave imagingDP-FSM Dual port free space measurement

Performance and non-destructive evaluation methods of airborne radome and stealth structures

Topical Review

IOP

2018

1361-6501

1361-6501/18/062001+29$33.00

https://doi.org/10.1088/1361-6501/aaa8aaMeas. Sci. Technol. 29 (2018) 062001 (29pp)

Topical Review

2

ECPT Eddy current pulsed thermographyEMC Electromagnetic compatibilityEMI Electromagnetic interferenceFDTD Finite difference time domainFF-PE-UPI Full field pulse-echo ultrasonic propagation

imagingFFT Far field testFPGA Field programmable gate arrayFSL Free space lossFSM Free space measurementFSS Frequency selective surfaceGBT Gaussian beam testGDD Glow discharge detectorGRL Gated reflect lineHCT High current testHDPE High-density polyethyleneHVT High voltage testINDE In situ non-destructive evaluationISR Inverse source reconstructionLIE Lightning indirect effectLMS Laser mirror scannerMAS Microwave absorbing structure

ML-R&S Multilayer radar and stealthMM MillimeterMMS Metamaterial structureMM-WI Millimeter wave imagingMRCS Monostatic radar cross sectionNCCM Normalized cross correlation methodNDE Non-destructive evaluationNFT Near field testNRL Naval research laboratoryOERW Open-ended rectangular waveguidePEC Perfect electric conductorPE-UPI Pulse-echo ultrasonic propagation imagingQO-FSM Quasi optical free space measurementRadar Radio detection and rangingRCS Radar cross section

R&S Radome and stealthRHM Radome health managementSAR Synthetic aperture radarSE Shielding effectivenessSFSM Scanning free space measurement

SL-R&S Single layer radar and stealthSP-FSM Single port free space measurementSNR Signal-to-noise ratioSUT Specimen under testTEM Transverse electromagneticTHz-WI Terahertz wave imagingTLM Transmission line methodTRL Thruough-reflect-lineTRM Thruough-reflect-matchUEM Ultrasonic energy mappingUPI Ultrasonic propagation imagingVNA Vector network analyzer

1. Introduction

Since its advent during World War II, the radar (radio detection and ranging) has been inextricably intertwined with aircraft radome and stealth (R&S) structures. Both in the air and on the ground, it serves a number of purposes, from tracking enemy aircraft to watching for undesirable weather. R&S structures have been the subject of intensive invest igation for aerospace/defence related applications (Baskey et al 2017). This rapid progress in the field of R&S technology has explored high performance composite structures like curved, multilayered, frequency selective surface (FSS) impacted and sandwiched R&S structures (Varadan et al 2010, Carranza et al 2015, Yang et al 2016b, Baskey et al 2017, Szabo 2017, Zhang et al 2017). However, it is quite difficult to measure the performance of such heterogeneous structures due to their design complexities.

In general, the performance of R&S structures should be evaluated in terms of their electromagnetic performance and as well as non-destructive evaluation (NDE) for struc-tural damage and defects (i.e. load bearing capability). The transmission line method (TLM) and free space measurement (FSM) based techniques can be utilized to evaluate the electro-magnetic performance of R&S structures (Chen et al 2004, Ravuri et al 2008). TLM is basically a destructive microwave measurement technique, which can be further classified as waveguide and coaxial techniques. On the other hand, FSM is a non-destructive method, which can be categorized as sin-gle-port FSM, dual-port FSM, and arc type FSM. The com-pact range, far field, near field and Gaussian beam based test methods are some other widely used microwave measurement techniques for the performance evaluation of R&S structures (Chang et al 2004, Chen et al 2004, Kerouedan et al 2008). A series of other conventional electromagnetic techniques for R&S structures are available in the literature (Chen et al 2004, Li et al 2011b, Cheng et al 2012).

An R&S structure must not only provide good electro-magnetic characteristics, but also withstand various forces during operation throughout the flight envelope. In this man-ner, R&S structures have stringent requirements of electro-magnetic characteristics as well as physical properties like mechanical properties-to-weight ratio, thickness, rain erosion resistance, impact resistance, thermal stability, etc. R&S struc-tures possess discontinuities with size ranges from a few nm to a few mm, resulting in various undesired defects (Raja et al 2010). Damage in R&S structures can degrade their structural integrity and decreases strength, making electro magnetic sig-nal propagation through the damaged area difficult. However, appropriate NDE at the infant stage can reduce the risk of fail-ure and extend the lifetime of R&S structures (Shull 2002, Pagliarulo et al 2015, Shin et al 2017). Eddy current inspec-tion, giant magneto-resistive sensors, potential drop methods, shearography, pulsed TV holography, scanning laser-line thermography, laser-spot thermography, magnetic induction topography and embedded fiber optic sensors are a few of the conventional techniques adopted for NDE of composite

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

3

structures (Gu et al 1990, Huang et al 1992, Lu et al 1995, Bocherens et al 2000, Komura et al 2001, Cerny et al 2004, Saguy et al 2005, Francis et al 2010, Raja et al 2010, Singh et al 2008, Li et al 2011a).

Millimeter wave imaging (MM-WI) and terahertz wave imaging (THz-WI) techniques have been recognized as two of the effective techniques for composite structures (Kaatze 2013). Such advanced measurement techniques provide a more comprehensive picture of an inspection problem (Bardak et al 2014, Agarwal et al 2015, Alvarez et al 2015, Carranza et al 2015). Here, the reflected electromagnetic signals from the target are utilized for defect detection in R&S structures. Recently, NDE has been upgraded to in situ non-destructive evaluation (INDE) by enhancing the full-scale in-service aircraft inspection (Shin et al 2017). The laser ultrasonic propagation imaging (UPI) for Lamb and bulk waves is an example of such advanced NDE techniques employed for the NDE of R&S structures (Bae et al 2016). The coplanar strip line resonator, open ended rectangular waveguide (OERW), eddy current pulsed thermography (ECPT) and inverse source reconstruction (ISR) techniques are other adopted techniques by the researchers for NDE of R&S structures. The electro-magnetic performance and structural NDE of such complex shaped R&S structures require a deep understanding of con-ventional as well as advanced performance evaluation tech-niques, corresponding mathematical models and algorithms and associated challenges as well as issues. Therefore, in this article a new and useful insight into evaluation techniques for both the electromagnetic performance and structural integrity of modern R&S structures of different shapes and require-ments is presented in detail.

2. Theoretical background

An ideal radome structure should be designed to transmit all the incident electromagnetic energy and reflect nothing. A radome can be treated as merely a protective cover placed in front of the antenna. Thus, a radome should be mechani-cally strong and at the same time transparent at the operat-ing frequency of the particular aircraft antenna. On the other hand, stealth structures are employed to minimize the radar cross section (RCS) of aircraft, spacecraft, automobiles, ships, tanks and missiles. The mechanical strength of R&S structures also matters a lot in addition of their broadband frequency response. Nowadays, stealthy radome structures are also becoming the main choice of researchers. An ideal stealthy radome with band-pass characteristics can be employed to reduce the RCS of any antenna outside its operating frequency range (Kraus et al 1950). Here the antenna can freely receive and transmit electromagnetic signals in the frequency range where the radome is transparent. The RCS of antenna will not be affected in this particular frequency range; on the other hand, a stealthy radome will reduce the antenna RCS for out of band frequency range. It is desirable for R&S structures to function for different angles of incidence as well as polariza-tions. The resonance frequency of R&S structures is greatly affected by corresponding variation in the angle of incidence.

A design with a more stable resonant frequency can be pos-sible by the combination of dielectric and metallic surfaces.

The transmission and reflection characteristics of R&S structures can be easily understood with the help of a well known radar range equation and RCS equation (Kraus et al 1950, Skolnik 1990). The free space loss (FSL) factor repre-sents the losses due to the spherical spreading of the electro-magnetic energy of the antenna system. The RCS is basically a far-field parameter, which is used to characterize the scatter-ing property of a radar target. The RCS can be characterized as a monostatic or backscattering RCS (MRCS) and a bistatic RCS (BRCS). The MRCS of a target refers to the situation when both the transmitting and receiving antennas are situated at the same location. If the two antennas are not situated at the same location, then the BRCS can be measured. The RCS of a target is a function of the angle of incidence, the angle of the observation, polarization of the incident wave, target geome-try, target electrical properties, and the frequency of operation.

Accurate measurements are necessary to establish the actual performance of R&S structures in terms of their reflec-tion coefficient, transmission coefficient, polarization, band-width, thickness and resonance frequency. To accommodate such requirements, improved instrumentation and measur-ing techniques have been explored, such as tapered anechoic chamber, compact range, near field and far field automatic measurements. Most R&S structure measurements are carried out in an anechoic chamber or in free space in a reflectionless environment. The wall of anechoic chambers is printed with microwave absorbers in order to minimize spurious reflections from the target. The rectangular and tapered (or pyramidal) kind of anechoic chambers can be utilized for the measure-ment of R&S structures.

In general, the responses of R&S structures to electro-magnetic fields are closely determined by the displacement of free and bound electrons by electric fields and the corre-sponding orientation of their atomic moments by magnetic field intensity. The concept of R&S structure and electro-magnetic field interaction at microwave frequencies can be understood with the help of the well known Maxwell equa-tions (Balanis 1989). In general, complex dielectric permit-tivity (ε), complex magnetic permeability (µ) and conductivity (σ) of the R&S structure are three important factors that deter-mine the response of a composite structure under the influ-ence of an electromagnetic field. These parameters are quite helpful to determine the extent to which the electromagnetic field can penetrate into the R&S structure at a specified fre-quency range. The imposition of an electric field into an R&S surface will produce electrical currents, the amplitude and direction of which are proportional to the electric field inten-sity. The ways in which physical processes in an R&S struc-ture can affect the wave electric field are described through a parameter ε, which can be mathematically represented as ε = ε′ − jε″. The parameter ε describes the coupling of the electric component of the incident wave with the R&S sur-face. The real part of the complex dielectric permittivity (i.e. ε′) represent the electric energy storage; on the other hand, the imaginary part (i.e. ε″) represents the loss of the electric energy. Similarly, the magnetic characteristics of composite

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

4

structures at high frequencies can be mathematically repre-sented as µ = µ′ − jµ″. The real part of the complex magnetic permeability (i.e. µ′) represents the magnetic energy storage, on the other hand, the imaginary part (i.e. µ″) represents the loss of the magnetic energy. For ferrite based R&S structures, an applied magnetic field may align magnetic dipole moments in the magnetic structure to generate magnetic polarization (Balanis 1989). R&S structures are usually characterized by specifying their loss tangents at a certain frequency. The loss tangent is also known as the loss factor or power factor or dis-sipation factor. In general, low dielectric loss tangent values are required for the radome structures; on the other hand, effi-cient stealth structures are characterized on the basis of their high loss tangents.

The electromagnetic performance and NDE evaluation of R&S structures using microwave, MM-WI and THz-WI techniques is of great interest in the aerospace industries. The techniques based on electromagnetic field interactions with composite structures have developed rapidly due to their potential to provide a faster evaluation of R&S structures in terms of their electromagnetic performance as well as struc-tural integrity.

The transmission line theory can be utilized for the analy-sis of a surface coated with microwave absorbing materials. When an electromagnetic wave travels from one medium to another, reflections take place. The reflection coefficient (Γ) is used to represent these reflections. It is related to medium impedance and free space impedance as follows (Balanis 1989):

Γ = Ereflected/Eincident = |(Zin − Zo)/(Zin + Zo)| (1)

where Zo is the wave impedance of free space, which can be expressed as Zo = (µ0/ε0)1/2 = 376.73 Ω, and Zin is the input impedance at the boundary surface. If Zin = Z0, then zero reflections take place. This condition of Zin = Z0, is also called the impedance matching condition. The input impedance can be computed by assuming that the material layer thickness is small as compared to the wavelength (λ) of the incident trans-verse electromagnetic (TEM) wave (Balanis 1989):

Zin = Z0(µ/ε)1/2

tan h

j(2πfd/c)(µε)1/2

(2)

where f is the measurement frequency, d is the specimen under test (SUT) thickness, c is the velocity of light (3 × 108 m s−1), and µ and ε are the complex dielectric permit tivity and com-plex magnetic permeability of the SUT, respectively. It is clear from equations (1) and (2) that the reflection strongly depends on d, f, ε′, ε″, µ′ and µ″.

The phenomenon of electromagnetic induction can be explained on the basis of the skin effect or depth of penetration (Collin et al 1996). The depth of penetration strongly depends on f, µ and σ. The skin depth decreases with a corresponding increase in the f, µ and σ. Therefore, it is important to notice that microwave techniques are not suitable for evaluating the interiors of highly conducting surfaces unless the anomaly of interest is on or near the surface of the conductor or the sam-ple is to be evaluated is thin as compared to the skin depth (Sezer et al 2007).

The aircraft R&S structures present a unique set of difficul-ties and requirements for their performance evaluation. Any variation in paint or structural thickness as well as disconti-nuities within the composite structure may alter the transmis-sion and reflection characteristics of R&S structures. NDEs are generally carried out with an interrogating magnetic, elec-tric or acoustic signal. The basic principle is an interaction of signal with the SUT and the corresponding response provides the properties of the SUT. A range of well established and standard NDE methods are available in the literature for man-ufacturing and in-service defect detection and their growth (Shull 2002, Kharkovsky and Zoughi (2007). However, each method has its own merits, demerits and applications. Visual inspection, radiography, penetrant testing, infrared thermog-raphy and ultrasonic testing are a few of the traditional NDE techniques (Liang et al 2007). Visual inspection allows a fast and simple inspection of the surface by identifying the defects using the naked eye or magnifiers. However, quantita-tive assessment of internal defects using visual inspection is difficult. Radiography is another typical NDE technique that results in a photographic record by an influence of electro-magnetic radiations (such as x-rays or gamma rays) through an object onto a film, but this method requires special secu-rity measures because of exposure of harmful radiations. Ultrasonic testing techniques like the pulse echo technique and the through transmission technique are widely used tech-niques for the NDE of composite structures. However, ultra-sonic testing is a contact method that employs coupling agents like mineral oil, gel or water (Dragan et al 2009). In addition, it is quite challenging to evaluate the performance of curved structures using ultrasonic testing, but nowadays researchers have explored some advanced techniques like scanning micro-waves, MM-WI, THz-WI and laser UPI, which can be effec-tively utilized for the NDE of R&S structures.

3. Overview of radome and stealth wall configurations

The nose cone in an aircraft acts as a radome for the plane’s weather radar. As already discussed, an aircraft radome must be electrically transparent as much as possible with a high degree of uniformity as well. A substrate with low di electric constant such as a fiber glass sheet is generally used for the fabrication of the inner and outer surfaces of the radome, which is separated by a core. A radome must be fabricated based on the tight design specifications and the frequency band of the radar. Figure 1 depicts distinct types of radome wall configurations such as half-wave, monolithic, sand-wiched and multilayered structures (Kozakoff 2010). A half-wave radome can be designed by employing a dielectric sheet having electrical thickness equal to one-half wavelength (λ/2) inside the dielectric. This condition ensures that the reflections from the front and back will cancel each other out as they are equal in amplitude but with opposite phase (Kraus et al 1950). The R&S structures are available in different configu-rations, i.e. single layer R&S (SL-R&S), multilayered R&S (ML-R&S), metamaterial structures (MMSs) and FSS based

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

5

R&S structures. A brief description of such electromagnetic structures is depicted as follows.

3.1. Single layer radome and stealth structures

The transmission and reflection characteristics of R&S based structures are greatly affected by their thickness, dielectric, magnetic and loss tangent characteristics. As the name sug-gests, single layer R&S (SL-R&S) based structures are single layer composite structures. A variety of magneto-dielectric SL-R&S structures have been reported by researchers, but these SL-R&S structures have their own merits and demer-its. They usually provide better transmission and reflection properties only for a particular frequency range and coating thickness (i.e. matching frequency and matching thickness). The major issue with SL-R&S based structures are the thick-ness–bandwidth trade-off, which can be resolved by adopting advanced electromagnetic approaches like multi-layering and FSSs (Baskey et al 2017, Panwar et al 2017).

3.2. Multilayered radome and stealth structures

Multilayered R&S (ML-R&S) structures are a layer-by-layer arrangement of SL-R&Ss in order to achieve broad bandwidth. The layer discontinuity is the major reason for the good trans-mission and reflection properties of the ML-R&S structures. Figures 1(b)–(d) depict sketches of a few ML radome struc-tures. The TLM can be used for the development of ML-R&S structures. However, the optimization of such structures is still a challenge due to their complex nature. A multi-objective optim-ization approach can be adopted to solve the issue. The important design considerations include the selection of a suitable mat erial for an individual layer, layer preferences and corresponding coating thickness values. The fabrication of ML-R&S structures is usually complex and special care is required for that.

3.3. Metamaterial structures (MMSs)

MMSs are man-made artificial electromagnetic structures designed to have artificial dielectric and magnetic properties,

and are not available in nature (Szabo 2017). MMSs resemble natural crystals as they are made up of periodically arranged unit cells. Here the electromagnetic wave interacts with the unit cells consisting of small metallic resonators. The MMSs are defined based on their negative ε and µ values. Nowadays, researchers are using MMS in diverse applications such as stealth technology, radome, electromagnetic interference (EMI) suppression, electromagnetic compatibility (EMC), stopping light, optical cloaking, etc (Varadan et al 2010, Carranza et al 2015, Szabo 2017).

3.4. FSS based radome and stealth structures

An FSS is a periodic electromagnetic structure exhibiting dif-ferent transmission and reflection characteristics as a function of frequency (Kraus et al 1950). Such structures consist of an array of periodic metallic patches or apertures on a di electric substrate, designed to transmit, reflect and absorb electro-magnetic waves based on their spatial filtering operation (Munk 1995, 2000, Panwar et al 2017). The high-pass (induc-tive FSS) and low-pass (capacitive FSS) filtering operation of an FSS results in the exhibition of total transmission and reflection near the resonance wavelength. FSSs are employed in various applications in today’s technology. Perhaps the most important applications of FSSs are the design of R&S structures, metamaterial absorbers and antennas. FSS plays a significant role in order to develop a thin and a broadband R&S structure (Panwar et al 2017). Recently, a substantial amount of research has been conducted in the field of FSS analysis for distinct practical electromagnetic applications other than stealth, such as EMI and EMC. FSS based electromagnetic structures can be broadly classified as traditional FSSs, active FSSs, fractal FSSs, multilayered FSSs, sandwiched FSSs, etc. A critical analysis of these FSS based-smart electromagnetic structures are presented in one of our recent articles (Panwar et al 2017).

4. Defect classification and causes

R&S structures are easily damaged by internal manufacturing defects and various external forces. Any damage to the com-posite structure results in a change to the radome’s signature and RCS of stealth structures. The performance of R&S struc-tures is strongly affected by damage due to environ mental factors. Rain erosion may result in delamination and fracture of the walls of an aircraft radome subjected to extreme rain impact beyond its design load. The load is in general a func-tion of a maximum angle of impact, rainfall rate, mean rain drop size and the velocity of the aircraft (Kozakoff 2010). The disbond of fibreglass and foam layer results in a struc-tural weakness and sudden collapse. In addition, the void of the delamination and disbond can become filled with water (water intrusion), disrupting the antenna radiations (Duling et al 2009). The problem of rain erosion can be eliminated by employing rain erosion resistant paint like ‘Caapcoat’ (Kozakoff 2010). The rain erosion boot can also be applied to resolve the issue of rain erosion.

Figure 1. Different radome wall configurations. (a) Half-wave wall solid (monolithic) radome. (b) Thin-wall monolithic structures with a wall thickness equal to or less than 0.1 λ. (c) A sandwich multilayer wall. (d) Multilayer walled structures having five or more dielectric layers.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

6

Lightning strikes and static charge build-up on radome sur-faces are damage caused by atmospheric electricity (Kozakoff 2010). The lightning results in a rapid and a serious threat to an aircraft radome. A lightning puncture is a common type of damage caused in radome structures due to lightning. The lightning puncture damage results in water ingestion and/or moisture absorption over the aircraft radome surface. Such damage greatly affect the transmission characteristics of an aircraft radome. The radio frequency subsystem is also affected by lightning puncture damage. However, the use of an array of lightning diverters over an aircraft radome surface can solve the problem of lightning punctures. The traditional radome structures were protected by utilizing pitot tube and diverter strips.

R&S structures wetting is another critical issue. At micro-wave frequencies, the transmission and reflection characteris-tics of R&S structures are influenced by a thin layer of water on their surface. The presence of water may further cause dam-age to the composite structure due to expansion and contrac-tion of freezing and thawing water (Johnson 2008). The thin film of water may result in signal attenuation in radome struc-tures (Anderson 1975, Dietrich et al 1988). The application of a hydrophobic coating over the surface of the R&S structure can resolve the issue. The hydrophobic surface causes water to bead, and minimize the chances of signal attenuation during rain, etc. The damage due to hail and large bird strikes are also a serious kind of damage to radome structures. The internal manufacturing defects include foreign inclusions, resin rich or starved regions, inter-laminar voids and porosity, fiber mis-orientation, etc. Such defects result in further damage due to excess and fatigue loading (Shin et al 2017).

Figure 2 depicts an image of the orthogonal section of an antenna. Here the defect is represented by a resin-rich area evolving in between the fiber glass layers. Sometimes, low and high cycle fatigue cracks in the aircraft composite may take place due to repeated and excessive loading and vibrations (Shin et al 2016). The aircraft structures are exposed harsh environments (temperature and moisture variations), which may lead to corrosion hidden under the paint and primer. It is very difficult to visualize this kind of defects. The presence of corrosion precursor pitting provides information about the susceptibility to corrosion initiation.

5. Microwave based performance evaluation techniques

Microwave measurement techniques are available in the wide band; therefore, for a range of samples, the microwave fre-quency can be adjusted in order to measure the whole prod-uct. The microwave signal incident on the sample can be taken as a plane wave, and the properties of the sample can be obtained from the reflection from the sample and transmis-sion through the sample. The amount of field strongly depends on the electromagnetic properties of the SUT, coating thick-ness and operating frequency. In this manner, the transmis-sion and reflection methods are based on the measurement of transmitted and/or reflected power from the SUT illuminated by a TEM wave. Further, the frequency dependent dielectric, magn etic and loss tangent characteristics can be obtained using the scattering parameters (S-parameters), such as the transmission coefficient (S21) and reflection coefficients (S11), by employing certain algorithms.

5.1. Transmission line based measurement methods

The transmission and reflection measurements can be per-formed in closed measurement cells, open ended probes or even in a free space environment. Automatic vector network analyzers (VNAs) are the basic building blocks of a trans-mission and reflection measurement setup. VNAs allow fast determination of S11 and S21 of the SUT by simply measuring the absolute value and phase of the sample signal relative to a reference signal (Kaatze 2013). The SUT can be fitted inside a transmission line, either waveguide or coaxial line, in order to determine its frequency dependent dielectric and magnetic properties.

Figure 3 shows a transmission line measurement cell with two ports (i.e. port 1 and port 2). Here, the thickness of thespecimen inserted in the fixture is represented by ‘L’. The distance between the SUT and ports (1 and 2) is sym-bolized by ‘L1’ and ‘L2’. The response of a dual port micro-wave network can be described based on the input and output electro magnetic signals. Figure 4(a) depicts a dual port micro-wave network, which can be utilized to understand the phys-ics of reflected and transmitted electromagnetic signals. The concept of reflected and transmitted electromagnetic signals can be understood with the help of S-parameters. The scat-tering matrix (i.e. S-matrix) provides a complete description of a microwave network in terms of S-parameters as follows (Pozar 1989):

Figure 2. NDE of a cross-section of an antenna radome structure. Reprinted from Shin et al (2017), Copyright 2017, with permission from Elsevier.

Figure 3. Transmission line measurement cell. Reproduced from Krupka (2006). © IOP Publishing Ltd. All rights reserved.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

7

[S] =

[S11 S12

S21 S22

] (3)

where reflected electromagnetic signals are represented by S11 and S22. On the other hand, transmitted signals are rep-resented by S21 and S12. The electromagnetic signal gener-ated in the VNA is transmitted from the transmitting port. It strikes the SUT, is partly reflected and partly transmitted from it and comes back to the VNA either through the transmitting or reflecting ports. The VNA uses the magnitude and phases of these waves to compute the four S-parameters (i.e. S11, S22, S12 and S21). The symmetry of the transmission line gives rise to reciprocity, S11 = S22 and S12 = S21. Therefore, only the input reflection and forward transmission coefficients, S11 and S21, need to be measured. The S-matrix relates the volt-age wave’s incident on the ports to those reflected from the ports. Figure 4(b) shows that the parameters a1, a2, b1 and b2 may be voltage and current in a two port microwave network. The parameters a1 and a2 represents input waves at port 1 and port 2; on the other hand, the parameters b1 and b2 represents output waves at port 1 and port 2. The S-matrix (S) is related to the input wave (a) and output wave (b) by the following expression:

[b] = [S][a]. (4)

When port j is connected to a source and the other port i is connected to a matching load, the reflection coefficient at port j is equal to Sjj. On the other hand, when port j is connected to a source and port i is connected to a matching load, the trans-mission coefficient from port j to port i is equal to Sij

Γj = Sjj =bj

aj (5)

Γj→i = Sij =bi

aj. (6)

5.1.1. Waveguide based measurement techniques. The in situ measurement of the electromagnetic properties of the

perfect electric conductor (PEC) backed microwave absorb-ing structures (MASs) over a broad frequency range is an extremely challenging task for researchers. In cases where defects of the SUT is suspected, even a rough estimate of the SUT parameters is valuable to augment visible and mechani-cal inspection, and to correlate with the results from other field based methods (Dester 2010). Waveguide methods have been widely reported for the measurement of structural transmis-sion and reflection coefficient, in which a sample is placed in a suitable waveguide section (Bergo et al 2006, Pauli et al 2007, Guo et al 2011, Hyde et al 2011). Figure 5 shows a basic single port waveguide based-microwave measurement setup with a SUT of length L inserted inside the waveguide body. Figure 6 depicts two configurations of rectangular waveguide sections consisting the SUT. Here, electromagnetic signals incident on the SUT and are reflected back in terms of S11. The transmission and reflection based dual port waveguide methods can be employed for the simultaneous extraction of complex dielectric and magnetic properties of composite structures (Courtney 1998, Ma et al 1999, Al-Moayed et al 2008).

Figure 7 presents a dual port waveguide measurement setup for the extraction of frequency dependent ε′, ε″, µ′ and µ″ values of R&S materials at the X-band. The measurement setup consists of a VNA, Model No. Agilent N5247A PNA series, X-band waveguide calibration kit, sample holder/fix-ture and Agilent 85071E material measurement software. The material–epoxy composite mixture can be used for the prep-aration of a rectangular pellet SUT of dimensions 22.86 mm (length) × 10.16 mm (width) for the X-band. The S-parameters of rectangular SUT can be measured by placing the SUT in the WR-90 waveguide. The frequency dependent ε′, ε″, µ′ and µ″ values can be calculated from the measured S-parameters based on the ‘ref/trans µ and ε polynomial fit model’. It is also possible to determine ε of thin dielectric structures, which do not completely fill the entire sample holder (Hasar 2009). The method extracts the ε from reflection using only a single-port VNA system and provides an error-free and cost-effective solution as compared to dual-port systems.

5.1.2. Coaxial line measurement technique. The transmis-sion and reflection based coaxial line measurement technique can be employed for the electromagnetic characterization of

Figure 4. A dual port microwave network. (a) Representation of reflected and transmitted electromagnetic signals using S-parameters (S11 and S21). (b). Voltage and current representation.

Figure 5. Waveguide based measurement setup showing incident and reflected signals. Reproduced from Guo et al (2011). © IOP Publishing Ltd. All rights reserved.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

8

R&S structures (Ebnabbasi et al 2013, Song et al 2013). Fig-ure 8 depicts a coaxial line based measurement setup consists of a VNA, coaxial line (with inner and outer conductors) and software to compute the frequency dependent electro magnetic properties of R&S structures. A 50 Ω airline can be adopted as a coaxial sample holder. The diameter of the outer and inner conductors is 7.0 mm and 3.04 mm, respectively. The thickness of the SUT is represented by tm. This method can be utilized to measure the electromagnetic properties for a broad frequency range, but the major limitation is the sample preparation. The sample used for such measurement setup is toroidal in shape, which is comparatively difficult to manufacture.

5.2. Free space measurement techniques

The waveguide and coaxial methods suffer from narrow bandwidth and typically require a specific shape to accu-rately examine a particular sample. FSM has no tolerance requirements for fitting the samples in the waveguide or cav-ity, which can become very stringent at high frequencies. The FSM setup usually consists of two horn antennas facing each other at the open end either directly or through reflecting mirrors and connected to a VNA system from the back end. The diffraction effect at the edges of the sample and multiple reflections between horn antennas can cause inaccuracies

in the measurement. The inaccuracies due to diffraction from the SUT can be eliminated by satisfying the condition D ⩾ 3d, where D and d represents the minimum transverse dimensions of the plate and the beam width of the antennas at the focus, respectively (Ghodgaonkar et al 1989). Generally, the spot focusing horn lens antennas are preferred for the elimination of diffraction effects. On the other hand, multiple internal reflections can be reduced by employing time gat-ing and calibration. The selection of components is the first important aspect for the development of an FSM system. The selection is made based on the number of ports utilized (i.e. the one port or dual port measurement technique) and the operating frequency range.

5.2.1. Single port free space measurement method. The single port FSM (i.e. SP-FSM) method is employed for the measurement of the reflection coefficient of stealth structures. A schematic of a single-port FSM setup is shown in figure 9. The measurement setup shows the reflected signal coming out from the SUT with layer thickness (d). The SUT is backed with a PEC in order to avoid any transmission of electro-magnetic waves through the SUT. FSM imposes a limit on the minimal sample size. If the sample size is much smaller than λ of the electromagnetic wave, the response of the sample to microwaves will be similar to those of a particle object. To achieve convincing results, the size of the SUT should be larger than λ of the electromagnetic wave. FSM can be uti-lized for the NDE of planar as well as non-planar samples. Recently, a FSM set-up has been employed for measuring the shielding effectiveness (SE) of FSS based MASs for X-band applications (Nauman et al 2016).

5.2.2. Dual port free space measurement method. The dual port FSM (i.e. DP-FSM) setup consisting of a VNA, a pair of horn antennas, material measurement software and a comp-uter. Such a measurement strategy can be employed for the measurement of the frequency dependent S11, S21, ε′, ε″, µ′ and µ″, and loss tangent values of R&S structures at desired microwave frequencies (Ghodgaonkar et al 1989, 1990, Varadan et al 2007, Awang et al 2013). The bi-static RCS measurement of R&S structures can be carried out using a DP-FSM setup (Huang et al 2016, Pan et al 2016). The DP-FSM can also be employed to examine the effects of the sur-face curvature and beam defocusing on the characterization

Figure 6. Two configurations of waveguide sections containing SUTs: sample (1) filling the cross section of a rectangular waveguide with dielectric windows (2) and the slab-shaped sample (1). Reproduced from Kaatze et al (2010). © IOP Publishing Ltd. All rights reserved.

Figure 7. Dual port X-band (i.e. WR-90) waveguide based microwave measurement setup for the performance evaluation of R&S materials.

Figure 8. Coaxial line based microwave measurement method for R&S materials.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

9

of PEC backed MASs (Rothwell et al 2013). A compact, non-destructive DP-FSM setup has been developed at Opto-Electro-Structural Laboratory of Korea Advanced Institute of Science & Technology (KAIST), South Korea, as shown in figure 10. Here, the beam focused antennas mounted with metallic fixtures were interfaced via coaxial-to-waveguide adaptors with the VNA system. An absorber was used to absorb the electromagnetic waves in order to avoid multiple internal reflections, which may cause measurement uncertain-ties. Such a DP-FSM setup can also be used for the electro-magnetic characterization of curved structures. The far-field focusing ability of the beam focused horn antennas is helpful in the development of an NDE system for the microwave char-acterization of samples in high- and low-temperature environ-ments (Jose et al 2001). In addition, a DP-FSM setup can also be employed for the high temperature measurement of the dielectric properties of an SUT at microwave frequencies (Varadan et al 1991, Hollinger et al 2000, Varadan et al 2010). The measurement setup consists of a high temperature furnace (up to 850 °C) in addition to a VNA and horn antennas. The temperature of the furnace temperature can be controlled with the help of a computer. The quasi-optical FSM (QO-FSM) setup without time-domain gating can be utilized for compos-ite structural characterization at the W-band (Bourreau et al 2006). The Gaussian optic lens antennas can be employed for the signal transmission and reception in the QO-FSM setup.

The transmission losses in the cable and antenna mismatch in the FSM can cause various measurement uncertainties. Therefore, calibration methods are very much required in order to provide accurate transmission and reflection char-acteristics of the SUT (Rolfes et al 2004). The parallel and perpend icular reflection coefficients of planar SUTs at oblique incidence can be measured by employing a free space bi-static calibration (Umari et al 1991). A calibration algorithm with unknown line-series-shunt standards can be used for the broadband S-parameter measurements (Huang et al 2008). An LRR based self-calibration technique may be adopted for the calibration of the VNA system (Rolfes et al 2003). Here L represents a line element and R symbolizes a sym-metrical reflection standard. The 85071E software offers an optional free space calibration routine called gated reflect line (i.e. GRL). This calibration routine increases the ease of use and reduces the costs associated with through-reflect-match

(TRM) and through-reflect-line (TRL) calibration methods. The GRL calibration technique converts a coaxial or wave-guide dual-port calibration into a full dual-port free space cali-bration. This option also includes a gated isolation/response calibration, which reduces errors from diffraction effects at the sample edges and multiple residual reflections between the antennas. In the VNA, additional ease and time saving is provided with the use of electronic calibration, which fea-tures a guided calibration wizard and results in a fast and easy calibration process. The line-reflect-match (Eul et al 1988), line-reflect–reflect-match (Davison et al 1990), bi-static cali-bration (Umari et al 1991), line-series-shunt (Huang et al 2008), line-network-network (Orlob et al 2013) are some other calibration methods reported by researchers.

According to the orientation of the sample in the meas-urement system, the electromagnetic wave signal incident to the sample may be normal or oblique to the SUT. Therefore, the transmission and reflection coefficient can be measured in both the cases: (a) normal incidence and (b) oblique inci-dence (i.e. the input plane wave approaches the sample from a different angle of incidence, depending on the sample posi-tion). Most of the performance evaluation techniques of R&S structures are limited to normally incident plane waves. In this manner, the effect of variation in angle of incidence can be studied with respect to the transmission and reflection charac-teristics of the SUT.

5.2.3. Naval Research Laboratory arch free space measure-ment method. The Naval Research Laboratory (NRL) arch FSM is a well established measurement technique for evalu-ating the transmission and reflection characteristics of R&S structures over a broad frequency range. This technique was originally introduced by NRL for measuring the angu-lar dependent microwave absorbing performance of stealth structures. As the name of the technique suggests, here the

Figure 9. Single-port FSM setup for the measurement of S11.

Figure 10. A compact DP-FSM setup at OESL, KAIST, South Korea.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

10

measurement system is crafted out of a physical arch consist-ing both transmitting and receiving antennas. The transmitting antenna is used to transmit the electromagnetic signals over the SUT, while reflecting signals will be collected by a receiv-ing antenna. Recently, the NRL arch FSM has been employed for measuring the reflectivity of carbon nano-tube (CNT) based MASs in the range 2.0–18 GHz (Zhang et al 2017).

5.2.4. Far field test method. The radiation characteristic in the far field (or Fraunhofer field) is an important aspect as far as practical electromagnetic applications are concerned. The performance and NDE of R&S structures can be carried out in outdoor far-field ranges where

R 2D2

λ. (7)

Here D represents the diameter of the largest antenna and λ is the free space wavelength, which can be estimated based on the centre operating frequency. There are several advantages of far-field test (FFT) measurements (Kraus 1950).

• The measured field pattern is valid for any distance in the Fraunhofer region.

• The coupling and multiple reflections between the antennas are not significant.

• The chances of measurement errors due to the rotation of the SUT are comparatively less.

FFT can be utilized to measure the scattering strength of curved shaped R&S structures (Yang et al 2016b). Figure 11 shows an FFT setup: a meta-skin wrapped nylon rod is placed in between two horn antennas. The distance between two horn antennas (transmitting and receiving) from the sample (i.e. L) is equal to 80 cm. During the measurement, the transmit-ting antenna can be fixed, while the receiving antenna can be moved in order to obtain scattered signals from distinct angles, θ, w.r.t. the transmitting antenna. The source reconstruction of far-field data can be used for defect imaging in frequency selective radomes (Persson et al 2013). The major limitation associated with FFT is the required large distance between the antennas, leading to larger antenna ranges. Sometimes, this distance may be too large for R&S measurements in the

anechoic chamber or it may result in atmospheric attenuation. Under such conditions, it is advisable to carry out the meas-urements in the radiating near field region.

5.2.5. Near field test method. The near field region can be classified as radiating near field region and reactive or evanes-cent region. The radiating near field measurement is preferred as compared to reactive near field measurements, because in the reactive near field region the SUT will be located too near the antenna, which results in mutual impedance caused by reactive coupling between the antennas. Such an effect cre-ates complications during measurement of R&S structures. In the near field test (NFT), first the electromagnetic field is measured near the antenna system and then the data is math-ematically transformed to any arbitrary location (Kozakoff 2010). Basically, the NFT is a type of microwave hologra-phy, which is based on the Huygens’s principle. The near-field microwave, dual behavior resonator filters can be used for micro-crack detection and imaging (Kerouedan 2008). NFT measurement systems are comparatively small, portable, user friendly and require low microwave power. Near field probes are very helpful to provide high spatial resolution. Moreover, a bulky antenna is not required for this type of measurement (Zoughi 2000).

5.2.6. Compact antenna test range method. The compact antenna test range (CATR) is a facility that provides conve-nient testing of antenna systems at frequencies where obtain-ing far-field spacing to the antenna under test is infeasible using traditional methods. The CATR simulates an infinite range length by producing a flat phase front with a lens, holo-gram, horn, array or a reflector. CATR is generally installed indoors in an anechoic chamber, but very large CATRs are carried out at outdoor ranges only. CATR is one of the most useful methods utilized by researchers for diverse practical applications (Repjar et al 1982, Menzel et al 1984, Parini et al 1988, McKay et al 1991). The CATR method utilizes a source antenna, which radiates a spherical wave front and one or more secondary reflectors to collimate the radiated spherical wave front into a planar wave front (Kozakoff 2010).

A direct-broadcast satellite reflector antenna, without an edge treatment, has been utilized to generate the quiet zone (Chang et al 2004). The fields due to the edge diffractions, feed spill over and other stray signals can be gated out by the impulse time-domain antenna measurement system in order to enhance the performance of the quiet zone. In the same manner, the time synchronization and pulse integration of the impulse time-domain antenna measurement system can be used for the elimination of radio frequency interference in the environment. A hologram based CATR can overcome the limitations of conventional reflector type CATR like tight surface accuracy requirements for the reflectors and high cost. The hologram CATR can be employed for the characteriza-tion of antenna and RCS measurement (Hirvonen et al 2007, Koskinen 2005, Lonnqvist et al 2006, Capozzoli et al 2007). The performance of the CATR measurement setup can be assessed in terms of quiet zone size and field uniformity.

Figure 11. FFT setup for measuring the scattering from the meta-skin wrapped nylon rod. Reproduced from Yang et al (2016b). CC BY 4.0.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

11

CATR mechanically collimates the electromagnetic energy, thus creating a plane wave useful for testing anten-nas in the far-field. In general, the physical dimensions of a compact range setup should be less than the size required for a full-size, far-field anechoic chamber. The phase and ampl-itude variations across the antenna under test should be less than 10° and −0.5 dB, respectively, for the characterization of radomes. The main problem associated with this method is its cost. The cost of fabrication of the specially designed reflec-tor can be comparatively more due to the need for a precision reflecting surface.

5.3. Gaussian beam test method

Gaussian beam test (GBT) method can be utilized in the analysis of quasi-optical systems and reflector antennas. A Gaussian beam is a field whose exponential amplitude and phase have a Gaussian taper in the direction of wave propaga-tion. The computation of edge diffraction is the major issue in GBT. A modular GBT based on a 3D-diffraction technique can be employed to handle the scattering from the reflector edge (Lu et al 2015). The method consists of a frame based Gabor transformation, Gaussian beam reflection, and a novel 3D-diffraction method. A 160 GHz Gaussian beam microwave interferometer in low density RF plasma has been reported (Dittamann et al 2012).

5.4. Coplanar strip-line resonator

It is well known that ε of the structure greatly affects its electro magnetic performance. In addition, the knowledge of ε is also required for the NDE of R&S structures. There is no field ready method available to evaluate the electrical consis-tency of the composite structures. However, a coplanar strip-line resonator (CSR) based advanced NDE technique can be employed to solve the issue (Johnson 2008). A CSR consists of two parallel metal strips impacted over a printed circuit board. CSR is very sensitive to both the dielectric constant of the substrate as well as whatever material or other impu-rities settles over the surface of the metal strips. In general, CSR employs a sensor, which indicates a change in dielectric properties of the sample based on the shift in the resonant fre-quency. The shift in resonance frequency generally depends on the extent of the dielectric constant. The larger the value of the dielectric constant, the stronger the shift in the reso-nant frequency. In this manner, CSR can effectively inspect the defects in the R&S structures.

5.5. Open ended rectangular waveguide (OERW)

The OERW can be efficiently employed for electromagnetic performance evaluation as well as defect visualization of com-plex composite structures (Zoughi 2000). Such measurement strategies are popular due to various advantageous features. Firstly, such measurements can be carried out in both a con-tact as well as non-contact fashion. Images with high spatial resolution can be produced; therefore, it is comparatively easy to identify small sized pits. OERW can be employed

in non-uniform manual scanning for rapid microwave NDE imaging in order to produce complete synthetic aperture radar (SAR) images (Case et al 2013).

Open ended probes such as standard rectangular, tapered, dielectric slab loaded and dielectric waveguide structures can be used. Generally, tapered waveguides can provide high spa-tial resolution, but result in unwanted multiple internal reflec-tions, which greatly affect the radiation efficiency (Ghasr et al 2005). Figure 12 shows an OERW based NDE setup consist-ing of a VNA, open ended waveguide, scanner, scanner con-troller, SUT, GPIB and co-axial cables.

A MMW-OERW based probe can also used for the detec-tion of corrosion precursor pitting (Qaddoumi 1997, 2000, Fitzgerald 2001, Hughes 2001). When microwave signals interact with the defects, the phase variation of the reflected signal can provide information about the presence of defects in the composite structures. An OERW probe based NDE method can be used for the electromagnetic characterization of PEC backed MAMs (Dester 2010). A theoretical reflection coefficient can be obtained by a full-wave solution, and then an extrapolation scheme can be utilized to minimize the errors.

6. Advanced non-destructive evaluation techniques

6.1. Millimeter wave imaging technique

The MM wave covers the frequency range of 30–300 GHz with a corresponding wavelength range of 10−1 mm. MM waves possess numerous unique characteristics, which make them particularly suitable for subsurface sensing applica-tions (Bowring et al 2008). MM-WI waves can be utilized to interrogate and image the interiors of aircraft composite mat-erials as they can easily penetrate inside them. Unlike sound waves, MM waves do not experience severe attenuation while propagating in free space, thus, they result in non-destructive

Figure 12. An OERW based NDE setup. Reproduced from Yin et al (2015). © IOP Publishing Ltd. All rights reserved.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

12

measurements. Another important advantageous feature of MM-WI is images with relatively high spatial resolution, due to their small wavelengths. Unlike x-rays, electromagnetic radiation in the MM wave is non-ionizing and safe, and, hence, the associated imaging equipment and procedures can be read-ily deployed. Furthermore, unlike visible and infrared systems, passive MM-WI systems are not significantly hindered by atmospheric obscurants, like rain, smoke, fog, cloud cover and dust storms, and can even eliminate the impact of low-visibil-ity atmospheric conditions. Figure 13 shows the atmospheric attenuation of naturally emitted black-body radiation through 1.0 km of fog, showing how low-loss bands within the MM wave region allow passive imaging even in poor weather con-ditions. Recent developments in the field of MM wave technol-ogy and devices, along with their more prevalent commercial availability, have focused the attention of researchers towards this frequency range for imaging and NDE applications (Sheen et al 2001, Kharkovsky and Zoughi (2007).

The MM-WI based NDE techniques, using near-field, lens-focused, synthetic aperture focused and 3D (holographic) techniques, have been effectively utilized for various practical applications. Such methods have been applied toward inspect-ing a wide range of aerospace composite structures. In gen-eral, synthetic aperture MM-WI has shown great utility for inspecting the space shuttle external fuel tank spray-on foam insulation and its acreage heat tiles, as well as honeycomb composites similar to those used in aircraft radomes (Ghasr et al 2011). MM-WI techniques, based on synthetic aperture focusing can be used for effective NDE of aerospace struc-tures. MM-WI can detect and measure concealed di electric layers using reflection of coherent frequency scanned MM waves (Bowring et al 2008). This can be achieved by measur-ing Γ of the object as a function of frequency. This technique is able to detect and monitor the multi-layers of dielectrics through their microwave optical properties. A typical MM-WI measurement setup consists of a VNA, active harmonic gen-erator, horn antennas with a focusing dielectric lens and

corre sponding connecting accessories. The MM wave power can be directed to a sample surface at various angles of inci-dence. The rectified MM wave signals from the detector can be fed through a coupled amplifier to the data acquisition board input of a laptop. This MM wave scanning and detection sys-tem is able to scan at a rate of 1 ms per sample in order to obtain a dataset of 256 or more points. A typical 3D-scanning di electric probe MM-WI system consists of a millimeter wave GUNN diode oscillator, Schottky diode detector, dielectric probes, an xyz-stage, and an automatic scanning and image

Figure 13. Low-loss bands within the MM wave region allow passive imaging in adverse weather conditions. Reproduced with permission from Martin et al (2009).

Figure 14. Different MM-WI setups. (a) A 3D scanning dielectric probe MM-WI setup, (b) a reflective polarization measuring system, and (c) a reflective scanning dielectric probe MM-WI setup. Reproduced from Kume et al (2008). © IOP Publishing Ltd. All rights reserved.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

13

processing system, as shown in figure 14(a). The polarization of the radiated wave can be measured using a reflective polari-zation measuring system as shown in figure 14(b).

The scanning dielectric probe MM-WI measurement setup can be transformed into a reflection type scanning dielectric probe MM-WI, which can be employed for NDE of composite structures (Kume et al 2008). Figure 14(c) depicts the meas-urement setup for a reflection type scanning dielectric probe MM-WI system. A list of other MM-WI systems reported by researchers is presented in table 1. The MM-WI technique can also be employed to extract the ε data of lossy liquids in the range 26–110 GHz. Such data can be incorporated in a num-ber of numerical weather prediction models and microwave and MM-WI instruments (Lamkaouchi 2003). To obtain a high-resolution image of large area samples in a short dura-tion of time is very much required for a good imaging system. However, the use of a fast scanning platform, a fast-responding transceiver, and the proper communications software capable

of accommodating both, can be very helpful in this direc-tion. MM wave active imaging plays a significant role in the field of microwave engineering due to its variety of applica-tions. However, a serious problem associated with MM wave

Table 1. Different MM-WI configurations reported by researchers.

MM-WI technique utilized Frequency range utilized (GHz) Key features Reference

Near-field MM wave probe based MM-WI

26.5–40 and 50–75 • Used to detect very small pits on aircraft surfaces using raster scanning

Ghasr et al (2005)

• High level of detection sensitivity• High spatial resolution and high contrast

imagesMultiple-input-multiple-output based MM-WI

75–90 • Based on a hybrid concept with SAR and digital beam forming

Gumbmann et al (2011)

• Good compromise between lateral resolution and depth of penetration

Digital reflect array based terrestrial-MM-WI

220 • Application of a confocal Gregorian optics geometry

Dietlain et al (2012)

Interlaced Tx/Rx array based 1D-MM-WI

30 • Linear high resolution and rapid MM-WI Ghasr et al (2013)• Capable of fast image production• Reduced scan time as compared to raster

scanningConical scanning MM-WI radiometer

50.3, 52.8, 89 (dual-polarized), 165 (dual polarized), 183.3 ± 1, 183.3 ± 3, and 183.3 ± 7 GHz

• An airborne total-power radiometer Wang et al (2013)• Capable of measuring scattering signatures

from storm-associated hydrometeors in dual polarization

Reflectometer based 3D -holographic-MM-WI

26.5–40 • Broadband, small, and cost-effective reflectometer

Ghasr et al (2014)

• Enables production of image slices at different depths

Commercial plasma display based MM-WI

62–90 • Utilizes low cost plasma display as an imager

Brooker et al (2015)

• Performance is not as good compared to GDDs

• Effective for short range imaging3D-compressed sensing based- MM-WI

55–65 • Computation time reduction Alvarez et al (2015)• Reduction in the number of samples

Polarization based passive MM-WI

94 • A polarization based method is presented to acquire the object surface orientation information

Cheng et al (2016)

• Possible applications for object detections and terrain models

Active MM-WI system 75–110 • Raster scanning system with reflection and transmission modalities

Jia et al (2016)

• Better spatial resolution, high power and good penetration of W-band

Figure 15. General schematic of a 1D-imaging system.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

14

circuits is their lossy nature and costly amplifiers. Therefore, the envelope phase detection technique can be utilized, where MM waves can be detected without amplification just after being received at the antenna (Radenamad et al 2011).

Raster scanning is basically a slow process, and in addi-tion requires bulky, slow and expensive scanning platforms. However, such limitations can be eliminated with the help of an array of probes. A significant amount of time can be saved during the image acquisition process by using an imaging array instead of a single probe. In addition, cost-effectiveness is an important advantage of the 1D-imaging array as compared to a 2D-imaging array. The 1D-imaging array provides a cost-effec-tive and complexity-free solution for many NDE applications. Based on such advantageous features, a linear 1D-MM wave array based imaging system has been reported (Ghasr et al 2013). The basic schematic diagram of a 1D-imaging system is shown in figure 15. The imaging setup consists of a transmit-ter (Tx)/receiver (Rx) array configurations. The radio frequency circuitry generates the electric field, which is transmitted by the slot antennas in the Tx array one at a time, illuminating the object to be imaged. The slot antennas in the Rx array collect the scattered electric field synchronously with the Tx array.

Here the switching operation was performed using switch-ing elements like PIN diodes. The imaging array operates in quasi-mono-static reflection mode, which results in a coher-ent vector reflection coefficient data for the generation of high spatial resolution SAR images. The role of a field program-mable gate array (FPGA) is to generate the modulation signal and to control the sequential electronic scan on the array. The FPGA then transfers the collected data to a computer to per-form other required operations on the collected data such as calibration and image formation. Such techniques provide a fast electronic scan and greatly reduces the scan time as com-pared to raster scanning.

There is also the possibility of developing a cost-effective MM-WI system using a commercial plasma display (Brooker et al 2015). The detection mechanism for glow discharge detec-tors (GDDs) at MM wave bands can be explained by cascade ionization in which the energy of the electrons in the plasma increases due to the electromagnetic radiation, results in an increase in the current density. The strong applied electric field accelerates signal electrons sufficiently to generate an avalanche effect, which results in an amplification of up to 106. Table 2 lists distinct practical applications of MM-WI based systems.

6.2. Terahertz wave imaging technique

THz waves occupy a frequency band of about 0.1 THz to 10 THz lying between the microwave and infrared regions

(Hu et al 1995, Dobroiu et al 2006). This region of the electro-magnetic spectrum has been the least explored compared to other regions due to technical difficulties involved in mak-ing efficient and compact THz sources and detectors (Chen et al 2012, Wu 2012). Electronic detectors such as Schottky diode based-detectors provide the phase information, but suf-fer from the skin effect at THz frequencies. On the other hand, a photon detector such as a bolometer effectively measures thermal background noise, but does not providing phase infor-mation (Duling et al 2009). Nowadays, THz-WI has become a valuable tool for NDE as new laser sources and detectors are launched onto the market. The sub-millimeter wavelength, non-destructiveness and transparency of common materials, are a few unique characteristics of THz waves, which makes them suitable candidates for NDE (Siegel 2002, Juuti et al 2009, Hou et al 2011, Fukasawa 2015). The spectrum peak of the THz wave of most conventional THz sources such as photoconductive antennas or electro-optical emitters is about 1 THz. In addition, the spatial resolution of THz imaging is limited to several hundred micrometers (Chen et al 2001). The basic principle of THz-WI is based on the fact that the transmission and reflection characteristics of objects lead to THz scattering, which greatly affects the electric-field inten-sity distribution. The location of internal defects over the object can be identified on the basis of different luminance distributions (Chen et al 2012). THz-WI measurements can broadly be divided into active or passive THz-WI. The pas-sive THz-WI, employs a small 1D-array of detectors to mea-sure the THz radiation, in contrast, active THz-WI utilizes both a THz source and a detector. The pulsed or time domain based-active THz-WI measurements provide better target information, highest signal-to-noise ratio (SNR) without any range ambiguity. Pulsed THz-WI provides structural informa-tion and the spectral or material composition of the sample, owing to changes in the spectral contents of the reflected THz waves. The fibre-optic coupled pulsed THz-WI measurements provide a compact, rugged, powerful and highly flexible sys-tem for implementing a broad range of THz measurements (Duling et al 2009).

Pulsed THz-WI is widely adopted for NDE applications in the aerospace and military sectors including radome struc-tures (Duling et al 2009). It is possible to scan the wide area of radome (affected by delamination and water intrusion) using THz waves. In a THz–radome inspection system, the scanner is connected to the main control unit with the help of a fibre-optic cable. A scanner with 15 cm scan width can be rolled across the radome surface at a speed of 12.7 cm−1. In this manner, a THz–radome inspection system can be effectively adopted for the collection of regular health history of radome

Table 2. Distinct practical applications of MM-WI based systems.

Applications of MM-WI techniques Reference

Space shuttle external tank foam insulation Kharkovsky et al (2005, 2006)Radome composites Ravuri et al (2008)Corrosion pits, structural defects and beat damage Ghasr et al (2005), Kemp et al (2010) and Ghasr et al (2011)Extraction of ε of lossy liquids Lamkaouchi (2003)Short range applications Gumbmann et al (2011)

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

15

structures. NASA has also utilized reflection based time-domain THz-WI to examine the interface between the space shuttle external tank and spray-on-foam insulation. Pulsed THz-WI is not only utilized for the observation of the varia-tion in reflectivity of the pigments, but can also be employed for the mapping of the inter-layer structure of art/paints. It can reveal hidden structures, the examination of which is impos-sible without destroying the artwork.

THz-WI methods can also be classified as coherent and incoherent THz-WI methods. Tomography, electro-optic sampling, imaging and time-domain spectroscopy based methods fall under the category of coherent THz-WI, on the other hand, continuous wave THz-WI (CW-THz-WI) is an incoherent method (Hu et al 1990, Nahata et al 2002). Time domain measurements are well known more broad bandwidth and good resolution, but suffered from low intensity due to the longer scanning time for the objects. The CW-THz-WI system is a simple, fast, compact and cost effective method, but is unable to provide either time-domain or frequency-domain information of the target as a fixed-frequency source and a single detector is employed (Karpowicz et al 2005). In general, there is a need for fast, cost-effective and room-temperature operated THz detectors for THz-WI. Golay cells, Schottky diodes, and pyroelectric detectors are three impor-tant types of THz detectors, but are comparatively expensive (Maestrini 2004). However, a GDD based THz wave detector can efficiently be employed for THz-WI applications due to its advantageous features such as low cost, room temperature operation, broad spectral range, fast response time and large dynamic range. Earlier, GDDs were used for microwave and MM wave detection, but recently their applications for higher frequencies within the THz band of the electromagnetic spec-trum have been reported (Hou et al 2011).

Figure 16 depicts a schematic diagram of a GDD based CW-THz-WI setup (Hou et al 2011). A Gunn diode emitting a 0.2 THz beam has been utilized as a THz source. The beam is focused on the sample by a 66 mm focal length high-density polyethylene (HDPE) lens. After being transmitted through

the sample, the THz beam is focused on the electrode gap of the GDD by another 66 mm focal length HDPE lens. A metal reflector is used on the GDD surface, which increases the signal by reflecting and focusing the THz wave on the elec-trode gap. A direct current power supply is used for the bias-ing of GDD to break down the inert gas. The current through GDD changes with an increase in ionization rate of the gas due to the incident THz radiations. The lock-in-amplifier is employed to measure the alternating current signal separated from the direct current bias using an alternating current cou-pler. The signal quality can be improved by eliminating the noise using a filter and an amplifier. The raster scanning of the sample has been carried out using a motorized scanning stage.

The near-field THz-WI techniques can be used to improve the spatial resolution of THz-WI (Hunsche et al 1998). The high-pass filtering of the THz signal due to the waveguide effect, bandwidth issues are two major problems associated with such techniques. However, the use of a semiconductor dynamic aperture for near-field THz-WI can solve the prob-lem (Chen et al 2001). There is no requirement for the physi-cal aperture in the case of a semiconductor dynamic aperture based near-field THz-WI method. The near-field THz aperture can be realized using a transient photo-carrier layer induced by an optical gating beam. The size of the photo-carrier layer can be determined by the focal size of the optical beam. A CW spectroscopic THz-WI system can be developed using InGaAs bow-tie diodes at room temperature (Kasalynas 2013). Figure 17 depicts the measurement setup, consisting of an optically pumped molecular THz laser, reference detec-tor (D1), signal detector (D2) and mirrors (M1–M4). Table 3 shows various other THz-WI based systems reported by the researchers.

Figure 18 depicts a typical measurement setup for free-space coherent broadband THz time-domain spectroscopy (Han et al 2001). The major components of the measurement setup are a laser, a THz emitter, a THz detector, focusing and collimating parts for laser and THz optics, a chopper, a motorized delay line, a lock-in amplifier and a data acquisi-tion system. The laser pulses can be divided into the probe and pump pulses. A single cycle burst of 1 ps duration from an emitter can be produced with the help of pump pulses. In THz spectroscopy measurements, an object is exposed to the THz pulses. An image of the object can be collected by moving the object in the plane perpendicular to the THz beam. The laser probe and THz beam simultaneously arrive at an electro-optic sampling detection system. As per the Pockel effect, a THz wave electric field modulates the polarization ellipiticity of the optical pulses passing through the electro-optic crystal.

6.3. Eddy current pulsed thermography

Thermographic techniques play a significant role in the NDE of composite structures (Sfarra et al 2012, Pawar et al 2013, Sharath et al 2013, Boccardi et al 2016, Yang et al 2016a). Pulse and lock-in thermography are two common techniques reported by researchers (Meola et al 2004). An ECPT based measure-ment technique combines both eddy current and thermography

Figure 16. Schematic diagram of the GDD based THz-WI setup.

Figure 17. Schematic diagram of a InGaAs bow-tie diode based spectroscopic THz-WI measurement setup.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

16

Table 3. Various THz-WI methods reported by researchers.

Measurement technique Frequency Key features Reference

Semiconductor dynamic aperture based near-field THz-WI

0.9 THz • Provides a sub-wavelength spatial resolution of ~λ/10

Chen et al (2001)

• Free from the high-pass filtering effectHybrid CW demodulating multipixel THz-WI

2.0 THz • Provide a demodulating detector array for phase-sensitive multipixel THz detection

Friederich et al (2010)

GDD basedTHz-WI 0.2 THz • Provision for an efficient, rapid and cost effective detector

Hou et al (2011)

• Improved efficiency• Low noise

Optimization based THz-WI 0.2–0.9 THz • Based on nonlinear optimization that is not limited by weak scatter or low refractive index.

Tsai et al (2012)

• Faster convergence and improved image quality

Backward wave oscillator based THz-WI

450 and 890 GHz • High output power and good wave-front quality

Chen et al (2012)

• Tunability of working wavelength• Better image resolution• High SNR

2D-spatiotemporal THz-WI 0–2.98 THz • Based on non-collinear free-space electro-optic sampling

Yasui et al (2013)

• Provides high-speed image acquisition at much higher rates

InGaAs bow-tie diodes based spectroscopic THz-WI

0.585–2.52 THz • Bow-tie diodes are very robust, easy to produce, fast, and wideband

Kasalynas et al (2013)

• Good candidate for a compact all-solid-state spectroscopic imaging system

Schottky diodes based active THz-WI

1.0 THz and 650 GHz • No requirement of mirror/lenses to form THz images

Han et al (2013)

• Results in an affordable and portable fully-integrated CMOS-THz-imager.

Quantum cascade laser based 2D and 3D THz-WI

2.5 THz • A fast scanning mirror, and a sensitive Ge:Ga detector

Rothbart et al (2013)

• Images with a diameter of approximately 40 mm and a SNR of up to 28 dB can be produced within 1.1 s

THz digital off-axis holography imaging

0.495 and 0.710 THz • Imager captures lensless transmission holograms

Heimbeck et al (2015)

• Provide sub-millimeter resolution images of voids within visually opaque printed structures

Metamaterial based THz-WI 1–4 THz • An innovative, cost-effective, uncooled, metamaterial based THz focal plane array

Carranza et al (2015)

• Compact and real-time THz-WI systemChip based near-field THz-WI 2.25 and 2.06 THz • Based on a collection-mode imaging method Kawano (2016)

• Generate frequency-selective near-field THz images

Light-field THz-WI 0.53 and 0.65 THz • Provides sufficient accuracy to estimate the camera operation

Jain et al (2016)

THz dual-frequency interferometry based 3D surface imaging

330 and 335 GHz • Avoids the nonlinear error caused due to signal nonlinearity

Jiang et al (2016)

• Good tolerance with respect to frequency instability

Phase-sensitive single-pixel THz-WI

690 GHz • Cost-effective and accurate imaging system Saqueb et al (2016)• Measurement speed as fast as 0.2 ms/pixel• Accurate reconstruction of both intensity and

phase images using intensity-only measure-ments in the THz band

• Images with superior contrast

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

17

for the detection and separation of impact damage (Ramdane et al 2010, Wilson 2010, Yang et al 2011, Ren et al 2013, Yin et al 2013, Gao et al 2014, Liu et al 2014b). Here a high current electromagnetic pulse interacts with the sample, which results in heating of the sample due to the induction of eddy currents. The presence of defects distorts eddy cur rent propagation, which results in a variation in sample temper ature that can be emphasized by thermography (Meola et al 2004). The quantita-tive NDE of non-homogeneity of the sample can be evaluated using mixed phenomena of induction heating dominating the heating phase and the diffusion of such dominance in the cool-ing phase. The phenom enon of electro magnetic induction can be explained on the basis of the skin effect.

As compared to laser and flash thermography techniques, ECPT not only focuses the heat generation on defects on the surface, but also on the sub-surfaces. µ and σ are two basic parameters for the consideration of ECPT. This measure-ment technique also allows area imaging of defects without scanning. In addition, it is not only utilized for magnetic and non-magnetic metals, but also for reinforced composites with weak conductivities.

Figure 19 depicts the block diagram of an ECPT measure-ment setup. Here eddy currents are induced due to the current in the coil, resulting in the generation of heat in the conduct-ing material. The presence of any kind of defect in the con-ductive material will lead to obstruction of the heat diffusion phenomenon. The variation in the temperature distribution on the material surface will be captured by the infrared camera.

Fatigue damage characterization is a very challenging task; however, NDE of early contact fatigue at different intervals of the loading cycle can be carried out using an ECPT meas-urement setup. The degree of fatigue damage can be charac-terized using a thermo-optical flow entropy tracking method (Liu et al 2015). In addition, the quantitative information of defects, i.e. length, depth and angle are very much required for proper imaging of structures. The quantitative NDE plays a significant role for the proper visualization of defects with corresponding quantitative information about defect geom-etry. The nor malized cross correlation method (NCCM) can be employed for the visualization of interaction of heat with a defect (Al-Qubaa 2010). NCCM is basically a full-field image analysis technique and gives correlate pattern information

from image sequences. The relationship between the defect dimensions and orientation can be estimated using the direc-tion of heat propagation in the scan area. This method provides clearer and faster information about defect geometries. The influence of angular defects on heat distribution can be studied using NCCM in order to provide a quantitative defect char-acterization. Nowadays, researchers are investigating a real-time 3D-thermography method, based on 3D-scanners, for the NDE of composite structures (Hellstein et al 2016). Here, results are depicted in the form of 3D-thermographs, where the temper ature is mapped on a 3D-model reconstruction of the target. Similarly, other thermography methods such as scan-ning laser-line thermography and laser-spot thermography can be employed for crack imaging in the composite structures (Li et al 2011a).

The defect depth quantification is one of the useful applica-tions of quantitative NDE. The pulsed thermography method

Figure 18. Measurement setup for free-space coherent broadband THz time-domain spectroscopy. Reproduced from Han et al (2001). © IOP Publishing Ltd. All rights reserved. Figure 19. Block diagram of the ECPT measurement setup.

Figure 20. Schematic diagram of a single port SFSM system.

Figure 21. Sketch of a dual port SFSM setup.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

18

has been widely reported in the literature for the depth quanti-fication. The defect size is a one of the most important param-eters that affects the measurement performance of any system. There may be a change in heat diffusion rate due to disconti-nuities and defects (represented by cold or hot spots), which results in a corresponding variation in surface temperature distribution. Therefore, the effect of defect size can be stud-ied, which is very much required for a basic understanding of the heat diffusion mechanism. Based on the defect depth and peak time, various methods have been reported by researchers for depth quantification (Ringermacher et al 1998, Zeng et al 2012, Sharath et al 2013). In addition, a theoretical approach for the quantitative surface crack depth evaluation can be per-formed using heat transfer modeling (Streza et al 2013).

6.4. Scanning free space measurement setup for defect visualization

Often, it is important to identify the location of prominent scat-ters of the complex shaped targets in addition to the transmission and reflection coefficients. The target needs microwave imaging to find out the prominent scatters also known as hot spots of the target. The microwave imaging is basically the spatial dis-tribution of scattering in two dimensions to identify and isolate individual scattering centres. Such an approach can effectively be adopted in order to visualize the complete electro magnetic behavior of R&S structures. The FSM setup can be interfaced with a scanning system to measure and visualize graphically the electromagnetic response and defects in R&S structures. The schematics of a single port and dual port scanning FSM (SFSM) setup are shown in figures 20 and 21. Such apparatus can illustrate a potential application in characterizing both pla-nar as well as curved structures. From the measured mappings of the microwave field, the visualization of spatial and physi-cal process, and quantitative characterization of transmission and reflection coefficients can be obtained in a quite efficient manner. Using such an approach, one can not only visualize the significant effect of microwave absorbing coating on the propagation of the microwave, but also gives a very accurate quantitative test on its performance in a quite different way. The automated scanning and data acquisition process can be con-trolled by a computer via both a GPIB and RS-232 interface.

In order to have full imaging information, the sample can be scanned in two orthogonal directions, i.e. horizontal (along the width) and vertical (along the height), such that the sample is covered completely, as shown in the figure 20. The S-parameter readings of these two types of scans viz. A scan and B scan can be obtained. The complete C scan is a 3D matrix, with each cell entry representing intensity values as a function of particular down range, cross range and height index. Data acquired in this way undergoes several signal pro-cessing steps to extract full information of the target.

The antenna motion induced microwave imaging provides freedom to rotate the antenna. Based on horizontal and verti-cal polarization, the frequency dependent S-parameters can be measured as well as visualized. The movable driven carriage can be used for this purpose, having provision for configur-ing horn antennas for transmission and bistatic microwave measurements. The angle and position of the antennas can be assured with the help of an automatic driven positioner. As the antenna moves, at fixed intervals transmit, receive and store operations can be completed. The stored signals can be summed to generate signals equivalent to those received by a physical aperture. The microwave image is generated by pro-cessing and sampling of the received signals scattered from the target at distinct viewing angles.

The effect of sample tilt angle can be studied by rotating the sample at different positions, say angle θ1, θ2,… and so on, while keeping the antenna stationary. Such a study can pro-vide other research perspectives in terms of a complete evalu-ation (i.e. electromagnetic performance and NDE) of the R&S structures as reflectivity due to the different scatterers of the target can be considered here. The measurement can be made with reference to the target’s rotation and range. The single pixel of the image will be equivalent to the area covered by the diameter of the horn antenna. Such study can effectively be employed for both planar as well as curved R&S structures.

6.5. PZT sensing laser ultrasonic propagation imaging

Non-contact and remote ultrasonic generation, good com-patibility with ultrasonic sensors, high spatial resolution, no requirement of adjustment of the incident angle of laser beam, etc are a few advantageous features of laser based ultrasonic techniques. The PZT sensor based laser UPI plays a significant role in radome health management (RHM) (Chia et al 2012). The RHM system consists of in-flight damage detection and ground damage evaluation processes. Figure 22 shows a sche-matic of an RHM system for a jet fighter pod radome structure. The measurement system is divided into an on-board system and a ground subsystem. The on-board subsystem consists of an acoustic emission event logger and six acoustic emis-sion sensors (i.e. PZT based sensors) with a centre frequency of 300 kHz and an integrated amplifier. The function of the on-board subsystem is to detect and record acoustic emission events like impact and material failure in the R&S structures.

During the ground inspection operation, the laser pulses are launched from a Q-switched continuous laser (having a wavelength of 1064 nm) towards the laser mirror scanner (LMS). The LMS consists of two laser mirrors mounted on

Figure 22. Schematic diagram of an RHM system for a jet fighter pod radome made up of glass/epoxy and a honeycomb core. Reprinted from Chia et al (2012), Copyright 2012, with permission from Elsevier.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

19

a galvanometer. The function of the LMS is to deflect the laser pulses to the R&S structure. The laser beam impinge-ment on R&S results in the generation of an ultrasonic wave. The material on the R&S surface experiences local thermal expansion during the laser pulse illumination. On stopping the laser illumination, the R&S material cools down and contracts rapidly. The ultrasonic wave in the R&S structure is produced due to the expansion–contraction of the material (Chia et al 2012). The acoustic emission sensor connected to the ground subsystem acts as an ultrasonic receiver. These ultrasonic sig-nals can be captured using a digitizer and then processed with the help of certain algorithms like wavelet-transformed ultra-sonic propagation imaging, ultrasonic spectral imaging, etc.

6.6. Full-field pulse-echo laser ultrasonic propagation imaging

The pulse-echo ultrasonic propagation imaging (PE-UPI) is an advanced NDE technique that combines the pulse-echo

technique and the ultrasonic propagation imaging technique. A full-field PE-UPI (FF-PE-UPI) measurement system utilizes a laser Doppler vibrometer, a Q-switched continuous wave laser with a wavelength of 1064 nm, a two-axis translation stage and a laptop for control and image processing, as shown in figure 23(a). The scanning pattern and scan representation (A scan, B scan and C scan) is shown in figures 23(b) and (c). Such a non-contact, in situ NDE technique can be employed for the visualization of radome structural defects (Shin et al 2017).

The detailed excitation and sensing principle of the FF-PE-UPI technique is depicted in figure 24. It is shown that the two laser beams impinged by the laser Doppler vibrometer and Q-switched continuous wave laser scan along a raster-scanning pattern. The impinged laser beam results in a thermo-elastic wave on the object surface. These waves propagate through

(a)

(b)

(c)

Figure 23. FF-PE-UPI. (a) Measurement setup. (b) Scanning pattern. (c) A-scan, B-scan, and C-scan representation of a laser ultrasonic wave-field. Reprinted from Hong et al (2016), Copyright 2016, with permission from Elsevier.

(a)

(b)

Figure 24. (a) Concept diagram of a FF-PE-UPI system for composite UAV. The dashed black line represents a generation beam and the red solid line depicts a sensing beam. (b) Excitation and sensing principles of the FF-PE-UPI system. Reprinted from Shin et al (2017), Copyright 2017, with permission from Elsevier.

Figure 25. Geometry of the radome and the antenna. The center of rotation is located at the origin. © 2013 IEEE. Reprinted, with permission, from Persson et al (2013).

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

20

the thickness direction and reflect off the internal structures. The laser Doppler vibrometer can be used to receive the waves and in-line band-pass and numerical filters can be utilized to filter the acquire signals. The NDE of curved surfaces using FF-PE-UPI is a very difficult task. However, this particular issue can be tackled with the help of an image processing algorithm, i.e. ultrasonic energy mapping (UEM) and a cur-vature-compensating subtraction based imaging method (Shin et al 2017). UEM is a wave number based damage visuali-zation method, and visualizes the energy of scattered waves filtered as a single wave mode in the direction opposite to the incident propagation direction (Flynn et al 2013).

6.7. The ISR method

The ISR method has great potential in radome diagnostics (Persson et al 2013). In general, the field components of an antenna can be computed simply by using the sources of the electromagnetic fields. However, there is the possibility to uti-lize an inverse source problem, where an electromagnetic field is known, and the cause of the radiation is unknown. The key challenge is to reconstruct the sources, and to back propagate the measured field components close to the surface.

The ISR based NDE method is based on a surface int-egral representation together with the extinction theorem. A measured far field followed by an equivalent surface current approach can be adopted to obtain tangential field components on the surface of a frequency selective radome structure. The purpose of doing so is to localize the defects on the FSS lat-tice. Figure 25 sketches the radome geometry and standard 18in slot antenna setup. The polar and azimuthal angles are used to illuminate the surfaces of an aircraft radome. The com-putational complexity can be reduced by the presence of axial symmetry in the measurement setup. Here reconstruction is carried out on a fictitious surface in free space, located outside the physical surface of the radome. The ISR method is used to identify defects such as metal patches, dielectric patches and dislocations in FSS based radome structures. Any change in the electromagnetic field results in an unwanted deviation in the far field data, transmission characteristics, side lobe level and beam deflection. The causes of such deviations can be easily understood by the ISR method.

7. Lightning strike and bird strike test methods and facilities for airborne radomes

Airborne radomes are natural targets for lightning strikes and bird strike based major damage mechanisms because they are mounted on the extremities of the aircraft. Therefore, special attention must be given to preserve them from the harmful effects of lightning and bird strikes.

7.1. Lightning based damage mechanisms

The conducting radar antenna housed within non-conducting plastic radome is the primary source of ionized streamers generated at the aircraft nose due to a high stress field. Such

an intense electric field results in punctures and pinholes at the radome surface. Since the radome is usually constructed using low dielectric constant materials, the radome dielectric strength is not sufficient to withstand these lightning flashes, which results in serious damage. The damage mechanisms can be divided as direct and indirect damage. Direct dam-age refers to physical harm to the aircraft such as dielectric punctures, structure blasting, etc. These effects are gener-ally caused at lightning attachment points, along the current flow paths and hang on locations. Such effects include high voltage and high current related damage to the radome struc-ture. On the other hand, indirect damage is caused due to the strong electromagnetic fields of the lightning. Such damage refers to the lightning current flowing through the skin and the body of the aircraft structure inducing strong electro-magnetic fields. Such effects may result in permanent dam-age to the radome as well as other electrical and electronic components. A lightning strike at a distance away from the radome may also result in indirect damage. The strength of a local electric field is utilized to control these two damage mechanisms.

A standard lightning test setup with corresponding optical waveform generated by spark is shown in figures 26 and 27, respectively. On the other hand, antennas used to discriminate between lightning and non-lightning sources of the induced cur-rent are depicted in figure 28. The tests are one way to describe the aircraft resistance to the lightning effect. Generally, tests need to be conducted in a very effective manner. A minor issue may result in unrealistic results. A series of test methods are available for examining the lightning effects. High voltage tests (HVTs) and high current tests (HCTs) can be carried out to demonstrate the resistance to direct effects. On the other hand, equipment and system tests can be employed for demonstrating resistance to indirect effects.

7.1.1. High voltage tests (HVTs). HVTs are performed on a test configuration consisting of the radome, the antennas housed in the radome and a mock-up of the aircraft structure. Such tests are accomplished by exposing the radome to high voltage discharges. HVTs are carried out with standard volt-age waveforms, which depict the electric field variation during the lightning strikes. The main aim of such tests is to examine the dielectric parts of the aircraft (like an airborne radome) to possible damage. In addition, such tests can be employed to compute the zoning of the aircraft. The tests on radomes are carried out using two voltage waveforms.

• The slow waveform, depicting the slow variations of the electric field.

• The fast waveform, representing the fast variations of the electric field.

The detailed pictorial description of these waveforms is available in Lago (2014). In such tests, an arc of certain meters is produced. The current flowing in such arcs is limited to a few thousand of amperes. HVTs simulate the high electric fields that exist around the radome. The corona/streaming test and attachment point test are two widely used HVT based test-ing strategies.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

21

7.1.2. High current tests (HCTs). The aircraft forms a part of the total lightning path, when struck by lightning. Lightning is basically a pulse current flow between the attachment point and exit points of the aircraft body. This total event generally com-prises of several current pulses with different peak current, rise time and charge transfer duration. HCTs are performed accord-ing to the normative current waveforms. These waveforms allow reproduction of the power and energy corresponding to each zone for the evaluation of thermo-mechanical damage. The arc attachment and current conducting tests are common test cate-gories utilized for this purpose. In an arc attachment test, arcs are produced to examine the behavior of the structures. The current conducting test produces a flow of the current within a structure.

7.1.3. Lightning indirect effect (LIE) tests. Lightning indirect effect (LIE) tests apply to the systems for which protection is needed. The basic types of aircraft LIE tests are classified as impulse current injection tests and swept frequency tests. The main target of indirect effects tests is to find out the actual transient levels, which can be utilized to obtain the engineer-ing data to verify the lightning protection design. The main steps in the lightning protection design for an aircraft are as follows (Huang et al 2017):

• Lightning strike point identification. • Internal and external electromagnetic field environment

characterization. • Induced voltage and current identification. • Validation of aircraft lightning protection design.

A typical test configuration includes an aircraft, the impulse current generator, the ground plane and a 0.6 Ω resistor con-necting the aircraft with the ground plane (Huang et al 2017). The function of the resistor is to eliminate the quarter wave-length transmission line resonant response. The typical testing strategies incorporate distinct kinds of current entry and exit points. The primary goal of such tests is to find out the internal electromagnetic field. The knowledge of electro magnetic field characteristics produced by lightning discharges is required for examining the effect of the potentially deleterious cou-pling of lightning fields to radome. The measured electric and magnetic fields can be utilized for the estimation of distinct lightning parameters and for testing lightning models.

The airborne radome structures can be protected from light-ning effects by providing an electrostatic shield over a radome surface (Amason et al 1970). Such shielding can be provided with the help of conducting strips impinged on the outer sur-face of the radome. The function of these conducting structures is to reduce the steamers from the antennas enclosed inside the radome while providing a new source of streamers outside the radome. Lightning can then attach to these streamers and travel down the conducting strip to the fuselage without harm-ing the aircraft antennas. A thick metallic layer will be helpful to attain a reasonable resistance against lightning strikes and take care of lightning current without melting. Such a metallic screen also protects against the entrance of the electro magnetic energy into the electrical wires of the aircraft. In addition, some sophisticated lightning protection devices can also be engineered to cope with a direct strike of lightning.

7.2. Bird strikes testing and prevention strategies

The term ‘bird strike’ indicates the collision between an air-craft front facing part and a bird. Bird strikes are a major threat to airborne radome structures, and may result in serious struc-tural damage. Therefore, a lot of research is required in the development of bird strike testing and prevention approaches. An extensive review of computational techniques for bird strike simulations (i.e. virtual tests) has been presented by Heimbs (2011). The UGent bird-strike test set-up (or Ghent University bird strike set-up) consists of a projectile (or sabot) with a gelatine bird, the stripper, the pressure vessel and test chamber (Allaeys et al 2017a). In addition, there is a provi-sion of three lasers with photodiodes. These lasers are mounted on the opposite side of the test chamber. The set-up is able to shoot birds up to 42 kJ. The sabot is mounted in front of a pres-sure vessel and is released at the required pressure. Birds can be launched at speed up to 250 m s−1 with a weight of 4 lb. The same research group has introduced bird strike calibration tests (Allaeys et al 2017b). Such initial testing strategies provide an insight into the complex behavior of a bird. The bird strike tests have been performed on three different types of targets (i.e. a splitter, a wedge and a plate). Based on the momentum transfer, birds can be compared with distinct materials, masses and speeds, etc. The design and testing of a fiber–metal–lam-inate bird strike resistant leading edge has been introduced (Guida et al 2009). Liu et al. reported an experimental setup of bird impact with a flat plate (Liu et al 2014a). Two kinds of plate have been employed for the measurement: aluminium alloy (with thickness of 10 mm and 14 mm) and steel plates (with thickness of 4.5 mm and 8 mm). The total dimensions of the plate was 600 mm × 600 mm. Here plate was bolted to a clamping fixture. A total sixteen groups of experiments were conducted at bird impact velocities of 70 m s−1, 120 m s−1 and 170 m s−1, respectively. The displacement and strain mea-surements were captured at a sample frequency of 10 MHz using a 12-channel data acquisition system. Laser displace-ment sensors were incorporated for the displacement measure-ment. During the test, nine strain gauges were employed for the recording of displacement and strain. A killed domestic chicken with a mass of 1.8 kg was used as the bird.

Recently, a novel design has been introduced for reinforc-ing the aircraft tail leading edge structure against bird strikes (Liu et al 2017). The experimental test set-up consists of a bird, velocity sensors, pressure chamber, valve, air storage and data lab. The gas gun unit comprises a compressed air gun with corresponding compressor, instrumentation and control units. The bird strike tests was carried out using a dead chicken of 8 lb with a speed of 250 kts, which hit the leading edge bay in composite (Guida et al 2011). The test set-up consisted of an air cannon bore, velocity measurement device, test article, test bed, high speed camera, load cell and safeguard screen.

A high velocity impact test platform for bird strike has been introduced (Barriere et al 2015). This test platform consists of a large-diameter gas gun with fast opening valve. This gun can launch a 2 kg projectile at a speed of 190 m s−1 at less than 20 mbar pressure. A stereo-correlation device consisting of a high speed camera and a large scale target with the coarse speckle

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

22

pattern was also employed. An inexpensive simulated bird strike test method has been developed to predict the dynamic impact response (bird strike resistance) of aircraft (Colvin et al 1990). The test utilizes an instrumented pendulum with a weighted strik-ing head. The striking head was designed in order to maintain a constant moment arm distance between the striking point and the sample base. The energy transferred from the striker’s head to the test specimen was monitored with the help of two independent transducers. The strain in each test piece was measured using a strain gauge. An accelerometer was employed for the measure-ment of deceleration (negative acceleration) of the weighted pen-dulum. The high frequency resonant vibrations of the pendulum and test stand were eliminated by incorporating a rubber piece in between the accelerometer and striking pendulum.

Recently, a high resolution, MM wave W-band radar system has been designed for the detection and imaging of the airport movement area (Galati et al 2016). The bird strike preven-tion and intruder detection are more recent potential applica-tions of the MM-WI system. The better signal reconstruction and resolution is a result of increased sampling frequency and digital sampling width. An automated bird-monitoring system can be designed for the detection and categorization of bird species (Yoshihashi et al 2015). An image dataset can be con-structed as a benchmark for ecological investigations of birds. Barnes et al. reported an overlap scan method for improved 3D radome characterization (Barnes et al 2007). Special audi-ble radars were employed, which can help the birds to notice the aircraft sooner and get out of the way in order to avoid any collisions. It is also possible to utilize a ‘dizziness beam’ using special modulation on the radar.

8. Application of dedicated mathematical models and algorithms

Table 4 presents a list of distinct algorithms utilized for the NDE of composite structures. An artificial neural network

based sensor system can be utilized for the microwave char-acterization of thin samples and dielectric sheets (Panda et al 2016). The data reconstruction algorithms are very helpful in order to provide high quality SAR images without any back-ground noise and image artifacts (Case et al 2012, 2013). A 3D-SAR algorithm employs non-uniform fast Fourier trans-forms (Song et al 2006). The spectral estimation techniques can be utilized to reduce uneven brightness and image artifacts. A novel reconstruction method has been adopted that can pro-vide an inherent spatial information about the data, resulting in a superior data reconstruction and final SAR image. SAR based algorithms can be categorized as forward SAR, reverse SAR and inverse SAR based algorithms. The background clut-ters affect the defect detection; therefore how to decrease the background clutter is a very important aspect. The filter and regression methods are commonly employed background sup-pression methods (Yang et al 2004, Wang et al 2009, Gu et al 2010). The filter methods act in frequency and space domains, but such methods are interfered by background clutter fluc-tuations. Regression methods can be utilized for background prediction and suppression. Regression methods can be cat-egorized as linear and nonlinear methods. A robust Kernel regression algorithm can be utilized for the passive MM-WI target detection (Yang et al 2012). The performance evalua-tion of an imaging system can be carried out using an optical transfer function, a modulation transfer function and a point transfer function (Zhang et al 2012). Image quality can be evaluated using the image SNR and noise equivalence reflec-tivity difference. The principal component and independent component analysis techniques can be employed for impact damage detection (Cheng et al 2014). A finite element method can be used for the computation of defect depth by identifying the defect blind frequency (Valle et al 2015).

A curvature compensating subtraction algorithm consists of three main steps. The first step is to choose a reference freeze frame and a target freeze frame. The target time frame

Table 4. Models and algorithms used for NDE of R&S structures.

Models and algorithm utilized Reference

Reference-free micro defect visualization algorithm Yang et al (2016a)UEM (an image processing algorithm): wave number based damage visualization technique Bae et al (2016)Propagating mode extraction algorithm Yin et al (2015)ω-k based 3D-SAR algorithm Case et al (2003)Kernel regression algorithm Yang et al (2012)Anomalous wave propagation imaging algorithm Lee et al (2012)Variable time window amplitude mapping algorithm Lee et al (2012)Wave number filtering algorithm Flynn et al (2013) and Kudela et al (2015)Delay-and-sum algorithm Khodaei et al (2014)Compressed sensing algorithm Kajbaf et al (2013) and Valdes et al (2014)Principal component and independent component analysis Cheng et al (2014)Time-reversed finite difference time domain (FDTD) Bardak et al (2014)Statistics based adaptive algorithm Agarwal et al (2015)ISR method Persson et al (2013)Curvature-compensating, subtraction based algorithm Shin et al (2017)Simultaneous iterative reconstruction algorithm Schrapp et al (2013)Warped-basis pursuit algorithm Marchi et al (2010)Physical optics with FDTD Hirvonen et al (2007)

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

23

basically consists of the local damaged amplitude distribution. In the second step, some coefficients are determined, which are applied to the amplitude of the reference freeze frame in order to match with the target freeze frame. There is a strong relationship between the reference time frame, i.e. A (x, y, tref.)

and the signal amplitude of the target time frame, i.e. A (x, y, tdamaged) as per the following expression (Shin et al 2017)

A(x, y, tdamaged) = a × A(x, y, tref.) + b. (8)

Here a and b are coefficients, which can be obtained as

a =

(V(A(tdamaged))

V(A(tref.))

)0.5

(9)

b = E(A(tdamaged))− a × E(A(tref.)). (10)

In the final step of the curvature-compensating subtrac-tion algorithm, the shadow effect due to surface curvature is reduced by subtracting the signal amplitude map of the damage-detected target frame from the amplitude map of the matched reference frame. Recently, in 2016, the absorption cross section of an arbitrarily shaped electrically large object was carried out in a reverberation chamber (Calberg et al 2004, Xu et al 2016).

9. Issues associated with the performance and non-destructive evaluation of radome and stealth structures

The phase curvature and amplitude taper greatly affect the main beam of the antenna; on the other hand, ripples dete-riorates the side-lobe measurement accuracy. The insufficient distance between the antennas may result in the phase curva-ture and amplitude taper. The amplitude and phase ripples may take place due to the reflections from the surroundings. The ripple length may be of the order of a wavelength. Sometimes, even small reflected waves may result in a large measurement error because the fields of the wave are added. The impedance mismatching between antennas and instruments, and instru-mental errors (like imperfections of the transmitter, receiver and positioner) result in measurement uncertainties. The con-necting cables may leak and act as an antenna, which is also a major limiting factor for the measurement of R&S structures.

The atmospheric effects also affect the performance of R&S structures at larger measurement distances. The higher the atmospheric attenuation at MM and sub-MM frequen-cies is also an important issue associated with R&S structure measurement. Such high atmospheric attenuation may create amplitude variations during the measurement process. The scintillation and multipath propagation may take place due to a corresponding variation in the refractive index. The antenna misalignment may also result in a series of measurement errors (Kraus 1950).

The measurement of curved shaped R&S composite struc-tures is still a very challenging task for researchers. The dam-aged areas can be shadowed by curvature effects due to the change in beam incident angle and wave travelling distance. The beam incident angle and wave travelling distance increases in the curved part in the scan area, which results in difficulty of detecting damage in curved shaped R&S structures. Nowadays, researchers are incorporating FSSs with R&S structures for the performance enhancement of R&S structures. However, the incorporation of such advanced electromagnetic structures results in design complexity and, therefore, it is comparatively

Figure 26. Measurement set-up for the standard lightning test and set-up view (inset). Reproduced from Rosolem et al (2010). © IOP Publishing Ltd. All rights reserved.

Figure 27. Optical waveform generated by the sparks and spark electrodes (inset). Reproduced from Rosolem et al (2010). © IOP Publishing Ltd. All rights reserved.

Figure 28. Two antennas used to discriminate between lightning and non-lightning sources of induced current. Reproduced from Bennett (2013). © IOP Publishing Ltd. All rights reserved.

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

24

difficult to measure their performance using NDE techniques. Similar problems exist with the NDE of multilayered struc-tures (i.e. FSS sandwiched and FSS impacted structures). However, advancement in comp uter hardware and software algorithms has enabled significant improvements. R&S design requires the designer to create custom algorithms for specific geometries to predict performance.

10. Conclusions and future scope

The measurement methods for electromagnetic performance evaluation and non-destructive structural evaluation of R&S structures have been developed enormously over the past few decades. Considerable advancements have been made in the field, which include development of new measurement strategies, modeling approaches, signal processing tech-niques, etc. The experimental techniques in use are based on proven and tested principles to measure the performance of aircraft composite structures in terms of their electro-magnetic characteristics as well as load bearing capabil-ity. Microwave based measurement techniques such as TL and FSM have been found to be most useful approaches for evaluating the electro magnetic performance of R&S struc-tures. On the other hand, MM-WI, THz-WI, PZT sensing laser UPI and FF-PF-UPI techniques including SFSM are the most versatile NDE methods adopted for R&S struc-tures. The field of electromagnetic performance and NDE measurement of R&S structures has just begun and promises to bear much fruit in the coming years. Although in recent years attempts have been made to introduce useful NDE sys-tems, further attention of researchers is still needed for the performance evaluation of complex shaped structures like curved and non-planar R&S structures. In addition, there is a strong need to focus on exploring some other cost effective measurement techniques for the performance evaluation and NDE of R&S structures.

Acknowledgment

This research was supported by the National Research Foun-dation of Korea (NRF), Grant funded by the Ministry of Sci-ence and ICT (NRF-2017R1A5A1015311) and SERB, DST, Government of India under Early Career Research Grant (ECR/2017/000676). The authors would also like to acknowl-edge Inspection & Testing Engineers, Noida, Uttar Pradesh, India for their meaningful discussions about NDE techniques.

ORCID iDs

Jung Ryul Lee https://orcid.org/0000-0003-0742-4722

References

Agarwal S and Singh D 2015 An adaptive statistical approach for non-destructive underline crack detection of ceramic tiles using millimeter wave imaging radar for industrial application IEEE Sens. J. 15 7036–44

Allaeys F, Luyckx G, Paepegem W V and Degrieck J 2017a Development and validation of a set-up to measure the transferred multi-axial impact momentum of a bird strike on a booster vane Int. J. Impact Eng. 99 102–10

Allaeys F, Luyckx G, Paepegem W V and Degrieck J 2017b Characterization of real and substitute birds through experimental and numerical analysis of momentum, average impact force and residual energy in bird strike on three rigid targets: a flat plate, a wedge and a splitter Int. J. Impact Eng. 99 1–13

Al-Qubaa A R, Tian G Y, Wilson J, Woo W L and Dlay S S 2010 Feature extraction using normalized cross-correlation for pulsed eddy current thermographic images Meas. Sci. Technol. 21 115501

Alvarez Y, Vaqueiro Y R, Valdes B G, Rappaport C M, Las-Heras F and Martinez-Lorenzo J A 2015 Three-dimensional compressed sensing based millimeter-wave imaging IEEE Trans. Antennas Propag. 63 5868–73

Amason M P and Cassell G J 1970 Radome lightning protection techniques and their electromagnetic compatibility IEEE Electromagnetic Compatibility Symp. Record pp 286–304

Anderson I 1975 Measurements of 20 GHz transmission through a radome in rain IEEE Trans. Antennas Propag. 23 619–22

Awang Z, Zaki F A M, Baba N H, Zoolfakar A S and Bakar R A 2013 A free space method for complex permittivity measurement of complex bulk and thin film dielectrics at microwave frequencies Prog. Electromagn. Res. B 51 307–28

Bae D Y and Lee J R 2016 A health management technology for multisite cracks in an in-service aircraft fuselage based on multi-time-frame laser ultrasonic energy mapping and serially connected PZTs Aerosp. Sci. Technol. 54 114–21

Balanis C A 1989 Advanced Engineering Electromagnetics (New York: Wiley) pp 180–5

Bardak C, Zhu Y, Bernussi A A and Saed M 2014 Terahertz imaging with a time-reversed finite difference time-domain method IEEE Trans. Terahertz Sci. Technol. 4 240–6

Barnes W J, Yearyl M, Olivero K, Phillips J and Ibrahim T 2007 An overlapped scan method for enhanced 3D radome characterization Instrumentation and Measurement Technology Conference - IMTC 2007 (Warsaw, Poland, May 1–3, 2007)

Barriere L, Bouquet M and Ferrero J F 2015 Bird-Strike Shielding Materials: Development of a High Velocity Impact Test Platform (Amiens: SAMPE Europe)

Baskey H B and Akhtar M J 2017 Design of flexible hybrid nanocomposite structure based on frequency selective surface for wideband radar cross section reduction IEEE Trans. Microw. Theory Tech. 65 1–11

Bennett A J 2013 Identification and ranging of lightning flashes using co-located antennas of different geometry Meas. Sci. Technol. 24 125801

Bergo P, Carvalho R A, Sobral P J A, Bevilacqua F R S, Pinto J K C and Souza J P 2006 Microwave transmittance in gelatin based films Meas. Sci. Technol. 17 3261–4

Boccardi S, Carlomagno G M, Simeoli G, Russo P and Meola C 2016 Evaluation of impact-affected areas of glass fibre thermoplastic composite from thermographic images Meas. Sci. Technol. 27 075602

Bocherens E, Bourasseau S, Dewynter-Marty V, Py S, Dupont M, Ferdinand P and Berenger H 2000 Damage detection in a radome sandwich material with embedded fiber optic sensors Smart Mater. Struct. 9 310–5

Bourreau D, Peden A and Maguer S L 2006 A quasi-optical free-space measurement setup without time-domain gating for material characterization in the W-band IEEE Trans. Instrum. Meas. 55 2022–8

Bowring N J, Baker J G, Rezgui N D and Alder J F 2008 A sensor for the detection and measurement of thin dielectric layers using reflection of frequency scanned millimetric waves Meas. Sci. Technol. 19 024004

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

25

Brooker G and Johnson D G 2015 Low-cost millimeter wave imaging using a commercial plasma display IEEE Sens. J. 15 3557–64

Calberg U, Kildal P S, Wofgang A, Sotoudeh O and Orlenius C 2004 Calculated and measured absorption cross sections of lossy objects in reverberation chamber IEEE Trans. Electromagn. Compat. 46 146–54

Capozzoli A, D’ Elia G and Liseno A 2007 Phaseless characterization of compact antenna test ranges IET Microw. Antennas Propag. 1 860–6

Carranza I E, Grant J, Gough J and Cumming D R S 2015 Metamaterial based Terahertz imaging IEEE Trans. Terahertz Sci. Technol. 5 892–901

Case J T, Ghasr M T and Zoughi R 2012 Optimum 2D non-uniform spatial sampling for microwave SAR-based NDE imaging systems IEEE Trans. Instrum. Meas. 61 3072–83

Case J T, Ghasr M T and Zoughi R 2013 Non-uniform manual scanning for rapid microwave nondestructive evaluation imaging IEEE Trans. Instrum. Meas. 62 1250–8

Cerny I 2004 The use of DCPD method for measurement of growth of cracks in large components at normal and elevated temperatures Eng. Fract. Mech. 71 837–48

Chang D C, Liao C H and Wu C C 2004 Compact antenna test range without reflector edge treatment and RF anechoic chamber IEEE Antennas Propag. Mag. 46 27–37

Chen G, Pei J, Yang F, Zhou X Y, Sun Z L and Cui T J 2012 Terahertz-wave imaging system based on backward wave oscillator IEEE Trans. Terahertz Sci. Technol. 2 504–12

Chen L F, Ong C K, Neo C P, Varadan V V and Varadan V K 2004 Microwave Electronics Measurement and Materials Characterization (New York: Wiley)

Chen Q and Zhang X C 2001 Semiconductor dynamic aperture for near-field terahertz wave imaging IEEE J. Sel. Top. Quantum Electron. 7 608–14

Cheng G, Liu L, Cai D and Yuan C 2012 Microwave measurement of dielectric properties using the TM011 and TE011 modes excited by a generalized non-radiative dielectric resonator Meas. Sci. Technol. 23 115901

Cheng L, Gao B, Tian G Y, Woo W L and Berthiau G 2014 Impact damage detection and identification using eddy current pulsed thermography through integration of PCA and ICA IEEE Sens. J. 14 1655–63

Cheng Y, Hu F, Gui L, Wu L and Lang L 2016 Polarization-based method for object surface orientation information in passive millimeter-wave imaging IEEE Photon. J. 8 5500112

Chia C C, Lee J R and Park C Y 2012 Radome health management based on synthesized impact detection, laser ultrasonic spectral imaging, and wavelet-transformed ultrasonic propagation imaging methods Composites B 43 2898–906

Collin R E 1996 Foundations for Microwave Engineering (New York: McGraw-Hill)

Colvin G N and Falls J R 1990 Development of a simulated bird strike test method J. Met. 42 26–9

Courtney C C 1998 Time domain measurement of the electromagnetic properties of materials IEEE Trans. Microw. Theory Tech. 46 517–22

Davison R, Jones K and Strid E 1990 LRM and LRRM calibrations with automatics determinations of load inductance Proc. 36th ARFTG Conf. (Monterey, CA) pp 84–91

Dester G D 2010 Error analysis of a two-layer method for the electromagnetic characterization of conductor-backed absorbing material using an open-ended waveguide probe Prog. Electromagn. Res. B 26 1–21

Dietlein C R, Hedden A S and Wikner D A 2012 Digital reflectarray considerations for terrestrial millimeter-wave imaging IEEE Antenna Wirel. Propag. Lett. 11 272–5

Dietrich F J and West D B 1988 An experimental radome panel evaluation IEEE Trans. Antennas Propag. 36 1566–70

Dittmann K, Kullig C and Meichsner J 2012 160 GHz Gaussian beam microwave interferometry in low density RF plasmas Plasma Sources Sci. Technol. 21 024001

Dobroiu A, Otani C and Kawase K 2006 Terahertz-wave sources and imaging applications Meas. Sci. Technol. 17 161–74

Dragan K and Synaszko P 2009 In-service flaw detection and quantification in the composite structures of aircraft Fatigue Aircr. Struct. 1 37–41

Duling I and Zimdars D 2009 Terahertz imaging: revealing hidden defects Nat. Photon. 3 630–2

Ebnabbasi K, Mohebbi M and Vittoria C 2013 Coaxial line technique to measure constitutive parameters in magnetoelectric ferrite materials IEEE Microw. Wirel. Compon. Lett. 23 504–6

Eul H J and Schiek B 1988 Thru-match-reflect: one result of a rigorous theory for de-embedding and network analyzer calibration Proc. 18th European Microwave Conf. (Stockholm, Sweden) pp 909–14

Fitzgerald W C, Davis M N, Blackshire J L, Maguire J F and Mast D B 2001 Evanescent microwave sensor scanning for detection of subcoating corrosion J. Corros. Sci. Eng. 3 1–6

Flynn E B, Chong C Y, Jarmer G J and Lee J R 2013 Structural imaging through local wavenumber estimation of guided waves NDT&E Int. 59 1–10

Francis D, Tatam R P and Groves R M 2010 Shearography technology and applications: a review Meas. Sci. Technol. 21 102001

Friederich F, Spickermann G, Roggenbuck A, Deninger A, Weg C A, Spiegel W V, Lison F, Bolivar P H and Roskos H G 2010 Hybrid continuous-wave demodulating multipixel Terahertz imaging systems IEEE Trans. Microw. Theory Tech. 58 2022–6

Fukasawa R 2015 Terahertz imaging: widespread industrial application in non-destructive inspection and chemical analysis IEEE Trans. Terahertz Sci. Technol. 5 1121–7

Galati G, Piracci E G and Ferri M 2016 High resolution, millimeter wave radar applications to airport safety 8th Int. Conf. on Ultrawideband and Ultrashort Impulse Signals (Ukraine)

Gao B, Bai L, Woo W L, Tian G Y and Cheng Y 2014 Automatic defect identification of eddy current pulsed thermography using single channel blind source separation IEEE Trans. Instrum. Meas. 63 913–22

Ghasr M T, Case J T and Zoughi R 2014 Novel reflectometer for millimeter-wave 3D holographic imaging IEEE Trans. Instrum. Meas. 63 1328–36

Ghasr M T, Kharkovsky S, Bohnert R, Hirst B and Zoughi R 2013 30 GHz linear high-resolution and rapid millimeter wave imaging system for NDE IEEE Trans. Antennas Propag. 61 4733–40

Ghasr M T, Kharkovsky S, Zoughi R and Austin R 2005 Comparison of near-field millimeter-wave probes for detecting corrosion precursor pitting under paint IEEE Trans. Instrum. Meas. 54 1497–504

Ghasr M T, Pommerenke D, Case J T, McClanahan A, Aflaki-Beni A, Abou-Khousa M, Kharkovsky S, Guinn K, Paulis F D and Zoughi R 2011 Rapid rotary scanner and portable, coherent wideband Q-band transceiver for high-resolution millimeter-wave imaging applications IEEE Trans. Instrum. Meas. 60 186–97

Ghodgaonkar D K, Varadan V V and Varadan V K 1989 A free-space method for measurement of dielectric constants and loss tangents at microwave frequencies IEEE Trans. Instrum. Meas. 37 789–93

Ghodgaonkar D K, Varadan V V and Varadan V K 1990 Free-space measurement of complex permittivity and complex permeability of magnetic materials at microwave frequencies IEEE Trans. Instrum. Meas. 39 387–94

Gu J and Yu L Y 1990 The influence of geometric factors on crack depth measurement using the potential drop technique NDT&E Int. 23 161–4

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

26

Gu Y F, Wang C, Liu B and Zhang Y 2010 A Kernel based non parametric regression method for clutter removal in infrared small-target detection application IEEE Geosci. Remote Sens. Lett. 7 469–73

Guida M, Marulo F and Polito T 2009 Design and testing of a fiber-metal-laminate bird strike resistant leading edge J. Aircr. 46 2121–9

Guida M, Marulo F, Meo M, Grimaldi A and Olivares G 2011 SPH-Lagrangian study of bird impact on leading edge wing Compos. Struct. 93 1060–71

Gumbmann F and Schmidt L P 2011 Millimeter-wave imaging with optimized sparse periodic array for short-range applications IEEE Trans. Geosci. Remote Sens. 49 3629–38

Guo G, Li E, Li Z, Zhang Q and He F 2011 A test system for complex permittivity measurements of low-loss materials at high temperatures up to 2000 °C Meas. Sci. Technol. 22 045707

Han P Y and Zhang X C 2001 Free-space coherent broadband terahertz time-domain spectroscopy Meas. Sci. Technol. 12 1747–56

Han R, Zhang Y, Kim Y, Kim D Y, Shichijo H, Afshari E and Kenneth K O 2013 Active Terahertz imaging using schottky diodes in CMOS: array and 860 GHz pixel IEEE J. Solid-State Circuits 48 2296–308

Hasar U C 2009 Permittvity measurement of thin dielectric materials from reflection only measurements using one-port vector network analyzers Prog. Electromagn. Res. 95 365–80

Heimbeck M S, Ng W R, Golish D R, Gehm M E and Everitt H O 2015 Terahertz digital holographic imaging of voids within visibly opaque dielectrics IEEE Trans. Terahertz Sci. Technol. 5 110–6

Heimbs S 2011 Computational methods for bird strike simulations: a review Comput. Struct. 89 2093–112

Hellstein P and Szwedo M 2016 3D thermography in non-destructive testing of composite structures Meas. Sci. Technol. 27 124006

Hirvonen T, Ala-Laurinaho J P S, Tuovinen J and Raisanen A V 2007 A compact antenna test range based on a hologram IEEE Trans. Antennas Propag. 45 1270–6

Hollinger R D, Jose K A, Tellakula A, Varadan V V and Varadan V V 2000 Microwave characterization of dielectric materials from 8 to 110 GHz using a free space set up Microw. Opt. Technol. Lett. 26 100–5

Hong S C, Abetew A D, Lee J R and Ihn J B 2016 Three dimensional evaluation of aluminium plates with wall-thinning by full-field pulse-echo laser ultrasound Opt. Lasers Eng. 99 58–65

Hou L, Park H and Zhang X C 2011 Terahertz wave imaging system based on glow discharge detector IEEE J. Sel. Top. Quantum Electron. 17 177–82

Hu B B and Nuss M C 1995 Imaging with terahertz waves Opt. Lett. 20 1716–8

Hu B, Zhang X C and Auston D 1990 Free-space radiation from electro-optic crystals Appl. Phys. Lett. 506–8

Huang C C and Lin H C 2008 A novel calibration algorithm with unknown line-series-shunt standards for broadband S-parameter measurements IEEE Trans. Instrum. Meas. 57 891–6

Huang L, Gao C and Guo F 2017 Lightning indirect effects on Helicopter: numerical simulation and experimental validation IEEE Trans. Electromagn. Compat. 59 1171–9

Hughes D, Wang N, Case T, Donnell K, Zoughi R, Austin R M and Novack M 2001 Microwave nondestructive detection of corrosion under thin paint and primer in aluminum panels Subsurface Sensing Technol. Applicat. 2 435–51

Hunsche S, Koch M, Brener I and Nuss M C 1998 THz near-field imaging Opt. Commun. 150 22–6

Hyde M W, Havrilla M J and Bogle A E 2011 A novel and simple technique for measuring low-loss materials using the two

flanged waveguides measurement geometry Meas. Sci. Technol. 22 085704

Jain R, Grzyb J and Pfeiffer U R 2016 Terahertz light-field imaging IEEE Trans. Terahertz Sci. Technol. 6 649–57

Jia C, Geng G, Chang T, Yang C, Fan W Y, Guo Y, Sun Z, Liu L, Guo Q and Cui H L 2016 Millimeter wave imaging of cracks in bulk coal IEEE Trans. Terahertz Sci. Technol. 6 554–62

Jiang Y, Wang H, Qin Y, Deng B, Gao J and Zhuang Z 2016 A three-dimensional surface imaging method using THz dual-frequency interferometry IEEE Geosci. Remote Sens. Lett. 13 1651–5

Johnson D B 2008 Microwave nondestructive evaluation of aircraft radomes Retrospective Theses and Dissertations (Ames, IA: Iowa State University) p 264

Jose K A, Varadan V K and Varadan V V 2001 Wideband and non-contact characterization of complex permittivity of liquids Microw. Opt. Technol. Lett. 30 75–9

Juuti M, Tuononen H, Prykari T, Kontturi V, Kuosmanen M, Alarousu E, Ketolainen J, Myllyla R and Peiponen K E 2009 Optical and terahertz measurement techniques for flat-faced pharmaceutical tablets: a case study of gloss, surface roughness and bulk properties of starch acetate tablets Meas. Sci. Technol. 20 015301

Kaatze U 2013 Measuring the dielectric properties of materials. Ninety-year development of low-frequency techniques to broadband spectroscopy and high-frequency imaging Meas. Sci. Technol. 24 012005

Kaatze U and Hubner C 2010 Electromagnetic techniques for moisture content determination of materials Meas. Sci. Technol. 21 082001

Kajbaf H, Case J T, Yang Z and Zheng Y R 2013 Compressed sensing for SAR based wideband three-dimensional microwave imaging system using non-uniform fast Fourier transform IET Radar Sonar Navig. 7 658–70

Karpowicz N, Zhong H, Xu J, Lin K, Hwang J and Zhang X C 2005 Comparison between pulsed Terahertz time-domain imaging and continuous wave Terahertz imaging Semicond. Sci. Technol. 20 293–9

Kasalynas I, Venckevicius R and Valusis G 2013 Continuous wave spectroscopic Terahertz imaging with InGaAs bow-tie diodes at room temperature IEEE Sens. J. 13 50–4

Kawano Y 2016 Chip based near-field terahertz microscopy IEEE Trans. Terahertz Sci. Technol. 6 356–64

Kemp I, Peterson M, Benton C and Petkie D T 2010 Sub-mm wave imaging techniques for non-destructive aerospace materials evaluation IEEE Aerospace & Electronics Systems Magazine pp 17–9

Kerouedan J, Queffelec P, Talbot P, Quendo C, Blasi S D and Brun A L 2008 Detection of micro-cracks on metal surfaces using near-field microwave dual-behaviour resonator filters Meas. Sci. Technol. 19 105701

Kharkovsky S and Zoughi R 2007 Microwave and millimeter wave non-destructive testing and evaluation-overview and recent advances IEEE Instrum. Meas. Mag. 10 26–38

Kharkovsky S, Case J T, Abou-Khousa M A, Zoughi R and Hepburn F 2006 Millimeter wave detection of localized anomalies in the space shuttle external fuel tank insulating foam IEEE Trans. Instrum. Meas. 55 1250–7

Kharkovsky S, Hepburn F, Walker J and Zoughi R 2005 Nondestructive testing of the space shuttle external tank foam insulation using near-field and focused millimeter wave techniques Mater. Eval. 63 516–22

Khodaei Z S and Aliabadi M H 2014 Assessment of delay-and-sum algorithms for damage detection in aluminium and composite plates Smart Mater. Struct. 23 075007

Komura I, Hirasawa T, Nagai S, Takabayashi J and Naruse K 2001 Crack detection and sizing technique by ultrasonic and electromagnetic methods Nucl. Eng. Des. 206 351–62

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

27

Koskinen T, Ala-Laurinaho J, Saily J, Loonnqvist A, Hakli J, Mallat J, Tuovinen J and Raisanen A V 2005 Experimental study on a hologram based compact antenna test range at 650 GHz IEEE Trans. Microw. Theory Tech. 9 2999–3006

Kozakoff D J 2010 Analysis of Radome-Enclosed Antennas (Boston, MA: Artech House Publishers)

Kraus J D, Marhefka R J and Khan A S 1950 Antennas and Wave Propagation (New York: McGraw-Hill)

Krupka J 2006 Frequency domain complex permittivity measurements at microwave frequencies Meas. Sci. Technol. 17 55–70

Kudela P, Radzienski M and Ostachowicz W 2015 Identification of cracks in thin-walled structures by means of wave number filtering Mech. Syst. Signal Process. 50–1 456–66

Kume E and Sakai S 2008 Millimeter-wave radiation from a Teflon dielectric probe and its imaging application Meas. Sci. Technol. 19 115501

Lago F 2014 Lightning in aeronautics J. Phys.: Conf. Ser. 550 012001

Lamkaouchi K, Balana A, Delbos G and Ellison W J 2003 Permittivity measurements of lossy liquids in the range 26–110 GHz Meas. Sci. Technol. 14 444–50

Lee J R, Chia C C and Park C Y 2012 Laser ultrasonic anomalous wave propagation imaging method with adjacent wave subtraction: algorithm Opt. Laser Technol. 44 1507–15

Li T, Almond D P and Rees D A S 2011a Crack imaging by scanning laser-line thermography and laser-spot thermography Meas. Sci. Technol. 22 035701

Li Y, Bowler N and Johnson D B 2011b A resonant microwave patch sensor for detection of layer thickness or permittivity variations in multilayered dielectric structures IEEE Sens. J. 11 5–15

Liang W and Que P W 2007 Maximum non-Gaussianity parameters estimation of ultrasonic echoes and its application in ultrasonic non-destructive evaluation Meas. Sci. Technol. 18 3743–50

Liu J, Li Y and Gao X 2014a Bird strike on a flat plate: experimental and numerical simulations Int. J. Impact Eng. 70 21–37

Liu J, Li Y, Yu X, Tang Z, Gao X, Lv J and Zhang Z 2017 A novel design for reinforcing the aircraft tail leading edge structure against bird strike Int. J. Impact Eng. 105 89–101

Liu J, Ren W, Tian G Y, Gao B, Wang Y, Zhang J, Shaw B, Yin A and King-Alale N O 2015 Nondestructive evaluation of early contact fatigue using eddy current pulsed thermography IEEE Sens. J. 15 4409–19

Liu J, Tian G Y, Gao B, Ren W and Meng J S 2014b Investigation of thermal imaging, sampling frequency for eddy current pulsed thermography NDT&E Int. 62 85–92

Lonnqvist A, Tamminen A, Mallat J and Raisanen A V 2006 Monostatic reflectivity measurement of radar absorbing materials at 310 GHz IEEE Trans. Microw. Theory Tech. 54 3486–91

Lu Z J, Nicholas P J and Evans W J 1995 Calibration of an ACPD monitoring system for small crack growth in corner crack specimens Eng. Fract. Mech. 50 443–56

Lu Z, Liu X, Wang H, Chen X, Yao Y and Yu J 2015 A modular Gaussian beam analysis method based on 3D diffraction technique IEEE Antennas Wirel. Propag. Lett. 14 362–5

Ma Z and Okamura S 1999 Permittivity determination using amplitudes of transmission and reflection coefficients at microwave frequency IEEE Trans. Microw. Theory Tech. 47 546–50

Maestrini A, Ward J, Gill J, Javadi H, Schlecht E and Chattopadhyaym G 2004 A 1.7–1.9 THz local oscillator source IEEE Microw. Wireless Compon. Lett. 14 253–5

Marchi L D, Ruzzene M, Xu B, Baravelli E and Speciale N 2010 Warped basis pursuit for damage detection using Lamb waves IEEE Trans. Ultrason. Ferroelectr. Freq. Control 57 2734–41

Martin R, Schuetz C A, Dillon T E, Chen C, Samluk J, Stein E L, Mirotznik M and Prather D W 2009 Design and performance of a distributed aperture millimeter-wave imaging system using optical up conversion Proc. SPIE 7309 730908–1

McKay J P and Rahmat S Y 1991 Compact range reflector analysis using the plane wave spectrum approach with an adjustable sampling rate IEEE Trans. Antennas Propag. 6 746–53

Menzel W and Huder B 1984 Compact range for millimetre-wave frequencies using a dielectric lens Electron. Lett. 19 768–9

Meola C and Carlomagno G M 2004 Recent advances in the use of infrared thermography Meas. Sci. Technol. 15 27–58

Munk B A 1995 Frequency Selective Surfaces and Grid Arrays (New York: Wiley)

Munk B A 2000 Frequency Selective Surfaces: Theory and Design (New York: Wiley)

Nahata A, Yardley J T and Heinz T F 2002 Two-dimensional imaging of continuous-wave terahertz radiation using electro-optic detection Appl. Phys. Lett. 81 963–5

Nauman M, Saleem R, Rashid A K and Shafique M F 2016 A miniaturized flexible frequency selective surface for X-band applications IEEE Trans. Electromagn. Compat. 58 419–28

Orlob C, Reinecke T, Denicke E, Geck B and Rolfes I 2013 Compact unfocused antenna setup for X-band free-space dielectric measurements based online-network-network calibration method IEEE Trans. Instrum. Meas. 62 1982–9

Pagliarulo V, Rocco A, Langella A, Riccio A, Ferraro P, Antonucci V, Ricciardi M R, Toscano C and Lopresto V 2015 Impact damage investigation on composite laminates: comparison among different NDT methods and numerical simulation Meas. Sci. Technol. 26 085603

Pan W, Huang C, Pu M, Ma X, Cui J, Zhao B and Luo X 2016 Combining the absorptive and radiative loss in metasurfaces for multi-spectral shaping of the electromagnetic scattering Sci. Rep. 6 21462

Panda S, Tiwari N K and Akhtar M J 2016 Computationally intelligent sensor system for microwave characterization of dielectric sheets IEEE Sens. J. 16 7483–93

Panwar R and Lee J R 2017 Progress in frequency selective surface based smart electromagnetic structures: a critical review Aerosp. Sci. Technol. 66 216–34

Parini C G and Prior C J 1988 Radiation pattern measurements of electrically large antennas using a compact antenna test range at 180 GHz Electron. Lett. 24 1552–4

Pauli M, Kayser T and Wiesbeck W 2007 A versatile measurement system for the determination of dielectric parameters of various materials Meas. Sci. Technol. 18 1046–53

Pawar S S and Peters K 2013 Through-the-thickness identification of impact damage in composite laminates through pulsed phase thermography Meas. Sci. Technol. 24 11560

Persson K, Gustafsson M, Kristensson G and Widenberg B 2013 Source reconstruction by far-field data for imaging of defects in frequency selective radomes IEEE Antennas Wireless Propag. Lett. 12 480–3

Pozar D M 1989 Microwave Engineering 3rd edn (New York: Wiley) pp 73–7

Qaddoumi N, Handjojo L, Bigelow T, Easter J, Bray A and Zoughi R 2000 Microwave corrosion detection using open-ended rectangular waveguide sensors Mater. Eval. 58 178–84

Qaddoumi N, Shroyer A and Zoughi R 1997 Microwave detection of rust under paint and composite laminates Res. Nondestruct. Eval. 9 201–12

Radenamad D, Aoyagi T and Hirose A 2011 High-sensitivity millimeter-wave imaging front-end using a low-impedance tapered slot antenna IEEE Trans. Antennas Propag. 59 4868–72

Raja M K, Mahadevan S, Rao B P C, Behera S P, Jayakumar T and Raj B 2010 Influence of crack length on crack depth

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

28

measurement by an alternating current potential drop technique Meas. Sci. Technol. 21 105702

Ramdane B, Trichet D, Belkadi M and Fouladgar J 2010 3D numerical modeling of the thermo-inductive technique using shell elements IEEE Trans. Magn. 46 3037–40

Ravuri M, Abou-Khousa M A, Kharkovsky S, Zoughi R and Austin R 2008 Microwave and millimeter wave near-field methods for evaluation of radome composites AIP Conf. Proc. pp 975–6

Ren W, Liu J, Tian G Y, Gao B, Cheng L and Yang H 2013 Quantitative non-destructive evaluation method for impact damage using eddy current pulsed thermography Composites B 54 169–79

Repjar A and Kremer D 1982 Accurate evaluation of a millimeter wave compact range using planar near-field scanning IEEE Trans. Antennas Propag. 3 419–25

Ringermacher H I, Archacki R J Jr and Veronesi W A 1998 Nondestructive testing: transient depth thermography US Patent No. 5711603

Rolfes I and Schiek B 2003 LRR-A self-calibration technique for the calibration of vector network analyzers IEEE Trans. Instrum. Meas. 52 316–9

Rolfes I and Schiek B 2004 Calibration methods for microwave free space measurements Adv. Radio Sci. 2 19–25

Rosolem J B, Barbosa C F, Floridia C and Bezerra E W 2010 A passive opto-electronic lightning sensor based on electromagnetic field detection for utilities applications Meas. Sci. Technol. 21 094032

Rothbart N, Richter H, Wienold M, Schrottke L, Grahn H T and Hubers H W 2013 Fast 2D and 3D Terahertz imaging with a quantum-cascade laser and a scanning mirror IEEE Trans. Terahertz Sci. Technol. 3 617–24

Rothwell E J and Oladimeji K A A 2013 Effects of surface curvature and beam defocusing on the characterization of conductor-backed absorbing materials using a free-space system J. Electromagn. Waves Appl. 27 1473–82

Saqueb S A N and Sertel K 2016 Phase-sensitive single-pixel THz imaging using intensity-only measurements IEEE Trans. Terahertz Sci. Technol. 6 810–6

Schrapp M, Scharrer T, Goldammer M, Rupitsch S J, Sutor A, Ermert H and Lerch R 2013 Artifact reduction in non-destructive testing by means of complementary data fusion of x-ray computed tomography and ultrasonic pulse-echo testing Meas. Sci. Technol. 24 125403

Sezer H K, Li L, Wu Z, Anderson B and Williams P 2007 Non-destructive microwave evaluation of TBC delamination induced by acute angle laser drilling Meas. Sci. Technol. 18 167–75

Sfarra S, Ibarra-Castanedo C, Lambiase F, Paoletti D, Ilio A D and Maldague X 2012 From the experimental simulation to integrated non-destructive analysis by means of optical and infrared techniques: results compared Meas. Sci. Technol. 23 115601

Sharath D, Menaka M and Venkatraman B 2013 Effect of defect size on defect depth quantification in pulsed thermography Meas. Sci. Technol. 24 125205

Sheen D M, McMakin D L and Hall T E 2001 Three-dimensional millimeter-wave imaging for concealed weapon detection IEEE Trans. Microw. Theory Tech. 49 1581–92

Shin H J and Lee J R 2016 Development of a long-range multi-area scanning ultrasonic propagation imaging system built into a hangar and its application on an actual aircraft Struct. Health Monit. 1–15

Shin H J, Park J Y, Hong S C and Lee J R 2017 In situ non-destructive evaluation of an aircraft UHF antenna radome based on pulse-echo ultrasonic propagation imaging Compos. Struct. 160 16–22

Shull P J 2002 Nondestructive Evaluation: Theory, Techniques, and Applications (New York: Dekker)

Siegel P H 2002 Terahertz technology IEEE Trans. Microw. Theory Tech. 50 910–28

Singh W S, Rao B P C, Vaidyanathan S, Jayakumar T and Raj B 2008 Detection of leakage magnetic flux from near-side and far-side defects in carbon steel plates using a giant magneto-resistive sensor Meas. Sci. Technol. 19 015702

Skolnik M 1990 Radar Handbook 2nd edn (New York: McGraw-Hill) pp 6.44–5

Song J, Liu Q H, Kim K and Scott W R 2006 High-resolution 3D radar imaging through nonuniform fast Fourier transform (NUFFT) Commun. Comput. Phys. 1 176–91

Song N N, Ke Y J, Yang H T, Zhang H, Zhang X Q, Shen B G and Cheng Z H 2013 Integrating giant microwave absorption with magnetic refrigeration in one multifunctional intermetallic compound of LaFe11.6Si1.4C0.2H1.7 Nat. Sci. Rep. 3 2291

Streza M, Fedala Y, Roger J P, Tessier G and Boue C 2013 Heat transfer modeling for surface crack depth evaluation Meas. Sci. Technol. 24 045602

Szabo Z 2017 Closed form Kramers–Kronig relations to extract the refractive index of metamaterials IEEE Trans. Microw. Theory Tech. 65 1150–9

Tsai H R, Enderli F, Feurer T and Webb K J 2012 Optimization based Terahertz imaging IEEE Trans. Terahertz Sci. Technol. 2 493–503

Umari M H, Ghodgaonkar D K, Varadan V V and Varadan V K 1991 A free-space bistatic calibration technique for the measurement of parallel and perpendicular reflection coefficients of planar samples IEEE Trans. Instrum. Meas. 40 19–24

Valdes B G, Allan G, Vaqueiro Y R, Alvarez Y, Mantzavinos S, Nickerson M, Berkowitz B, Lorenzo J A M, Heras F L and Rappaport C M 2014 Sparse array optimization using simulated annealing and compressed sensing for near-field millimeter wave imaging IEEE Trans. Antennas Propag. 62 1716–22

Valle S H and Peters K 2015 Numerical simulation of phase images and depth reconstruction in pulsed phase thermography Meas. Sci. Technol. 26 115602

Varadan V V and Ji L 2010 Temperature dependence of resonances in metamaterials IEEE Trans. Microw. Theory Tech. 58 2673–81

Varadan V V and Ro R 2007 Unique retrieval of complex permittivity and permeability of dispersive materials from reflection and transmitted fields by enforcing causality IEEE Trans. Microw. Theory Tech. 55 2224–30

Varadan V V, Hollinger R D, Ghodgaonkar D K and Varadan V K 1991 Free-space, broadband measurements of high- temperature, complex dielectric properties at microwave frequencies IEEE Trans. Instrum. Meas. 40 842–6

Wang J R, Jackson G M S, Schwaller M R, Johnson C M, Monosmith W B and Zhang Z 2013 Observations of storm signatures by the recently modified conical scanning millimeter-wave imaging radiometer IEEE Trans. Geosci. Remote Sens. 51 411–24

Wang P, Tian J W, Gao C Q and Martin A 2009 Infrared small target detection using directional high-pass filters based on LS-SVM Electron. Lett. 45 156–8

Wilson J, Tian G Y, Abidin I Z, Yang S and Almond D 2010 Modelling and evaluation of eddy current stimulated thermography Nondestruct. Test. Eval. 25 205–18

Wu T Y 2012 High dynamic range terahertz-wave transmission loss measurement at 330–500 GHz Meas. Sci. Technol. 23 085904

Xu Q, Huang Y, Xing L, Tian Z, Zhou J, Chen A and Zhuang Y 2016 Average absorption coefficient measurement of arbitrarily shaped electrically large objects in a reverberation chamber IEEE Trans. Electromagn. Compat. 58 1776–9

Yang H, Hu F, Chen K, Li D, Yi G and Jin R 2012 A robust regularization Kernel regression algorithm for passive

Meas. Sci. Technol. 29 (2018) 062001

Topical Review

29

millimeter wave imaging target detection IEEE Geosci. Remote Sens. Lett. 9 915–9

Yang J, Choi J, Hwang S, An Y K and Sohn H 2016a A reference-free micro defect visualization using pulse laser scanning thermography and image processing Meas. Sci. Technol. 27 085601

Yang L, Yang J and Yang K 2004 Adaptive detection for infrared small target under sea-sky complex background Electron. Lett. 40 1083–5

Yang S, Liu P, Yang M, Wang Q, Song J and Dong L 2016b From flexible and stretchable meta-atom to metamaterial: a wearable microwave meta-skin with tunable frequency selective and cloaking effects Nat. Sci. Rep. 6 21921

Yang S, Tian G Y, Abidin I Z and Wilson J 2011 Simulation of edge cracks using pulsed eddy current stimulated thermography J. Dyn. Syst. Meas. Control 133 011008

Yasui T, Jewariya M, Yasuda T, Schirmer M, Araki T and Abraham E 2013 Real-time two-dimensional spatio-temporal terahertz imaging based on noncollinear free-space electrooptic sampling and application to functional terahertz imaging of moving object IEEE J. Sel. Top. Quantum Electron. 19 8600110

Yin A and Ren H 2015 A propagating mode extraction algorithm for microwave waveguide using variational mode decomposition Meas. Sci. Technol. 26 095009

Yin A, Gao B, Tian G Y, Woo W L and Li K 2013 Physical interpretation and separation of eddy current pulsed thermography J. Appl. Phys. 113 064101

Yoshihashi R, Kawakami R, Lida M and Naemura T 2015 Construction of a bird image dataset for ecological investigations IEEE Int. Conf. on Image Processing pp 4248–52

Zeng Z, Zhou J, Tao N, Feng L and Zhang C 2012 Absolute peak slope time based thickness measurement using pulsed thermography Infrared Phys. Technol. 55 200–4

Zhang D, Hao Z, Qian Y, Huang Y, Bizeng, Yang Z and Qibai W 2017 Simulation and measurement of optimized microwave reflectivity for carbon nanotube absorber by controlling electromagnetic factors Nat. Sci. Rep. 7 1–8

Zhang L, Parini C G and Hao Y 2012 Millimetre-wave-imaging system at 95 GHz quantitative performance evaluation IET Microw. Antennas Propag. 6 869–79

Zoughi R 2000 Microwave Non-Destructive Testing and Evaluation (Dordrecht: Kluwer)

Meas. Sci. Technol. 29 (2018) 062001