Identification of periodic structure target using broadband polarimetry in terahertz radiation

Transcription

1 Identification of periodic structure target using broadband polarimetry in terahertz radiation Yuki Kamagata, Hiroaki Nakabayashi a), Koji Suizu, and Keizo Cho Chiba Institute of Technology, Tsudanuma, Narashino, Chiba , Japan a) Abstract: This study demonstrates precision non-destructive inspection of periodic linear targets using polarimetry technique used in radars with frequency bands from P to W (0.3 to 110 GHz) and broadband terahertz pulse. The polarimetry technique is effective to identify targets in radar field, and terahertz waves are effective for non-destructive inspection due to its broadband and special penetration characteristics. In this study, we combine the both techniques to enhance the inspection accuracy. Our analytical and experimental results confirmed that the combination of the polarimetry technique and the broadband terahertz pulse wave was effective for the precision inspection of the targets. Keywords: terahertz, broadband, radar polarimetry, non-destructive inspection Classification: Sensing References [1] Y. Yamaguchi, Rader Polarimetry from Basic to Applications: Radar Remote Sensing Using Polarimetric Information, The Institute of Electronics, Information and Communication Engineers, Tokyo, 2007 (in Japanese). [2] D. Grischkowsky, S. Keiding, M. V. Exter, and C. Fattinger, Far-infrared timedomain spectroscopy with terahertz beams of dielectrics and semiconductors, J. Opt. Soc. Am. B, vol. 7, no. 10, pp , Oct DOI: / JOSAB [3] B. B. Hu and M. C. Nuss, Imaging with terahertz waves, Opt. Lett., vol. 20, no. 16, pp , Aug DOI: /OL [4] W. L. Chan, J. Deibel, and D. M. Mittleman, Imaging with terahertz radiation, Rep. Prog. Phys., vol. 70, pp , DOI: / /70/8/ R02 [5] A. Kanno, K. Inagaki, I. Morohashi, T. Sakamoto, T. Kuri, I. Hosako, T. Kawanishi, Y. Yoshida, and K. Kitayama, 40 Gb/s W-band ( GHz) 16- QAM radio-over-fiber signal generation and its wireless transmission, European Conference and Exhibition on Optical Communication (ECOC), pp. 1 3,

2 1 Introduction At present, imaging technology used mainly remote sensing and underground radar has been developed to obtain high resolution and identification of complicated shapes. In particular, polarimetry technique is effective for identification of targets in the field of remote sensing [1]. The technique is a method which transmits and receives two orthogonal polarization components, and identifies targets by the analysis of the components. In general, the technique is used to survey the material and the size of the target, and is carried out by selecting specified narrow band. However, if the analysis by polarimetry is expanded to a broadband, it permits the identifications of the material and the size of the target from the characteristics all over the wide frequency. We focused on the uniqueness of broadband terahertz sensing, and expand the narrowband polarimetry technique to broadband. In general, terahertz waves have been using for the identification of target materials and communications which transmits enormous data in short range [2, 3, 4, 5]. For the terahertz sensing, terahertz time domain spectroscopy (THz-TDS) which acquires information in a wide band by radiating a short pulse is often used. In this study, we introduce the radar polarimetry to the THz-TDS measurement, and verified the effectiveness of the method in the precision identification of a target shape. 2 Target identification In radar polarimetry, the scattering matrix is used for target identification. The scattering matrix [S] represents the scattering state of the target by four components, S HH, S HV, S VH, and S VV, and is obtained from the following equation " # " EH s ¼ S #" # HH S HV E t H ; ð1þ EV s S VH S VV EV t where E t ¼½E t H;E t VŠ T is the electric field vector for transmitting, E s ¼ ½E s H;E s VŠ T is the vector for reception, H and V denote of the horizontal and vertical polarization components, respectively, and T represents the transpose. In addition, the S HV is the scattering coefficient received as the horizontal component when vertical polarized wave is transmitted. The obtained scattering matrix is converted into the polarization signature to evaluate the target clearly. The polarization signature shows the received power characteristic from the target when the transmitting and the receiving electric field vectors, E t and E s, are assumed to be the arbitrary polarizations, and it is possible to identify the rough shape of the target from the signature. When assuming E s ¼ E t, the received power P in the co-polarization is given by P ¼jE tt ½SŠE t j 2 ¼jVj 2 ; where ð2þ " #" #" #" # cos sin SHH S HV cos sin cos " V ¼½cos "; j sin "Š ; ð3þ sin cos S VH S VV sin cos j sin " τ is the tilt angle, ε is the ellipticity angle in the transmitting electric field vectors E t. In this study, the polarization signatures calculated by Eq. (2) are used to identify the target shapes. 267

3 3 Polarimetric calibrations On radar polarimetry, three major error factors are well known, the antenna alignment, antenna crosstalk, and channel imbalance. The inaccuracy of the antenna alignment causes the phase difference on the received wave mainly between the different orthogonal polarization components. The antenna crosstalk produces the interference to the other polarization component. So it affects S HV and S VH. The channel imbalance causes the level difference between the orthogonal polarization components. So it mainly affects S HH and S VV. Calibration is performed to compensate the errors. In this study, the difference in the scattering matrix between the measured and theoretical values of the known target is used as the calibration quantities. To decide the calibration quantities, we used a paper printed metallic ink uniformly. Since the scattering characteristic of the paper printed metaric ink is almost the same as that of the metal plate, the components of the theoretical scattering matrix is assumed as S HH ¼ S VV ¼ 1, and S HV ¼ S VH ¼ 0. Furthermore, the scattering matrix measured in the printed paper is assumed as " # " # S 0 HH S 0 HV ¼ A0 e j0 HH B 0 e j0 HV ; ð4þ S 0 VH S 0 VV C 0 e j0 VH D 0 e j0 VV where A 0, B 0, C 0, and D 0 are the amplitudes, and 0 HH, 0 VH, 0 HV, and 0 VV are the phases of the components in the measured scattering matrix. As S HH is equal to S VV, and S HV is equal with S VH in the theoretical scattering matrix of the metallic plate, the calibration quantities are defined as HV ¼ 0 HH 0 HV ; VH ¼ 0 HH 0 VH ; VV ¼ 0 HH 0 VV ; and X ¼ A 0 D 0 : These quantities are decided beforehand by preliminary experiments. The scattering matrix applied the calibration quantities, Eq. (5) (8), is obtained as " # " S HH S HV Ae j HH ðb B 0 pffiffiffi # Þe jð HV HV Þ ¼ S VH S VV ðc C 0 pffiffiffi Þe jð VH VH p Þ ðd X ffiffiffi ; ð9þ Þe jð VV VV Þ where A, B, C, and D are the amplitudes, and HH, VH, HV, and VV are the phases of the components in the scattering matrix measured the target, and σ is the scattering cross section ratio. 4 Measurement setup and target ð5þ ð6þ ð7þ ð8þ A landscape of the measurement equipment is shown in Fig. 1(a). The measurements are performed by using a THz-TDS. The distance from both the transmitter and receiver to the target is 150 mm. The incident angle of the wave is 30 degrees from the normal direction of the planar surface, and the receiver is located in the direction of the specular reflection. Also, a linearly polarized pulse wave tilted 45 degrees to the ground is radiated from the transmitter, and is divided into the vertical and horizontal components by a polarizer. The frequency range of pulse is 268

4 from 0.2 to 1 THz. The radiated wave is transformed to the plane wave by a collimator. The vertical and horizontal components of the scattering wave are separately received by placing another polarizer in front of the receiver. The targets are parallel line patterns printed on glossy paper with metallic ink as shown in Fig. 1(b). The target is set in a state that the parallel lines are inclined 45 degrees from the base line. The line widths are equal to the blank widths. Two types of the line width are tested, one is 0.3 mm and the other one is 0.2 mm. (a) Measurement. (b) Magnified target. Fig. 1. Landscapes of measurement and target. 5 Experimental results The polarization signatures of the measured co-polarization component at 0.2, 0.4, 0.6 and 0.8 THz were shown in Fig. 2 and Fig. 3. Fig. 2 and 3 show the results when the line widths are 0.3 mm and 0.2 mm, respectively. In Fig. 2(a), the received power is maximum at the ellipticity angle of 0 degrees and the tilt angle of about 45 degrees. Therefore, we can understand that the linear lines are inclined 45 degrees. In Fig. 2(b), it can be seen that the power is maximum at the tilt angle of about 45 degrees. The power is also slightly high at 45 degrees as indicated by the yellow region. In Fig. 2(c) and (d), the power is high at the ellipticity angle of 0 degrees and the tilt angle of from 90 to 90 degrees. Therefore, the target is observed as a metallic plate at 0.6 and 0.8 THz. Even for the same target, the shape identification depends on observing frequency as described above. In Fig. 3(a) and (b), the powers are maximum at the tilt angle of 45 degrees, so we can understand that the linear lines are inclined at 45 degrees. Furthermore, comparing Fig. 3(b) with Fig. 2(b), the power in Fig. 3(b) is lower than that in Fig. 2(b) at the tilt angle of 45 degrees and the ellipticity angle of 0 degree. The power in Fig. 3(c) is somewhat lower than that in Fig. 2(c) at tilt angle of from 20 to 20 degrees. The signature in Fig. 3(d) shows the same characteristic as that in Fig. 2(d). It is just like the characteristic of a metallic plate. From the above results, the frequency response of the polarization signature depends on the metal line width. It is considered that the response is related to the metal line width corresponding to the wavelength of the radiated wave. Therefore, 269

5 IEICE Communications Express, Vol.7, No.7, Fig. 2. Fig. 3. (a) 0.2 THz (b) 0.4 THz (c) 0.6 THz (d) 0.8 THz Polarization signature in target the line width of which is 0.3 mm. (a) 0.2 THz (b) 0.4 THz (c) 0.6 THz (d) 0.8THz Polarization signature in target the line width of which is 0.2 mm. 270

6 the broadband observation of the signature supplies us new information that does not supply in the narrowband observation. 6 Conclusion We introduced the radar polarimetry to the THz-TDS measurement, and verified the effectiveness of the method. The measurements using the periodic linear targets were carried out, and it is demonstrated that the observed polarization signature depended on the frequency. Using the frequency dependence of the signature, it becomes possible not only to roughly identify the shape of the target but also the size or the thickness. Although this study investigated a basic target, the results show that the combination of radar polarimetry and THz-TDS is effective for the precise identification of the targets. In the future, the promotion of the combination will bring the advance of the identification in the terahertz sensing. Acknowledgments This project is supported by MEXT-Supported Program for the Strategic Research Foundation at Private Universities (S ). 271