超寬頻訊號於複雜環境的傳播
圖十三顯示的是量測到的插入損失頻譜，發現到接收量隨著頻率往下掉的幅 度很大，從2 GHz 的-30 dB 掉到 6 GHz 的-50 dB 以下，能量至少掉了一百倍以 上。若仔細看，不難發現在頻率前半段，「Add Ground」曲線比較高一點，約 3 dB，表示從地面的反射量跟直接傳播量大小差不多。從圖十四結果可發現各個 包含有金屬箱的結果都比「No Box」也就是先前定義的「Add Ground」結果小 很多，頻率前半段有差到10 dB 以上，且出現位置疊在一起，幾乎一致，隨頻率 的上升也有往下掉的趨勢，從低頻到高頻共掉了約10 dB 左右。
圖十三：有無金屬接地面的傳播係數頻譜
圖十四：有無金屬箱障礙的傳播係數頻譜
第二階段，我們利用傅立葉反轉換將頻域測量結果轉換到時域。頻域與時域 的關係可以由以下來解釋：一個在時域上集中的訊號，如一脈衝訊號，經傅立葉
轉換便可觀察到在頻域上的廣泛分布。同樣的，一個在時域上連續不斷，如一正
量值一樣並出現在同一位置。「Add Ground」與 Case1 模擬結果隨頻率升高，一 個掉的快、一個慢符合實驗結果。比對傳輸係數的差異，在頻率為 2 GHz 處，
「Add Ground」較「No Ground」為大，「No Ground」與各個 Case 大小差約 9 dB 與實測結果相近。在時域的轉換結果上，最大值比例接近實際量測值，且出現位 置也非常接近。
圖十五： 有無金屬箱障礙的時域能量傳輸結果
(a) H=19cm, W=26cm, L=29cm (b) H=19cm, W=29cm, L=26cm
(c) H=26cm, W=19cm, L=26cm (d) H=26cm, W=26cm, L=19cm 圖十六： 有金屬箱障礙的時域能量傳輸結果比較
(a) 頻域傳輸係數頻譜 (b) 時域轉換結果 圖十七： 模擬的頻域與時域傳輸結果
在實際測量結果中，以最大值來說，「Add Ground」是「No Ground」的 1.77 倍，接收能量增加近兩倍，符合頻率域結果，且 Case1~6 的最大值約是「Add Ground」的 1/20 倍以下，接收能量大幅減少，也能符合頻率域結果。以 Case1~6 的時域結果來說，幾乎各個Case 的最大值都出現在距離 100~120 cm 之間，能量
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5
|E'(theta,refraction & diffraction)|, dB
Y axis (m)
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5
反射的訊號量，在接收天線的位置會減少或消失，而車體尾部的左右邊緣，則會
[1] Z. N. Chen, “UWB antennas: design and application,” Information,
Communications & Signal Processing, 2007 6th International Conference on,
2007, pp. 1-5.[2] P. Siripornnarachseema, S. Promwong, P. Sithiyopaskul, R, Muanghlua, N.
Sukutamtanti, “The effect of laptop computer shadowing on UWB propagation in an indoor environment,” Communications and Information Technology, 2005.
ISCIT 2005. IEEE International Symposium on, vol. 2, 2005, pp. 975–978.
[3] F. Berens, H. Dunger, S. Czarnecki, T. Bock, R. Reuter, S. Zeisberg, J. Weber, J.F.
Guasch, “UWB car attenuation measurements,” Mobile and Wireless
Communications Summit, 2007, 2007 , pp 1-5.
[4] Y. Suzuki, T. Kobayashi, “Ultra wideband signal propagation in desktop environments,” Ultra Wideband Systems and Technologies, 2003 IEEE
Conference on, 2003, pp. 493–497.
[5] T.-G. Ma, S.-J. Wu, “Ultrawideband Band-Notched Folded Strip Monopole Antenna,” Antennas and Propagation, IEEE Transactions on, vol. 55, no. 9, 2007, pp. 2473–2479.
[6] N. P. Agrawall, G. Kumar, and K. P. Ray, “Wide-band planar monopole antennas,” IEEE Trans. Antennas Propag., vol. 46, pp. 294–295, Feb. 1998.
[7] H.G. Schantz, G. Wolenec, and E.M. Myszka, “Frequency notched UWB antenna,” Proc. IEEE UWBST Conf., pp.214–218, Reston, VA, Nov. 2003.
[8] A. Kerkhoff and H. Ling, “Design of a planar monopole antenna for use with ultra-wideband (UWB) having a band-notched characteristic,”IEEE AP-S Int.
Symp. Dig., vol.1, pp.830–833, Columbus, OH,June 2003.
[9] K. L. Chan, S. R. Judah, “A beam scanning frequency modulated continous wave radar,” IEEE Trans. Instruction and Measurement, vol. 47, 1998, pp. 1223-1227.
[10] G. S. Woods, D. L. Maskell, M. V. Mahoney, “A high accuracy microwave ranging system for industrial applications,” IEEE Trans. Instruction and
Measurement, vol. 42, 1993, pp. 812-816.
[11] J. A. McGregor, E. M. Poulter, “Switching system for single antenna operation of an s-band fmcw radar,” IEEE Proc.-Radar, Sonar Naving, vol. 141, no. 4, 1994, pp. 241-248.
[12] K. V. Chet, C. B. Kuan, C. H. Teik “Development of a ground-based radar for scattering measurements,” IEEE Antennas and Propagation Magazine, vol. 45, no.
2, 2003, pp. 36-42.
計畫成果自評 (electromagnetic interference)問題。
二、達成目標：
本計劃所研究之工作項目主要有兩項:其一為評估車用防撞雷達操作時的 FMCW 回波特徵變化；另一項則是研發利用電磁脈衝技術進行無線通訊所需的超 寬頻天線，並評估超寬頻脈衝訊號在車輛引擎室內與車廂內的傳播特性，該研究 可因應車內各式感應器與車內微控制器(micro control unit, MCU)的通訊需 求。在第一項工作中，我們利用高頻數值電磁模擬方法(NEC-BSC code)，模擬出
四、期刊發表：
本案所發展之汽車回波特徵模擬程序與超寬頻訊號在車內環境傳播特性評估結 果，將於整理後，投稿期刊發表。
五、主要發現：
經本研究發現在車輛防撞雷達的操作環境下，以多個接收天線位置達成空間多樣 性，則可比對各觀察點所接收的回波量大小與相位變化判斷出目標物的縱向距 離、橫向位移，甚至是車身的姿態角度，作為未來發展防撞雷達目標物定位、定 向與行進方式推估演算法之用。
出席國際學術會議心得報告
會議名稱 2008 IEEE International Symposium on Antennas and Propagation and the 2008 USNC/URSI National Radio Science meeting
發表論文題目 Target Imaging using Multistatic Impulse Radar
一、參加會議經過
本人自 97 年 7 月 4 日啟程，於 5 日抵達美國聖地牙哥市，在完成登記後，及與會場 間先前熟識的各國教授與合作人員交換研究近況與心得。於 7 日到 11 日間主要參加下列 幾個 session:
#103 Metamaterials I: New Effects, Materials, and Devices
#126 Special Session: Recent Small Antennas and Sensors: Design and Applications
#224 Metamaterials and Left-Handed Antennas
#303 Characterization and Analysis of Metamaterials I
#330 Beamforming and Switched-Beam Antennas for Wireless Communications
#405 EBG Structures
#436 Electromagnetic Material Property Measurements
#503 Metamaterials and Antenna Applications
#526 Ultra Wideband Antennas and Systems II
本人並於 8 日上午 11:20 在 Session 206 “Antennas and Microwave Components for UWB Communications”中，發表”Target Imaging using Multistatic Impulse Radar”的論文報 告，在 Q&A 時段，有多位聽眾提問，並交換聯絡方式，持續進行意見交換。
Target Imaging using Multistatic Impulse Radar
Wen-Jiao Liao*The Department of Electrical Engineering, National Taiwan University of Science and Technology
43, Sec. 4, Keelung Rd., Taipei 106, Taiwan
E-mail: [email protected] Fax: 886-2-2737-6699
In this work, a radar signal processing method is developed to construct 2-D images of targets using RCS data collected from a multistatic broadband impulse radar. The advantages of using such a radar scenario include detecting stealthy target using multistatic RCS, concealing the presence of the active radar system with the spread spectrum impulse signal, and facilitating target recognition using the 2-D radar images with a high resolution. The implementation comprises the synthesis of radar return signals and the transform of RCS data in polar coordinate to 2-D target image using correlation operation with a point scatter reference in the spatial domain.
The radar signal returns are simulated with a high frequency numerical method based on geometric theory of diffraction. In order to generate the response of a time domain impulse, simulations a frequency domain sweep is performed in a loop fashion. By collecting back scattered field intensities at various frequencies, the time domain response can be derived using the Fourier transform. Then a radial distance-angle map as shown in Figure 1(a) can be formulated from returns collected at different angles.
Next, by assuming that the RCS profile of a geometrically complicated target can be formed with superposition of point scatter responses, the 2-D radar image is produced with a spatial domain correlation operation. Here, the delta function is used to represent the ideal point scatter response. A reference radial distance-angle map is formulated according to the prescribed radial distance and distance-angle.
The reference is next convoluted with the polar RCS profile of a complicated target to yield a correlation value. Finally, the transform is completed by stitching correlation values derived at different positions to form the correlation map shown in Figure 1(b), which is derived from an aircraft model. The transformed target image exhibits a rough aircraft contour and indicates that the tail and wing tips of the airplane are relatively strong scatters. For practical implementation considerations, parametric studies of system parameters such as the angular and radial distance resolution are performed to their effects on the radar imaging quality.
Figure 1(a) Synthesized bistatic returns, (b) Transformed target image