目標辨識在雷達工程中是非常重要的一環。一般來說，SAR（合成孔徑雷達）或ISAR（逆合成孔徑雷達）圖像常用於識別雷達目標。但是SAR或ISAR圖像很難獲得數據。RCS（雷達截面）與SAR或ISAR圖像相比，其數據更容易獲得。本研究中使用的雷達目標辨識的數據是取自RCS（雷達截面積），且包括雜訊效應。依據量測時天線的接收方式可分為單向雷達雷達截面積（monostatic RCS）及雙向雷達截面積（bistatic RCS）。當發射天線與接收天線置於同方位時，稱之單向雷達截面積；反之則為雙向雷達截面積。本文採用雙向雷達截面積，目地是比較未知和已知目標之間的相似性。
文中將討論不同角度方向蒐集而來的RCS數據如何進行辨識。基於現實中不同角度的RCS數據需花費較長時間才能完整取得，所以我們使用商用電磁軟體Ansys HFSS 模擬而得。角度分集與頻率分集RCS技術雖已有數種方法可成功用於雷達辨識，但RCS訊號在加入雜訊過後許多方法的成功辨識率都會有顯著地下降情況，為了克服這些情況，我們嘗試著使用其他的新方法去對RCS訊號做辨識。
文中，所使用的第一個方法是深度神經網路（Deep Neural Networks,DNN），第二個方法為卷積神經網路（Convolutional Neural Networks,CNN），雖然我們使用的RCS數據量並不多，但不影響比較未知目標物與已知目標物的相似機率與討論，且為了對兩方法做更進一步的比較，在共同結構部份的參數均設為一樣。最後，利用不同的數據集進行分析預測，討論雜訊對模型辨識的影響與比較，並探討此兩種方法所預測結果的準確性。

SUMMERY
The main purpose of this study is to achieve target recognition using the diversity radar section. The radar corss section can be divided into several different ways of collecting data. Each ways has its own characteristics, which will affect the effectiveness of target recognition, so we use this as a starting point to do this research.
This study used angular diversity radar and frequency diversity radar to collect data,
and use deep neural networks and convolution neural networks as models to train the data, classify unknown data, and compare the results to discuss the feasibility of this method.
The results of the simulation are also very exciting. Whether it is a deep neural network or a convolutional neural network, most of the cases have very good classification performance, that is, using deep learning to do radar. Target recognition is a very good method, at the end, also compares angle diversity with frequency diversity. Angle diversity can achieve better recognition rate, and frequency diversity is an advantage when collecting data.
This research has achieved its original purpose, and later hopes to perform radar target recognition in more different ways.

[1] R. G. S. Matthew and D. C. Hughes, “Getting at deep learning: a problem-based approach”, IET Journals & Magazines Engineering Science and Education Journal, Volume: 3, Issue: 5, pp. 234 - 240, Oct, 1994
[2] K. Charalampous, I. Kostavelis, A. Amanatiadis and A. Gasteratos, ”Sparse deep-learning algorithm for recognition and categorisation”, IET Journals & Magazines Electronics Letters, Volume: 48, Issue: 20, pp. 1265 - 1266, Oct, 2012
[3] G. Hajduch, J. M. L. Caillec and R. Garello, “Airborne high-resolution ISAR imaging of ship targets at sea”, IEEE Transactions on Aerospace and Electronic Systems, Vol. 40, Issue: 1, pp. 378-384, Jan, 2004.
[4] S. Musman, D. Kerr and C. Bachmann, “Automatic recognition of ISAR ship image”, IEEE Trans. Aerospace Electronic Systems, Vol. 32, Issue 4, pp. 1392-1403, Oct, 1996.
[5] M. Menon, “Automatic ship classification system for ISAR imagery”, SPIE Proc. Applications and Science of Artificial Neural Networks, Vol. 2492, pp. 373-388, 1995.
[6] H. Osman, S. Blostein and L. Gagnon, “Classification of ships in airborne SAR imagery using backpropagation neural networks”, SPIE Proc. Radar Processing, Technology, and Applications II , Vol. 3161, pp. 126-136, Aug, 1997.
[7] M. Tello, C. L. Martinez and J. J. Mallorqui, “A novel algorithm for ship detection in SAR imagery based on the wavelet transform”, IEEE Geoscience and Remote Sensing Letters, Vol. 2, No. 2, pp. 201–205, Apr, 2005.
[8] G. T. Ruck, D. E. Barrick, W. D. Stuart and C. K. Krichbaum, Radar Cross Section Handbook, Vol. 1, Plenum, New York, 1970.
[9] J. W. Crispin and A. L. Maffett, “Radar cross-section estimation for simple shapes”, IEEE Journals & Magazines Proceeding of the IEEE, Vol. 53, No. 8, pp. 833-848, Aug, 1965.
[10] H. Xingyu, T. Ningning and H. Xiaowei, ”Automatic recognition of ISAR images based on deep learning”, IEEE Conferences, CIE International Conference on Radar (RADAR), Guangzhou, China, pp. 1-4, Oct, 2016
[11] L. Jarmo and K. Visa, “Deep learning for HRRP-based target recognition in multistatic radar systems”, IEEE Radar Conference (RadarConf), Philadelphia, PA, USA, pp. 1-6, May, 2016
[12] Z. Fan, H. Chen, Y. Qiang, L. Wei and H. C. Li, “Multi-aspect-aware bidirectional LSTM networks for synthetic aperture radar target recognition”, IEEE Journals & Magazines IEEE Access, Vol: 5, pp. 26880 - 26891, 2017
[13] S. Chen, H. Wang, F. Xu and Y. Q. Jin, “Target classification using the deep convolutional networks for SAR images”, IEEE Transactions on Geoscience and Remote Sensing, Vol: 54, Issue: 8, pp. 4806 - 4817, Aug, 2016
[14] A. E. Housseini, A. Toumi and A. Khenchaf, ” Deep learning for target recognition from SAR images”, IEEE Conferences Seminar on Detection Systems Architectures and Technologies (DAT), Algiers, Algeria, pp. 1-5, Feb, 2017
[15] C. D. Gupta, R. B. Ahmad and I. A. Safar, “Neural network study of two proposed structures: 1. truncated conical monopole: 2. fan-beam structure fed by a metal-dielectric feeder”, IEEE Conferences International Waveform Diversity & Design Conference , Lihue, HI, USA, pp. 1-4, Jan,2006
[16] P. Sermanet, S. Chintala and L. C. Yann, “Convolutional neural networks applied to house numbers digit classification”, IEEE Conferences Proceedings of the 21st International Conference on Pattern Recognition, Tsukuba, Japan, pp. 3288 - 3291, Nov, 2012
[17] D. Ciresan and D. Pescaru, “Off-line recognition of handwritten numeral strings composed from two-digits partially overlapped using convolutional neural networks”, IEEE Conferences 2008 4th International Conference on Intelligent Computer Communication and Processing, Cluj-Napoca, Romania, pp. 53 - 60, Aug, 2008
[18] E. J. Rothwell, K. M. Chen, D. P. Nyquist, J. E. Ross and R. Bedermeyer, “A radar target discrimination scheme using the discrete wavelet transform for reduced data storage”, IEEE Transactions on Antennas and Propagation, Vol. 42, Issue 7, pp. 1033-1037, Jul, 1994.
[19] J. L. Rasmussen, R. L. Haupt, J. Michael and D. Walker, “RCS feature extraction from simple targets using time-frequency analysis”, SPIE Proc. Radar Processing Technology, and Applications, Vol. 2845, pp. 66-74, 1996.