衛星通訊最初為軍事和科學研究而發展。現今，全世界有釵h國家提出商用Ka-band寬頻衛星系統計劃並進行建設。這些系統將提供多媒體及網際網路等通訊服務給全球的個人及企業使用。中華衛星一號是臺灣的第一顆科學衛星，它的成敗攸關日後衛星科技的發展，因此深受國人的關注。
對於低軌道衛星而言，系統在低仰角不僅遭受大的自由空間損失，且遭受嚴重的物理衰減。從驗證結果知，雨衰是實驗中影響最大的因素，隨著雨率愈大、頻帶愈高，或仰角愈低，雨衰值亦隨之增加。因為頻帶關係，所以上鏈(28.25GHz)比下鏈(18.45GHz)雨衰值大
對於衛星通訊實驗，建立一個趨近真實情況的通道模型是非常重要。在過去的兩年中，我們已經成弘竣F釵h針對中華衛星一號Ka頻帶的通道模型相關的研究和模擬。本論文中，我們做了釵h提升錯誤更正碼的編碼效能的架構：迴旋碼(convolutional code)、連鎖碼(concatenated code)、晶格編碼調變(trellis coded modulation)、渦輪碼(turbo code)、渦輪晶格編碼調變(turbo-trellis coded modulation)加入模型中。渦輪晶格編碼調變(turbo-trellis coded modulation)有很好的編碼效能，但是其複雜度高且須要花大量的時間。
然後，我們討論中華衛星一號鏈結之所需及在不同路逕的相關變化。因此，我們選擇中華衛星一號的一個路逕，便對於其模擬的結果作完整的描述。最後，對於中華衛星一號實驗模擬結果給予一些結論及建議。

Satellite communications were originally developed for scientific and military researches. Nowadays, a number of nations have begun their plans to implement commercial Ka-band satellite systems in the world. These systems will offer multi-media and Internet communications services to individual and corporate customers around the world. ROCSAT-1 is Taiwan’s first scientific satellite. Because its success or failure will affect the future development of satellite technology, it is an object of great concern for the people of Taiwan.
For LEO satellites, systems at low elevation angle suffer not only large free space loss but also great physical attenuations. From the verifications, rain attenuation is the dominant propagation impairment in our experiment. The larger rain rate, the higher frequency or the lower elevation angle is, the larger rain attenuation value will become. Rain attenuation induced in the uplink direction (28.25GHz) are much larger than that induced in the downlink direction (18.45GHz) since the uplink utilizes a high frequency band.
For satellite communication experiments, it is very significant to establish a channel model as close as possible to the real situation. In the past two years, we have successively made some researches on channel characteristics and channel simulations of ROCSAT-1. In this thesis, we make many researches to improve the system performance by various FEC coding schemes: convolutional code, concatenated code, trellis coded modulation, turbo code, and turbo trellis coded modulation scheme. From the results, Turbo trellis coded modulation scheme has better performance, but its decoding structure demands more complexity and takes more time.
We discuss the specifications of ROCSAT-1 link need and how they vary through different paths. For this reason, we choose a real ROCSAT-1 pass as an example to make a complete description of the simulation procedure. Finally, we present the simulation process and give some suggestions from the simulation results.

[1] Louis J. Ippolito Jr., “Radiowave Propagation in Satellite
Communications”.
[2] ITU Recommendation ITU-R RPN.618-4, 1996
[3] Asoka Dissanayake, Jeremy Allnutt, and Fatim Haidara, ” A prediction model that combines rain attenuation and other propagation impairment along earth-satellite paths ”, IEEE, Trans. on antenna and propagation, vol45, No 10,pp 1546-1558, 1997.
[4] Yen-Hsyang Chu and Shun-Peng Shih, “Ka Band Propagation Experiment for ROCSAT-1:Preliminary Results,” 中華衛星一號科學成果研討會大會手冊及論文摘要集, pp.111~112, 1999.
[5] Robert K. Crane, “Prediction of Attenuation by rain,” IEEE Trans. on Communications, Vol. Com-28, No. 9, pp1717~1733 Sept. 1980.
[6] Robert K. Crane, “A Global Model for Rain Attenuation Prediction,”EASCON’78 Record, IEEE Pub, 78CH 1354-4 AES, Arlington, VA, Sept. 1978.
[7] Crane, R. K., “Electromagnetic Wave Propagation Through Rain,”
Wiley, 1996.
[8] ITU-R Rec. 837-1, International Telecommunication Union.
[9] CCIR Rep.564-4
[10] Leon W., Digital and Analog Communication Systems.
[11] Rolf Johannesson, Kamil Sh. Zigangirov, ”Fundamentals of Convolutional Coding,” IEEE Press, 1999.
[12] G. D. Forney, Concatenated Codes, Cambridge, Mass.: MIT Press,
1967.
[13] Stephen B. Wicker, Vijay K. Bhargava, Reed-Solomon Codes and Their applications, Piscataway, pp. 25 ~ 38, NJ : IEEE Press, 1994.
[14] G. Ungerboeck and I. Csajka, “On improving data link performance by increasing the channel alphabet and introducing sequence coding,” in Proc., IEEE Int. Symp. On Inform. Theory, June 1976.
[15] G. Ungerboeck, “Channel Coding with Multilevel/Phase Signals”, IEEE Trans. Inform. Theory, vol. IT-28, no. 1, pp.55-67, Jan. 1982.
[16] Matthew C. Valenti, “Iteractive Detection and Decoding for Wireless Communication”, Viginia Polytechnic Institute and State University, July 8, 1999.
[17] J. Hagenauer, “Iterative decoding of binary block and convolutional codes,” IEEE Trans. Infrom. Theory, vol. 42, pp.429-445, Mar. 1996.
[18] Matthew C. Valenti, “Iteractive Detection and Decoding for Wireless Communication”, Viginia Polytechnic Institute and State University, July 8, 1999.
[19] B. Vucetic,”Iterative decoding algorithm,” in PIMRC’97, Helsinki, Finland, Sep. 1997, pp. 99-120.
[20] Branka Vucetic, Jinhong Yuan,” TURBO CODES Principles and Applications ”.
[21] P. Robertson and T. Woerz, “Bandwidth efficient turbo trellis-coded modulation using punctured component codes,” IEEE Journal on Selected Areas in Communications, vol. 16, no. 2, Feb. 1998, pp. 206-218.
[22] Gray D. Gordon and Walter L. Morgan, Principles of Communications Satellites, Jonh Wiley & Sons, Inc, 1993.
[23] Yi-Jeng Liau, “ Simulation & Preliminary Verification of Direct
Broadcast Experiments Via ROCSAT-1,” Master Thesis, National
Tsing-Hua University.
[24] Kuo-Wei Chao, “Noise Figure Modification and Simulation Results,”quarterly report to NSPO, September, 2000.
[25] Le-Gen Shi, “Channel Simulation of LEO ROCSAT-1 Using Ka-Band,” Master Thesis, National Tsing-Hua University.
[26] Pei-Kai Liau, “Channel Characteristics Analysis and ECP Experiment Planning for LEO ROCSAT-1 Using Ka-Band,” Master Thesis, National Tsing-Hua University, June 1999.
[27] Leo Wang, “System Performance Analysis and Simulation for ROCSAT-1 Direct Broadcast Experiment of Digital TV signals Using Ka-Band,” Master Thesis, National Center University.