• 沒有找到結果。

先導訊號輔助式之正交分頻多工通信系統通道估測設計

N/A
N/A
Protected

Academic year: 2021

Share "先導訊號輔助式之正交分頻多工通信系統通道估測設計"

Copied!
14
0
0

加載中.... (立即查看全文)

全文

(1)國 立 交 通 大 學 電子工程學系 電子研究所碩士班 碩 士 論 文. 先導訊號輔助式之正交分頻多工通信系統 通道估測設計. Pilot-Aided Channel Estimations for OFDM Systems. 研 究 生:吳清泉 指導教授:陳紹基 博士. 中 華 民 國 九 十 三 年 六 月.

(2) 先導訊號輔助式之正交分頻多工通信系統 通道估測設計 Pilot-Aided Channel Estimations for OFDM Systems 研 究 生:吳清泉. Student:Chin-Chuan Wu. 指導教授:陳紹基 博士. Advisor:Sau-Gee Chen. 國 立 交 通 大 學 電子工程學系. 電子研究所碩士班. 碩 士 論 文. A Thesis Submitted to Department of Electronics Engineering & Institute of Electronics College of Electrical Engineering and Computer Science National Chiao Tung University in Partial Fulfillment of the Requirements for the Degree of Master of Science in Electronics Engineering. June 2004 Hsinchu, Taiwan, Republic of China. 中華民國九十三年六月.

(3) 先導訊號輔助式之正交分頻多工通信系統 通道估測設計 學生:吳清泉. 指導教授:陳紹基 博士. 國立交通大學 電子工程學系 電子研究所碩士班. 摘. 要. 在本篇論文中,我們研究了正交分頻多工通信系統的通道估測及 等化。在論文的第一部分,我們比較了不同通道內插估測方法在 IEEE 802.16a 和 DVB-T 通信系統中的表現效能包括線性、Lagrange、cubic spline、cubic Bspline、DFT-based 和 DCT-based 內插器;而發現線 性及 cubic Bspline 內插器有著較佳的位元錯誤比率。如果同時考量 運算複雜度的話,通道估測採用線性內插器是一個較佳的選擇。在論 文的第二部分,我們研究了由快速衰減通道和載波頻率偏差造成的頻 道之間相互干擾對正交分頻多工通信系統的影響。根據通道線性變化 的假設,參考現有的方法,我們提出一個能夠有效減低頻道之間相互 干擾的估測方法。.

(4) Pilot-Aided Channel Estimations for OFDM Systems Student: Chin-Chuan Wu. Advisor: Sau-Gee Chen. Department of Electronics Engineering & Institute of Electronics National Chiao Tung University. Abstract. In this thesis, channel estimation and equalization for the Orthogonal Frequency Division Multiplexing (OFDM) system are investigated. In the first part of the thesis, we compare the performance of different channel interpolation schemes in IEEE 802.16a and DVB-T systems inclusive of linear, Lagrange, cubic spline, cubic Bspline, DFT-based, and DCT-based interpolators. As a result, linear and cubic Bspline interpolators show better bit error rate. Concerning the computational complexity as well as the performance, the linear interpolator is considered as a proper choice for channel estimation. In the second part of this thesis, we investigate the effect of inter-carrier interferences (ICI) caused by fast fading channels and carrier frequency offsets., According to the linear assumption of channel impulse response, we propose an estimation method that can reduce the ICI effect efficiently based on a current channel estimation method..

(5) 誌 謝 對於能夠順利完成我的碩士學位,首先要感激的是我的指導教授 陳紹基博士,在這兩年中對於我的課業研究著實提供了許多幫助,在 我感到困惑時,適時的引導正確的方向,就像迷航的船隻驚見遠方明 亮的燈塔那般;另外在生活上也使得我成長不少,懂得該如何應對進 退、待人處事,在此獻上由衷的感激。 另外要感謝的就是 429 實驗室的夥伴們,伴我度過兩年的研究所 生涯,陪同我一起歡笑、一起玩樂、以及一起苦悶。我不能想像如果 這兩年沒有你們這群朋友,我的生活將會是怎麼樣?尤其是卓卓學 長、明秀、小紀、A 貓、以及昆蟲,兩年之中,我們大概有四分之三 的時間是在一起努力的,謝謝你們;雖然要畢業了,但是希望能夠在 未來多聚聚。 最後,要感謝這兩年默默支持我的家人,給我許多呵護跟包容, 使得我能夠順利的完成學業,謝謝。.

(6)

(7) Contents Chapter 1 Introduction 1.1. 1. Motivation of the Thesis. 1.2 Organization of the Thesis. Chapter 2 Fundamentals of OFDM Systems 2.1. 2.2. OFDM System Model. 2 3. 4 5. 2.1.1. Continuous-Time Model. 5. 2.1.2. Discrete-Time Model. 7. 2.1.3. Effect of Cyclic Prefix. 8. Channel Characteristics in Wireless Environments. 8. 2.2.1. Flat Fading and Frequency Selective Fading. 11. 2.2.2. Slow Fading and Fast Fading. 12. Chapter 3 Physical Layer of IEEE 802.16a and DVB-T Standard 14 3.1. IEEE 802.16a System 3.1.1. 3.1.2. 3.1.3. 3.2. 14. OFDMA Frame Structure. 16. 3.1.1.1. OFDMA Symbol Structure. 16. 3.1.1.2. OFDMA Frame Structure. 17. OFDMA Carrier Allocations. 18. 3.1.2.1. Downlink Carrier Allocations. 19. 3.1.2.2. Uplink Carrier Allocations. 21. Carriers Modulation. 23. 3.1.3.1. Data Modulation. 23. 3.1.3.2. Pilot Modulation. 24. DVB-T System. 25.

(8) 3.2.1. DVB-T System Overview. 26. 3.2.2. Frame Structure. 29. 3.2.3. Carrier Allocations. 30. 3.2.3.1. Continual Pilots. 30. 3.2.3.2. Scattered Pilots. 30. 3.2.3.3. Transmitted Parameter Signaling. 31. Chapter 4 Channel Estimations for 802.16a and DVB-T Systems 33 4.1. 4.2. Fundamentals of Channel Estimations for OFDM Systems. 34. 4.1.1. Pilot Arrangement. 34. 4.1.2. The LS Channel Estimation. 36. 4.1.3. The LMMSE Channel Interpolation. 36. 4.1.4. Some Popular Channel Interpolation Techniques. 37. 4.1.4.1. Piecewise Linear Interpolation. 38. 4.1.4.2. Lagrange Interpolation. 38. 4.1.4.3. Spline Interpolation. 39. 4.1.4.4. Bspline Interpolation. 41. 4.1.4.5. DFT-Based Interpolation. 42. 4.1.4.6. DCT-Based Interpolation. 45. Channel Estimation for 802.16a and DVB-T Systems. 49. 4.2.1. Channel Estimations for 802.16a Uplink. 49. 4.2.2. Channel Estimations for 802.16a Downlink. 50. 4.2.3. 4.2.2.1. Pilot Arrangement and Guard Band Effect. 50. 4.2.2.2. Simulations and Performance Evaluation. 51. Channel Estimations for DVB-T System. 59.

(9) 4.2.4. 4.2.3.1. Pilot Arrangement and Guard Band Effect. 59. 4.2.3.2. Simulations and Performance Evaluation. 59. Computational Complexities. 65. Chapter 5 Channel Estimations and Equalization with Fast Fading. 5.1. Channels and Carrier Frequency Offsets. 67. Introduction of System Model. 69. 5.1.1. Effect of Inter Carrier Interferences in OFDM Systems 69. 5.1.2. System Model of Fast-Fading Channels with Carrier Frequency Offsets. 5.2. 5.3. The Proposed Estimation Method for Fast-Fading Channels. 69 72. 5.2.1. Proposed Estimation Scheme. 72. 5.2.2. Proposed Simplified Estimation Scheme. 77. Simulation Results. 79. Chapter 6 Conclusion. 85. Bibliography. 87.

(10) List of Tables Table 3.1. 802.16a OFDMA DL carrier allocation ..................................................21. Table 3.2. 802.16a OFDMA UL carrier allocation ..................................................22. Table 3.3. DVB-T OFDM parameters for 8K & 2K modes, with 8MHz channel ..29. Table 3.4. TPS signaling and format of DVB-T system..........................................32. Table 4.1. Characteristics of ETSI “Vehicle A” channel environment ....................53. Table 4.2. Relative power, phase and delay values for portable channel P1 ..........62. Table 4.3. Computational complexity of various key interpolation schemes..........68. Table 5.1. Relative power and delay values for multipath channels........................82.

(11) List of Figures Figure 2.1. Cyclic prefix of an OFDM symbol ........................................................5. Figure 2.2(a) Continuous-time OFDM baseband modulator.......................................6 Figure 2.2(b) Continuous-time OFDM baseband demodulator ...................................6 Figure 2.3. Spectrum of OFDM signal.....................................................................6. Figure 2.4. Discrete-Time OFDM system model.....................................................7. Figure 2.5. The equivalent discrete impulse responses ..........................................11. Figure 2.6. The frequency responses for two channels ..........................................11. Figure 2.7. Fading type as a function of baseband signal bandwidth ....................13. Figure 3.1. 802.16a OFDMA symbol time structure..............................................17. Figure 3.2. 802.16a OFDMA frequency description..............................................18. Figure 3.3. 802.16a OFDMA TDD time frame......................................................19. Figure 3.4. Carrier allocation of 802.16a OFDMA DL..........................................20. Figure 3.5. Carrier allocation of 802.16a OFDMA UL..........................................22. Figure 3.6. QPSK, 16-QAM, 64-QAM constellations of 802.16a.........................23. Figure 3.7. PRBS for pilot modulation of 802.16a ................................................24. Figure 3.8. Function block diagram of DVB-T system..........................................26. Figure 3.9. Hierarchical transmissions with non-uniform 64-QAM modulation of DVB-T system .....................................................................................27. Figure 3.10. Continual pilot locations of DVB-T system ........................................30. Figure 3.11. Scattered pilot locations of DVB-T system .........................................31. Figure 4.1. (a)Comb-type pilot arrangement (b) Block-type pilot arrangement of an OFDM system .................................................................................35. Figure 4.2. Lagrange interpolation .........................................................................39. Figure 4.3. Cubic spline interpolation ....................................................................40.

(12) Figure 4.4. Construction of B-Splines....................................................................42. Figure 4.5. Regular equal-spaced pilot placement .................................................43. Figure 4.6. Aliasing effect of channel impulse response due to down-sampling in frequency domain.................................................................................44. Figure 4.7. DFT-based channel estimator...............................................................44. Figure 4.8. DCT/EIDCT-based channel estimator .................................................49. Figure 4.9. Equivalent DCT-based and DFT-based channel estimators.................49. Figure 4.10. BER performance of various interpolation schemes with f d = 40 Hz , one-symbol pilot case ..........................................................................54. Figure 4.11. BER performance of various interpolation schemes with f d = 40 Hz , two-symbols pilot case.........................................................................54. Figure 4.12. BER performance of various interpolation schemes with f d = 40 Hz , four-symbols pilot case ........................................................................55. Figure 4.13. BER performance of linear interpolation versus pilot symbol number with f d = 40 Hz ..................................................................................55. Figure 4.14. BER performance of various interpolation schemes with f d = 40 Hz , two-dimensional interpolation .............................................................56. Figure 4.15. BER performance of various interpolation schemes with f d = 120 Hz , one-symbol pilot case ..........................................................................56. Figure 4.16. BER performance of various interpolation schemes with f d = 120 Hz , two-symbols pilot case.........................................................................57. Figure 4.17. BER performance of various interpolation schemes with f d = 120 Hz , four-symbols pilot case ........................................................................57. Figure 4.18. BER performance of linear interpolation versus pilot symbol number with f d = 120 Hz ................................................................................58. Figure 4.19. BER performance of various interpolation schemes with f d = 120 Hz ,.

(13) two-dimensional interpolation .............................................................58 Figure 4.20. BER performance of linear interpolation versus Doppler frequencies .............................................................................................................59. Figure 4.21. BER performance of various interpolation schemes with f d = 40 Hz , one-symbol pilot case ..........................................................................63. Figure 4.22. BER performance of various interpolation schemes with f d = 40 Hz , four-symbols pilot case ........................................................................63. Figure 4.23. BER performances of linear interpolation versus pilot symbol number with f d = 40Hz ....................................................................................64. Figure 4.24. BER performance of various interpolation schemes with f d = 40 Hz , two-dimensional interpolation .............................................................64. Figure 4.25. BER performance of various interpolation schemes with f d = 120 Hz , one-symbol pilot case ..........................................................................65. Figure 4.26. BER performance of various interpolation schemes with f d = 120 Hz , four-symbols pilot case ........................................................................65. Figure 4.27. BER performances of linear interpolation versus pilot symbol number with f d = 120 Hz ................................................................................66. Figure 4.28. BER performance of various interpolation schemes with f d = 120 Hz , two-dimensional interpolation .............................................................66. Figure 4.29. BER performances versus Doppler frequencies of linear interpolation .............................................................................................................67. Figure 5.1(a) Time variation of the CIR for different relative Doppler frequency changes within a block symbol. ...........................................................76 Figure 5.1(b) Magnitude responses for CIR versus Doppler frequencies..................76 Figure 5.2. The shape of G 1 (ICI effect for 1-th subcarrier) ...............................80. Figure 5.3. The shape of G 3 (ICI effect for 3-th subcarrier) ...............................81.

(14) Figure 5.4. BER comparison with f d T = 0.0384 and f o =0.5 carrier spacing...84. Figure 5.5. BER comparison with f d T = 0.0576 and f o =0.5 carrier spacing...84. Figure 5.6. BER comparison with f d T = 0.0786 and f o =0.5 carrier spacing...85. Figure 5.7. BER comparison with f d T = 0.0384 and f o =0.5 carrier spacing...85. Figure 5.8. BER comparison with f d T = 0.0576 and f o =0.5 carrier spacing...86. Figure 5.9. BER comparison with f d T = 0.0768 and f o =0.5 carrier spacing...86.

(15)

參考文獻

相關文件

How does drama help to develop English language skills.. In Forms 2-6, students develop their self-expression by participating in a wide range of activities

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix

Optim. Humes, The symmetric eigenvalue complementarity problem, Math. Rohn, An algorithm for solving the absolute value equation, Eletron. Seeger and Torki, On eigenvalues induced by

 Number of spectral averages:頻譜 平均的數量,設定為1。. 教育部資通訊科技人才培育先導型計畫

Microphone and 600 ohm line conduits shall be mechanically and electrically connected to receptacle boxes and electrically grounded to the audio system ground point.. Lines in

傳統的 RF front-end 常定義在高頻無線通訊的接收端電路,例如類比 式 AM/FM 通訊、微波通訊等。射頻(Radio frequency,

The purpose of this thesis is to investigate the geometric design of curvic couplings and their formate grinding wheel selection, and discuss the geometric