跳到主要內容

臺灣博碩士論文加值系統

(18.206.76.226) 您好!臺灣時間:2021/07/30 21:13
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:顏景池
研究生(外文):Chin-Tsu Yen
論文名稱:運用類神經網路於數位通信系統之通道等化器設計
論文名稱(外文):Design of channel Equalizer Using Artificial Neural Networks for Digital Communication Systems
指導教授:翁萬德
指導教授(外文):Wan-de Weng
學位類別:博士
校院名稱:國立雲林科技大學
系所名稱:工程科技研究所博士班
學門:工程學門
學類:綜合工程學類
論文種類:學術論文
論文出版年:2004
畢業學年度:92
語文別:中文
論文頁數:70
中文關鍵詞:類神經網路MLPFLANNRDF-FLANN通道等化器
外文關鍵詞:Neural networksMLPFLANNRDF-FLANNChannel
相關次數:
  • 被引用被引用:0
  • 點閱點閱:161
  • 評分評分:
  • 下載下載:31
  • 收藏至我的研究室書目清單書目收藏:0
在本論文中,我們主要探討數位通信系統之通道等化器設計,在眾多的通道等化設計方法中,我們採用函數連結類神經網路 (functional link artificial neural networks, FLANN) 為基礎架構之網路去設計通道等化器,且將它分別應用在2PAM與4QAM之數位通信系統中。主要原因是FLANN利用非線性函數近似的方式,成功地將原先在較低輸入信號空間維度下之線性不可分類的問題,擴展至較高信號空間維度之超平面上 (hypersurface),藉以形成線性可分類,解決通道等化問題。由於FLANN本身透過函數展開 (functional expansion)的功能,取代Multilayer Perceptron Networks(MLP)的隱藏層,所以其具有較簡單的網路結構與較低之計算複雜度 (computational complexity),同時擁有較快的收斂速度。另外,我們針對函數連結類神經網路提出一個改善的類神經網路架構,我們稱之為簡化回授函數連結類神經網路 (reduced decision feedback functional link artificial neural networks, RDF-FLANN),在此架構中我們除利用FLANN原有的優點之外,並兼顧節省硬體成本的考量下,我們選擇在原來的FLANN網路內部加入輸出端之回授信號,藉由此回授補償的功能,使RDF-FLANN通道等化器在網路訓練期間比FLANN通道等化器收斂更快速,更能符合現代通信之即時需求。從模擬結果顯示,在相同之符元錯誤率 (symbol error rate, SER) 情況之下,RDF-FLANN通道等化器在信號雜訊比 (signal-to-noise ratio, SNR) 方面的表現比FLANN通道等化器改善 2~3 dB,而其硬體成本卻只增加不到3%。
The design of a channel equalizer in digital communication systems is discussed in this thesis. Among the various methods for realizing channel equalizers, we have chosen functional link artificial neural networks (FLANN) for the implementation. The design has been successfully applied to digital communication systems transmitting 2PAM or 4QAM modulated signals. The FLANN structure has the advantage of being able to expand the lower dimensional signal space onto a higher dimensional hypersurface by using nonlinear functional approximation. This expansion can convert the linear inseparable problem into a separable one, which makes the FLANN have pretty simple network structure and low computational complexity. Because the FLANN does not need the hidden layers, which are existed in most MLP-based equalizers, it generally exhibits high speed of convergence. Moreover, in order to further improve the performance of FLANN, we have proposed a reduced decision-feedback functional link artificial neural networks (RDF-FLANN) for the design of a nonlinear channel equalizer in digital communication systems. In this RDF-FLANN we add a local feedback signal to the input layer directly from the output. This new architecture not only preserves the advantages of the traditional FLANN, but also significantly saves the hardware cost. Besides, the decision feedback mechanism utilized in the RDF-FLANN structure can greatly speed up the convergence of network settings during the training process. As it can easily meet the real-time processing requirement of modern communication systems, the RDF-FLANN is more suitable for implementing a nonlinear channel equalizer than FLANN. Simulation results demonstrate that the RDF-FLANN outperforms FLANN by about 2 to 3 dB and its hardware cost is only about 1.3% more than that of FLANN.
第一章 緒論 1
1.1回顧 1
1.2研究動機與目的 2
1.3研究方法與內容 3
1.4本論文之貢獻 3
1.5各章提要 3
第二章 類神經網路於通信系統之應用 4
2.1類神經網路簡介 4
2.2類神經網路在通道等化器之應用 10
第三章 FLANN之通道等化器之設計 12
3.1 FLANN架構 12
3.2軟體模擬 13
3.2.1系統架構 13
3.2.2 LIN結構 14
3.2.3 模擬結果 15
3.2.3.1 計算複雜度之比較 16
3.2.3.2 MSE效能比較 17
3.2.3.3 SER效能比較 17
3.2.3.4 眼狀圖的比較 20
3.3硬體設計 21
3.3.1 2PAM通道等化器 22
3.3.1.1硬體結構 22
3.3.1.2 誤差效能比較 23
3.3.1.2.1時序圖 23
3.3.1.2.2 BER效能 23
3.3.1.2.3 學習曲線圖 25
3.3.1.3 硬體考量 25
3.3.2 4QAM通道等化器 27
3.3.2.1硬體結構 27
3.3.2.2 誤差效能比較 29
3.3.2.2.1時序圖 29
3.3.2.2.2 SER效能 31
3.3.2.2.3 學習曲線圖 32
3.3.2.3 硬體考量 32
第四章RDF-FLANN之通道等化器之設計 35
4.1 RDF-FLANN架構 35
4.2學習演算法之推導 36
4.3軟體模擬 37
4.3.1 計算複雜度之比較 38
4.3.2 MSE效能比較 39
4.3.3 SER效能比較 39
4.3.4 眼狀圖的比較 39
4.4硬體設計 43
4.4.1 2PAM通道等化器 43
4.4.1.1 硬體結構和需求量 44
4.4.1.2 誤差效能比較 46
4.4.1.2.1 BER效能 46
4.4.1.2.2 時序圖 46
4.4.2 4QAM通道等化器 47
4.4.2.1 硬體結構和需求量 47
4.4.2.2 誤差效能比較 49
4.4.2.2.1時序圖 49
4.4.2.2.2 SER效能 50
第五章 結論與未來研究方向 52
第六章 參考文獻 53
[1] S. Haykin, Communication Systems, 4th ed., New York: Wiley, 2001.
[2] A. Cichocki and R. Unbehauen, Neural Networks for Optimization and Signal Processing, New York: Wiley, 1993.
[3] M. T. Hagan, H. B. Demuth and M. Beale, Neural Network Design, Boston: PWS, 1996.
[4] R. Hecht-Nielsen, Neurocomputing, Reading, MA: Addison Wesley, 1990.
[5] B. Widrow and Lehr M. A., “30 years of adaptive neural networks: Perceptron, Madaline, and Backpropagation,” Proc. IEEE, vol. 78, no. 9, pp. 1415-1442, 1990.
[6] J. J. Hopfield, “Neural networks and physical systems with emergent collective computational properties,” Proc. NAS, vol. 79, pp. 2554-2558, 1982.
[7] R. Hecht-Nielsen, “Counterpropagation Networks,” IEEE ICNN (San Diego), vol. 2, pp. 19-32, 1987[b].
[8] D. S. Broomhead and Lowe D., “Multivariable functional interpolation and adaptive networks,” Complex Systems, vol. 2, pp. 321-335, 1988.
[9] S. Chen, G. J. Gibson and C. F. N. Cowan, “Adaptive channel equalization using polynomial-perceptron structure,” IEE Proceedings, vol. 137, Pt. I, no. 5, pp. 257-264, Oct. 1990.
[10] Y. H. Pao and Y. Takefuji, “Functional-link net computing: theory, system architecture, and functionalities,”Computer, Volume: 25 Issue: 5, pp. 76-79, May 1992.
[11] M. A. Hearst, S. T. Dumais, E. Osman, J. Platt and B. Scholkopf, “Support vector machines,” IEEE Intelligent Systems, Volume: 13 Issue: 4 , pp. 18-28, July 1998.
[12] B. Widrow, Adaptive Signal Processing, Englewood Cliffs, N. J., Prentice-Hall, 1985.
[13] S. Haykin, Adaptive Filter Theorys, 3th ed., N. J., Prentice-Hall, 1996.
[14] D. E. Rumelhart and J. L. McClelland, eds., Parallel Distributed Processing: Explorations in the Microstructure of Cognition, vol. 1, Cambridge, MA : MIT Press, 1986.
[15] D. C. Dracopoulos, Evolutionary Learning Algorithms for Neural Adaptive Control, N. Y., Springer, 1997.
[16] D. G. Luenberger, Optimization by Vector Space Methods, New York: Wiley, 1969.
[17] C. Charalambous, “Conjugate gradient algorithm for efficient training of artificial neural networks,” IEE Proceedings, vol. 139, no. 3, pp. 301-310, 1992.
[18] M. T. Hagan and M. Menhai, “Training feedforward networks with the Marquardt algorithm,” IEEE Trans. Neural Networks, vol. 5, no. 6, 1994.
[19] D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning representations by backpropogating errors,” Nature, vol. 323, pp. 533-536, 1986.
[20] M. Stinchcombe and H. White, “Universal approximation using feedforward networks with non-sigmoid hidden layer activation functions,” International Joint Conference on Neural Networks, 18-22 June 1989.
[21] G. J. Gibson and C. F. N. Cowan, “On the decision regions of multilayer perceptrons,” Proceedings of the IEEE , Volume: 78 , Issue: 10, pp. 1590-1594, Oct. 1990.
[22] R. Brad, I. Mihu and M. Breazu, “Statistical analysis of multilayer perceptrons performances ,” Proceedings of International Joint Conference on Neural Networks, vol. 4, pp. 2794-2799 ,15-19 July 2001.
[23] H. Czap, “Construction and interpretation of multi-layer-perceptrons,” 2001 IEEE International Conference on Systems, Man, and Cybernetics, vol. 5, pp. 3349-3354, 7-10 Oct. 2001.
[24] S. Siu, G. J. Gibson and C. F. N. Cowan, “Decision feedback equalization using neural network structures and performance comparison with standard architecture,” Proc. Inst. Elect. Eng., vol. 137, pt. 1, pp. 221-225, Aug. 1990.
[25] M. Meyer and G. Pfeiffer, “Multilayer perceptron based decision feedback equalisers for channels with intersymbol interference,” IEE Proceedings I, Communications, Speech and Vision, Volume: 140, Issue: 6, pp. 420-424, Dec. 1993.
[26] Huang R. C. and Chen M. S., “Adaptive equalization using complex-valued multilayered neural network based on the extended Kalman filter,” 5th International Conference on Signal Processing Proceedings, WCCC-ICSP 2000, Volume: 1, pp. 519-524, 21-25 Aug. 2000.
[27] D. Erdogmus, D. Rende, J.C. Principe and T.F. Wong, “Nonlinear channel equalization using multilayer perceptrons with information-theoretic criterion,” Proceedings of the 2001 IEEE Signal Processing Society Workshop on Neural Networks for Signal Processing XI, pp. 443-451, 10-12 Sept. 2001.
[28] A. Kantsila, T. Haverinen, M. Lehtokangas and J. Saarinen, “On equalization with complex MLP network in the GSM environment,” IFSA World Congress and 20th NAFIPS International Conference, vol. 5, pp. 2687-2692, 25-28 July 2001.
[29] S. U. H. Qureshi,“Adaptive Equalisation,”Proc. IEEE, vol. 73, no. 9, pp.1349-1387, Sept. 1985.
[30] Chen-Yi Lee, Hsuan-Yu Liu, Jyh-Neng Yang, Shuenn-Der Tzeng, Terng-Ren Hsu and Terng-Yin Hsu,“A MLP/BP-based equalizer for NRZ signal recovery in band-limited channels,”Proceedings of the 43rd IEEE Midwest Symposium on Circuits and Systems, vol. 3, pp. 1340 –1342, 2000.
[31] A. Zerguine, A. Shafi and M. Bettayeb, “Multilayer perceptron-based DFE with lattice structure,” IEEE Trans. Neural Networks, vol. 12, no. 3, pp. 532-545, May 2001.
[32] C. Lu, B. Shi and L. Chen, “BG: a hybrid algorithm for MLP,” The 2000 IEEE Asia-Pacific Conference on Circuits and Systems, pp. 360-363, Dec. 2000.
[33] F. Albu, A. Mateescu, J. C. M. Mota and B. Dorizzi, “Adaptive channel equalization using neural network,” SBT/IEEE International Telecommunications Symposium, Volume: 2, pp. 438 –441, Aug. 1998.
[34] Q. Gan, P. Saratchandran, N. Sundararajan and K. R. Subramanian, “A complex valued radial basis function network for equalization of fast time varying channels,” IEEE Trans. Neural Networks, Volume: 10, Issue: 4, pp. 958-960, July 1999.
[35] I. Cha and S. Kassam, “Channel equalization using adaptive complex radial basis function networks,” IEEE J. Select. Area Commun., vol. 13, pp. 122-131, Jan. 1995.
[36] S. H. Heo, S. K. Park and S. W. Nam, “Channel equalization for severe intersymbol interference and nonlinearity with a radial basis function neural network,” IJCNN ''99 International Joint Conference on Neural Networks, Volume: 6, pp. 3992-3995, July 1999.
[37] J. Lee, C. D. Beach and N. Tepedelenlioglu, “Channel equalization using radial basis function network,” IEEE International Conference on Neural Networks, Volume: 4, pp. 1924 –1928, June 1996.
[38] B. Mulgrew, “Applying radial basis functions,” IEEE Signal Processing Magazine, Volume: 13 Issue: 2, pp. 50 –65, March 1996.
[39] D. Jianping, N. Sundararajan and P. Saratchandran, “Communication channel equalization using complex–valued minimal radial basis function neural networks,” IEEE Trans. Neural Networks, vol. 13, no. 3, pp. 687-696, May 2002.
[40] T. Kohonen, “The self-organizing,” Proc. IEEE, vol. 78, no. 9, pp. 1464-1480, 1990.
[41] S. Chen, A. K. Samingan and L. Hanzo, “Optimal decision feedback equaliser for M-PAM signals using support vector machine solution,” Electronics Letters, Volume: 36 Issue: 20, pp. 1742-1744, Sep. 2000.
[42] D. J. Sebald and J. A.Bucklew, “Support vector machine techniques for nonlinear equalization,” IEEE Transactions on Signal Processing, Volume: 48 Issue: 11, pp. 3217 –3226, Nov. 2000.
[43] F. Schwenker, “Hierarchical support vector machines for multi-class pattern recognition,” Fourth International Conference on Knowledge-Based Intelligent Engineering Systems and Allied Technologies, Volume: 2, pp. 561 -565 Aug. 2000.
[44] J. A. K. Suykens, “Nonlinear modelling and support vector machines,” IEEE Instrumentation and Measurement Technology Conference, Volume: 1, pp. 287 –294, May 2001.
[45] Choi S. and Hong D., “Equalization using the bilinear recursive polynomial perceptron with decision feedback,” Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks, Volume: 5, pp. 366-71, July 2000.
[46] R. G. Bartle and D. R. Sherbert, Introduction to Real Analysis, 3nd ed., New York: Wiley, 2000.
[47] K. Hornik, M. Stinchcombe, and H. White,“Multilayer feedforward networks are universal approximators,”Neural Networks, vol. 2, no. 5, pp. 359-366, 1989.
[48] W. S. Gan, J. J. Soraghan, and T. S. Durrani,“New Functional-link based Equalizer,”Electronics Letters, vol. 28, no. 17, pp. 1643-1645, Aug. 1992.
[49] J. C. Patra, R. N. Pal, B. N. Chatterji and G. Panda, “Identification of nonlinear dynamic systems using functional link artificial neural networks,” IEEE Trans. Syst., Man, Cybern. B., vol. 29, no. 2, pp. 254-262, Apr. 1999.
[50] J. C. Patra, R. N. Pal, R. Baliarsingh and G. Panda, “Nonlinear channel equalization for QAM signal constellation using artificial neural networks,” IEEE Trans. Syst., Man, Cybern. B., vol. 29, no. 2, pp. 262-271, Apr. 1999.
[51] N. Sadegh, “A perceptron based neural networks for identification and control of nonlinear systems,” IEEE Trans. Neural Networks, vol. 4, no. 6, pp. 982-988, Nov. 1993.
[52] C. You and D. Hong, “ Adaptive equalization using the complex backpropagation algorithm,” IEEE International Conference on Neural Networks, vol. 4, pp. 2136 –2141, Jun. 1996.
[53] J. G. Proakis, Digital Communications, 3rd ed., New York: McGraw-Hill, 1995.
[54] J. de Villiers and E. Barnard, “Backpropagation neural nets with one and two hidden layers,” IEEE Trans. Neural Networks, vol. 4, no. 1, pp. 136-141, Jan. 1992.
[55] http://www.mathworks.com.
[56] C. M. Wu, R. T. Wang and C. T. Yen, “Adder-based SIMD-systolic architectures and VLSI chip for computing the Winograd small FFT algorithms,” Int. J. Electronics, vol. 76, no. 6, pp. 1135-1149, 1994.
[57] L. Chen and B. Shi, “CMOS PWM VLSI implementation of neural network,” Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks, Volume: 3, pp. 485-488, July 2000.
[58] Y. Xu, Z. Li, Ruan Ming, H. Luo and W. Song, “VLSI design and implementation of WCDMA channel decoder,” Canadian Conference on Electrical and Computer Engineering, Volume: 1, pp. 241-245, May 2001.
[59] K. Chadha and J. R. Cavallaro, “A reconfigurable Viterbi decoder architecture,” Conference Record of the Thirty-Fifth Asilomar Conference on Signals, Systems and Computers, Volume: 1, pp. 66-71, Nov. 2001.
[60] R. Cambio and D. C. Hendry, “Low-power digital neuron for SOM implementations,” Electronics Letters, Volume: 39, Issue: 5, pp. 448-450, March 2003.
[61] D. P. Sharma and J. Singh, “Design and implementation of emission filter for soft modem on TM9320C50 DSP chip,” 2002 Asia-Pacific Conference on Circuits and Systems, Volume: 2, pp. 499-503, Oct. 2002.
[62] C. W. Murphy and D. M. Harvey, “Reconfigurable hardware implementation of BinDCT,” Electronics Letters, Volume: 38, Issue: 18, pp. 1012-1013, Aug. 2002.
[63] L. Wu, Z. Yen, M. Zhou, J. Zhang and L. Zheng, “An approach to TMS320C50 high-speed data processing and transmission system,” 1998 Fourth International Conference on Signal Processing Proceedings, vol.1, pp. 53-56, Oct. 1998.
[64] R. H. Yang, S. B. Chiah and W. Y. Chan, “Design and implementation of a digital audio tone control unit using an efficient FIR filter structure,” IEEE TENCON. Digital Signal Processing Applications, Volume: 1, pp. 273-277, Nov. 1996.
[65] J. Singh, A. Singh, D. Sharma, H. P. Singh and S. S. Bhatti, “Soft computing in FSK modem using DSP,” 2000 IEEE International Conference on Systems, Man, and Cybernetics, Volume: 4, pp. 2333-2337, Oct. 2000.
[66] W. D. Weng and W. P. Yang, “The CPLD implementation of Viterbi algorithm in grand alliance ATSC systems,” IEEE Transactions on Industrial Electronics, Volume: 48, Issue: 5, pp. 898-903, Oct. 2001.
[67] J. Wang and Z. Dong, “An FPGA approach of channel decoder for DAB test receiver,” 2nd International Conference on ASIC, pp. 214-217, Oct. 1996.
[68] B. Pandita and S. K. Roy, “Design and implementation of a Viterbi decoder using FPGAs,” Twelfth International Conference on VLSI Design, pp. 611-614, Jan. 1999.
[69] E. Lord, M. Devlin, N. Harold and C. Sanderson, “Accelerating BER analysis using an FPGA based processing platform,” Proceedings MILCOM 2002, Volume: 1, pp. 218-223, Oct. 2002.
[70] T. H. Lay, Y. H. Lee, L. C. Lo, C. Y. Leu and P. M. Gong, “Design and implementation of channel encoder/decoder for the Digital Audio Broadcasting (DAB) system,” Fifth Asia-Pacific Conference on Communications, Volume: 1, pp. 680-683, Oct. 1999.
[71] 顏景池,翁萬德 “類神經網路通道等化器的設計與實現,” The Seventh Conference on Artificial Intelligence and Applications, Taichung, Taiwan, pp. 789-794, Nov. 2002.
[72] http://www.altera.com.
[73] C. T. Yen, W. D. Weng, and Y. T. Lin, “FPGA realization of a neural network based nonlinear channel equalizer,” IEEE Transactions on Industrial Electronics, vol. 51, no. 2, pp. 472-479, Apr. 2004.
[74] W. D. Weng and C. T. Yen, “A reduced decision feedback FLANN based nonlinear channel equalizer for digital communication systems,” IEE Proceedings-Communication (to be published).
[75] W. D. Weng and C. T. Yen, “An improved soft feedback FLANN channel equalizer,” submitted to Telecommunication Systems.
[76] A. Hussain, J. J. Soraghan and T. S. Durrani, “A new adaptive functional-link neural-network-based DFE for overcoming co-channel interference,” IEEE Transactions on Communications, vol. 45, no. 11, Nov. 1997.
[77] B. C. Kuo, Automatic Control Systems, 7th ed., Englewood Cliffs, NJ: Prentice-Hill, 1995.
[78] R. J. Schalkoff, Artificial Neural Networks, New york: McGraw-Hill, 1997.
[79] 顏景池,翁萬德 “以回授函數連結類神經網路實現通道等化器,” 2003 Workshop on Consumer Electronics, Tainan, Taiwan, pp. 42, Nov. 2003.
[80] W. D. Weng and C. T. Yen, “Realization of a reduced decision feedback FLANN based nonlinear channel equalizer,” Journal of the Chinese Institute of Electrical Engineering (accepted).
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top