臺灣博碩士論文加值系統

本論文提出兩種簡單準確的通用公式，主要是用來鑑定正交分頻多工系統中的非線性通道。正交分頻多工系統是一種被廣泛採用的寬頻通訊應用技術。正交分頻多工系統訊號很容易受到通訊網路中非線性通道的影響。使得正確估計OFDM 系統非線性通道的任務變得很重要。非線性通道通常是由VolterraSeries 來表示，非線性通道鑑定的任務就是確定Volterra Kernel。當非線性系統階數增加時，這項任務會變得非常困難。本論文，論述OFDM 信號的高階統計特性和多層分解特性，以克服遇到的瓶頸並推導出解決Volterra Kernel 鑑定的通用方法。這兩種方法各有優缺點。OFDM 信號的高階統計特性法是利用高階自動矩為頻域Volterra Kernel 提出閉式表達式。使用高階統計特性這個方法可以推導出上至5 階非線性通道之簡單公式。多層分解的方法是透過設計具有時域或頻域頻譜凹陷的不同輸入信號來激發非線性系統，這種方法理論上可適用於任意階之非線性通道，且不限於OFDM 系統，但會產生較大的計算複雜度。基於這兩種方法，本論文推導準確簡單的方法來鑑定高階非線性正交分頻多工系統系統Volterra Kernel 的通用公式。理論分析和模擬結果顯示，所得到的解決方法都達到了最小均方誤差。

The dissertation proposes two general formulas that are simple to derive and applicable to the identification of nonlinear channels in orthogonal frequency division multiplexing (OFDM) systems. OFDM is a widely used wideband communication application technology. As nonlinearities in the communication network can easily impact OFDM signals, the task of correctly estimating the nonlinear channel of an OFDM system is of critical importance. Since the nonlinear channel is usually represented by the Volterra series, the task in nonlinear channel identification is to find and solve for the Volterra kernels. However, this task becomes very difficult when the order of the nonlinear system increases. This dissertation discusses the high-order statistical properties of OFDM signals and multilayer decomposition properties, to overcome related bottlenecks and derive a general method for Volterra kernel identification, respectively. Both methods have their own advantages and disadvantages. The high-order statistics method of OFDM use the high-order moments to propose a closed-form expression for the frequency domain Volterra kernels. This method of high-order statistics can enable the derivation of simple formulas for nonlinear channels up to 5th order. In contrast, the multilayer decomposition method excites the nonlinear system by designing different input signals with time-domain or frequency-domain spectral notches, which theoretically can be applied to nonlinear channels of any order. This method is not limited to ODFM systems. However, it does produce greater computational complexity. By drawing upon these two methods, we derive an accurate and simple method to identify Volterra kernels of high-order nonlinear OFDM systems. The theoretical and simulated results show that the proposed solutions achieve the minimum mean square error (MMSE).

Contents
1 Introduction 1
1.1 Research motivation and purpose . . . . . . . . . . . . . . . . . 1
1.2 Literature Discussion . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Solution . . . . . . . . . . . . . .. . . . . . . . . . . . . . . 4
1.4 Dissertation Outline . . . . . . . . . . . . . . . . . . . .. . . 5
2 Concept Discussion 7
2.1 Baseband Signal and Passband Signal . . . . . . . . . . . . . . . 7
2.1.1 Baseband signal . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.2 Passband signal . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 Nonlinear Channel and Volterra Series . . . . . . . . . . . . . . 8
2.2.1 Volterra series for baseband channel . . . . . . . . . . . . . .9
2.2.2 Volterra series for bandpass channel . . . . . . . . . . . . . .11
2.2.3 Symmetry Properties of Volterra Kernels . . . . . . . . . . . . 14
2.3 The MMSE Approach for Volterra Kernel Estimation . . . . . . . . .16
2.3.1 Time-domain kernel estimation for real nonlinear systems . . . .16
2.3.2 Time-domain kernel estimation for complex nonlinear systems . . 19
2.3.3 Frequency-domain kernel estimation for real nonlinear systems . 22
2.3.4 Frequency-domain kernel estimation for complex nonlinear. . . . 24
3 Identification of Nonlinear Channels with OFDM Inputs . . . . . . . 27
3.1 OFDM Signal and Its Spectral Properties . . . . . . . . . . . . . 27
3.1.1 Higher-order auto-moment spectra . . . . . . . . . . . . . . . .30
3.2 Identification of Volterra Kernels for Nonlinear Bandpass Channels 31
3.3 Derivation of the Formula for 5th-Order Nonlinear OFDM Channels. . 38
3.4 Design of Pseudo Random Test Sequences . . . . . . . . . . . . . . 53
4 Identification by Decomposition . . . . . . . . . . . . . . . . . . .56
4.1 Time-Domain Decomposition . . . . . . . . . . . . . . . . . . . . .56
4.2 Frequency-Domain Decomposition . . . . . . . . . . . . . . . . . . 61
5 Multilayer Decomposition . . . . . . . . . . . . . . . . . . .. . . 66
5.1 An Additional Layer of Decomposition . . . . . . . . . . . . . . . 66
5.2 Multiple Layers of Decomposition . . . . . . . . . . . . . . . . . 71
6 Computer Simulations . . . . . . . . . . . . . . . . . . . . . . . . 73
6.1 Verification of the method derived in Chapter 3 . . . . . . . . . 73
6.2 Verification of the method derived in Chapter 4 . . . . . . .. . . 78
7 Conclusion 86

List of Figures
2.1 The K-th order frequency-domain Volterra model of a nonlinear baseband
system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 Frequency-domain Volterra model of a nonlinear bandpass channel. . . 14
2.3 The estimated area of the quadratic Volterra kernel in the time-domain. . 17
2.4 The estimated area of the cubic Volterra kernel in the time-domain. . . . 18
2.5 (a) The input x(n) and its output y(n) with a maximal amplitude A.
(b) The notched input x(¯p)(n) and its output y(¯p)(n). (c) The difference
signal y(p)(n) between y(n) and y(¯p)(n). . . . . . . . . . . . . . . . . . 20
2.6 Baseband OFDM system block diagram. . . . . . . . . . . . . . . . . . 23
2.7 Block diagram of passband OFDM system. . . . . . . . . . . . . . . . 25
3.1 The 16-QAM constellation is divided into 4 subsets of 4-QAMs with
different graphic symbols. 16-QAM is used for the OFDM subcarrier
in the simulation. Each data symbol is marked with its 4-digit representation
and the symbol number in parentheses. . . . . . . . . . . . . . . 29
3.2 The feedback shift register corresponds to the original polynomial p(x)
in (151) and is used to generate the test sequence. . . . . . . . . . . . . 55
4.1 (a) A hypothetical spectrum of x(n). (b) The spectrum of x(¯p)(n). . . . 62
4.2 (a) A hypothetical output spectrum, Y (m), for the inputX(m) in Fig. 4.1(a).
(b) A hypothetical output spectrum Y (¯p)(m) for the input X(¯p)(m) in
Fig. 4.1(b). (c) The difference spectrum Y (p)(m) between Y (m) and
Y (¯p)(m). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.1 Multilayer decomposition diagram. . . . . . . . . . . . . . . . . . . . . 70
6.1 The shift register corresponds to the original polynomial p(x) = x20 +
x3 + 1 in (189). The 20-bit sequence in the sliding window represents
the initial state of the shift register. . . . . . . . . . . . . . . . . . . . . 75
6.2 Simulation of the power spectrum of the input and output signals of a
nonlinear satellite channel. . . . . . . . . . . . . . . . . . . . . . . . . 77
6.3 Estimated NMSE and SNR of linear, cubic, 5th-order, and total frequencydomain
Volterra kernels. . . . . . . . . . . . . . . . . . . . . . . . . . 79
6.4 Estimation of the NMSEs of the time-domain Volterra kernel obtained
by the T-FEX (denoted by the symbol ‘2’) and T-NEX (denoted by the
symbol ‘◦’) methods at SNRs of 20, 30, and 40 dB with different numbers
of data frames. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.5 Actual time-domain linear Volterra kernel (T-ACT) and its estimated
values obtained by the T-FEX and T-NEX methods. . . . . . . . . . . . 81
6.6 (a) Actual time-domain quadratic Volterra kernel and its estimated value
obtained by (b) T-FEX and (c) T-NEX methods. (d) Actual time-domain
cubic Volterra kernel and its estimated values obtained by (e) T-FEX and
(f) T-NEX methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
6.7 F-FEX (denoted by the symbol ‘2’) and F-NEX (denoted by the symbol
‘◦’) methods use different numbers of data frames to estimate the
NMSEs of the Volterra kernels in the frequency domain at an SNR of
20, 30, and 40 dB. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
6.8 The amplitude of the actual frequency domain linear Volterra kernel (FACT)
and its estimated value obtained by the F-FEX and F-NEX methods. 84
6.9 (a) The amplitude of the actual frequency-domain quadratic Volterra
kernel and its estimated values obtained by (b) F-FEX and (c) F-NEX
methods. (d) The actual frequency-domain cubic Volterra kernel amplitude
obtained by (e) F-FEX and (f) F-NEX methods. . . . . . . . . . . 85

List of Tables
6.1 Time-domain Volterra kernels of a satellite communication system. . . . 74
6.2 States of the shift register in Fig. 6.1. . . . . . . . . . . . . . . . . . . . 75
6.3 Corresponding 16-QAM symbols for the output sequence in Table 6.2. . 76
6.4 NMSEs of the linear, cubic, 5th-order, and total frequency-domain Volterra
kernels as realized by applying the proposed method under various SNRs. 78

References
[1] A. R. S. Bahai, B. R. Saltzberg, and M. Ergen, Multi-Carrier Digital Communications
Theory and Applications of OFDM, 2nd ed., New York, NY: Springer, 2004.
[2] ITU-T, “Asymmetric digital subscriber line (ADSL) transceivers,” ITU-T Recommendation G.992.1, Geneva, 1999.
[3] S. Galli and O. Logvinov, “Recent developments in the standardization of power
line communications within the IEEE,” IEEE Commun. Mag., vol. 46, no. 7, pp. 64-71, Jul. 2008.
[4] Y. Li, M. Zhang, W. Zhu, M. Cheng, C. Zhou and Y. Wu, ”Performance evaluation for medium voltage MIMO-OFDM power line communication system,” in China Communications, vol. 17, no. 1, pp. 151-162, Jan. 2020, doi: 10.23919/ JCC.2020.01.012.
[5] IEEE, “Wireless LAM medium access control (MAC) and physical layer (PHY) specifications: High-speed physical layer in the 5 GHz band,” IEEE Standard 802.11a-1999, Sep. 1999.
[6] S. P. Stapleton and F. C. Costescu, “An adaptive predistortion for a power amplifier based on adjacent channel emissions,” IEEE Trans. Veh. Technol., vol. 41, no. 1, pp. 49-56, Jan. 1992.
[7] I. Iofedov and D. Wulich, “MIMO-OFDM with nonlinear power amplifiers,” IEEE Trans. Commun., vol. 63, no. 12, pp. 4894-4904, Dec. 2015.
[8] A. Mohammed, M. Shehata, H. Mostafa and A. Nassar, ”Peak-to-Average Power Ratio Suppression using Companding schemes in OFDM Systems,” 2018 IEEE 61st International Midwest Symposium on Circuits and Systems (MWSCAS), 2018, pp. 933-936, doi: 10.1109/MWSCAS.2018.8623875.
[9] Xingjian He, Fengjun Yang and Ran Xi, ”Peak-to-average power ratio reduction in OFDM signals via self-adaptive EVM method,” 2016 IEEE Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC), 2016, pp. 676-679, doi: 10.1109/IMCEC.2016.7867295.
[10] Qijia Liu, R. J. Baxley and G. Tong Zhou, ”Free subcarrier optimization for peakto- average power ratio minimization in OFDM systems,” 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, 2008, pp. 3073-3076, doi: 10.1109/ICASSP.2008.4518299.
[11] R. Deshpande and D. J. Pete, ”Optimized Peak to Average Power Ratio (PAPR) Reduction Technique for OFDM,” 2017 International Conference on Computing, Communication, Control and Automation (ICCUBEA), 2017, pp. 1-5, doi: 10.1109/ICCUBEA.2017.8463796.
[12] A. J. Redfern and G. T. Zhou, “Nonlinear channel identification and equalization for OFDM systems,” Proc. IEEE Int. Conf. Acoust. Speech, Signal Process., vol. 6, pp. 3521-3524, May 12-15, 1998.
[13] A. F. Sheta and A. H. Abel-Wahab, ”Identification of nonlinear communication channel using evolutionary Volterra time-series,” Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406), 1999, pp. 229-235 Vol. 1, doi: 10.1109/CEC.1999.781930.
[14] W. Qiang, Z. Jiashu and Y. Jing, ”Identification of Nonlinear Communication Channel Using an Novel Particle Swarm Optimization Technique,” 2008 Inter-ational Conference on Computer Science and Software Engineering, 2008, pp. 1162-1165, doi: 10.1109/CSSE.2008.1228.
[15] Fang Yangwang and Jiao Licheng, ”Blind identification of nonlinear FIR Volterra channels,” WCC 2000 - ICSP 2000. 2000 5th International Conference on Signal Processing Proceedings. 16th World Computer Congress 2000, 2000, pp. 294-297 vol.1, doi: 10.1109/ICOSP.2000.894494.
[16] C.-H. Tseng and E. J. Powers, “Identification of nonlinear channels in digital transmission systems,” Proc. IEEE Signal Processing Workshop on Higher-Order Statistics, South Lake Tahoe, California, June, 1993, pp. 42-45.
[17] C.-H. Tseng, “Identification of nonlinear systems driven by random multisine signals,” Proc. IASTED Int. Conf. Signal and Image Process., Honolulu, Hawaii,
USA, Aug. 20-22, 2007.
[18] K. G. Gard, H. M. Gutierrez, and M. Steer, “Characterization of spectral regrowth in microwave amplifiers based on the nonlinear transformation of a complex Gaussian process,” IEEE Trans. Microw. Theory Tech., vol. 47, no. 7, pp. 1059-1069, July 1999.
[19] G. T. Zhou, H. Qian, L. Ding, and R. Raich, “On the baseband representation of a bandpass nonlinearity,” IEEE Trans. Signal Process., vol. 53, no. 8, pp. 2953-2957, Aug. 2005.
[20] G. T. Zhou and J. S. Kenney, “Predicting spectral regrowth of nonlinear power amplifiers,” IEEE Trans. Commun., vol. 50, no. 5, pp. 718-722, May 2002.
[21] Y. Shen and J. L. Tauritz, ”On the analysis of quasistatic bandpass nonlinearities using Volterra series,” 60th ARFTG Conference Digest, Fall 2002., 2002, pp. 59-66, doi: 10.1109/ARFTGF.2002.1218686.
[22] S. Benedetto and E. Biglieri, Principles of Digital Transmission with Wireless Applications, Norwell, MA: Kluwer, July 1999.
[23] B. Fehri and S. Boumaiza, “Baseband equivalent Volterra series for digital predistortion of dual-band power amplifiers,” IEEE Trans. on Microwave Theory and Techniques, vol. 62, no. 3, pp. 700-714, Mar. 2014.
[24] C. Tsai et al., “Incoherent laser heterodyned long-reach 60-GHz MMWoF link with Volterra filtered 16-QAM OFDM beyond 13 Gbps,” IEEE Journal of Selected Topics in Quantum Electronics, vol. 27, no. 2, pp. 1-11, Mar.-Apr. 2021.
[25] A. Hamza et al.,“A code-domain, in-band, full-duplex wireless communication link with greater than 100-dB rejection,” IEEE Tran. Microwave Theory and Techniques, vol.‘69, no. 1, pp. 955-968, Jan. 2021.
[26] C. Lee, Y. Lin and W. Lin, “Investigation of linearity in the high electric field region for SiGe HBTs based on Volterra series,” IEEE Transactions on Device and Materials Reliability, vol. 14, no. 4, pp. 1049-1055, Dec. 2014.
[27] X. Y. Z. Xiong, L. J. Jiang, J. E. Schutt-Aine and W. C. Chew, “Volterra seriesbased time-domain macromodeling of nonlinear circuits,” IEEE Transactions on Components, Packaging and Manufacturing Technology, vol. 7, no. 1, pp. 39-49, Jan. 2017.
[28] W. J. Rugh, Nonlinear System Theory - The Volterra/Wiener Approach, Baltimore, MD: Johns Hopkins University Press, 1981.
[29] T. Koh and E. J. Powers, “Second-order Volterra filtering and its application to nonlinear system identification,” IEEE Trans. Acoust., Speech, Signal Processing, vol. 33, no. 6, pp. 1445-1455, Dec. 1985.
[30] P. Koukoulas and N. Kalouptsidis, “Nonlinear system identification using Gaussian inputs,” IEEE Trans. Signal Process., vol. 43, no. 8, pp. 1831-1841, Aug. 1995.
[31] T. C. Hsia, System Identification: Least-Squares Methods, Massachusetts: Lexington Books, 1977.
[32] V. J. Mathews and G. L. Sicuranza, Polynomial Signal Processing, New York: John Wiley & Sons, 2000.
[33] H. Zhao, J. Wang and X. Li, “Modeling of low noise amplifier based on Volterra series with recursive least square algorithm,” International Applied Computational Electromagnetics Society Symposium (ACES), Suzhou, China, 2017, pp. 1-2.
[34] M. Usman, M. S. Ibrahim, J. Ahmed, S. S. Hussain and M. Moinuddin, “Quantum calculus-based Volterra LMS for nonlinear channel estimation,” International Conference on Latest Trends in Electrical Engineering and Computing Technologies (INTELLECT), Karachi, Pakistan, 2019, pp. 1-4.
[35] C. Evans, D. Rees, L. Jones, and M. Weiss, “Periodic signals for measuring nonlinear Volterra kernels,” IEEE Trans. Instrum. Meas., vol. 45, no. 2, pp. 362-371, April 1996.
[36] J. S. Bendat, Nonlinear System Analysis and Identification From Random Data, New York: Wiley, 1990.
[37] S. W. Nam and E. J. Powers, “Application of higher order spectral analysis to cubically nonlinear system identification,” IEEE Trans. Signal Processing, vol. 42, no. 7, July 1994.
[38] C.-H. Tseng, “Identification of cubically nonlinear systems using undersampled data,” IEE Proc.-Vis. Image Signal Process., vol. 144, no. 5, pp. 267-277, Oct. 1997.
[39] P. J. Lawrence, “Estimation of the Volterra functional series of a nonlinear system using frequency response data,” IEEE Proc. D, vol. 128, no. 5, pp. 206-210, Sept. 1981.
[40] S. Boyd, Y. S. Tang, and L. O. Chua, “Measuring Volterra kernels,” IEEE Trans. Circuits and Systems, vol. 30, no. 8, pp. 571-577, Aug. 1983.
[41] L. O. Chua and Y. Liao, “Measuring Volterra kernels (II),”Int. J. Circuit Theory, Applicat., vol. 17, pp. 151-190, 1989.
[42] Z. A. Khan, E. Zenteno, P. Handel and M. Isaksson, “Extraction of the third-order 3×3 MIMO Volterra kernel outputs using multitone signals,” IEEE Trans. Microw. Theory Techn., vol. 66, no. 11, pp. 4985-4999, Nov. 2018.
[43] ETSI, “Digital audio broadcasting (DAB), guidelines and rules for implementation and operation, part 1: system outline,” ETSI Technical Report, TR 101 496-1 v1.1.1, Nov. 2000.
[44] ETSI, “Digital video broadcasting (DVB), framing structure, channel coding and modulation for digital terrestrial television,” ETSI European Standard, EN 300 744 v1.2.1, Jul. 1999.
[45] C.-H. Cheng and E. J. Powers, “Optimal Volterra kernel estimation algorithms for a nonlinear communication system for PSK and QAM inputs,” IEEE Trans. Signal Processing, vol. 49, no. 1, pp. 147-163, Jan. 2001.
[46] C.-H. Tseng and E. J. Powers, “Identification of cubic systems using higher order moments of I.I.D. signals,” IEEE Trans. Signal Process., vol. 43, no. 7, pp. 1733- 1735, Jul. 1995.
[47] G. Mileounis, P. Koukoulas, and N. Kalouptsidis, “Input-output identification of nonlinear channels using PSK, QAM, and OFDM inputs,” Signal Process., vol. 89, no. 7, pp. 1359-1369, July 2009.
[48] C.-H. Tseng, “Estimation of cubic nonlinear bandpass channels in orthogonal frequency-division multiplexing systems,” IEEE Trans. Commun., vol. 58, no. 5, pp. 1415-1425, May 2010.
[49] J.-H. Cheng, Y.-H. Lin and C.-H. Tseng, “Identification of Volterra kernels for nonlinear communication systems with OFDM inputs,” International Conference on Systems and Informatics (ICSAI 2014), Shanghai, China, 2014, pp. 860-865.
[50] J. C. Pedro and N. B. Carvalho, “Designing multisine excitations for nonlinear model testing,” IEEE Trans. Microwave Theory and Techniques, vol. 53, no. 1, pp. 45-54, Jan. 2005.
[51] J. Zhang et al., “Decision-feedback frequency-domain Volterra nonlinear equalizer for IM/DD OFDM long-reach PON,” Journal of Lightwave Technology, vol. 37, no. 13, pp. 3333-3342, July 1, 2019.
[52] W. Chung, “Channel estimation methods based on Volterra kernels for MLSD in optical communication systems,” IEEE Photon. Technol. Lett., vol.2 2, no. 4, pp. 224-226, Feb. 2010.
[53] D. R. Morgan et al., “A generalized memory polynomial model for digital predistortion of RF power amplifiers,” IEEE Trans. Signal Process., vol. 54, no. 10, pp. 3852-3860, Oct. 2006.
[54] J. Reina-Tosina et al., “Behavioral modeling and predistortion of power amplifiers under sparsity hypothesis,” IEEE Trans. Microw. Theory Techn., vol. 63, no. 2, pp. 745-753, Feb. 2015.
[55] J. A. Becerra, M. J. M. Ayora, J. Reina-Tosina and C. Crespo-Cadenas, “Sparse identification of Volterra models for power amplifiers without pseudoinverse computation,” IEEE Trans. Microw. Theory Techn., vol. 68, no. 11, pp. 4570-4578, Nov. 2020.
[56] J. -H. Cheng and C. -H. Tseng, “Identification of Nonlinear Channels in Bandpass Communication Systems with OFDM Inputs,” Journal of Marine Science and Technology, vol. 29, Iss. 4.
[57] S. C. Cripps, RF Power Amplifiers for Wireless Communications, Norwood, MA: Artech House, 1999.
[58] M. Schetzen, The Volterra and Wiener Theories of Nonlinear System. New York: Wiley, 1980.
[59] J. Tsimbinos and K. V. Lever, “Input Nyquist sampling suffices to identify and compensate nonlinear systems,” IEEE Trans. Signal Processing, vol. 46, no. 10, pp. 2833-2837, Oct. 1998.
[60] W. A. Frank, “Sampling requirements for Volterra system identification,” IEEE Signal Processing Letters, vol. 3, no. 9, pp. 266-268, Sept. 1996.
[61] A. Papoulis, Probability, Random Variables, and Stochastic Processes, New York: McGraw-Hill, 1984.
[62] G. H. Golub and C. F. Van Loan, Matrix Computations, 2nd ed., Baltimore, MD: Johns Hopkins University Press, 1989.
[63] C.-H. Tseng, “A mixed-domain method for identification of quadratically nonlinear systems,” IEEE Trans. Signal Process., vol. 45, no. 4, pp. 1013-1024, April 1997.
[64] C. -H. Tseng and J. -H. Cheng, ”Nonlinear Channel Estimation for Minimally Sampled Baseband OFDM Systems,” IEEE Transactions on Communications, doi: 10.1109/TCOMM.2021.3104031.
[65] G. T. Zhou and G. B. Giannakis, “Nonlinear channel identification and performance analysis with PSK inputs,” Proc. IEEE Signal Proc. Advances for Wireless Commun., vol. 1, pp. 337-340, Paris, France, Apr. 16-18, 1997.
[66] S. Haykin, Adaptive Filter Theory, 2nd ed., Englewood Cliffs, New Jersey: Prentice-Hall, 1991.
[67] D. R. Brillinger, “Identification of polynomial systems by means of higher-order spectra,” J. Sound, Vibration, vol. 12, pp. 301-313, 1970.
[68] F. J. Macwilliams and N. J. A. Sloane, “Pseudo-random sequences and arrays,” IEEE Proc. vol. 64, no. 12, pp. 1715-1729, Dec. 1976.
[69] R. E. Blahut, Theory and Practice of Error Control Codes, Reading, MA: Addison- Wesley, 1983.
[70] D. M. Burton, Elementary Number Theory, 4th ed., Boston, MA: Allyn and Bacon, 1989.
[71] C.-H. Tseng, “Characterization of nonlinear channels in OFDM systems using signals with spectral notches,” Proc. IEEE Int. Symp. Power Line Commun. and Its Applications, pp. 249-254, Rio de Janeiro, Brazil, March 28-31, 2010.
[72] R. B. J. T. Allenby and A. Slomson, How to Count: An Introduction to Combinatorics, 2nd ed., Boca Rton, FL: CRC Press, 2011.
[73] A. A. M. Saleh, “Frequency-independent and frequency-dependent nonlinear models of TWT amplifiers,” IEEE Trans. Coommun., vol. COMM-29, pp. 1715-1720, Nov. 1981.