The author has granted a nonexclusive licence dowing the National Lihary of Canada to reproduce, loan, distri'bute or sell copies of this thesis in microform, paper or electronic formats.
L'auteur a accord une licence non exclusive pemettant a la Bibliothque nationale du Canada de reproduire, prter, distribuer ou vendre des copies de cette thse sous la forme de microfiche/film, de reproduction sur papier ou sur format 1ectronique.
L'auteur conserve la proprit du droit d'auteur qui protge cette thse. Ni la thse ni des extraits substantiels de celle-ci ne doivent tre imprimes ou autrement reproduits sans son autorisation.
The author retains ownership of the copyright in this thesis. Neither the thesis nor substantial extracts fiom it may be printed or othenvise reproduced without the author's permission.
This thesis is about the simulation of optical fiber communication systems with the effect of PMD (Polarization Mode Dispersion) and the application of adaptive channel equalizers, CPI (Cross Polarization Interference) cancellers and echo cancellers which are used to mitigate PMD.
The basic structure of this system is that two QAM signals are transmitted through
a long optical fiber with PMD and they are received by a coherent receiver which consists
of adaptive electrical channel equalizers and adaptive electrical CPI cancellers. The echo in a bidirectional transmission system is studied and adaptive echo cancellers are used to cancel it. The objective of this thesis is to create the system structure, to simulate the optical
to fiber chamel with PMD, simulate the adaptive channel equalizers, CPI cancellers and
echo cancellers and to evaluate the systern performance.
iii
ACKNOWLEDGEMENTS
I would iike to express my profound gratitude and high regards to professor Brent R Petersen. His encouragement, fiiendship and suggestions during the course of this research have played a vital role.
1also would like to express the deepest feeling of gratitude to my parents, Yong
Wei Lu and Zhao Yun Ding for their support and love.
...
2.1 System Mode1 ................................................................................................. 7 2.2 Baseband Simulation and Coherent Demodulation ....................................... 13 2.3 Baseband System Mode1 ................................................................................ 15 2.3 System Development Tools: UL\TLAB%~ SIMULINK! ......................... 18
22 3.2 Principal States of Polarization ................................................................... 3.3 Waveplate Mode1 ........................................................................................ 26 . . .......................................... 28 3 -4 Cornputer Simulation of PMD .................... 3.5 Channel Simulation using MATLAB~ SIMULINK? ........................... 31 and
3.1.1 Birefringence and PMD ............................................................. 19 3.1.2 Mode Coupling in Long Fiber ....................... .......................21
. . .
Chapter 4 Adaptive Channel Equalizers and CPI Cancellers ............... 33 3 4.1 Adaptive Channel Equalization Technique ............................................ 3 4.1.1 Fundarnentals of Equalization ...................... . . ...................... 33 4.1.2 Adaptive Equalization ................... ................-............................... 35
4.2 Channel Equalization and CPI Cancellation ................................................. 39 4.3 Simulation of Channel Equalizea and CPI Cancellers .................................46 4.3.1 Simulation Structure of Channel Equalizers and CPI Cancellers ....46 4.3.2 Performance of Channel Equalizm and CPI Cancellea .................48
6.2 Echo Cancellation ......................................................................................... 63 6.3 Power Budget .............................................................................................. 64 6.3.1 Pcwer Budget of the System without Echo Cancellers .................. 65 6.3.2 Power Budget of the System with Echo Cancellen ....................... 66 6.4 Simulation of Echo Cancellen Using SIMULMK" .................................... 67
70 71
References ...................................................................................................................
LIST OF TABLES
Table 5.1 Table 5.2 Table 5.3 Table 6.1 Table A.1 Table A.2 Table A.3 Table A.4 Table A.5 Table A.6 Table A.7 Table A.8 Table A.9 Table A .10 Table A .1 1 Table A.12
J,, (dB) when DGD is 500 ps ................................................................. 56 WD 1515 senes: 153811558nrn Bidirectional Coupler ............................ 61 . Parameters of System Model ............................................................... 7 5 Parameten of Transrnitter-A/I3 ............................................................... 76 Parameters of Transmitter........................................................................ 77 Parameters of Channel ............................................................................ 78 Parameters of Impulse Response ............................................................. 80 Parameters of Gaussian Noise ................................................................. 81 Parameters of Baseband Demodulation................................................. 82 Parameten of Adaptive Channel Equalizers. CPI Cancellers and Echo Canceliers ............................................................................................... 84 Parameters of Normalized LMS Adaptive Filter...................................... 85 .. . . Parameters of Decision Circuit................................................................. 86 Parameters of Performance Meter........................................................... 87 Parameters of Square Error Trend ......................................................... 8 8
vii
LIST OF FIGURES
Figure 2.1 Figure 2.2 Figure 2.3 Figure 3.1 Figure 3.2 Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 Figure 3.9 Figure 3.10 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Figure 4.6 Figure 4.7 Figure J.8a Figure 4.8b Figure 4 . 8 ~ Figure 4.9a Figure 4.9b Figure 5 .l Figure 5.2 Figure 5.3 Figure 5.4 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.1 Figure 6.5 Figure 6.6 Figure 6.7 Figure 6.8 Figure 6.9 Figure A .1 Figure A.2 Figure A.3 Passband System Model ...........................................................................8 Impulse Response of RC Filter................................................................. 12 Baseband System Mode1 .......................................................................... 16 PMD and DGD ........................................................................................ 20 PDF of A t ................................................................................................25 Waveplate Mode1 ..................................................................................... 27 Magnitudes of Frequency Response ........................................................ 29 Phases of Frequency Response ............................................................. 29 Magnitudes of Impulse Response ............................................................ 30 Phases of Impulse Response ................................................................... 30 Interferometric Measurement of the 10-ps PMD Emulator ..................... 31 Optical Fiber Channel Simulation ............................................................ 32 Simulation of Baseband Impulse Responses .......................................... 32 Equalization ............................................................................................ 34 Categorization of Adaptive Equalization ................................................. 36 Linear LMS Transversal Equalizer ........................................................ 37 Channel Equalizers and CPI Cancellers ................................................ 39 Vertical Channel Equdizen and CPI Cancellers ..................................... 41 LMS Adaptive Filter .............................................................................. 47 Adaptive Channel Equalizers and CPI Cancellers ................................... 48 J (dB)when ~ 0 . ................................................................................... 49 2 J (dB)when ~ 0 . ................................................................................... 49 4 J (dB) when ~ 0 . .................................................................................. 19 6 J (dB) when Filter Length= 16 ................................................................. 49 J (dB) when Filter Length=8 .................................................................. 49 Fully Open Eye Pattern ........................................................................... 50 Eye Pattern (SN R=30) ............................................................................. 51 Eye Pattern ( S M = 15) .............................................................................. 51 BER Plot ............................................................................................ 5 8 Bidirectional Transmission System ...................................................... 9 5 Fiber Coupler ........................................................................................... 60 A Possible Backscatter Plot ......................................................................62 Echo Cancellation .................................... . .............63 . . Structure of Echo Cancellers ................................................................... 64 Power Budget ........................................................................................... 64 Simulation of Echo Cancellers ............................ ......... . . .. 67 Eye Pattern without Echo Cancellen ..................................................... 68 Eye Pattern with Echo Cancellers ........................................................68 System Mode1 ................................................................................... 71 Transmitter-A .................................... . . . . . ...................72 Transmitter .............................................................................................. 73
viii
Figure A.4 Figure AS Figure A.6 Figure A.7 Figure A.8 Figure A.9a Figure A.9b Figure A .l Oa Figure A .1O b Figure A.11 Figure A .12 Figure A.13 Figure A .14
Channel .................................................................................................... 74 Opticd Fiber Channel with PMD ........................................................... 75 Impulse response ................................................................................... 76 Gaussian Noise ....................................................................................... 77 Baseband Demodulation .................................................................-....... 78 Receiver with Echo Cancellers ................................................................ 79 Receiver without Echo Cancellers ........................................................... 79 Adaptive Channel Equaliters and CPI Cancellers .................................. 80 Echo Cancellers ..................................................................................... 80 Normalized LMS Adaptive Filter ............................................................81 Decision Circuit ....................................................................................... 82 Performance Meter ................................................................................... 83 Square Error Trend ..................................................................................84
LIST OF SYMBOLS
real part of the tap weight c(n). See Eqs. (4.20,4.21). input PAM signal in the horizontal channel. input PAM signal in the vertical channel. imaginary part of tap weight c(n). See Eqs. (4.20,4.21). bit rate. bandwidth of the message signal. light speed. See Eq. (3.1). filter coefficients. estimated of the equalizer tap weights at step n. See Eq. (4.13). impulse response of the equalizer. See Fig. (4.2). channel equalizers . See Fig. (4.4). CPI cancellen. See Fig. (4.4). estimated tap-weight of the equalizer c,,.(n). See Eq. (1.44). estimated tap-weight of the CPI canceiler c,.,,(n).See Eq. (4.45). complex coefficients of PSPs. See Eq. (3.9). Fourier transfomi of c(r). 'desiredn signal. See Eq. (4.8). differential of *. phase retarder. See Eq. (3.1 7) and Fig. (3 3). estimated error. See Eq. (4.8). estimated error vector. See Eq. (4.18). input signal power. See Eq. (5.1). mathematical expectation of input optical field vector. See Eqs. (3.6.3.7). amplitude of the input PSPs in frequency domain. See Eq. (3.7). amplitude of the output PSPs in frequency domain. See Eq. (3.8). complex unit vector specifjhg the PSPs. See Eq. (3.7). complex unit vector specifj4ng the PSPs. See Eq. (3.8). amplitude of the time-varying input field. See Eq. (3.9). amplitude of the time-varying output field. See Eq. (3.9). output optical field in frequency domain. See Eqs. (3.6,3.8). unit vectors specifying the two output PSPs. See Eq. (3.9). carrier Frequency. See Eq. (2.3). sampling frequency. receiver filter. See Eqs. (2.10,2. 11) and Figs. (2.1.2.2). overall complex baseband impulse response of the trammitter, channel and matched filter. See Fig. (4.1). Fourier transform of& (I). See Fig. (4.1). transmitter filter. See Eqs. (2.2.2.11) and Figs. (2.1.2.2). combined impulse response of the transmitter, channel, matched fter and
equalizer. See Eq. (4.3) and Fig. (4.1). Planck's constant, 6 . 6 3 1 JlHz. See Eq. (6.8). ~ m & of the baseband impulse responses of the optical channel. See a Eq. (4.1 1). complex baseband impulse response. See Eq. (3.2 1). matrix o f the passband channel impulse responses. See Eq. (2.4). baseband horizontal to horizontal impulse response. See Eq. (2.5). baseband horizontal to vertical impulse response. See Eq. (2.5). baseband vertical to horizontal impulse response. See Eq. (2.5). baseband vertical to vertical impulse response. See Eq. (2.5). matrix of the baseband horizontal to horizontal channel impulse responses. See Eq. (2.24). matrix of the baseband horizontal to vertical channel impulse responses. See Eq. (2.23). in-phase of passband horizontal to horizontal impulse response. See Eq. (2.5). in-phase of passband horizontal to vertical impulse response. See Eq. (2.5). in-phase of passband vertical to horizontal impulse response. See Eq. (2.5). in-phase of passband vertical to vertical impulse response. See Eq. (2.5). passband horizontal to horizontal impulse response. See Eq. (2.4) and Fig. (2.1). passband horizontal to vertical impulse response. See Eq. (2.4) and Fig. (2.1). passband vertical to horizontal impulse response. See Eq. (2.4) and Fig. (2.1 ). passband vertical to vertical impulse response. See Eq. (2.4) and Fig. (2.1 ). quadrature of passband horizontal to horizontal impulse response. See Eq. (2.5). quadrature O f passband horizontal to vertical impulse response. See Eq. (2.5). quadrature of passband vertical ta horizontal impulse response. See Eq. (2.5). quadrature of passband vertical to vertical impulse response. See Eq. (2.5). matrix of the baseband vertical to horizontal channel impulse responses. See Eq. (2.22). matrix of the baseband vertical to vertical channel impulse responses. See Eq. (2.21). transfer matrx of the fiber channel in the waveplate model. See Eq. (3.20). imaginary number, satisQing ' = - 1. j
minimum MSE. See Eq. (5.10). mean-squared error vector. See Eq. (4.1 9). coupling length. fiber length. DGD generator. See Eqs. (3.16,3.17) and Fig. (3.3). system loss. effective index of refiaction for fast mode. See Eq. (3.1). effective index of refraction for slow mode. See Eq. (3.1 ). randorn Gaussian noises. See Eq. (5.2). number of the transmitter or receiver filter taps. See Eq. (2.2). transfer maaix of the segment k in a single mode fiber. See Eq. (3.17). receiver sensitivity (average nurnber of photos per bit). probability density function of DGD. See Eq. (3.12). bit error rate. power in the starting mode. See Eq.(3.3). average power in the orthogonal polarization mode. See Eq(3.3). crosstalk. See Eq. (6.3). directivity. See Eq. (6.4). excess loss. See Eq. (6.1). optical power launched into the fiber. See Eq. (6.6). insertion loss. See Eq. (6.2). Rayleigh backscattered opticai power. See Eq. (6.6). received sensitivity (dBm). See Eq. (6.8). Rayleigh scatter loss. symbol error probability for Clevel QAM signal. See Eq. (5.4). split ratio. See Eq. (6.5). total power in the fiber. See Eq.(3.3) power of Trammitter-B. See Fig. (6.6). power of Transmitter-A. See Fig. (6.6). signal power after Coupler-A. See Fig. (6.6). signal power received by Coupler-B. See Fig. (6.6). signal power received by Receiver-B. See Fig. (6.6). echo signal power due to coupler crosstalk and the Rayleigh scatter. See Fig. (6.6). cross-correlation vector between the input vector to the equalizer uJn) and the desired response x,(n). See Eq. (4.29). instantaneous estimate of P,,,(n). See Eq. (4.40). cross-correlation vector between the input vector to the equalizer uJn) in the vertical polarjzation channel and the input vector in the vertical polarization channel u,(n). See Eq. (4.3 1). instantaneous estimate of Q&). See Eq. (4.42). Q-function of filter rolloff factor. See Eq. (2.1 1) and Fig. (2.2).
0.
xii
phase rotator. See Eq. (3.7). auto-correlation matnx of the input vector to the equalizer uv(n). See Eq. (4.30). instantaneous estimate of F&,,(n). See Eq. (4.41). baseband signal vector in horizontal channel. See Eq. (2.19) and Fig. (2.3). baseband signal vector in vertical channel. See Eq. (2.19) and Fig. (2.3). modulated signal after the modulator in the horizontal channel. See Eq. (2.3) and Fig. (2.1). in-phase signal after the transmitter filter in the horizontal channei. See Eq. (2.2) and Figs. (2.1,2.3). in-phase signal after the trammitter filter in the vertical channel. See Eq. (2.2) and Figs. (2.1,2.3). quadurate signal d e r the transmitter filter in the horizontal channel. See Eq. (2.2) and Figs. (2.1, 2.3). quadurate signal after the transmitter filter in the vertical channel. See Eq. (2.2) and Figs. (2.1, 2.3). modulated signal after the modulator in the vertical channel. See Eq. (2.3) and Fig. (2.1). fraction of captured optical power. See Eq. (6.6). complex baseband waveform of the modulated signal in the vertical channel. See Eq. (2.13). three-component unit Stokes vector. See Eq. (3.10). transmitter filter sampling time. See Eq. (2.2). optical fiber complex transfer matnx. See Eq. (3.4). discrete baseband input signal to the equalizer. See Fig. (4.3). K dimensional input discrete signal vector at step n. See Eq. (4.9). output signal vector of the fiber channel. See Eqs. (2.6,J. 12). received baseband signal vector in the horizontal channei. See Eq. (2.25) and Fig. (2.3). received baseband in-phase signal in the horizontal channel. See Eq. (2.25) and Fig. (2.3). received baseband in-phase signal in the vertical channel. See Eq. (2.25) and Fig. (2.3). received baseband quadrature signal in the horizontal channel. See Eq. (2.25) and Fig. (2.3). received baseband quadrature signal in the vertical channel. See Eq. (2.25) and Fig. (2.3). received baseband signal vector in vertical channel. See Eq. (2.25) and Fig. (2.3). output of the horizontal channel. See Eqs. (1.6,4.11). output of the vertical channel. See Eqs. (2.6,4.11). input signal vectors to the equalizer in the vertical and horizontal channels.
xiii
u ~ ( o )~ ,
~(0 ' 2
VA{)
See Eq. (4.15) and Fig. (4.4). ( cornplex quantities of the transfer matrix, T(o). 0 ) Gaussian noise signal. See Eq. (2.8) and Fig. (2.1). different group velocity. See Eq. (3.2). Gaussian noise signai in the horizontal channel. See Eq. (2.8) and Fig. (2.1). Gaussian noise signal in the vertical channel. See Eq. (2.8) and Fig. (2.1). angular carrier frequency. See Eq. (3.2 1). input optical pulse width. See Fig. (6.6). 'desiredm signal vector. See Eq. (4.17). input QAM signal vector for the horizontal channel. See Eq. (2.1) and Fig. (2.1). in-phase signal in the horizontal polarization channel. See Eq. (2.1) and Fig. (2.1). in-phase signal in the vertical polarization channel. See Eq. (2.1) and Fig. (2.1). quadurate signal in the horizontal polarization channel. See Eq. (2.1) and Fig. (2.1). quadurate signal in the vertical polarization channel. See Eq. (2.1) and Fig. (2.1 ). input QAM signal vector for the veniczl channel. See Eq. (2.1) and Fig. (2.1). output of the adaptive equalizer. See Eq. (1.7). output signal vector. See Eq. (4.10) and Fig. (4.4). output of the equalizer. See Fig. (4.1). output of receiver in the horizontal channel. See Eq. (2.9) and Fig. (2.1). demodulated in-phase signal in the horizontal channel. See Eq. (2.10) and Fig. (2.1). demodulated in-phase signal in the vertical channel. See Eq. (2.10) and Fig. (2.1). demodulated quadurate signal in the horizontal channel. See Eq. (2.10) and Fig. (2.1 ). demodulated quadurate signal in the vertical channel. See Eq. (2.10) and Fig. (2.1). output of the receiver in the vertical channel. See Eq. (1.9) and Fig. (2.1). received PAM signals in the horizontal channel. See Fig. (2.1). received PAM signals in the vertical channel. See Fig. (2.1). average differential goup delay time. See Eq. (3.12). fiber loss. propagation constant for the fast mode. See Eq. (3.1) and Fig. (3.1 ). propagation constant for the slow mode. See Eq. (3.1) and Fig. (3.1 ). received SNR. See Eq. (5.5). Rayleigh scattering coefficient. See Eq. (6.6). s s e SNR. See Eq. (5.12). ytm xiv
atenuation coefficient. See Eq. (6.6). bitefngence. See Eq. (3.1). differential index of fraction. See Eq. (3.1). DGD. See Eq. (3.2) and Fig. (3.1). nns value of the differential group delay d r. See Eq. (3.14). gradient of J,(n)to c,,(n). See Eq. (4.23). gradient of JJn) to c,,,(n). See Eq. (4.22). independent unifonly distributed random variables, distnbuted frorn -nto r See Eqs. (3.18,3.19). . wavelength. step-size, an adaptation constant. See Eq. (4.8). variance of the channel Gaussian noise. variance of n,. variance of n,. Mauwe;elliandistributed DGD. See Eq. (5.16). arrivd times in the two polarization States. See Eq. (3.9). average DGD in each segment of a fiber. See Eq. (3.15). initial phases in the horizontal channel. See Eq. (2.18). initial phases in the vertical channel. See Eq. (2.17). polarization-independentphase. See Eq. (3.4). Ludolphian number, 3.1415926 --. angular frequency. central angular fiequency. phase of the optical field. See Eqs. (3.7.3.8). dispersion vectors, indicates rate and direction of rotation. See Eq. (3.10). Hermitian transpose of *. matrix transpose of a. complex conjugate of *. convolution of and +. multiplication of and + average of absolute value of a.
0.
CPI
DFP
DGD DSP EDFA FIR IF IM-DD ISI LMS LTE MSE OTDR PAM PDF PbID
PN
PSPs
QAM
RLS
RMS
SNR
Z F
Bit Error Rate Cross Polarization Interference inverse Discrete Fourier Transformation Differential Group Delay Digital Signal Processing Erbium-Doped Fiber Amplifier Finite Impulse Response Intemediate Frequency Intensity Modulation with Direct Detection Inter-Symbol Interference Least Mean Square Linear Transversal Equalizer Mean Squared Error Optical Time Domain Reflectomew Pulse Amplitude Modulation Probability Density Function Polarkation Mode Dispersion Pseudonoise Principal States of Polarization Quadrature Amplitude Modulation Recursive Least Squares Root Mean Square Signal Noise Ratio Zero Forcing
Chapter 1
Introduction
1.1 Introduction to the Research
Optical fiber communication systems have attracted more attention in the recent
years, because of the outstanding advantage of optical fiben. The m s significant ment ot
makes the dispersion worse since the dispersion effects accumulate over the muitiple
amplifier stages. Therefore, the current task of optical fiber communication system
design is to solve the fiber dispersion problem, especially PMD. There are many
techniques developed for PMD compensation in both the optical domain and the electrical domain [2]. Early strategies to reduce PMD were focused on reducing the intrinsic PMD of the fiber by altering the manufacturing process [60,611. This led to the low and stable values of PMD in the new generation of single-mode fibea being manufactured [62]. More recent efforts have been examining ways in which lightwave
and thus make full use of the existing embedded fiber base. Such strategies indude
reduction of ISI (Inter-Symbol Interference) by electronic equalization in the receiver
[40,41.431, and optical equalization using an automatic polarization controller at the
[42]. He used a polarkation controller to adjust the polarization into a fiber to one of the PSPs (Principal States of Polarization) of the fiber. The receiver detects only one of the
PSPs to eliminate first-order PMD. A method to track the changing PSPs using a
gradient search algorithm was also presented in his paper. The advantage of optical techniques is that they can be bit rate independent, while for electrical techniques, the complexity and difficulty will increase dramatically as the bit rate increases. With the development of high speed DSP (Digital Signal Processing) and Si/SiGe epitaxial transistors (64,651, an electrical equalizer operating at 10 GHz may be realized [j11. A
hi& speed electncal equaiizer becornes an attractive solution for PMD mitigation due to
its compact and cost effective implementation.
This thesis proposes a novei optical fiber transmission system where two 4-QAM
(Quadrature Amplitude Modulation) signals are transmitted separately over vertical and horizontal polarization channels. This strategy has a clear advantage, more system capacity and higher bit rate. However, it aiso causes CPI between the two signals in the two polarization channels. Actually, CPI is the result of PMD due to the power coupling.
which are applied to compensate the PMD and CPI in an optical fiber communication system. It shows the potential application of electrical adaptive equalizen and CPI cancellers in an optical fiber communication system and demonstrates a bright funue for
ihis strategy. The structure of the channel equaiizea and cancellen is described and analyzed. The contribution that the equalizen and CPI cancellers offer is demonstrated by the performance improvement they create. The system performance such as the eye pattern and bit error rate are presented. The bit error rate is an important parameter to measure the performance of an optical fiber communication system. It shows the PMD,CPI and the noise effects on the optical fiber communication system as well as and the contribution of adaptive channel equalizers and
CPI cancellers in the receiver. Also, the bit error rate reveals the relation between PMD
and the received SNR (Signal Noise Ratio). and helps us determine the requirement for
presented. The crosstalk in the coupler and the backscatter [1] in the fiber causes a retumed echo in this system. To eliminate this echo, echo cancellers in the receiver might
be used. The power penalties when there are echo cancellers and no echo cancelles [19,28] are compared to show the contribution of the echo cancellers.
requirements of the baseband simulation model are proposed and explained. Finally, the simulation tools, MATL.AB' and
-4-
SIMULINKD,are introduced.
Chapter 3
PMD concepts and the PSPs are discussed first. Based on the
theory of PSPs, a waveplate mode1 is presented, which is used to
simulate the frequency responses and complex baseband impulse responses. The computer simulation of PMD is achieved by using
the complex baseband impulse responses. nie method of
simulation using MATLAB~ sIMULMK' is also shown in and this chapter. Chapter 4: At first, the adaptive channel equalizer technique is introduced, which includes hdamental equalizer theory and the type of
adaptive equalizers. Then. the method to determine the adaptive
tap weights of the equalizer and CPI canceller using the normalized
The system bit error rate curve is dso shown. The discussion of the system performance is presented. Chapter 6: Bidirectional transmission in a single fiber is proposed. The coupler crosstalk and Rayleigh backscatter as well as their effect on
the echo are introduced. Then, the power penalty is calculated. and
Chapter 7 :
The conclusion of this research and the future work are presented.
The input signals A,(n) and A,,(n)are 4 level PAM (Pulse Amplitude
Modulation) signals. They are uniformly distributed random integer signals ranging from
1 to 4. These P A ! signals are then converted to QAM signals x,.(n) and x,(n). by
mapping,
where x,,(n), xh(n) are QAM signal vecton in the vertical and horizontal channels.
x,,(n) is the in-phase signal in the vertical polarization channel,
x1Jn), x,(n), x,,(n)and x,,(n) are the values of 1 +j, 1+j, I -j and I -j. And j is the
The signals after transmitter filter, g(r), are s,,,(t). s&), sJr) and s&),
where sI&) is the in-phase signal after transmitter filter in the vertical channel,
sJt)
s,(t) is the in-phase signal afier transmitter filter in the horizontal channel,
s,,(t)
g(t) is the transmitter filter, which can be selected by the designer to satisfy the limitations on transmitted power and bandwidth,
do*
where sJt) is the modulated signal afier the modulator in the vertical channel.
sh(t)is the modulated signal after the modulator in the horizontal chamel. and
impulse responses,
In Eq. (2.4), the passband channel impulse responses are related to the baseband channel
impulse responses by,
where Re[.] denotes the real part of m. The subscripts of h(t) in Eq. (2.5) are of the fom
s, s2s,. The first subscnpt, s, is one of =pn, "in or "qn, denoting the type of impulse "bn, response. 'passband". 'cornplex baseband", "in-phase baseband" or "quadrature baseband", respectively. The second and the third subscripts, s2 and s,, will be either "vn or "hn. denoting "horizontal" and "verticaln respectively. The subscript s, denotes the input to the channel and the subscript denotes the output of the channel.
Having known the modulated signals and channel responses, the output of the channel. u(t),can be obtained [25],
i fis the Gaussian noise signal vector with the components in both vertical and )
The noise signais vv(t)and v,,(t) are zero mean uncorrelated with variances ofof.
In the receiver, u(t) is first demodulated and then passed through receiver filters.
The outputs of the receiver are y,(t) and yh(t),
and we have,
y&) is the demodulated in-phase signal in the horizontal channel. y,,(!) is the demodulated quadrature signal in the horizontal channel,
#.and A are the phase offsets between the trammitter and the receiver in the
vertical and horizontal channels.
At) is the receiver filter that is used to cut off the fiequency components near 2f,
by demapping the received QAM signals y,(t) and yh(t). &(t) a d Zh(t)are similar in
In digital transmission, signals are ohen represented by ideal steps, and they have
infinite bandwidth. ' o restrict the spectnim of the source signal, an RC (Raised Cosine) filter is commonly used [Il]. In this system, the RC filter is divided into two parts, one is the transmitter filter g(t) and the other is the receiver filterf(t). Each part is the square
root of the magnitude response of the RC filter in the frequency domain. nie impulse response of g(t) and f(t) is [Il],
g(t) = /(O
where r is the rolloff factor, O 5 r s 1. The impulse response of RC filter is s h o m in Fig. (2.2) [Il].
requires that a simulation sampling frequencyf; must be at least twice the maximum fiequency being modeled. Since the optical carrier fiequency is usually very high, around
200 THz,modeling passband communication systems involves high computational loads.
complex envelope of a passband signal. As an example, consider the vertical channel in Fig. (2.1). The equation for modulating the in-phase signal s,,(t) and the quadrature signal
S , ~ , . ( I by )
S . ( t ) = JS[S,,.
larger than 2B,. In the baseband simulation, the signal bandwidth is always assumed to be much smaller than the carrier fiequency, or B ,
modulator and the input to a baseband demodulator are cornplex signals. Demodulation can be divided into RVO methods, coherent and noncoherent demodulation [Il]. Coherent demodulation requires the knowledge of the frequency and phase to recover the received message signal. Noncoherent demodulation permits the recovery of the message signal without knowing the fiequency and phase.
The coherently demoduiated signal in the vertical channel is. y.([),
Suppose,
= (ubn.(t) jubpv(t))e-'A. +
( 2.17 )
where ub,,,(t) the received baseband in-phase signal in the vertical channel, and is
ubP(t)is the received baseband quadrature signal in the vertical channel.
where
ubth(t) the received baseband in-phase signal in the horizontal channel, and is
u4,(t)
constnicted which includes a trammitter, a receiver and a chamel?they are shown in Fig. (2.3).
In the baseband model, the complex signals can be expressed as the mauix forms.
Thus, in the trammitter, the baseband modulated signals s,,(t) and s,(r) are,
where
channel.
h,.,,(r) is the baseband impulse response matrix, fiom the vertical to the horizontal
channel ,
hdt) is the baseband impulse response mauix. fiom the horizontal to the vertical
channel,
h,(r) is the baseband impulse response matrix, fiom the horizontal to the
horizontal channel,
In the receiver, the received baseband signai vectors uh(t) and uhh(t)are,
- sin 4,.
cos#,.
][ ]
tibfv( t )
zlw(t)
and
simulation system.
Signal Processing Toolbox is a collection of tools built on MATLAB" [14]. This toolbox
supports a wide range of signal processing operations, from waveform generation to filter design and irnplernentation, parametric modeling, and spectral analysis [15,161. SIMULINK% a s o h a r e package in MATLAB' for modeling, simulating, and analyzing dynamical systems [IO]. It supports linear and nonlinear systerns, modeled in continuous time, sarnpled time, or a hybrid of the two. sIMuLMK' is a powerful tool
for real-time simulation. The communications Toolbox and DSP BIockset are the most
usehl tools for simulation of telecommunication systems. The Communications Toolbox is a collection of computation functions and simulation blocks for research, development. system design. analysis, and simulation in the communications area [I Il. The DSP Blockset is a collection of block libraries which have been designed specifically for DSP applications. and include operations such as classical, multirate, and adaptive filtering, compler and matrk arithmetic, tmnscendental and statiaical operations, convolution, and Fourier transforms [ 121.
In single-mode fibers there are two different polarization modes, the fast mode and the slow mode, due to birefnngence which is caused by intemal or external stresses or by non-perfect circularity of the fiber core. Each polarization mode has its own propagation constant PT 4,which are shown in Fig. (3.1). Usually the state of and poiarization of an arbiirary optical field can be represented by the vector sum of the field components aligned with the two polarkation modes.
where A p is birefnngence.
- t 9-
where d(m)denotes the differential of 0 . The different group velocities of the modes limit the bandwidth of the optical
fiber communication system, because they cause pulse broadening. Thus the difference
in the group delays of the fast and slow modes is,
where d r is also called the differential group delay (DGD). L is the fiber length. and
1
I
Polarization Mode Dispersion (PMD). And the delay between the two components is
In Eq. (3.2), for a specific Anef,DGD is proportional to the fiber length, L. However, this phenornenon occurs only in short fiber. In long-distance optical fiber transmission systems, there is a polarization mode coupling due to power of one polarization mode leaking into the other mode. The mode coupling is a random process, but the average power will grow with the distance.
A coupling length, 1, is defined to be the length at which the average power in the
orthogonal polarization mode, Ph((.) ,is within l/e2 of the power in the starting mode. Pu
The coupling length. [,c be used to distinguish between a long fiber and a shon m
fiber. When a fiber length is less than l,, it can be considered to be a short-length fiber. In a short-length fiber, the polarization effects are deterministic and PMD grows linearly
with the fiber length. When a fiber length is much larger than [ we can regard it as a ,
long-length fiber. In a long-length fiber, the fiber shows a statistical variation in the polarization due to both PMD and mode coupling. Aiso, the DGD becomes proportional to the square root of the fiber length [57].
which the corresponding output states of polarization are orthogonal and show no dependence on wavelength to first order*. Such states are called principal states of polarization (PSPs). Under the assumptions of this model, the linear medium can be described by a complex transfer matrix T(o)[67],
and u,(o) are complex quantities, and they satisfy the relation,
1% (dl2
0'
+ [ri,
(dl21, and =
0.
Suppose the input optical field vector is E ( ) and the output opticai field is u@,
E ( O )= E, (o)eJPU (O), , ,
( 3.7 )
(o) the complex unit vector specifj6ng the input PSPs, and , is
Eh( O )is the complex unit vector specifying the output PSPs.
The Eq. (3.6) is a frequency model used to describe the PSPs. The important
contribution of this model is that it points out the property of polarization invariance with frequency for the fint order, that an optical pulse aligned with a principal state at the input of a fiber will emerge at the output with its spectral components al1 having the same state of polarkation [52]. This means that the only distortion in the pulse is the phase distortion and it will not change the shape of the pulse. Because of the existence of PSPs, any input signal can be expressed as the sum of
two states. In either of these polarization states there is no pulse shape distonion. but the
two states have different time delays. Thus we also describe the model in the time domain as [54],
where E,(t) is the output PSPs in the tirne domain. Eu@)are the amplitudes of the time-varying input PSPs i the time domain, n
c . ,c-are complex coefficients, which are the fraction of the power of the input
fiber. In the long fiber spans, these states are deterrnined by the cumulative effects of the
birefringence over the entire span [2]. In frequency domain, the polarization mode dispersion is manifested as a frequency dependent state of polarization at the output of a fiber [54,55]. Over a narrow frequency range, it takes the fonn of a rotation of the polarization on the Poincar sphere,
where S is the three-component unit Stokes vector. representing the states of polarization
and + .
The relationship between the fiequency domain and the time domain behaviors is
0, that the rate of rotation, 11is equal to the differential group delay time experienced by
spectrally narrow pulses having polarzation aligned at the input with two orthogonal
PSPs [ S I ,
Using the theory of PSPs, the statistical properties of DGD were studied. The
10
otherwise,
where 4,
with zero mean and identical variance [56]. Therefore, to generate a Mawellian distributed variable, first we should pnerate three independent Gaussian random variables, then take the square root of sum of three squared Gaussian random variables.
DGDs in every segment of a fiber whose average DGD through the whole fiber is 100 ps,
and the number of segments, K,is 100. Thus, the average DGD in each segment, ( r ) is,
excited by one of its PSPs defined at the optical angular frequency o behaves as a simple .
delay. Thus if we suppose the input field is an input reference frarne which consists of
two orthogonal input PSPs, and the output field is an output reference frarne which
consists of two orthogonal output PSPs, then the phase shifi of a fiber can be described by
a Jones matrix, iCl,(o) which we cal1 the "DGDgenerator",
where N,(o) is the transfer matrix of the segment k in a single mode fiber, and
0, and pkare independent unifomly distributed random variables. distributed
f o - 75 to IC. rm
nie overall transfer matrix, H(w),is the concatenation of ail the segments s h o w in Fig. (3.3),
frequency,/,
simulation performed in this thesis, to balance the computing speed and the accuracy, we used 128 frequency points in the frequency domain, which means the sampling frequency,
f;, is 128xB,where B
is the signal bit rate. Thus a passband Frequency is in the range from
/; - j 7 2 t0f;.~1;/2.
In this simulation, the bit rate, B. is 2.5 GHz,the DGD is 100 ps, and K is 100 segments. Fig. (3.1) shows the magnitudes of the frequency responses. H(o), and Fig. (3.5) shows the phases of these frequency responses.
O r objective is to get the baseband impulse responses of the optical fiber channel u
with PMD. To achieve this objective, the passband frequency responses should be shified
to baseband. Then the baseband impulse responses cm be obtained from, h ~ t=)DFT-I[H(~a,)], where h,(t) is the complex baseband impulse response,
Thus we can get four complex baseband impulse responses, hbbv(t), hb,(t), h,,(t) and hbM(t).Fig. (3.6) shows the magnitudes of impulse responses, and Fig. (3.7) shows
phases of impulse responses.
Frequency Response: Hw
14
14
Frequency Response: Hw
4
14
14
Frequency Response: f v i h
Impulse Response: Hw
0.25
lm pulse Response:
~ k O-" h
0.25
0.25
a
O .2
aJ
0.2
3 .- 0.15 c
0.1 5 0.05
O
3 -- 0.15 c
I
O
Rf
0.1 0.05
O -2
-2
lmpulse Response: Hw
4 n
-fO
Impulse Rerponss:
HRJ~-'O
Fig. (3.8) shows the interferometric measurement of PMD (1= 1.3 pm)of a 10 ps (Intensity Modulation emulator in a IM-DD
-20
20
fiber channel simulation is constructed in Fig. (3.9). It is just the implementation of the
fiber model in Fig. (2.3).
In Fig. (3.9). V,, H,, and represent the inputs to the vertical and horizontal
represent the outputs of the vertical and horizontal channels. channels. and V , , and Ho,, The impulse responses are complex. and the implementation of these impulse responses
is shown in Fig. (3.10). A complex impulse response is comprised of four real impulse
responses. Convolution of the input signals in Fig. (3.10) with these real impulse function,fdter(), as in responses is realized by calling the MATLAB~
y(t 1= fi[te&(t)? x ( t ) ) ,
h vv(t)
..
Impulse Raseonse
hhh(t)
vout
Hout
-8i
0*lR.
In1
SplH
Join
is the data input, andf,(t) is the overall complex baseband impulse response of the transmitter, channel and matched filter. The signal received by the equalizer is,
y ( [ )=
w*j,W
+ v u ),
where v(t) is the baseband Gaussian noise signal. If the impulse response of equalizer is c(t), then the output of the equalizer is
~q(09
( 4.2 )
where g,,(t) is the combined impulse response of the transmitter, channel, matched filter
H) l
Mstcbed
yEqualizer
-*) :( '
x,W
1
40
Channel
Mt)
film -
d0
Decision
-circuit
->m
fd(')
x2(t)
and equalizer,
g.O) = f.(O* ~ ( 0 .
( 4.3 )
where N is the number of filter taps. Ideally, the desued output of equalizer, y,(t), should be equal to the original input
x(t).
If v(t) is equal to zero, then to force y,(t) equal to x(r), Eq. (4.3) is used to get
F,(f)C(f)=
19
The result we get in Eq. (4.6) shows that when the baseband noise v ( t ) is ignored, the equalizer filter CV)actually is an inverse filter of F,CT)which is the fiequency response of the combination of fiber channel, transmitter and matched filter.
41.2 Adoptive Equalizatioa
The charactenstic of the time varying opticd fiber channel requires the use of an
adaptive equalizer. An adaptive equalizer needs the knowledge of the desired input so as
to compare it with the output of the equalizer to get the error signal which is used in the
adaptive algorithm. However, because of the different locations of trammitter and receiver, it is difficult to get the original data input to the transmitter in the receiver.
There are two methods which cm generate a "desired" signal as that in the transmitter [ ] 7.
1. The first method is called training. In this method, before data transmission. a sequence of PN (pseudonoise) signals are sent fiom transmitter as test signals. In the
receiver, this PN signal is pre-stored and it is the sarne as that in the transrnitter. The PN signal is called a training sequence. and the time to transmit the training sequence is called the training period. In the training perod, the adaptive equalizer c m adjust its filter coefficients and reaches a steady state where the smallest enor is achieved. Mer the aining period, the filter coefficients are fixed. Sometimes. the training sequence will
be resent perodically to allow the adaptive equalizer to adapt to the tirne-varying channel.
In Fig. (4.1), when the switch is switched to "1". it means the receiver is using the
-35-
feedback control loop. In Fig. (4. 1), afier the training period, the switch is then switched
to "2"from " 1".and x,(t) is used to estimate the error e(t).which will be used to update
the tap weights.
Zcm Forcing
LMS
Gradient RLS
LMS
RLS
GriuiicarW
LMS RLS
RLS
Algorithm
-3 6-
Equalizers can be divided into two general categories, linear and nonlinear equalizers. In Fig. (4.1), there is no feedback fiom the output of receiver, z(t), thus it is a linear equalizer. On the other hand, if z(t) is fed back in order to control the equalizer, the equalizer is nonlinear. In this thesis, a discrete L E (Linear Transversal Equalizer) structure is used which is s h o w in Fig. (4.3). A discrete LTE is made up of K tapped delay lines which have a symbol period TV, the number of taps is K. In Fig. (4.3), the output of the LTE, and
y(n) is,
Mean Squares) and RLS (Recursive Least Squares). In this thesis, four linear LMS
transversal adaptive filters are used as cbannel equalizers and CPI cancellen. The complex normalized LMS algorithm is descnbed mathematically as [7,12],
where
an
...
ic(n -
K+
[)Ir.
( 49 .
c(n) is the K dimensional vector of estimated of the equalizer tap weights at step n.
a is a small positive constant (e-'9 used to overcome the potential numerical ins~ability the tap weight update. in
ISI equalization.
Fig. (4.4) shows the discrete baseband structure of the channel equdizers and CPI
cancellen, where cw(n)and cVl(n) the channel equalizers and we cal1 c d n ) and c,(n) are are the CPI cancellers.
1
Figure 4.4 Channel Equalizers and CPI Cancellers
The complex inputs to the vertical and horizontal polarization channels are xv(n)
and x,(n), and the outputs of channels are u l n ) and u,(n). Because this is a baseband
model, the IF (Intemediate Frequency) stage and the rnodulator/demodulator are not shown. nius the channel is followed by channel equalizen and CPI cancellers whose outputs are yJn) and yh(n). We can express the relationship between these signals in matrix form,
where y(n) is the output signal vector, c(n)is the matrix of adaptive channel eqdizers and CPI cancellers,
u,(n) and u,(n) are the input signal vectors in vertical and horizontal channels to
the equalizer,
eT
h(n) is the matrix of the complex baseband channel impulse responses, see
Eq. (2.21). The error, e(n), is the difference between the desired response r(n) and output of
To simpliQ the analysis, consider only the vertical polarization channel equalizer
c,(n) and the CPI canceller c,.(n) fiom the horizontal to the vertical channel, as s h o w in
Fig. (4.5). Suppose the complex filter tap weight vector for channel equalizer c,(n), is,
where a(n) is the real part and b(n) is the imaginary part of tap weight c(n). The --dient
of J,.(n)with respect to c,(n) and c,,,,(n)is [7],
where
where P,,.(n) is the cross-correlation vcctor between the input vector to the equalizer u,(n)
RJn) is the auto-correlation matnv of the input vector to the equalizer u,.(n)
R,,.(n) = ~ [ ~ , - ( n ) u % o ] .
( 4.30 )
Q,.,,(n) is the cross-correlation vector between the input vector to the equalizer
uJn) in the vertical polarization channel and the input vector in the
According to the method of steepest descent, the updated value of the tap-weight vectors for equaiizer cJn) and canceller c d n ) are,
and
where p is called step size, an adaption constant which controls the size of the incremental correction applied to the tap-weight vector [q. Subnituting in V , ~ ( J , ( ~ ) ) and
In Eqs. (4.38,4.39), the adaptive tap-weights for channel equalizer c,(n) and CPI canceller cAn) are established. In reality, however, this method will not work because the exact measurement of the gradient vector V (~,(n)) and ,
( ~ ~ ( n ) impossible is )
since it requires the correlation matrix R(n) and cross-correlation matrixes P(n) and Q(n). Therefore, O ( ~ , . ( n ) ) O .( ~ ~ ( nmust be estimated from the available data. This idea , and , , , ))
is realized in the LMS algorithm.
In the LMS algorithm. the instantaneous estimates are used instead of the
correlation matrix R(n) and cross-conelation matrices P(n) and Q(n), and for channel
equalizer cJn). they are defined by, respectively.
P, ( n ) = u, w x : (4 ,.
( 4.43 )
where S,(n) is the estimated tap-weight of channel equalizer cJn), and ,(n) is the estimated tap-weight of CPI canceller cvh(n).
V &,An)) in Eq. (4.36), gives get the adaptive tap-weight in the LMS algorith,
k ( n + 1) = , ( n ) + pu,(n)[x:(n)- u:(n),(n)- u : ( n ) ~ , ( n ) ] ,
and similar to Eq. (4.43),
( 4.44 )
t,(n
When the input u(n) is large, the LMS algorithm experiences a gradient noise
amplification problem [7]. Therefore. the normalized algorithm is used to overcome this problem. The normalized algorithms of Eqs. (4.41 -4.44) are,
6,Jn
+ 1 ) = ,(n) +
PU.^
a + u:(n)u,(n)
[xi ( n ) - u (n)Qhhn ) - u : ( :
(n)t,,( n ) ],
( 4.50 )
Based on the Eqs. (4.48-4.5 l), the Nomalized LMS channel equalizers or CPI cancellen can be constnicted using SIMULMK~ shown in Fig. (4.6). as
Fig. (4.6) shows the model of the nonnalized LMS filter. In this model, the first
input is u(n), the second input is error e(n)and the output is y@). The buffer is used to generate input signal vector u(n).
Figure 4.6 LMS Adaptive Filter The overall adaptive equalization and cancellation system is s h o w in Fig. (4.7). Complex delays, 2 and ?,are " used to synchronize the output of the equalizer and the
desired input. The factors that cause the delay include the mansrnitter filter. receiver filter and channel delay. So the desired input h m the information generator has to be delayed by the same time. The input signals are sarnpled at the rate 1; = B which is 128 times lower than the channel sampling rate before they go into the equalizer. T w switches are used to select the training or decision-directed rnethod. At the beginning of this simulation, the nvitches are set to the training method. In every bit period, the tapweights are adjusted to minimize the enoa. After the MSE reaches the steady state,
switches are then set to decision-directed method where output signals of the decision circuits are used as the desired input.
I
av
From Eqs. (4.48-4.5 we can see that the value of step-size p is very important. 1),
Fig. (4.8) shows three leaning c w e s wth erron (dB) on the ordinates, when p is 0.2.0.4
and 0.6 respectively. By comparing these three cases, it is clear we have to tradeoff the
training time versus the residuai error. When p is 0.2, the error is the smallest. but it
takes longer time to train. When p is 0.6, the equalizer can reach the steady state quickly,
but the error is the largest. Thus we choose p equal to 0.4, which is the best compromise.
Also, the number of taps in the equalizer and canceller is the factor that affects the
perfotmance and accuracy of the receiver. In the simulation system, ive choose 8 iaps for
equalizers and cancelles rather than 16 taps, because when we use 8 taps, the filter takes less time to reach steady state and has smaller variance. The errors (dB) of the equalizer and canceller when number of filter number of taps is 8 and 16 are show in Figs. (4.9~4.9b).
.... ....
.........
.......... . & . . . . . . . .
.............
.,. . . . . . . . . . . , .
II'
: . . . . . . . . . . . ;.. . . ...............................
, !
4 ,
R ' I
'.
.:
.:
* ; , a
Through the eye pattem, Ive can roughly mesure the performance of the received
signai in the receiver, which suffers fiom the PMD and noise. When the input SNR is
30 dB, the eye pattem in the receiver is obtained as s h o w in Fig (5.2). By companng
these two figures, it is clear that the width of eye pattern d2 i Fig. (5.2) is smaller than n dl, which indicates the latter traces deviate fiom the completely open eye pattern.
Fie. (5.3) is the eye pattern when input SNR is 15 dB. The width of eye pattern d3 is
much smaller than d2, so the enor probability is larger, because there is a smaller space to distinguish between the two levels.
The simplest way to measure BER is error counting. or counting the number of
error bits in a penod. This is simple, but it may not be accurate. Also because the BER
it requires a
where E is the input signai power. The correspondhg received signal in receiver, r(t),is
where n, and + are the error signais with zero mean and the same variance, d,
where
~ and
detected bit is due to ISI, CPI and noise. To simpliQ the calculations. the distortion is treated as Gaussian noise. For high signal-to-noise ratios, this assumption will produce
where
b ( 1 - pb)Z =
c.
Thus, we have,
The relationship between the teceived SNR, y, and minimum MSE, J,,
is [8],
Therefore, it is possible to measure the minimum received MSE J,, when the channel
equalizen and CPI cancellers reach the steady state, and to get the BER according to
Eq. (5.1 1).
In this simulation system, the channel and receiver thermal noises are treated as
Gaussian noises, the input signal power is O dB. The signal bit rate is 2.5 GHz. The system SNR, y , is from 4 to 32 dB, where.
y, =
t -, and .
2 q
is the variance of channel Gaussian noise. The DGD is 10 ps, 100 ps and 500 ps, separately. The number of the adaptive filter taps is 8, and step-size, p is 0.4. The training time is longer than 3 x 1O4 sec for 1O-ps PMD,5 x I O4 sec for 100-ps PMD and
10 104 sec for 500-ps PMD. We measure the J,,
y,. Each t h e , the J,,
in vertical in-phase signal, vertical quadrature signal, horizontal inare obtained. The experimental results of J,,, are
Table 5.1
1
Vertical In-phase Vertical Quadurate -7.84
-12.0
-1 1.9 -1 1.1
Avg.
Jmtn
Avg.
(dB)
-10.2
-6.78
-10.1
2
3 4
1
- 12.3
-13.1 -1 1.8 -13.9
-8.5 1
-7.22 -8.94 -12.7
-9.50
- 10.3
- 10.4 .
-9.94 -12.4 -11.6
I
-13.1
2
3
- 10.2
-1 1.3
- 10.5
-1 1.6 -1 3.1
- 10.2
-1 1.4
- 10.7
-13.0
-1 1.3
-14.9
-12.3 -10.8
4
1
3
4
-12.8
-12.5
- 12.9
-14.7
-15.1
- 13.6
- 14.3 - 15.8
-1 1.8
- 12.8
-12.2
- 14.0
1
-13.9
-13.5 -16.6 -14.4 -16.6 -17.7 -19.2
-
2
3
4
- 14.0 - 16.9
-13.5 -18.8
-1 7.0
- 13.5 - 16.4
- 14.0
-17.1
- 14.7
-16.2
-15.3
-17.0 -13.0
- 17.3
-13.9
- 16.4
-17.3 -1 8.9
- 19.2
-22.6
-19.7 -18.3
- 17.3
-18.0
-18.5
-17.6
3
4
- 17.3 -16.9
-24.1
- 17.0 - 17.7
-23-6
-25.5
- 16.0
-21.9 -21.7 -30.1
-1 7.55
- 18.7
-24.3
-26.3
-23.6
-27.7
-36.5
-23.1
YS
NO.
r
Ji.(dB)
(DGD = 100 ps)
Avg.
~m
(dB)
Vertical I n-phase
1
Vertical Quadurate
-8.3 1
-6.77 -7.58
Horizontal In-phase
-8.22 -5.54
Horizontal Quadurate
-9.47 -5.5 1 -7.94 -7.43
-8.9 1
Avg.
-8.89
(dB)
-7.75
1
-6.32
-7.83 -7.67
-10.4
3
1
-7.5 1
-7.77 -1 1.5
4
1
-7.23
-9.05
-9.42
-9.13
- 10.4
-8.05
-9.66
-8.77 -7.20
-1 1.8
- 10.2
-8.13
-7.84
3
4
-8.41
-8.89
-7.70 -1 1.0
12
- t 2.6
- 12.3
-1 1.0
-11.5
1
-10.8
2
3
4
- 10.3
-9.3 1
-9.89
-10.5
- 10.2
-9.57
-1 1.2
-IO.L
- 10.8
- 12.7
-14.2
16
I
- 10.2 - 10.5
-13.2
- 10.2
-11.6
- 12.5
-1 1.5
-12.5
-1 1.0
- 12.5
Table 5.3
From the above tables and Eq. (5.1 1), the BER plot is show in Fig. (5.4). This figure shows the BER when DGD is 10,100 and 500 ps. From the BER plot, to make the
BER smaller than 1c9, system SNR, y,, should be larger than 17 dB when DGD is 10 the
ps, larger than 21 dB when DGD is 100 ps, and larger than 27 dB when DGD is 500 ps. Also the BER plot shows the trends of'BER for these three cases. With the increase of y,,
the plot of BER when DGD is 10 and 100 ps decrease steeply, however. when DGD is
500 ps, it decreases slowly.
Reccivcr
Coupler
Coupler
- Recriva
Ttranscciver
. -.
Ttransceivcr
-
Clearly, the bidirectional transmission system can improve the synern capacity,
and Save fiber resources. However, there are some problems which need to be solved in
the bidirectional transmission system. The most important limit is the echo ahich is
-59-
caused by the use of a fiber coupler and the existence of Rayleigh backscatter in fiber.
6.1.1 Fiber Coupler
Fig. (6.2) is a basic structure of a four-port fiber coupler [l]. The optical power is
launched into port 1, and cornes out of ports 3 and 4. A very small part of the launched power is echoed to port 2. The powen in these ports are P,, P?, and P. P, ,
p, = lOlog,, 4
4+4'
The insertion loss, P,,, is the loss obtained for a particular port to port path. For example, the insertion loss of port 1 to port 4, is,
The crosstalk, P,,is the ratio of the backscatter power received at the second input port to
the input power. Crosstaik provides a mesure of the directional isolation. It contributes
The split ratio, P,, is the percentage division of the optical power between the output
Parameter
r
Specification
< 2.5 dB bidirectional system
,
a
> 60 dB
of the refractive index. Light scattering in such a medium is known as Rayleigh scattering. The backscattered opticai power, P,(t), can be obtained from the relationship
[ i l 9
where
leflected Power
A
'-
Backscatter
.
Reflection
Reflectometry). The reflected power measured by a OTDR is illustrated in Fig. (6.3). The slope in this figure represents the loss caused by Rayleigh backscatter, and pulses are caused by reflections. The typical value of Rayleigh scatter loss is -45 dB to -60 dB [Il.
Transmitter
Laser
Echo canceller
v
+
. --
Coupler
Detector A
--
Y-
<
Figure 6.4 Echo Cancellation The structure of the echo cancellea in the simulation syaem is shown in
Fig. (6.5). The echo cancellea are located afier the channel equalizers and CPI cancellea. The structure of the echo cancellea is very similar with that of the channel equalizers and CP[ cancellers, and they also use the normalized LMS adaptive algorithm. -63-
i g z g
y-?
A
Pi
T&S-ite-
Rayleigh scatter
_ _ -y #
PO
,
.-
--
.--. - s - -- e_
-:-
Tra?s_mmite^
->
--:& P3-_:-
/ -
Receber-A - -
'AHCoupler_A
-- -
maintain reliable performance, such as a BER being less than 10". The power budgets of
two bidectional transmission schemes are discussed, which are the bidirectional
transmission system without echo cancellers and the bidirectional transmission system with echo cancellers.
6.3.1 Power Budget of the System without Echo Cancellen
Fig. (6.6) shows the bidirectional transmission system without echo cancellers.
R = 1550 nrn?
Po = O dBm,
P, = O dBm,
P,,=3 dB.
P ,= -50 dBT
Pm,= -50 dB,
rr, = 0.2 dB/km,
hr. = 6 dB,
B= 2.5 G H z
Fiber length: L, measured in km for these calculations. Therefore, we can get, the signal power after couple-, P,, is,
P3 = P 2 -q-xL-M,=-9-0.2xLdBm.
n i e signal power received by receiverJ3, P,, is,
To detect the signal without the effect of echo, the received signal P, should be
larger sufficiently than the echo signal P,. From Fig. (5.4). to make the BER smaller than 1U9when DGD is 100ps. the S N R is assurned to be 2 1 dB. that means,
L = 70 km.
6.3.2 Power Budget of the System witb Echo Ciacellers
When the echo canceller is introduced in this bidirectional transmission systern, the echo signal will be negligible compared with the received signal, and much smaller
than the receiver sensitivity, P,Pm,.Therefore, only the received signal needs only to
The discussion of power budget shows the improvement in transmission distance because of the echo canceller.
in Fig. (6.7). The received signals are UV U,, which include the echo signals. 4, and and
, d, are desired signals which are used for training. E and Ehare transmitted signals which
cause the echo.
lb
Lh
PMD is 100 ps, and the system S y,, is 30 dB. The signal bit rate, B, is 2.5 GHz. The m
number of the adaptive filter taps is 8, step-size, p is 0.4. The training time is larger than
5 104sec. Two schemes, without echo cancellers and with echo cancellen, are
investigated. The eye pattern of these two schemes are shown in Figs. (6.8,6.9). In
Fig. (6.8), because there are no echo cancellen, the eye is almost closed, which indicates that the fiber length can not exceed the maximum fiber length, 70 km. In Fig. (6.9), is a it open eye that shows the contributions of echo cancellers.
Cancellers
Chapter 7 Conclusions
7.1 Conclusions
This thesis has presented the simulation of optical fiber communication systems. For the first time, a baseband QAM signal transmission system in dually polarized fiber channels with PMD was simulated. This baseband simulation system included the transmitter, the fiber channel with PMD, the Gaussian noise simulator, the baseband coherent demodulator, channel equaiizers, CPI cancellers and echo cancellers.
The waveplate mode1 has been studied to obtain the frequency and impulse
responses of the fiber channel with PMD.s h o w in Figs. (3.4-3.7). The impulse responses showed the broadened pulse, which agree with previous experimental results. in Fig. (3.8). Through the study of the nomalized LMS algorithm applied to adaptive channel
l), equaiizers and CPI canceliers, shown in Eqs. (4.48-4.5 for the first time, a PMD
compensation technique using the combination of channel equalizen and CPI cancellers
was presented. The BER has been calculated when PMD is 10 ps, 100 ps and 500 ps, and
the BER plot in Fig. (5.4) showed the contributions of this PMD compensation technique.
Moreover, this thesis presented, for the fim tirne, adaptive echo cancellers applied in a bidirectional optical fiber communication system, with PMD compensation, to overcorne the coupler echo and Rayleigh backscatter. For a transmission system with 100-ps PMD, the improvement in fiber length was 1 17 km when echo cancellen were
used.
experiments require optical transmission and reception equipment, a PMD emulator [68], and an integrated circuit design to adapt the tap weights of the adaptive filters. Only such
a prototype would allow one to identify al1 of the cornplex issues associated with an
irnplementation.
References:
John M. Senior, Optical Fiber Communications, Priwiples and Practice, 2nd edition, Prentice Hall, 1992. Ivan P. Kaminow and Thomas L. Koch, Optical Fiber Telecommunications IIL1, Academic Press, 1997. Govind P. Agrawal, Fiber-optic CommunicationSystems, New York, Wiley, 1997. Edward A. Lee and David G. Messerchmitt, Digital Communication, Kluwer Academic Publishers, Boston, 1988. Theodore S. Rappaport, Wireless Communications: Principles and Practice, N.J., Prentice Hall PTR, 1996. Jerry D. Gibson, Principles of Digital and Analog Communicutions, 2"d edition, Macmillan Publishing Company. New York, 1993. Simon Haykin, Adaptive Filter Theory, 2"*edition, Prentice Hall, 1991. John G. Proakis. Digital Communication, 3" edition, McGraw Hill, 1995. San.it K. Mitra, Digiraf Signal Processing, A Compter-Bmed Rpproach, McGraw Hill. 1998. [1 O SIiCIULINKR - Dynumic System Simulation for ICIAT U B R ,MathWorks, 1999. ] [I 1] Weizheng Wang, SIL~IULINK' Communications Toolbox MathWorks, 1999. [12] DSP Blockser Fur Use ivifhSMULI.VK" MathWorks, 1999. [13] rCMTLAB" The Langrrage of Technical Comptrting ,MathWorks. 1999. [14] Signal Processing Toolbox For Use rvith MA TLAB" MathWorks, 1999. [15] Vinay K. Ingle and John G.Proakis, Digital Signal Processing using i\.UTLABR K 4 , PWS Publishing Company, Boston, 1997. TL4Bu. [16] Samuel D. Stearns and Ruth A. David, Signal Processing Algorithms in Prentice Hall, New Jersey, 1996. [17] Paulo Marques and Fernando Sousa, Designing an Echo Canceller Sysrem Using ~ h e TitlS32OCjO DSP, TEXAS INSTRUMENTS,1996. [18] David Smalley,Eqzrulization Concepis: ri Tutorial, T E X A S INSTRUMENTS.1 994. [191 Echo Cancellation Sofnvare for the TMS320Cj-lx, TEXAS INSTRUMENTS,1997. . [20] Ellion D Hoole, Channel Equalizutionfor the IS-54 Digital Cellular Sysfem With the T~ClS320Cjx, TEXAS INSTRUMENTS,1994. [2 11N. Amitay and J. Salz, "Linear Equalization Theory in Digital Data Transmission Over Dually Polarized Fading Radio Channels," AT&T Bell Lab. Tech. J., 63, No. 10, pp. 2215-2259, 1984. [22] J. Salz, "Digital Transmission Over Cross-Coupled Linear Channels," A T& &T Bell Lab. Tech. J., 64, No. 6, pp. 1147-1159, 1985. [23] Michel Borgne, "A New Class of Adaptive Cross-Polarization Interference Cancellers for Digital Radio Systems," IEEE J. Select. Areas Cummrrn., Vol. SAC5, No. 3, pp. 484-492, 1987. [24] L. J. Greenstein, "Anaiysis/Simulation study of Cross-Polarization Cancellation in
Dual-Polarization Digital Radio:' AT& T Bell Lob. Tech. J. ,64, No. 10, pp. 226 12280,1985. [25] M. Kavehrad and J. Salz, Tross-Polarization Cancellation and Equalization i n Digital Transmission Over Dually Polarized Multipath Fading Channels," AT&T Bell Lab. Tech J.,64, No. 10, pp. 2211-2245, 1985. [26] M. Kavehrad, "Baseband Cross-Polarization Interference Cancellation for MQuadrature Amplitude- Moddated Signals Over Multipath Fading Radio Charnels," A TdiT Bell Lab. Tech. J., 64, No. 8, pp. 1913- 1926,1985. [27] J. Salz, "A Unified Theory of Data-Aided Equalization," B.S. T.J., Vol. 60, No. 9, pp. 2023-2038, 1981. [28] T. L. Lim and M. S. Mueller, "Adaptive Equalization and Phase Tracking for simultaneous Analog/Digital Data Transmission," B.S. T.J.,Vol. 60, No. 9, pp. 20392063,1981. [29] Gersho and T. L. Lim. "Adaptive Cancellation of Intersymbol Interference for A. Data Transmission," B.S. T.J., Vol. 60, No. 11, pp. 1997-2021, 1981. [30] D. D. Falconer, "Jointly Adaptive Equalization and Carrier Recovery in TwoDimensional Digital Communication Systems," B.S. T.J., Vol. 55, No. 3, pp. 3 1 7334,1975. [3 11 David G. Messerschmitt, "Minimum MSE Equalization of Digital Fiber Optic Systems," IEEE Trans. Commun., vol. COM-26, No. 7, pp. 1 1 10-1 1 18, 1978. [32] David G. Messerschmitt, "Echo Cancellation in Speech and Data Transmission," IEEE J. Select. Areas Commun. Vol. SAC-2. No. 2. pp. 283-297, 1984. [33] Kazuo Murano, Shigeyuki Unagami, Fumio Amano. "Echo Cancellation and Applications," IEEE Communications Magazine, pp. 49-54, Jan.. 1990. [34] Hideaki Matsue, Hiroyuki Ohtsuka and Takehiro Murase, "Digitalized CrossPolarization Interference Canceller for Multilevel Digital Radio," IEEE J. Select. Areas Commun., Vol. SAC-5, No. 3, pp. 493-501, 1987. [35] B. L. Kasper, "Equalization of Multimode Optical Fiber Systems," B.S. T.J.. Vol. 61. NO. 7, pp. 1367-1387. 1982. [36] M. Kavehrad, "Cross-Polarization Interference Cancellation and Nonminimum Phase Fades," AT& T Bell Lab. Tech. J .64, No. 1 ,pp. 2247-2259, 1985. 0 [37] Aly F. Elrefaie, J. Keith Townsend, Malcolm B. Romeiser and K. Sam Shanmugan. "Cornputer Simulation of Digital Lightwave Links," IEEE J. Select. Areas Commun., Vol. 6, NO. 1, pp. 94- 104, 1988. [38] Donald G. Duff, "Cornputer-Aided Design of Digital Lightwave Systems," IEEE J. Select. Areas Commun., Vol. SAC-2, No. 1, pp. 171- 185, 1984. [39] Dietrich Marcuse and Curtis R Menyuk, '%nulation of Single-Channel Optical Vol. System at 100 Gbh," Journal ofLighnvave Technolo~, 17, No. 4, 1999. [40] J. H. Winten and Richard D. Giltin., "Electncal Signal Processing Techniques in Long-had, Fiber-optic Systems," IEEE Trans. Commun., vol. 38, pp. 1439- 3, 145 1990. [41] J. H. Winters and M. A. Santoro., "Experimental Equalization of Polarization Dispersion," IEEE Photon Tech. Lett., pp. 59 1-593, 1990. [42] Jack H. Winten, Zygmunt Haas, Mario A. Santoro, and Alan H. Gnauck, "Optical
Equalization of Polarization Dispersion," Proc. Soc. Photo-Opt. Instr. Eng., vol. 1787, pp. 3469357,1992. [43] Jack H. Winters, "Equalization in Coherent Lightwave Systems Using a Fractionally Spaced Equaiizer," Journal o Lighnvuve Technology, Vol. 8, No. 10, 1990. f [44] A. Elrefaie, R E. Wagner, D. A. Atlas and D. G. Daut, "Chromatic Dispersion Limitations in Coherent Optical Fiber Transmission Systerns," Elecfron Lett., Vol. 23, pp. 756-758, 1987. [45] Frank Bruyere, "Impact of Fim- and Second-Order PMD in Optical Digital Transmission Systems," Optical Fiber Technology 2, pp. 269-280, 1996. [46] Lothar Moller and Herwig Kogelnik, "PMD Emulator Restricted to First and Second Order PMD Generation," ECOC '99, pp. 26-30, 1999. [47] William K. Burns, Robert P. Moeller and Chin-Lin Chen, "DepoI~zationin a Single-Mode Optical Fiber," Journal o Lighhvuve Technology, Vol. LT-1, No. 1, f pp.4-49, 1983. [48] Fabnno Coni, Andrea Galtarossa and Luca Palmieri, "Polarization Mode Dispenion Characterization of Single-Mode Optical Fiber Using Backscattering Technique," Journal o Lightwave Technology, Vol. 16, No. 10, pp. 1832- 1843. f
1998.
[J9] E. Iannone. F. Matera, A. Galtuossa, G. Gianello and M. Schiano. "Effect of Polarkation Dispersion on the Performance of IM-DD Communication Systems." IEEE Photonics Technology Lerrers, Vol. 5. No. 10, pp. 1247- 1249,1993. [SU] F. Bmyere and 0. Audouin, *' Assessrnent of System Penalties Induced by Polarization Mode Dispersion in a 5 Gbfsoptically Amplified Transoceanic Link," IEEE Photonics Technology Letters, Vol. 6, No. 3, pp. 4 3 - 4 5 ? pp. 455468, 1994. [5 11 H.Bulow, R Ballentin, W. Baumert, G. Maisonneuve, G. Thielecke and T-Wehren. "Adaptive PMD Mitigation at 10GbiVs Using an Electronic SiGe Equaiiser I , C" Proc. ECOC '99. ivice, France, vol II, pp. 1 38- 139, 1999. [jZ] C.D.Poole, R.E.Wagner, "Phenomenological Approach To Polarisation Dispersion in Long Single-Mode Fiben", Eleciron. Leir., vol. 22, no. 19, pp. 1029-1030, 1986. [53] C. D. Poole and C. R. Giles, "Polarkation-Dependent Pulse Compression and Broadening Due to Polarization Dispersion in Dispersion-Shified Fiber." Oprics Letters, Vol. 13, No. 2, 1988. [54] C. D.Poole, Neal S. Bergano, R. E. Wagner and H. L Schulte, "Polarization Dispersion and Principal States in a 147-km Undersea Lightwave Cable," Joiirnal o f Lighhvave Technology, Vol. 6, No. 7, pp. 1185-1 190, 1988. [55] C. D. Poole. "Measurement of Polarization-Mode Dispersion in Single-Mode Fibers with Random Mode Coupling," Optics Letters, Vol. 14, No. 10, pp. 523-525, 1989. [56] C. D. Poole, J. H. Winters and J. A. Nagel, "Dynamical Equation for Polarization Dispersion," Optics Letters, Vol. 16, No. 6, pp. 372-374, 1991. [57] F. Curti, B. Daino, Q. Mao, F. Matera, and C. G. Someda, "Concatenation of Polarisation Dispenion in Single-mode Fibres," Electron. Lett., vol. 25, pp. 190-292, 1989. [58] F. C r i B. Daino, G. De Marchis and F. Matero, "Statistical Treatment of the ut, Evolution of the Principal States of Polarization In Single-Mode Fibers,'? IEEE J. Lightwave Tech, Vol. 8, pp. 1162-1 166.
[59] J. P. Elbers, C. Glingener, M. Duser and E. Voges, "Modelling of Polarisation Mode Dispersion in Single Mode Fibres," Electron Lerr., vol. 33, No. 22, pp. 1894- 1895, 1997. [60] A. M. Vengsarkar, A. H. Moesle, L. G. Cohen and W. L. Mamrnel, "Polarization Mode Dispersion in Dispersion Shifled Fibers: A Exact Analysis," Processings of the OFCC, ThJ7, 1993. [61] A. J. Barlow, J. J. Ramskov-Hansen and D. N. Payne, "Birefnngence and Polarization Mode-Dispersion in Spun Single-Mode Fibers," Appl. Opr, Vol. 20, pp. 296202968,198 1. [62] A. F. Judy, "Improved PMD stability in OptiFiben and Cables," Processing of 4"' International Wire & Coble Symposium, pp. 658-664, 1994. [63] T. Takahashi ,T. Imai and M. Aiki, "Automatic Compensation Technique For Timewise Fluctuating Polarization Mode Dispersion in In-line Amplifier Systerns," Electron. Lert., vol. 30' pp. 348-319, 1994. [64] D. L. Harame, J. H. Comfort, J. D. Cressler, E. F. Crabbe, J. Y. -C. Sun, B. S. Meyerson and T. Tice, %/SiGe Epitaxial-Base Transistors-Part 1: Materials, Physics, and Circuits," IEEE Transactions on Efecrron Devices, Vol. 12, NO. 3, pp.455-469, 1995. [65] D. L. Harame, J. H. Comfon, J. D. Cressler, E. F. Crabbe, J. Y. O . Sun, B. S. C Meyerson and T Tice, "SiBiGe Epitauial-Base Transistors-Part II: Process . Integration and Analog Applications," IEEE Transactions on Electron Devices, Vol. 42, No. 3, pp.469-482, 1995. [66]1. P. Kaminow, "Polarization in Opticai Fiben," IEEE J. Qimntitm Electron., QE- 1 7. pp. 15-22, 198 1. [67]R. C. Jones, "A New Calculus For the Treatment of Optical Systern I.," J. Op?.SOC. Am., Vol. X, pp. 488-503. [68] C. H. Prola Jr.. J. A. Pereira da Silva. A. O. Da1 Fomo, R. Passy, J. P. Von der Weid. N. Gisin. "PMD Emulaton and Signal Distortion in 2.48-Gbls IM-DDLightwave Systems," IEEE Phoronics Technology Letters, Vol. 9, No. 6. 1997. [69] "WD15 15 Series 1538/1558 nm Bi-directional Couplen," JDS Uniphase, h~p://www.jdsunph.com/htmUcatdog!productmenusrch.ch?fnd 15.html. 15 [70] James J. Refi. "Optical Fibers for Optical Networking", http://www. lucent.corn/minds/techj 1999lpdVpaper 13.pdf.
~~~~~~~~mar
25e9
100
Fiber Segments
K
Trms
td
ts
1 Frequency Points
PMD
Bit Period
System Sampling Time
100e-12 1/B
td/I
ts 1
tdil6
0.0 1
4
Noise Ievel
QAM State
A
NO
Generator
Sour#, Data
Transni tterl
Block
Variable
Source Data
Generator
I Initiai output
Io
A 3 System Model\Transmitter-A(B)\T~mitter
4 ] e1
bised &ne
filter
u
Z ~ X I
Figure A 3 Transmitter
Blocks
QASK S-amp
Parameters
M-ary number
Values M
td td
tsl
0.8
Filter type
'FIR
1
QAM baseband
1O
p p p
Block
Parameter
Variable
Value
(rand(lJ)*2-l)*pi
td* t 00000
Random Phase
Shifi
Io
Impulsa Responsa
Vin
9
++.
hZZ(1)
*+
Jotn
: a
Blocks
channel
W B
-J
Butfer2
S-FunctionJ
Partial
UnPuffeR
Values
Buffer
Buffer overlap
1
ts
1 Initial condition
Buffer size First output index
Unbuffer
10
ISK
I+ 1
1%
1 Number of channels
Input sampling time
11
t s * ((K-1 )) ~
J
S-Function3
S-function parameters
I,K, 1,i,ts*(I*(k-1 ))
i=1 J,3,4
Generator
Blocks
l
randn(t,l).*NO
td
Cyclic control
Initial o u t ~ u t
Re
*' im /
Random P h a w fhiftl
Random P h a w shift
BIocks
Parameters
Values
1 Variable
Randorn Phase ShiR
'
Initial output
Digital Clock
1 Sampling time
1 ts
+
Figure A.9a Receiver with Echo Cancellers
Echo Cancellers
Figure AlOb Echo Cancellen Table A.8 Parameters of Adaptive Cbannel Equalizers, CPI Cancellers and Echo Cancellers
-
Blocks
Nonndized LMS Adaptive filter
Parameters
Values
8
0.4
zeros(l,8)
1 td
A.ll System ModelWeceiver with Echo Canceller\Adaptive Cbannel Equalers, CPI Cancellen and Echo cacellen Wormalized LMS Adaptive Filter
1
BIOC~S
* g P+
Compluc
?+
*+
*+*
1 - *-
Complu Add
1 paramoters
Nom
1 values
Squared 2-nom le-10
Normalization bias
Buffer size
Buffer overlap Input sarnpiing time
I
n
n- l
k a d ~
O
I
Cornplex Unit Delay
Initial condition
-
Initiai condition
ic
A 1 System model\ReceiverU)ecUion .2
Parameten
Values
Complex delay
2*I*(K- l)*ls/td+6+6+1
1 0 p
QADM baseband
Il
1
1 I
O
-
tsl
QASK square-
M
td
M td
demap
QASK square-
map
Sum4 Pmdu
r
5 2
I
W
Imagine Pan
Blocks
Parameters
[Syrnbol interval, plot offset, decision offset]
Values
Bloch
Parameten
Values
50
O
Buffer size
Buffer
Buffer overlap
Input sampling time
b
td
ones( 1,50)+ones(l.50)* i
Initial condition
% % % % % %
% O % % % %
Purpose: Implements the impulse responses of optical fiber channels % t Parameters: length: length of x type: real or imagine part of impulse responses i: channel bumber Ts: sampling time % u:input signal, %
% % %
%
%%%%%%%%%%%%%%%%%%%b%%%%%%~IbO%%%%%%%%%%%%%%%%%%%%%
switch flag,
%tt%%%%%%?%%%%%tt% % Initialization %
%%%%%%Y%$$%%%%%%%% case 0 ,
%%%%a%%%%%
% Update % %%%%%%%%%% case 2 ,
case
9,
sys = il;
otherwise error([Iunhandled flag = ltnum2str(flag)l); end
if type==O xo=real(B(:,i)' 1 ; elseif type==l xO=imag(B(:,i)' 1 ; end xO=abs (x0 ; str = [1 ; ts = [Ts O];
%
end mdlInitializeSizes
% %
% % %
%
Function: FiberImp ( ) Purpose: Simulate the channel of optic fiber in the effect of PMD Output:
Frequency response H ( f )
&
t
% 0 % %
% %
format long
%%W%%W%kt%%%W%
rand-angle=(rand(2,~) -0.5)*2*pi; Wtt%%tttt Random angle tb%%%%%%tt randghase=(rand(2,K)-0.5)*2+pi; ttZ%t%%%%O Random phase %t%b%%%t%%
t%%tt%tW%
k) Ml= [exp(j*randghase (i, +co (rand-angle (Il k) exp(j*rand_phase(l,k)1 +sin(rand-dngle(1,k)); -exp(-j+rand~hase(i,k))+sin(randdangle(l,k exp(j*rand_phase( 1,k) ) +cos (rand-angle( 1,k) ) ;
MO=MO*Mk; end
end
%%%%Wb%%%%%%%%%
t%%%%%%t Magnitude of frequency response figure subplot(221),plot(f,abs(M(:J )1 ) title ( a Frequency Response : Hl1 ' ) ; ylabel ( Magnitude ' ; subplot ( 2 2 2 ) ,plot(f,abs (~(:,2) ) title('Frequency Response: H12'); ylabel ( Magnitude ' 1 ; subplot(223) ,plot(f,abs(M(:,3) 1 ) t it le ( Frequency Response : H2 1 a ) ; ylabel( 'Magnitude ; subplot(224),plot(f,abs(M(:, 4 ) 1 ) title('Frequency Response: H22') ;
l
%%%%%%%%%%%%%%
%%%%%%%% Phase of frequency response %%lttZttWt%*Z figure subplot(221),plot(f,angle(M(:,l))) ; title('Frequency Response: H1lt) ylabel( phase ; subplot(222),plot(f,angle(M(:,2) ) 1 title(Ifrequency Response: H12'); ylabel ( 'phase' ; subplot (223),plot(f,angle(M(:,3) ) 1 t itle ( ' Frequency Response: H2 1 ' ) ; ylabel ( phase ' 1 ; subplot(224),plot(f,angle(M(:,4) ) 1 title(tFrequencyResponse: H22') ; ylabel( 'phaseo ;
%%%%%%%%
Real part of frequenq response figure subplot (221) ,plot (f,real(M(:,l) 1 ) title('Frequency Response: H1lt); ylabel ( ' real ; subplot (222) ,plot (f, real(M(:, 2 ) 1 ) title ( ' Frequency Response: Hl2 ' ) ; ylabel( ' real 1 ; subplot(223),plot(f,real(~(:,3)) 1 title('Frequency Response: H21'); ylabel( real ; subplot(224),plot (f,real(M(: 1 ) ,4) title(tFrequencyResponse: H22'); ylabel( reall ;
%%%%%%%%%%%%%%
%%%%%%%%
Imagine part of Frequency response figure subplot(221)lplot(f,imag(M(:,T) ) 1 title('Frequency Response: Hllt) ; ylabel( ' imagine ' ; subplot(222)lplot(f,imag(M(:,2) ) 1 title('Frequency Response: K12'); ylabel( ' imagine ' 1 ; subplot(223),plot(f,imag(~(:,3)) 1 title('Frequency Response: H21'); ylabel ( 'imagine ' ; subplot(224),plot(f,imag(M(:,Q))) title(tFrequencyResponse: ~ 2 2 ' ) ; ylabel ( l imagine ) ;
%%%%%%%%%%%%%%
t W W 1 1 1 2 t t t Covert frequency responses to impulse responses t W t t t % 0 0 0 ~=[~(1/2+1:1, ; M(1:1/2, : ) ] ; :) m=if f t (M); m=[m(I/2+1:1, : ) ; m(l:I/2,:)1;
%%%%%%%%%%%%%%
figure subplot(221) ,plot(t,abs(m(:,l))) ; title('Imp1use Response: H1lt) ylabel ( 'Magnitude ; subplot(222),plot(tlabs(m(:,2))) title ( ' Impluse Response: Hl2 ' 1 ; ylabel ( 'Magnitudem ; subplot(223),plot~t,abs(m(:,3))) title(tImpluse Response: H21t) ; ylabel ('Magnitude') ; subplot(224) ,plot(t,abs(m(: 4 ) 1 ) , title ( ' Impluse Response : H22 ' 1 ; ylabel ( 'Magnitudet ; )
Phase of impulse response figure subpiot(221),plot(tlangle(m(: 1 ) ,l) title(lImpluse Response: H1lt) ; ylabel( 'phase') ; subplot(222),plot(t,angle(m(:,2)1 ) t i tle ( Impluse Response : H l 2 ' ) ; ylabel( ' phase ) ; subplot(223),plot(t,angle(rn(:,3)) title( 'Impluse Response: H21m) ; ylabel ( ' phase ' 1 ; subplot (224),plot(t,angle(m(:4 ) 1 ) , title(aImpluse Response: H22'1; ylabel ( 'phase' 1 ;
%%%%%%%%
%%%%%%%%
%%%%%%%%%%%%%%
Real part of impulse response figure subplot(221),plot(t,real(m(:,l) 1 ) title ( ' Impluse Response : Hl1 ' ) ; ylabel ( ' real ; s~bplot(222) ,plot(t,real(m(:,2) 1 ) title('1mpluse Response: H12'); ylabel( real ' 1 ; subplot (223),plot (t,teal(m(:, 3 ) 1 ) title('fmp1use Response: H2It) ; ylabel( real ' 1 ; subplot(224),plot (t,real(m(: , 4 ) ) ) title(afmpluseResponse: H22') ; ylabel( ' real ' ;
%SZ%%0%%%t%tt%
ttSIIISt Imagine part of impulse response figure subplot(221),plot(t,img(m(:,l))) title('Imp1use Response: H U f ) ; ylabel ( l imagine ' ; subplot (222),plot(t,imag(m(: 1 ) ,2) title('Imp1use Response: H12'); ylabel( imagine ' ; subplot(223),plot(t,imag(m(:,3) ) 1 title('Imp1use Response: H21r); ylabel ( ' imagine ; subplot (224),plot(t,imag(m(: 4 ) 1 ) , title(lImpluse Response: H22') ; ylabel( ' imagine ' 1 ; %%%t%t%%W%%%% Save filter coefficients dlmwrite ( FiberImp .dat', m) ;
t%tWebtSWtt%
%%%%%%%%%%%%%%%
function z=DGD(delta,N) t%%t%%tO8tttt%ttt%%tttttOttttt%tttt%ttbbO%%~%%%%%%OtOt%%%~OO%O%O%%O%OO%~ O t O Function: DGDO t % Purpose: Generate Maxiwellian random DGDs t % t % t O Copyright (c) 2000 by BN LU. Al1 Rights Reserved. t % Revision: 2.0
%%%ttwtt82t%ttttS~OtO%Ott%t0%ttttttt%ttt%*t%%%%O%ttOtOtt%%O%%%O%%t%~%t~
[O:N]
almaga=randn( 3, N) *de-
%%%%%%%% Maxiwellian distributed DGD %%%%%b%%%%%%%%%% y=almagacA2; zO=(y(l,: ) + y ( 2 , : ) + y ( 3 , : ) ). ^ o . s ; zl= [ ] ; for l=O:N zl= [zl sum(z0 (1,:) > l & zO (1, :) cl+l)/NI ; end z=zO*le-12;
%%%%%%%%
figure
stem(x, 21) ;