339 340

10.2 Crosstalk Between Lines Similarly, the near-end and far-end signals on the driven line are found by adding,
instead of subtracting, the even- and odd-mode terms:
When only line-1 is energized, that is, 1 0 2 0, the coupling between the lines
induces a propagating wave in line-2, referred to as crosstalk, which also has some minor 1 2 1
1 0 1 1 2
influence back on line-1. The near-end and far-end crosstalk are the values of 2 at 2 1
0 and , respectively. Setting 2 0 in (10.1.32), we have from (10.1.31):
2 1 1 2 1
1 1 1 1 1 1 1 1 2
2 0 2 2 2
2 1 2 1 1
(10.2.5)
(10.2.1)
1 1 1
1 1 1 1 1 1 2 2
2 2 2 1 1 2
2 1 2 1 2 0

where we defined 1 2 and introduced the -transform delay variables

1 2 2
. Assuming purely resistive termination impedances , we may use 1 2
2
Eq. (9.15.15) to obtain the corresponding time-domain responses: 0

These expressions simplify drastically if we assume weak coupling. It is straightfor-

1 1
2 0 1 1 2 ward to verify that to first-order in the parameters 0 0 , or equivalently, to
2 1 first-order in , we have the approximations:

1 1 0 0 0 0 0
1 1 2
2 (10.2.6)
1
(10.2.2) 0
2
1
2 1 1 2 where 0 . Because the s are already first-order, the multiple reflection terms
2 0
in the above summations are a second-order effect, and only the lowest terms will con-
1 tribute, that is, the term 1 for the near-end, and 0 for the far end. Then,
1 1 2
2 0
1 1
2 0 2 2
2 2
where 1 2. Because 0 , there will be multiple reflections even when
1
the lines are matched to 0 at both ends. Setting 0 , gives for the reflection 2
coefficients (10.1.30): 2
0 Using a Taylor series expansion and (10.2.6), we have to first-order:
(10.2.3)
0
In this case, we find for the crosstalk signals: 2 2 2 ˙ 2 ˙

˙
1 2 1
2 0 1 1 2
2 1 Therefore, 2 2 ˙ 2 , where we
ignored the second-order terms ˙ . It follows that:
1 2 1
1 1 2 1
2 1 2 0 2 2
(10.2.4) 2
1 1
1 2 2
2 2
˙ ˙
2
2 2
0
These can be written in the commonly used form:
1 2 2
1 2
2 2 0 2
0
(near- and far-end crosstalk) (10.2.7)
is the signal that would exist on a matched line-1 in the absence of line-2, 0 1 0
2
1 2, provided 0.
 341 342

where are known as the backward and forward crosstalk coefficients: The interaction between the two lines is seen better in the MATLAB movie xtalkmovie.m,
which plots the waves 1 and 2 as they propagate to and get reflected from
0 0 their respective loads, and compares them to the uncoupled case 0 0 .
0 0 (10.2.8)
2 4 0 2 0 The waves 1 2 are computed by the same method as for the movie pulsemovie.m
of Example 9.15.1, applied separately to the even and odd modes.
where we may replace 0 . The same approximations give for line-1, 1 0
and 1 . Thus, to first-order, line-2 does not act back to disturb line-1.
10.3 Weakly Coupled Lines with Arbitrary Terminations
Fig. 10.2.1 shows the signals 1 0 1 2 0 2 for a pair of
coupled lines matched at both ends. The uncoupled line impedance was 0 50 Ω. The even-odd mode decomposition can be carried out only in the case of identical lines
both of which have the same load and generator impedances. The case of arbitrary
Lm/L0 = 0.4, Cm/C0 = 0.3 Lm/L0 = 0.8, Cm/C0 = 0.7
terminations has been solved in closed form only for homogeneous media [471,474]. It
1.2 1.2
has also been solved for arbitrary media under the weak coupling assumption [481].
1 1
Following [481], we solve the general equations (10.1.7)–(10.1.9) for weakly coupled
0.8 0.8
line 1 near end line 1 near end lines assuming arbitrary terminating impedances , with reflection coefficients:
0.6 line 1 far end 0.6 line 1 far end
line 2 near end line 2 near end
0.4 line 2 far end 0.4 line 2 far end
1 2 (10.3.1)
0.2 0.2

0 0 Working with the forward and backward waves, we write Eq. (10.1.7) as the 4 4
0.2 0.2
matrix equation:
0 1 2
t/T
3 4 5 0 1 2
t/T
3 4 5
1 1 0
2 2 0
Near- and far-end crosstalk signals on lines 1 and 2. 1 0 1
2 0 2
For the left graph, we chose 0 0 4, 0 0 3, which results in the even and odd
The weak coupling assumption consists of ignoring the coupling of 1 1 on 2 2.
mode parameters (using the exact formulas):
This amounts to approximating the above linear system by:
70 71 Ω 33 97 Ω 1 01 0 1 13 0
1 0 0 0
0 17 0 19 0 99 0 88 0 175 0 05
2 0
ˆ ˆ (10.3.2)
0 0 1 0
The right graph corresponds to 0 0 8, 0 0 7, with parameters:
0 2
122 47 Ω 17 15 Ω 1 36 0 1 71 0 ˆ ˆ
0 42 0 49 0 73 0 58 0 375 0 05 Its solution is given by 0 , where the transition matrix can be
expressed in closed form as follows:
The generator input to line-1 was a rising step with rise-time 4, that is, 1
0 0 0
1 ˆ
ˆ
ˆ 1 2 2 ˆ 1 2
0 1 2
1
2
0 0 1
0 ˆ
The weak-coupling approximations are more closely satisfied for the left case. Eqs. (10.2.7) 1 2
ˆ 1 2
0 ˆ 1 2 2

predict for 2 0 a trapezoidal pulse of duration 2 and height , and for 2 ,a

ˆ
rectangular pulse of width and height 0 2 starting at : The transition matrix may be written in terms of the -domain delay variables
, 1 2, where are the one-way travel times along the lines, that is,
2 . Then, we find:
1
1 1 0 0 0 1 0
These predictions are approximately correct as can be seen in the figure. The approxima- ˆ 1 1 1
ˆ 1
0 0
2 1 2 2 1 2 2
tion predicts also that 1 0 and 1 , which are not quite true— (10.3.3)
1 0 0 1 0 1 0
the effect of line-2 on line-1 cannot be ignored completely. 1
2 ˆ 1 2 0 ˆ 1 2 2 2 0
 343 344

These must be appended by the appropriate terminating conditions. Assuming that

only line-1 is driven, we have: 2 2 2
1 2 1 1 1 2
2 0 2 2 20
1 0 1 1 0 1 1 1 1
1 1 1 1 2 2
(10.3.9)
2 0 2 2 0 0 2 2 2 1 2 1
1 2 1
1 2 1 2
2 2 2 2
1 1 1 1 2 2
which can be written in terms of the waves:
The corresponding time-domain signals will involve the double multiple reflections
1 0 1 1 0 1 1 1 1 2 1 arising from the denominators. However, if we assume the each line is matched in at
1 1 1 (10.3.4)
2 0 2 2 0 0 2 2 2 1 2 least one of its ends, so that 0, then the denominators can be
1 1 2 2
1
eliminated. Replacing by the time-derivative and each factor by a delay by
Eqs. (10.3.3) and (10.3.4) provide a set of eight equations in eight unknowns. Once
, we obtain:
these are solved, the near- and far-end voltages may be determined. For line-1, we find:
2 0 1 2 1 2
˙ 2
2
1 1 1 1
1 0 1 0 1 0
2 1 1 1
2 1 2 2 1 2 2 4
1
(10.3.5) (10.3.10)
1 ˙ ˙
1 1 1 1 2 1 2 1 2 3
1 1 1 2
2 1 1 1 1
1 2 1 2 3
where 1 1 1 2 1 1 1 1 . For line-2, we have:
where 1 1 1 2, and we used the property 2 2 0 to simplify the
¯ 1
1
2
1
1 1
1
2 2
1
¯ 1 1
1
2
1
1 1 2 1
1
2
1 expressions. Eqs. (10.3.10) reduce to (10.2.7) when the lines are matched at both ends.
2 0 2 2 20
1 1 1 1 1 2 2 2
1 1 1 1 1 1 1 1
¯ 1 2 1 1 2 1 2 ¯ 1 1 2 1 1 2 2 10.4 Coupled-Mode Theory
2 2 2 2
1 1 1 1 1 2 2 2
(10.3.6) In its simplest form, coupled-mode or coupled-wave theory provides a paradigm for the
where 20 1 2 1 2 1 1 1 2 and 2 1 2 , and we interaction between two waves and the exchange of energy from one to the other as
defined ¯ ¯ by: they propagate. Reviews and earlier literature may be found in Refs. [482–503], see also
[333–352] for the relationship to fiber Bragg gratings and distributed feedback lasers.
2 2 1 There are several mechanical and electrical analogs of coupled-mode theory, such as
¯ ˆ 2
1 1 1 2 1 2 2 1 a pair of coupled pendula, or two masses at the ends of two springs with a third spring
(10.3.7) connecting the two, or two circuits with a coupling capacitor between them. In these
2 2 1 examples, the exchange of energy is taking place over time instead of over space.
¯ ˆ 2
1 1 1 2 1 2 2 1 Coupled-wave theory is inherently directional. If two forward-moving waves are
In the case of identical lines with 1 2 0 and 1 2 0, we must strongly coupled, then their interactions with the corresponding backward waves may
take the limit: be ignored. Similarly, if a forward- and a backward-moving wave are strongly coupled,
1 2
then their interactions with the corresponding oppositely moving waves may be ignored.
lim 1 1

2 1 1 2 1 Fig. 10.4.1 depicts these two cases of co-directional and contra-directional coupling.
Then, we obtain:

1 1
¯ 1 2 0
2 0
(10.3.8)
0
¯ 0
4 Directional Couplers.
0

where were defined in (10.2.8). Setting 1 2 , we obtain the Eqs. (10.1.7) form the basis of coupled-mode theory. In the co-directional case, if
crosstalk signals: we assume that there are only forward waves at 0, that is, 0 0 and 0 0,
 345 346

then it may shown that the effect of the backward waves on the forward ones becomes Co directional coupler, δ /κ = 0 Co directional coupler, δ /κ = 0.5
1 1
a second-order effect in the coupling constants, and therefore, it may be ignored. To
see this, we solve the second of Eqs. (10.1.7) for in terms of , assuming zero initial 0.8 0.8
conditions, and substitute it in the first:
0.6 0.6
P1 (z) P1 (z)
0 0 P2 (z) P2 (z)
0.4 0.4
The second term is second-order in , or in the coupling constant . Ignoring this
term, we obtain the standard equations describing a co-directional coupler: 0.2 0.2

1 1 1
(10.4.1) 0 0
2 2 2 0 0.5 1 1.5 2 0 0.5 1 1.5 2
σ z /π σ z /π
For the contra-directional case, a similar argument that assumes the initial conditions
2 0 1 0 0 gives the following approximation that couples the 1 and 2 waves: Power exchange in co-directional couplers.

1 1 1
(10.4.2)
2 2 2 10.5 Fiber Bragg Gratings
The conserved powers are in the two cases: As an example of contra-directional coupling, we consider the case of a fiber Bragg
2 2 2 2 grating (FBG), that is, a fiber with a segment that has a periodically varying refractive
1 2 1 2 (10.4.3)
index, as shown in Fig. 10.5.1.
The solution of Eq. (10.4.1) is obtained with the help of the transition matrix :

cos sin sin

(10.4.4)
sin cos sin

where
1 2 1 2 2 2 (10.4.5) Fiber Bragg grating.
2 2
Thus, the solution of (10.4.1) is:
The backward wave is generated by the reflection of a forward-moving wave incident
cos sin sin on the interface from the left. The grating behaves very similarly to a periodic multilayer
1 1 0
(10.4.6) structure, such as a dielectric mirror at normal incidence, exhibiting high-reflectance
2 2 0
sin cos sin bands. A simple model for an FBG is as follows [333–352]:

Starting with initial conditions 1 0 1 and 2 0 0, the total initial power will
2 2 (10.5.1)
be 1 0 2 0 1. As the waves propagate along the -direction, power is
exchanged between lines 1 and 2 according to:
where 2 is the Bloch wavenumber, is the period, and represent the
2 forward and backward waves. The following transformation removes the phase factor
1 1
2
cos 2 sin 2
2 from the coupling constant:
2
(10.4.7)
2 2
2
2
sin 2 1 1
2
0 2
2 2 (10.5.2)
0
Fig. 10.4.2 shows the two cases for which 0 and 0 5. In both cases,
maximum exchange of power occurs periodically at distances that are odd multiples of
2 . Complete power exchange occurs only in the case 0, or equivalently, (10.5.3)
when 1 2 . In this case, we have and 1 cos 2 , 2 sin 2 .
 347 348

where 2 is referred to as a detuning parameter. The conserved power is given Fiber Bragg Grating, κ l = 3 Fiber Bragg Grating, κ l = 6
2 2
by . The fields at 0 are related to those at by: 1 1

0
with (10.5.4) 0.8 0.8
0
0.6 0.6

|Γ|2

|Γ|2
The transfer matrix is given by:
0.4 0.4

cos sin sin

11 12 0.2 0.2
(10.5.5)
sin cos sin 12 11
0 0
4 3 2 1 0 1 2 3 4 4 3 2 1 0 1 2 3 4
δ /κ δ /κ
where 2 2 . If , then becomes imaginary. In this case, it is more
convenient to express the transfer matrix in terms of the quantity 2 2:
Reflectance of fiber Bragg gratings.

cosh sinh sinh 2 2

at the center of the band, 0, is given by max tanh . The reflectance at the
(10.5.6)
sinh cosh sinh asymptotic band edges is given by:
2
2
2 2 2
at
The transfer matrix has unit determinant, which implies that 11 12 1. 1
Using this property, we may rearrange (10.5.4) into its scattering matrix form that relates
The zeros of the reflectance correspond to sin 0, or, , which gives
the outgoing fields to the incoming ones: 2 2 , where is a non-zero integer.
The Bragg wavelength is the wavelength at the center of the reflecting band, that
0 0 12 12 1
(10.5.7) is, corresponding to 0, or, 2, or 2 4 2 .
11 11 11
By concatenating two identical FBGs separated by a “spacer” of length 4
where are the reflection coefficients from the left and right, respectively, and is 2, we obtain a quarter-wave phase-shifted FBG, which has a narrow transmission
the transmission coefficient. We have explicitly, window centered at 0. Fig. 10.5.3 depicts such a compound grating. Within the
spacer, the waves propagate with wavenumber as though they are uncoupled.
sin
1
(10.5.8)
cos sin cos sin

If there is only an incident wave from the left, that is, 0 0 and 0, then Quarter-wave phase-shifted fiber Bragg grating.
(10.5.7) implies that 0 0 and 0 .
A consequence of power conservation, 0 2 0 2 2 2
, is The compound transfer matrix is obtained by multiplying the transfer matrices of
the unitarity of the scattering matrix, which implies the property 2 2
1. The the two FBGs and the spacer: FBG spacer FBG, or, explicitly:
2
reflectance may be expressed in the following two forms, the first being appropriate 0
11 12 11 12 11 12
when , and the second when : (10.5.10)
12 11 12 11 0 12 11
2
2 2
2
sin 2 2
sinh where the are given in Eq. (10.5.5). It follows that the matrix elements of are:
1 2 2 2 2 2 2 (10.5.9)
cos sin 2 cosh 2 sinh 2 2
11 11 12 12 12 11 11 (10.5.11)
2
Fig. 10.5.2 shows as a function of . The high-reflectance band corresponds to
The reflection coefficient of the compound grating will be:
the range . The left graph has 3 and the right one 6.
As increases, the reflection band becomes sharper. The asymptotic width of the 12 12 11 11
comp (10.5.12)
band is . For any finite value of , the maximum reflectance achieved 11
2
11 12
2 2
 349 350

where we replaced 12 and 11 1 . Assuming a quarter-wavelength spacing

2
4 2, we have 2. Replacing
, we obtain:

comp 2
(10.5.13)

At 0, we have 1 cosh , and therefore, comp 0. Fig. 10.5.4 depicts

2 2
the reflectance, | comp , and transmittance, 1 comp , for the case 2.

Compound Grating, κ l = 2 Compound Grating, κ l = 2

Forward and backward intensities in stratified medium.
1 1

0.8 0.8
forward. Similarly, the backward intensity, going from to , will be decreased by

Transmittance
Reflectance

and increased by . Thus, the incremental changes are:

0.6 0.6

0.4 0.4

0.2 0.2

or, written in matrix form:

0 0
4 3 2 1 0 1 2 3 4 4 3 2 1 0 1 2 3 4
δ /κ δ /κ
(10.6.1)
Quarter-wave phase-shifted fiber Bragg grating.
This is similar in structure to Eq. (10.5.3), except the matrix coefficients are real. The
Quarter-wave phase-shifted FBGs are similar to the Fabry-Perot resonators discussed solution at distance is obtained in terms of the initial values 0 by:
in Sec. 5.5. Improved designs having narrow and flat transmission bands can be obtained
by cascading several quarter-wave FBGs with different lengths [333–353]. Some appli- 0
with (10.6.2)
cations of FBGs in DWDM systems were pointed out in Sec. 5.7. 0

The transfer matrix is:

10.6 Diffuse Reflection and Transmission
cosh sinh sinh
11 12
(10.6.3)
Another example of contra-directional coupling is the two-flux model of Schuster and sinh cosh sinh 21 22
Kubelka-Munk describing the absorption and multiple scattering of light propagating in
a turbid medium [504–518].
where and 2 2 2 . The transfer matrix is unimodular,
The model has a large number of applications, such as radiative transfer in stellar
that is, det 11 22 12 21 1.
atmospheres, reflectance spectroscopy, reflection and transmission properties of pow-
Of interest are the input reflectance (the albedo) 0 0 of the length-
ders, papers, paints, skin tissue, dental materials, and the sea.
structure and its transmittance 0 , both expressed in terms of the output,
The model assumes a simplified parallel-plane geometry, as shown in Fig. 10.6.1.
or background, reflectance . Using Eq. (10.6.2), we find:
Let be the forward and backward radiation intensities per unit frequency interval
at location within the material. The model is described by the two coefficients
21 11 sinh cosh sinh
of absorption and scattering per unit length. For simplicity, we assume that are
22 12 cosh sinh
independent of . (10.6.4)
Within a layer , the forward intensity will be diminished by an amount of 1
due to absorption and an amount of due to scattering, and it will be increased by 22 12 cosh sinh
an amount of arising from the backward-moving intensity that is getting scattered
These are related to the normalized Kubelka [509] variables , .
 351

The reflectance and transmittance corresponding to a black, non-reflecting, back-

ground are obtained by setting 0 in Eq. (10.6.4):

21 sinh
0
22 cosh sinh
(10.6.5)
1
0
22 cosh sinh

The reflectance of an infinitely-thick medium is obtained in the limit :

2
1
(10.6.6)
2 2

For the special case of an absorbing but non-scattering medium ( 0 0), we

have and the transfer matrix (10.6.3) and Eq. (10.6.4) simplify into:

0 2
(10.6.7)
0

These are in accordance with our expectations for exponential attenuation with dis-
tance. The intensities are related by 0 and 0 . Thus, the
reflectance corresponds to traversing a forward and a reverse path of length , and the
transmittance only a forward path.
Perhaps, the most surprising prediction of this model (first pointed out by Schuster)
is that, in the case of a non-absorbing but scattering medium ( 0 0), the trans-
mittance is not attenuating exponentially, but rather, inversely with distance. Indeed,
1
setting and taking the limit sinh as 0, we find:

1 1 1
(10.6.8)
1 1 1

In particular, for the case of a non-reflecting background, we have:

1
0 0 (10.6.9)
1 1

10.7 Problems
10.1 Consider the practical case in which two lines are coupled only over a middle portion of
length , with their beginning and ending segments being uncoupled, as shown below:

Assuming weakly coupled lines, how should Eqs. (10.3.6) and (10.3.9) be modified in this
case? [ Hint: Replace the segments to the left of the reference plane and to the right of
plane by their Th évenin equivalents.]