62 6. BINARY SWITCHING NETWORK SPECTRA
Figure 6.3: Butterfly diagram for Example 6.1 calculation.
6.4.2 WALSH TRANSFORM FOR VECTOR-VALUED SWITCHING
FUNCTIONS
When a switching network is modeled in the vector space, the Walsh transform can also be applied
as in the previous section. For the r D 2 binary case, the values h0jand h1j can be mapped to values
along the unit circle in an analogous manner to that of the scalar-valued case. is mapping is
obtained by applying the H
1
transformation to each value so that it may be represented in the
spectral domain.
h0jH
1
D
1 0
1 1
1 1
D
1 1
h1jH
1
D
0 1
1 1
1 1
D
1 1
e vector space model also allows for the values ht j and h¿j. ese values in the Walsh
domain are expressed as follows.
6.4. WALSH TRANSFORM 63
h¿jH
1
D
0 0
1 1
1 1
D
0 0
htjH
1
D
1 1
1 1
1 1
D
2 0
In this manner, the Walsh transform matrix H
1
can be viewed as the collection of row
vectors representing the spectral values corresponding to h0j and h1j. e Hasse diagram of the
vector-valued constants in the switching domain and the Walsh spectral domain are given in
Figure 6.4.
Figure 6.4: Hasse diagram of values in the switching and Walsh spectral domain for r D 2.
e Walsh spectrum of a switching network when modeled in the vector space can analo-
gously be computed as in the case of the more conventional scalar model described in the previous
section. A column vector of all possible output values can be formulated and multiplied with the
Walsh transform matrix. However, each component of the column vector is also a row vector
since a row vector corresponds to each distinct output response. erefore, the column vector
containing all network output responses is itself a matrix, and this matrix is exactly the transfer
matrix T
f
characterizing the switching network. e transfer matrix row vectors are mapped to
corresponding values as given in the Hasse diagram in Figure 6.4 for consistency with the defi-
nition.
eorem 6.2 Walsh Spectrum of Switching Network
e Walsh spectrum of switching network modeled in the vector space is equivalent to the product of the
mapped transfer matrix T
s
and the Walsh transformation matrix H
n
.
Proof. By definition, the spectrum of a switching function is the product of the Walsh transfor-
mation matrix with a column vector whose components are all possible function values. By the
property of truth table isomorphism, the transfer matrix characterizing a switching network con-
sists of row vectors that are all possible output responses of the network. By the definition of the
spectrum of network, the spectrum of the network is
64 6. BINARY SWITCHING NETWORK SPECTRA
S
f
D H
n
T
s
:
Example 6.3 Walsh Spectrum of Example Function
Consider the single output switching characterized by the truth table that contains both the scalar
model and vector model values.
x
2
x
1
x
0
f hf j
0 0 0 1
0 1
0 0 1 0
1 0
0 1 0 1
0 1
0 1 1 0
1 0
1 0 0 1
0 1
1 0 1 1
0 1
1 1 0 0
1 0
1 1 1 1
0 1
From the definition of the Walsh spectrum, a column vector is formed with components
equivalent to the truth table values of the function hf j. is column vector is equivalent to the
transfer matrix characterizing hf j.
T
f
D
2
6
6
6
6
6
6
6
6
6
6
6
4
0 1
1 0
0 1
1 0
0 1
0 1
1 0
0 1
3
7
7
7
7
7
7
7
7
7
7
7
5
e mapped version of the transfer matrix is denoted as T
s
:
6.4. WALSH TRANSFORM 65
T
s
D
2
6
6
6
6
6
6
6
6
6
6
6
4
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
3
7
7
7
7
7
7
7
7
7
7
7
5
e Walsh spectrum is then computed as S
f
D H
f
T
s
:
S
f
D
2
6
6
6
6
6
6
6
6
6
6
6
4
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
3
7
7
7
7
7
7
7
7
7
7
7
5
2
6
6
6
6
6
6
6
6
6
6
6
4
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
3
7
7
7
7
7
7
7
7
7
7
7
5
D
2
6
6
6
6
6
6
6
6
6
6
6
4
8 2
0 2
0 2
0 2
0 C2
0 6
0 C2
0 C2
3
7
7
7
7
7
7
7
7
7
7
7
5
eorem 6.2 leads to the following definition.
Definition 6.4 Spectral Response Matrix
e spectral response matrix T
s
characterizes a switching network in the spectral domain and is
equivalent to the transfer matrix T with each row vector mapped to their corresponding spectral
representations.
Each spectral coefficient can be calculated individually through a multiplicative operation
with a specific row vector. e row vector represents a switching network input stimulus expressed
where each individual primary input value is expressed in the spectral domain. If the input stim-
ulus is represented in the switching domain, the spectral response matrix yields a corresponding
output response with each primary output expressed in the spectral domain. Example 6.5 illus-
trates these two uses of the spectral response matrix.
Example 6.5 Spectral Response Matrix Calculations
Consider the example single-output switching network characterized by the following truth table.

Get Modeling Digital Switching Circuits with Linear Algebra now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.