في نظرية النظم و علم الضبط و في تعاملنا مع نظم ذات عدة مداخل و عدة مخارج أي نظم من نوع ميمو (mimo = multiple input multiple output) فإن مفهوم مصفوفة التحويل transfer matrix يحل محل مفهوم دالة تحويل . وهي صياغة على شاكلة مصفوفة بلوك-توپلتس لمعادلة في متغيرين، تميّز refinable functions . Refinable functions تلعب دوراً هاماً في نظرية المويجات ونظرية العناصر المتناهية .
For the mask
h
ℎ
{\displaystyle{\displaystyle h}}
, which is a vector with component indexes from
a
𝑎
{\displaystyle{\displaystyle a}}
to
b
𝑏
{\displaystyle{\displaystyle b}}
,
the transfer matrix of
h
ℎ
{\displaystyle{\displaystyle h}}
, we call it
T
h
subscript
𝑇
ℎ
{\displaystyle{\displaystyle T_{h}}}
here, تـُعرَّف كالتالي
(
T
h
)
j
,
k
=
h
2
⋅
j
-
k
.
subscript
subscript
𝑇
ℎ
𝑗
𝑘
subscript
ℎ
⋅
2
𝑗
𝑘
{\displaystyle{\displaystyle(T_{h})_{j,k}=h_{2\cdot j-k}.}}
وحرفياً
T
h
=
(
h
a
h
a
+
2
h
a
+
1
h
a
h
a
+
4
h
a
+
3
h
a
+
2
h
a
+
1
h
a
⋱
⋱
⋱
⋱
⋱
⋱
h
b
h
b
-
1
h
b
-
2
h
b
-
3
h
b
-
4
h
b
h
b
-
1
h
b
-
2
h
b
)
.
subscript
𝑇
ℎ
subscript
ℎ
𝑎
absent
absent
absent
absent
absent
subscript
ℎ
𝑎
2
subscript
ℎ
𝑎
1
subscript
ℎ
𝑎
absent
absent
absent
subscript
ℎ
𝑎
4
subscript
ℎ
𝑎
3
subscript
ℎ
𝑎
2
subscript
ℎ
𝑎
1
subscript
ℎ
𝑎
absent
⋱
⋱
⋱
⋱
⋱
⋱
absent
subscript
ℎ
𝑏
subscript
ℎ
𝑏
1
subscript
ℎ
𝑏
2
subscript
ℎ
𝑏
3
subscript
ℎ
𝑏
4
absent
absent
absent
subscript
ℎ
𝑏
subscript
ℎ
𝑏
1
subscript
ℎ
𝑏
2
absent
absent
absent
absent
absent
subscript
ℎ
𝑏
{\displaystyle{\displaystyle T_{h}={\begin{pmatrix}h_{a}&&&&&\\
h_{a+2}&h_{a+1}&h_{a}&&&\\
h_{a+4}&h_{a+3}&h_{a+2}&h_{a+1}&h_{a}&\\
\ddots&\ddots&\ddots&\ddots&\ddots&\ddots\\
&h_{b}&h_{b-1}&h_{b-2}&h_{b-3}&h_{b-4}\\
&&&h_{b}&h_{b-1}&h_{b-2}\\
&&&&&h_{b}\end{pmatrix}}.}}
The effect of
T
h
subscript
𝑇
ℎ
{\displaystyle{\displaystyle T_{h}}}
can be expressed in terms of the downsampling operator "
↓
↓
{\displaystyle{\displaystyle\downarrow}}
":
T
h
⋅
x
=
(
h
*
x
)
↓
2
.
⋅
subscript
𝑇
ℎ
𝑥
ℎ
𝑥
↓
2
{\displaystyle{\displaystyle T_{h}\cdot x=(h*x)\downarrow 2.}}
الخصائص
Actually not
n
-
2
𝑛
2
{\displaystyle{\displaystyle n-2}}
convolutions are necessary, but only
2
⋅
log
2
n
⋅
2
subscript
2
𝑛
{\displaystyle{\displaystyle 2\cdot\log_{2}n}}
ones, when applying the strategy of efficient computation of powers. Even more the approach can be further sped up using the Fast Fourier transform .
From the previous statement we can derive an estimate of the spectral radius of
ϱ
(
T
h
)
italic-ϱ
subscript
𝑇
ℎ
{\displaystyle{\displaystyle\varrho(T_{h})}}
. It holds
ϱ
(
T
h
)
≥
a
#
h
≥
1
3
⋅
#
h
italic-ϱ
subscript
𝑇
ℎ
𝑎
#
ℎ
1
⋅
3
#
ℎ
{\displaystyle{\displaystyle\varrho(T_{h})\geq{\frac{a}{\sqrt{\#h}}}\geq{\frac%
{1}{\sqrt{3\cdot\#h}}}}}
where
#
h
#
ℎ
{\displaystyle{\displaystyle\#h}}
is the size of the filter and if all eigenvalues are real, it is also true that
ϱ
(
T
h
)
≤
a
italic-ϱ
subscript
𝑇
ℎ
𝑎
{\displaystyle{\displaystyle\varrho(T_{h})\leq a}}
,
where
a
=
∥
C
2
h
∥
2
𝑎
subscript
norm
subscript
𝐶
2
ℎ
2
{\displaystyle{\displaystyle a=\|C_{2}h\|_{2}}}
.
انظر أيضاً
المراجع
Strang, Gilbert (1996). "Eigenvalues of
(
↓
2
)
H
fragments
fragments
(
↓
2
)
H
{\displaystyle{\displaystyle(\downarrow 2){H}}}
and convergence of the cascade algorithm". IEEE Transactions on Signal Processing . Vol. 44. pp. 233–238.
Thielemann, Henning (2006). Optimally matched wavelets (PhD thesis). (contains proofs of the above properties)