0% found this document useful (0 votes)
125 views436 pagesPub - Wavelets and Filter Banks PDF
Uploaded by
Kamalendu C MenonCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
125 views436 pagesPub - Wavelets and Filter Banks PDF
Uploaded by
Kamalendu C MenonCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 436
Preface
‘This book is about filter banks and wavelets. Those are new ways to see and represent a signal.
They are alternatives to the Fourier transform, using short wavelets instead of long waves. We
will explain the advantages (and disadvantages!) of the new methods. ‘The final decision will
depend on the application itself, the actual signal, and its bandwidth,
‘To design and understand wavelets we still use Fourier techniques —the connections be-
tween time and frequency. This idea remains at the center of signal processing. Wavelets are
“alternatives” rather than replacements. ‘The classical transforms will survive very well. But
other ideas have come quickly forward, to be understood and applied
In a word, the new transforms are much more local. An event stays connected to the time
when it occurs. Instead of transforming a pure “time description” into a pure “frequency de-
scription”, the new methods find a good compromise —a time-frequency description. This is
like a musical score, with specified frequencies at specified times. Remember that the Heisen-
berg uncertainty principle stands in the way of perfection! We lose accuracy in time when we
gain accuracy in frequency. The musician cannot and would not change frequencies instantly
But our eyes and ears succeed to give location as well as frequency, and the new wavelet trans-
forms have the same purpose.
‘The extreme case is an instantaneous impulse, with all frequencies in equal amounts. Its
Fourier transform has constant magnitude over the whole spectrum. By contrast, a wavelet trans-
form will involve only a small fraction of the wavelets —those that overlap the impulse. Fig-
ure 0.1 shows a sum of two extreme cases—an impulse and a pure wave 2cos wn. The Fourier
transform spreads the impulse while it concentrates the wave (at frequencies @ and —w of
et +e), The wavelet transform in the third figure is large at the rime of the impulse and
also large atthe frequency of the wave.
Purpose of the Book
‘There are already good books on this subject. The ideas and applications are beautiful, and the
‘word has spread. The bibliography lists many of those books, which have special strengths. Our
purpose is different, We believe that a textbook is needed. Our text explains filter banks and
‘wavelets from the beginning —in several ways and at least two languages. The examples and
«exercises come from our courses at M.L'T. and Wisconsin, which brought students from all over
‘engineering and science.
‘Whether you are working individually or in class, we hope this book clarifies the central
ideas. Implicit in that goal isthe recognition that we cannot describe every filter and prove every
theorem. The book concentrates on the underpinning of the subject, which is now stable. There
isa special “glossary” to organize and define the terms that are constantly used, some from signal
processing and others from mathematics. The central idea isa perfect reconstruction filter bank,
with properties and purposes selected from this list: x Wavelets and Fitter Banks
WT
|
Ta
te TT. te TT
Figure 0.1: Impulse plus sinusoid, in time and frequency and time-frequency/time-scale).
Properties: Orthogonality, symmetry, short length, good attenuation.
Purposes: Audio/video compression, echo cancellation, radar, image analysi
cations, medical imaging, ....
Design and analysis techniques: Time domain, frequency domain, z-transform.
If the reader is willing, we would like to develop specific examples in this preface. Tradi-
tionally, the opening pages thank those who helped to create the book. Our debt to friends will
soon be very gratefully acknowledged. But first, we go directly to transforms.
communi
‘Transforms
Start with the basic idea and its purpose. The transform of a signal (a vector) is a new repre=
sentation of that signal. The components x(0), x(1),x(2),x(3) of a four-dimensional signal are
replaced by four other numbers. Those numbers (0), »(1), ¥(2).»(3) are combinations of the
+'s. Our transforms ate linear, so these are linear combinations — for example sums and differ.
10) = xO +21) 92) = 2Q) +20)
HD = x@)-x(1) 3) = 2).
‘What is the purpose of the y's? And can we get back to the x's? The second question is easier
and we answer it first.
‘This transform can be inverted. If you add the equations for (0) and y(1), you get 2x(0).
‘Subtracting those equations yields 2x(i). The inverse transform uses addition and subtraction
‘ike the original), and then division by 2:
(0) 0560) +y()) x02) 0502) +y(3))
=) 0500) -¥)) xB) = 05 H2)—¥H). Preface
‘The y's allow perfect reconstruction of the 2's, by the inverse transform. Looking ahead for a
brief moment, we divide transforms into three groups:
(@) Lossless (orthogonal) transforms (orthogonal and unitary matrices)
(b) _Invertible (biorthogonal) transforms (invertible matrices)
© Lossy transforms (not invertible)
‘A lossless unitary transform is like a rotation. The transformed signal has the same length as
‘the original. This is true of the Fourier transform and all its rel versions (DCT = discrete cosine
transform, DST = discrete sine transform, HT'= Hartley transform). The same signal is measured
slong new perpendicular axes.
For biorthogonal transforms, lengths and angles may change. The new axes are not nec
cessarily perpendicular, but no information is lost. Perfect reconstruction is still available. We
just invert. Orthogonal wavelets give orthogonal matrices and unitary transforms, biorthogo-
nal wavelets give invertible matrices and perfect reconstruction. These transforms don’tremove
any information (or any noise), they just move it around — aiming to separate out the noise and
sumz(0)
sums y(0),9(2)
iference 2(2)
Ly ditterences y(1),¥(3)
‘This two-step wavelet transform is executed by a product of two matrices:
1 1 L020 1 tad
1 Aiet 080 1-1 0 0
1 -1 Ore O edad fee) le lee lead
| [sO Osean Oe On tat xiv Wavelets and Filer Banks
‘That is invertible! To draw the low-graph ofthe inverse, just reverse the arrows. To invert the
matrix on the right, multiply the inverses ofthe matrices on th left (in reverse order of course).
The wavelet transform becomes unitary as before, when we divide sums and differences by
2. You wil se this amber appearing throughout the book, to compensate for scale changes
Perhaps one more transform could be mentioned. Itis the Walsh-Hadamard transform, which
iterates also on the differences y(1) and y(3). Instead of the “logarithmic tree” for wavelets, we
havea complete binary tree for Walsh-Hadamard:
> sumz(0) =0
‘sums y(0) and y(2)
> difference 2(2) = 4.4
p> sumz(1) = 0.4
‘+ differences y(1), ¥(3) —|
L+ difference 2(3) = 0
This also counts as a wavelet packet (those include every binary tree). You would not want
{o miss the “Hadamard matrix” for this transform, which is exceptional. All entries are 1 and
=I and all rows are orthogonal:
1 1 1 Poro4ao4
i 1 1 a} to-too1 -t
1 -1 Tice eee eed sueeegaees penta)
' 1-1 fede teeny
When divided twice by -/2 this matrix is orthogonal and also symmetric. The original x's can
bbereconstructed from the y's or the w's or the wavelet outputs y(1), ¥(3),2(0),z(2), From com.
pressed outputs we get approximate (but good) reconstruction,
Don't forget the disadvantage compared to wavelets! All sixteen entries are nonzero. The
complete tree takes more computation than the wavelet tree. For signals of length n (a power of
(0) the Hadamard matrix has n? nonzeros and the wavelet matrix has only n+ logan. Those
are the costs without recursion, jumping from the original x's directly to the final z's
‘When the computations ae recursive, as they always should be, we have a product of very
sparse matrices. The flow-graph gives 2n log, n for the complete Walsh-Hadamard transforrs
exactly matching the Fast Fourier Transform). This is good but wavelets are slightly better
‘The transform to 2(0), y(1), 2(2), »(3) computes the sum and difference of n — | pairs. ‘This
needs only 2 — 2calls to memory.
‘The wavelet transform achieves the Holy Grai of complexity theory (or simplicity theory).
The transform is an O(n) computation. But does it separate the true signal from noise? Does it
allow compression of 4:1 or 8:1 or 20:1? This sum-difference transform is analyzed further in
Chapter 1 —itis such a good example — but we admit here that itis too simple. Better filters
ae needed, and they lead to better wavelets (still with O(n) operations). That is the subject of
‘our book. Protece ~
Applications.
‘he choice of waves or wavelets, Fourier analysis or filter analysis, depends onthe signal. It
tmust Signals coming from diferent applications have different characteristics. Ii helpful to
see a broad picture:
Audio: Use many subband filters or a windowed (short time) Fouries transform.
‘Speech: ‘The time variation is irregular and requires nonuniform imervals,
Images: Finite length s
‘B0es with symmetric filters.
‘Video: Use motion prediction (optical low of images) or space-time filtering,
‘The best choice for medical imaging is not clear. The legal questions arising from lossy com-
Pression are not clear either. Identification from synthetic aperture radar (SAR images) is an
romeo problem. So is de-noising, which is atthe heart of signal representation, Ultimately
we are trying to choose a good basis,
‘The problem ist represent typical signals with a small number of conveniently computable
Aanctions: The tational bases (Fourier, Bessel, Legendre, ...) come from differential equa.
fons, Wavelets do not come from differential equations! One reason is, those equations don't
include dilation. A dilation equation $() = 5. 2h(K)$(2t — k) involves two time seales, ie
Selution #(¢) is nonzero only on a finite interval. Then @ (2) is nonzero on half ofthat interval,
‘The basis is localized. It is quickly (and recursively) computed from the numbers h(t). ‘Those
‘ue the coefficients in the corresponding lowpass filter.
‘This relation of filters to functions isthe heart of the book.
Acknowledgments
“Many friends have created this theory, A lot of them also helped us to write about it. The bibli-
ceraphy points to their original work (we wish it were possible to be complete — that is beter
attempted electronically). Our personal thanks must begin with Vasily Strela and Chris Heil,
‘They read most of the words and discussed all the ideas. Their unselfish help is dceply appreci.
ated,
Othe fiends will sec, at specific places in the book, exactly where their ideas and sugges-
tions made a difference. We are enormously grateful to Kevin Amaratunga, Ross Barmish,
Christopher Brislawn, Charies Chui, Albert Cohen, Adriannus Djohan, Dave Donoho, Jefe
Geronimo, Doug Hardin, Peter Heller, Tom Hopper, Angela Kunoth, Seng Luan Lee. Michse]
{Gghtstone, Bric Majani, Stéphane Mallat, Henrique Malvar, Ricardo de Queiror ianhong Shen,
Wim Sweldens, Pankaj Topiwala, Steffen Trautmann, Chi Wah Kok, Victor Wickerhauser, and
Zhifeng Zhang. There is no way to thank you enough for such generous help.
We want to recognize separately the outstanding leadership of four authors, P. P. Vaidya:
nathan and Martin Vettrl in signal processing and Ingrid Daubechies and Yves Meyer in mathe
matics. The two subjects are thoroughly mixed, thanks to these four! Their work has been
‘model in every way, personal ((o us) as well as scientific (for all.
Our writing of the book was shared in a natural way— words mostly from the fist author,
fier designs and applications from the second author, ideas from both. Vasily Strela and Robert
Becker helped to get Versions 1.0 to 9.9 into TEX, and Martin Stock made it more beautifal
‘The whole project grew out of our courses and workshops (which will continue). Those were
tremendous sources of inspiration, thanks to the participants, xvi Wavelets and Filer Banks
‘This subject is not just theory. The ideas have to be implemented, and MATLAB is the out-
standing way to do that. [ts a pleasure to have this book closely linked to the Wavelet Toolbox
offered by MathWorks. Exercises that use this Toolbox are in the book: the reader can see the
wavelets and the multiresolution they produce. A more extensive Wavelet Manual will come
from the same sources: Wellesley-Cambridge Press and MathWorks,
For correspondence about this whole subject, and for comments and corrections of any kind,
the authors thank the readers. Especially we thank Jill and Thuy Duong for their patience and
support. ICs finished at last! We hope you enjoy this book.
Gilbert Stang and Truong Nguyen
M.LT. and Wisconsin, November 1995
[email protected] and nguyendece.wisc.edu
hetp: //saigon.ece.wisc.edu/“waveweb/Tutorials/book.html Guide to the Book
‘This book has a two-part subject. One partis discrete the other is continuous. In discrete time
‘we develop the idea and applications of filter banks. In continuous time we have scaling func-
sions 6(¢) and wavelets w(t). By a natural limiting process, iteration ofthe lowpass filter leads to
the scaling function. One highpass filter then produces a wavelet. Our goal is to make this con-
nection clear. We find the conditions on the diserete coefficients that lead to good filter banks
‘and good wavelets.
Historically and mathematically, the filters come first. Perfect reconstruction filter banks
were developed in the early 1980's. The excitement around wavelets started later (and grew
quickly). This excitement was not universal— designers of filter banks naturally asked what
‘was new. Part of the answer is precisely in that process of iteration. For a filter to behave well
in practice, when it is combined with subsampling and repeated five times, it must have an extra
property —not built into earlier designs. This property expresses itslf inthe frequency domain
by a sufficient number of “zeros at 2”. Then the frequency band can be successfully separated
into five octaves.
The underlying problem is to choose a good basis. We want to represent a signal well, by
‘a small number of basic signals. These can be sinusoids and they can be wavelets. On a dis-
crete grid, » = 7 is the highest frequency at which a signal can oscillate, Those oscillations
x(n) = e"* = (1)? are stopped by the lowpass filter with a “zero at x”, The highpass fil-
ter lets fast oscillations through, and the synthesis filters can reconstruct the exact input. But
compression may come between analysis and synthesis. Frequencies that are barely represented
will be intentionally ost. That mostly means high frequencies but the filterbank is impartial —
it keeps the basis functions that are important to the specific signal. We want to show when,
and why, filter banks and wavelets are effective in reconstruction and signal representation and
compression.
Filter Banks
Some readers will begin this book with Chapter 1. Others will jump forward toa topic of partic-
ular interest. This brief guide is for both, especially to tell the fist readers what is coming and
the second group where to look. We are pointing to places where preparation and explanation
come together, to design and study new structures,
For filter banks, that place is Section 4.1. There we identify the two conditions for perfectre-
construction (in the absence of lossy compression). One condition removes distortion, the other
condition removes aliasing. The anti-distortion condition applies tothe products FoHfo and Ft
along the channels of the filter bank. Then the anti-aliasing condition controls how those prod-
‘ucts can be separated into the foor filters.
“The design of a perfect reconstruction filterbank isa choice of Foff and then a factorization.
‘To understand the conditions on distortion and aliasing, we apply the techniques of multirate
fikering. Those techniques are explained in Chapters 1-3, with examples throughout. Of course vill Wavelets and Filter Banks
filters are to be defined! But we can go forward even now, to illustrate a filter bank that gives
perfect reconstruction
‘The analysis bank is on the left. It has a lowpass filter Ho, and a highpass filter H;, and dec-
imation by (| 2)—which removes the odd-numbered components after filtering. The analysis
‘bank yields two “half-length” outputs. Then the synthesis bank on the right begins with the up-
sampling operation (t2)— which inserts zeros in those odd components:
LHe J2;- ow 442 Pa
input x La pL. Sole U output §
‘Two-channel filer bank: Separate the input into frequency bands (iter and downsample).
‘Then reassemble (upsample and filter).
‘The gap in the center indicates where the subband signals are compressed or enhanced, The
applications of this structure are extremely widespread. We believe that any reader interested in
signal processing (and image processing) will find that filter bank analysis is extremely useful.
‘The filters Hy, Hi, Fo, and F are liness and time-invariant. The operators (1.2) and (2) are
not time-invariant. These multirate operations are responsible for aliasing and for imaging —
they create undesirable and extraneous signals thatthe filters must cancel, To understand how
that happens, and to design good filters. we use the tools developed in Chapters 1-3: especially
transformation tothe frequency domain and z-domain. We try toexplain the analysis of multirate
filtering, with (J 2) and (2), clearly and memorably.
“The theory and the design of filter banks and wavelets will dominate Chapters 4-6. This is
the heart of the book. The structure of an orthogonal bank is very special, and the next figure
shows how the filters are related. For length 4 all filters use the four coefficients a, b,c, that
Daubechiies derived:
ared [12 12} dena
x) x0~
‘The form of an omthogonal filter bank with Four coefficients.
How did she choose a, b, c,d? Part of the answer will have to wait, but here is the essential
idea. The product along the top channel gives a particular “halfband filter”:
@, b,c, d) # @,, 8, a)
(1, 0, 9, 16, 9, 0, 1/16,
‘This convolution is a multiplication of two polynomials, when a, b,c, d are the coefficients:
fat be bee? +e) d bee! + bz tae) =
(-1 +927? + 1627 + 92-4 = 2°9)/16,
The four coefficients are pleasant to calculate. The serious job in Chapter 4 is to explain what is
special about that 6"-degree polynomial in which z~! and z~® are missing. ‘Guide to the Book xix
A filter bank also gives perfect reconstruction if itis biorthogonal, This design is less re-
stricted, The product FoHo must skip the same odd powers of z!, but Fo does not have to be
the transpose (the flip) of Ho. Here are specific numbers forthe filter coefficients —not the only
choice and maybe not the best. They show how the filters Fo and Fon the synthesis side are
related to the analysis filters Hi and Ho (by alternating signs)
Fo
Pi2p tz oe
Ff voxe-0)
Lf t2H n2-621
me %
‘A biorthogonal filter bank: Perfect reconstruction with 3 delays.
For filters, we stop here. This time Fo(z) = | + 22~' + 2°?. Multiplied by Ho(2) it gives the
same important 6"-degree polynomial as before. To understand why the zero coefficients are
necessary in that polynomial, and why —7 and & are desirable, I am afraid that you have 10
read the book!
‘Our discussion went this far further than we intended) so as to make a basic and encouraging
point: The construction of new filter banks need not be complicated. This subject is accessible
to new ideas and experiments.
Wavelets
Wavelets are localized waves. Instead of oscillating forever, they drop to zero, They come from
the iteration of filters (with rescaling). The link between discrete-time filters and continuous-
time wavelets is inthe limit of a logarithmic filter tree:
.- Scaling function (1)
dH] <
dD Ho ‘
Hp aay * Wavelet w(t)
fo 2 By
2H -—
‘Scaling function and wavelets from iteration ofthe lowpass filter,
Scaling functions and wavelets have remarkable properties. They inherit orthogonality, or
biorthogonality, from the filter bank. Because of the repeated rescaling that produces them,
wavelets decompose a signal into details at all scales. The wavelet w(t) and its shifts w(t—k) are
atunit scale. The wavelets w(2/t) and w(2/1 — k) are at scale 2~/. The biorthogonal functions
(0) and G1) come from iterating the synthesis bank.
Wavelets produce a natural multiresolution of every image, including the all-important
edges. Where the low frequency part of the Fourier transform is often a blur, the output from
the lowpass channel is a useful compression. Xx Wavelets and Filer Banks
Sections 5.5 and 6.2 study the particular wavelets created by Ingrid Daubechies, They are
orthogonal, with the advantages and limitations that this property brings. Sections 4.1 and 6.5
study biorthogonal alternatives, which come from different factorizations of the same polyno-
‘ial (as above). This polynomial corresponds to a “maxtlat halfband filter”, and we hope you
will like the connections.
More than that, we hope you enjoy the whole book. This subject isa beautiful combination of
mathematical analysis and signal processing applications. The analysis and the applications are
based on designs that give perfect reconstruction. To explain both sides of this subject, we need
words from mathematics and words from digital signal processing. The Glossary at the end is a
dictionary of their meanings. Above all we need ideas from both sides, and from a tremendous
range of application areas. It is to the understanding of filters and wavelets, and the growth of
successful applications, that this book is dedicated,
‘Summary of the Theory
‘There are four conditions that play a central part in this book. Because of their importance we
/highlight them here. They apply directly to the coefficients in the filter banks — and the conse-
‘quences are felt (after iteration!) inthe scaling functions and wavelets. Here are the four condi-
tions — some might say in decreasing order of importance:
PR Condition Perfect reconstruction.
‘The synthesis bank inverts the analysis bank, with ¢ delays.
Biorthogonal banks with no aliasing and no distortion.
Orthogonality.
‘The analysis bank is inverted by its transpose.
‘The wavelets are orthogonal (o all their dilates and translates.
ip Accuracy of onder p for approximation by scaling functions #(t — k).
P vanishing moments in the wavelets,
pth order decay of wavelet coefficients for smooth f(t)
Condition E Eigenvalue condition on the cascade algorithm.
Determines convergence to (t) and smoothness of wavelets.
Equivalent to stability ofthe wavelet basis.
One step further and this Guide is ended. ‘The four fundamental conditions will be stated
explicitly for a two-channel filter bank, with the sections in which they appear. We continue to
use the polynomials Ho(z), H(z), Fo(z), and F\(z), whose coefficients come directly from the
filters. By convention, these are polynomials in z~", and the lowpass analysis filter is represented
by Hole) = (0) + A(z“! +--+ + A(W)2-™, Here are the conditions that give filters and
wavelets with good properti
1. Perfect Reconstruction (PR condition in Section 4.1)
Fo(2) Ho(2) + Fiz) H(z) = 22" and F(z) Ho(~z) + Ful) Hy(—2) = 0.
‘The second equation gives the anti-aliasing choices F(z) = Hy(—z) and F,(z) = — Ho(—2). Guide to the Book xxi
2 Orthogonality (Condition 0 in Section 5.3)
‘The filter coefficients are reversed by F(z) = 27 Ho(e~') and Fy(e) -+ 27" Hy(z~!), Then
perfect reconstruction depends on the “double-shift orthogonality” ofthe lowpass coefficients
Aw:
LAG) he + 2n) = 6(n).
In terms of the polynomials this is Ho(2) Ho(2~)) + Ho(—2) Ho(—
3. Accuracy of order p (Condition A,, in Sections 5.5 and 7.1)
‘The lowpass filter has a zero of order p at z = 1
Lte
2
Hye) = (7) 08).
4. Convergence and Stability (Condition E in Section 7.2)
‘The transition matrix 7 has 4 = 1 as simple eigenvalue and all other }A(T)| <1
Final note: The sixth degree polynomial in the examples above has four zeros at z =
149277 + 16279 + 92-4 =~ 127) (14 4 *)
‘These zeros give flat responses near « = x and also w = 0. The absence of z~? and 2-5 is the
‘key 1 perfect reconstruction. Polynomials of higher degree, also with zeros at : = —1 and also
with only one odd power, factor into F(z) Ho(2) to give the best filters for iteration. In the limit,
‘of the iterations, these filers give good wavelets. Filters and Matrices
f= (W(0), ..., A(N)) Causal FIR lowpass filter (impulse response) with > ACE)
‘Toeplitz matrix with A(R) onthe kth diagonal: Hy = AC ~ j)
= Daw ete (frequency response in reduced notation)
SAC eH (signal processing notation)
H(@=ScMk)z- Transfer function in the z-domain with z = e/*
y = Hx = hx corresponds to ¥(z) = H()X(z) Convolution Rule
S Delay gives (Sh)(k) = A(k—1) Time Invariance is SH = HS
5-1 Advance gives (S-!)(k) = A(k-+ 1) Multiply transform by
(42) Downsampling operator (J 2)h = even = (H(O), WC), (4), «
(12)8-1h = hess = (HCD), AGB), H(S), «-.) Odd phase of
Hy(2) = (Heyen(2) Hoss(2)1 = [LA2H2-* CAAK+ e+] Polyphase representation
(12, Upsampling operator (+2)» = [»(0) 0 »(1) 0 v(2) 0 +++] has z-transform V (22)
(12)42) = [x(0) 0 (2) 0 (4) 0} as z-transform LX (2) + X(—2)] = Xevan(@)
H()_Teanspose filter with coefficients (—K): anticausal matrix A?
MH) Flip toA(N —&) produces (H(N), ..., 2), ACL), (0))
H(-2) Alternating sign (—1)*%(&) produces (8(0), —A(I), #2), —AG). «-»)
2" H(-27!) Alternating flip (-1)*4(V — k) produces (A(N), ... ,(-1)%h(0))
K(k) =h(N =k) Symmetric filter (W symmetry for odd NV and H symmetry for even NV) Filter Banks
L= (1200 = (42)W72Ho_Lowpass analysis channel: Filter and downsample
B=({2)D =(}2)v2H, _ Highpass analysis channel: Filter has Cink) =
VIo(t2) Lowpass synthesis channel: Upsample and filter with Fo(e) = 1%)(~2)
V3F\(12)—Highpass synthesis channel: Upsample and filter with F(z)
—He(-2)
Fo(z)Ho(z) ~ Fo(—z)Ho(—2) = Po(z) — Po(—z) = 22-€ No distortion: € delays (odd €)
af Hocven(2) Hoos) ] ana pyce) =| Faeven(@) Fiewa(=) ] Polyphase
mo=[ Fieven(2) | Fe) =[ Fooa(2) feed matrices
Hn(2) =
M@ Mi(-2) Fol-2) Fi(-2) matrices
Perfect Reconstruction with no delays
[ ee Ho(—2) ] and ne Fo) F(z) ] Modulation
Fp(2)H (2)
Fa )Hm(2) = [ ope ] Perfect Reconstruction with £ delays (odd é)
HEH (2)
1 Hp(2) is paraunitary and the filter bank is orthogonal
Cho(n)ho(n + 2k) = 4k) and Ho(z) Ho(e~!) + Ho(—z) Ho(—z
)
LCI holk) = Ofor0
