8th International Symposium on Image and Signal Processing and Analysis (ISPA 2013) September 4-6, 2013, Trieste,
September 4-6, 2013, Trieste, Italy
IIR Filter Approximation of an
Innovative Digital Audio Equalizer
Marco Virgulti, Stefania Cecchi, Francesco Piazza
A3lab - DII - Università Politecnica delle Marche
Via Brecce Bianche 1, 60131, Ancona, Italy
Email: m.virgulti@univpm.it
Abstract—Digital audio equalization is used to enhance the
listening experience introducing corrections in the auditory
frequency response. The equalizer performance depends on
its implementation which includes optimal filters with sharp
transition bandwidth development and a low computational
complexity. Starting from a previous FIR implementation based
on multirate systems and filterbanks theory, an optimized digital
audio equalizer is derived. The proposed approach employs an
approximation of the FIR structure using IIR filters, in order to
improve the filterbanks structure developed to avoid interferences
between adjacent bands. The effectiveness of the optimized
implementation is shown comparing it with the previous ap-
proaches of the state of the art. The experimental results confirms
that the presented solution has several advantages increasing
the equalization performance in terms of low computational
complexity, low delay, and uniform frequency response.
I. I NTRODUCTION
Digital audio equalization is one of the most common
operations in the acoustic field and it is used in order to
enhance the listening experience by the correction of room
and loudspeakers frequency response. It is based on the use
of equalization filters bank connected in serial, in parallel or
in a mixed connection form [1]. The equalization filters are
developed in the frequency domain and they can boost or
attenuate specific part of the frequency spectrum. They are
characterized by three parameters: the center frequency, the
gain and the quality factor, that gives the relation between the
filter bandwidth and the center frequency. The equalizer bank
is called graphic equalizer when the center frequencies and
the quality factor are fixed, while the gain is adjustable [1]. In
this context, different type of graphic equalizer can be found
in the literature, that differ from filter development and final
implementation.
Generally, a tree structured filter bank was used, as de-
scribed in [2]: the analysis filter bank was built with equal
stages splitting the input signal in two subbands while the
synthesis filter bank had to recombine back the bands. The
drawback of such a structure is the too high delay that
exponentially increases with the number of subbands. A similar
approach has been described in [3]; a linear phase digital filter
is implemented using frequency masking technique. This is one
of the most computationally efficient techniques for the design
of sharp low-pass, high-pass, band-pass, and band-stop filters
with arbitrary passband but the introduced delay dramatically
increases with stricter constraints. Apart from too high delay
introduced in the filtering operation, the main disadvantages
of these approaches are the appearance of extra ripples in the
overall frequency response, when adjacent bands are added
Signal Processing
Speech, Music, and Audio Processing
togheter. Indeed, due to the fact that the filters have a non-
zero bandwidth, when the filters are connected together, their
frequency responses will interfere each other [1], resulting in
undesirable distortions of the overall frequency response that
will differ from the user’s gains setting. This problem has
been deeply investigated in [1], [4], [5]. An iterative approach
that involves the implementation of the filters using Remez
algorithm is presented in [6]. When the response of adjacent
bands are added together, if the composite frequency response
shows an unacceptable error deviations, a new filter with a new
stopband cut-off frequency has to be designed. This procedure
is iterated until the deviation becomes acceptable resulting
in a too high computational complexity. A method based on
the opposite filter concept has been proposed in [1]: a set of
simple filters are adequately parametrized and serially inserted
into the equalizer bank, in order to reduce the interference
effects. Also a frequency domain algorithm was presented in
[7]. The equalization consists of a complex multiplication of
the input spectrum with the frequency equalization function
that, when transformed in time domain, has all the properties of
a FIR filter. The computational complexity is very low but the
algorithm efficiency is strictly bound to frequency resolution;
large ripples at bands edges are easily observed.
A different solution could arise from the multirate tech-
niques applied to adaptive systems. In this context, the prob-
lem of aliasing cancellation when an adaptive filtering is
included in a filterbank with perfect reconstruction is well-
known. Different approaches were presented such as cross
terms between subbands [8] or extra terms taking into ac-
count adjacent bands [9]. Starting from these approaches, an
innovative digital audio equalizer has been introduced in [10],
[11]. Taking into consideration multirate systems and their
property, the idea was to realize a linear phase FIR equalizer
that overcame the well-known drawbacks. Therefore, a first
optimized version of this algorithm has been proposed in
[12]: it is based on the use of IIR filters capable to reduce
the required computational complexity preserving the audio
quality [13], [14]. As for the FIR approach, the overall scheme
of the algorithm derives from two analysis and two synthesis
cosine modulated filter banks properly combined in order to
have the desired response when all bands are added together,
exploiting multirate properties. However, the method proposed
in [12] shows some disadvantages: it is not possible to ex-
ploit the lattice structure behaviour considering the polyphase
components in the filtering procedure, in order to have an
efficient filter bank structure. In this context, a more feasible
structure for the filter coefficients development is presented,
in order to fully exploit the filter bank structure, reducing the
410
8th International Symposium on Image and Signal Processing and Analysis (ISPA 2013) September 4-6, 2013, Trieste, Italy

computational complexity of the final equalizer. The validity of

the proposed approach has been proved considering an analysis
of the computational complexity and a quality evaluation of the
desired equalization curve.
The paper is organized as follows. Section II describes the
proposed algorithms: in particular, the previous approach based
on the use of FIR filters will be reviewed in Section II-A while
the proposed approach based on IIR filters will be discussed
in Section II-B. After a description of the computational
complexity (Section II-C), the proposed algorithm validation
is presented in Section III, showing the main advantages of
this approach with reference to a simple equalizer. Finally, the
conclusions are reported in Section IV.

II. P ROPOSED ALGORITHMS

The main idea of the proposed approach is to realize an
innovative equalizer using a particular filter bank structure,
capable of reducing ripples in the frequency response, when Fig. 1. Overall structure of the proposed equalizer
adjacent bands are added togheter. This technique is derived
from the subband adaptive filtering structure presented in
[9], [15], where a double analysis/synthesis filter banks is [17], because of the very low computational cost. This tech-
combined employing multirate properties. Starting from this, nique modifies the 6-dB cut-off frequency of the filter in order
it is possible to obtain a similar structure for an innovative to obtain the 3-dB cut-off frequency placed approximately at
equalizer having all the advantages of this particular solution π/2M . The function minimized by the algorithm is reported
[10], [11]. Fig. 1 shows the final structure used to create the below:
equalizer where the filter banks have been realized considering 1
uniform cosine modulated technique as described in [16]. ξ = ||H(ejπ/2M )| − √ |, (3)
The impulse responses of the analysis/synthesis filters are the 2
following: being H the frequency response of the following filter
L
hk (n) = 2h0 (n)cos(ωk (n − ) + θk ) sin((n − L/2)ωc,6dB )
2 (1) h[n] = w[n], (4)
L π(n − L/2)
fk (n) = 2h0 (n)cos(ωk (n − ) − θk ),
2
with where w[n] is a Kaiser window and ωc,6−dB is the cut-off
π frequency.
θk = −(1)k
4
π
ωk = (k + 0.5), (2) B. IIR-based equalizer
M
where k is the subband index defined between 0 and M − 1, The IIR Equalizer has been designed still considering
M is the number of subband, and L is the order of the the cosine modulation approach and following the scheme
prototype filter h0 (n). In [10], [11], this structure has been proposed in Fig.1, but considering an IIR filter prototype.
used considering FIR prototype: it has been demonstrated Differently from [12] where a lattice structure has been used, a
that using this particular structure, it is possible to signifi- FIR filter approximation has been considered for the IIR-based
cantly reduce the ripple amounts difference in the transition graphic equalizer. The approximation has been performed tak-
band between adjacent filters. Since this scheme derives from ing into consideration the Vector Fitting (VF) methodologies
multirate systems, it is easily extended to higher number of in the discrete time domain as described in [18].
bands. However, increasing the number of subbands, the FIR
length has to be increased in order to have good performance The FIR prototype h0 (n) of order L defined in Section
at low frequency bands. For this reason, a new version of II-A can be written in the Z domain as follows:
the algorithm has been proposed taking advantage of IIR
filters, especially for a great number of bands, using the same L
X
structure of Fig.1. h0 (z) = fn z −n , (5)
n=0
A. FIR-based equalizer
where fn ∈ ℜ. The objective is to obtain a stable and causal
The FIR Equalizer has been designed using Eq.(1) and IIR filter of order N that approximates h0 (z), defined as
developing h0 (n) has a FIR filter prototype. The design of this follows:
filter is a very important issue to have a perfect reconstruction PN −n
filter bank: in this case, we have considered a near perfect p=0 bp z
p0 (z) = PN , (6)
reconstruction prototype realized by Kaiser Windows method q=0 an z
−q

Signal Processing
Speech, Music, and Audio Processing 411
8th International Symposium on Image and Signal Processing and Analysis (ISPA 2013) September 4-6, 2013, Trieste, Italy

where bp , an ∈ ℜ. The target IIR filter can be written in a x(n)

dk
simple fractional basis form [18] as follows:
NC 
X  X NR
cm cm rm H0,k +
p0 (z) = d+ −1 − α
+ −1 − α
+ −1 − α
,
m=1
z m z m m=1
z m
H1,k +
| {z } | {z }
σC (z) σR (z)
(7) H2,k +

where αm and αm are reciprocal poles, cm and cm are the

residues of the complex poles, rm is the residue of the real HNs,k +
z(n)

poles, NC is the number of complex conjugated pole pairs of

the function σC (z), and NR is number of real poles of the Fig. 2. Scheme of the single analysis filter bank Hk using the proposed IIR
function σR (z). The total amount of poles is N = 2NC + NR . filter structure.
Then, following the steps described in [18], it is possible
to find the parameters cm , αm , rm and d so that Eq.(7)
represents the IIR approximation of the FIR filter expressed in b̂0,p,k + b̂1,p,k z −1 + b̂2,p,k z −2 + b̂3,p,k z −3 + b̂4,p,k z −4
Hp,k =
Eq.(5). However, considering the partial fraction decomposi- 1 + â1,p,k z −1 + â2,p,k z −2 + â3,p,k z −3 + â4,p,k z −4
tion obtained from the V F algorithm where a pseudo-parallel (18)
IIR form is obtained, the second order sections formulation where
for a parallel structure has to be derived from Eq.(7) before
applying cosine modulation to the IIR prototype. The functions b̂0,p,k = 2b0,p cos(θka )
σC (z) and σR (z) can be expressed considering the following b̂1,p,k = 2b0,p a1,p cos(ωk − θka ) + 2b1,p cos(ωk + θka )
decomposition: b̂2,p,k = 2b0,p a2,p cos(2ωk − θka ) + 2b1,p a1,p cos(θka )+
−1 −2
b0,m + b1,m z + b2,m z + 2b2,p cos(2ωk + θka )
SOSm (z) = . (8)
1 + a1,m z −1 + a2,m z −2 b̂3,p,k = 2b1,p a2,p cos(ωk − θka ) + 2b1,p a2,p cos(ωk + θka )
The coefficients ar,m and br,m can be easily derived from b̂4,p,k = 2b2,p a2,p cos(θka ) (19)
cm , rm , αm . For the function σC (z), the following parameters
are obtained:
â0,p,k =1
cm α m + cm αm 2 Re cm
b0,p = − 2 b1,p = 2 b2,p = 0 (9) â1,p,k = 2a1,p cos(ωk )
kαm k kαm k
â2,p,k = a21,p + 2a2,p cos(2ωk )
2 αm 1
a1,p = 2 a2,p = 2. (10) â3,p,k = 2a1,p a2,p cos ωk
kαm k kαm k â4,p,k = a22,p (20)
For the function σR (z), the following parameters are obtained:
−1
b0,p = −αm rm b1,p = 0 b2,p = 0 (11) dˆk = 2d cos(θka ) (21)
a1,p = −1
−αm a2,p = 0. (12) with k = 1, · · · , M . Analogous formulas can be found for
the synthesis filter bank, but considering θkb instead of θka in
At this point, a cosine modulation has to be applied to the Eq.(19). Therefore, starting from the FIR prototype approxi-
IIR filter prototype for using it in the analysis and synthesis mation with a lower order IIR filter, a formulation for the IIR
filterbank as shown in Fig.1. Starting from Eqs.(1) and (2), the cosine modulated filters bank has been determined.
cosine modulation for the IIR filter prototype in the Z domain
is defined as follows:
C. Computational complexity
a a
Hk (z) = H(ze−jωk )ejθk + H(zejωk )e−jθk (13) The computational complexity has been calculated in terms
s s
Fk (z) = F (ze−jωk )ejθk + F (zejωk )e−jθk , (14) of the needed number of multiplications and additions. Table I
reports a comparison of the computational complexity between
where
θka = −ωk + θk (15) TABLE I. C OMPUTATIONAL C OMPLEXITY OF THE FIR EQUALIZER OF
θks = −ωk − θk . (16) [11], THE IIR EQUALIZER OF [12] AND THE PROPOSED APPROACH ,
WHEREIN L IS THE FIR PROTOTYPE LENGTH , Ns IS THE NUMBER OF IIR
FOURTH ORDER SECTION , N IS THE IIR PROTOTYPE LENGTH , M IS THE
The IIR filter prototype realized with the parallel structure
NUMBER OF SUBBAND , Nl IS THE NUMBER OF IIR LATTICE SECTION AS
permits to apply the modulation defined in Eq.(13) and Eq.(14) DESCRIBED IN [12].
directly on the second order sections of Eq.(8), eliminating
the coefficients round off and truncation error, that was find in Multiplications Additions Delay
[12]. The final filter for the analysis filter bank is described as
FIR equalizer of [11] L L−1 L/2 + 1
follows:
XNs IIR equalizer of [12] 4M Nl + 1 2M Nl + 2 M Nl
Hk (z) = dˆk + Hp,k (17)
Proposed approach 10Ns 9Ns − 1 Ns
p=0

Signal Processing
Speech, Music, and Audio Processing 412
8th International Symposium on Image and Signal Processing and Analysis (ISPA 2013) September 4-6, 2013, Trieste, Italy

TABLE II. D ISTORTION I NDEX FOR UNIFORM M BAND EQUALIZER

WITH A FIR PROTOTYPE OF LENGTH 1024 AND A IIR PROTOTYPE : ∆t
0 a b c d e
REPRESENTS THE TIME INTERVAL IN SECOND TO FILTER 4096 SAMPLES .
Magnitude [dB]

FIR of [11] IIR proposed approach

−50
DI ∆t Order DI ∆t Order

−100 M =8 0.9979 0.287s 1024 1.0043 0.351s 50

M = 16 0.9979 0.635s 1024 1.0043 0.372s 34

−150 M = 32 0.9979 1.276s 1024 1.0020 0.406s 24

0 0.005 0.01 0.015 M = 64 0.9979 2.603s 1024 1.0003 0.568s 18

Frequency [Normalized]
TABLE III. D ISTORTION I NDEX FOR UNIFORM M BAND EQUALIZER
Fig. 3. IIR filter prototype approximations of 4096 taps FIR prototype for WITH A FIR PROTOTYPE OF LENGTH 2048 AND A IIR PROTOTYPE : ∆t
64 subbands equalizer: (a) FIR of 4096 samples, (b) IIR of order N = 10, (c) REPRESENTS THE TIME INTERVAL IN SECOND TO FILTER 4096 SAMPLES .
IIR of order N = 24, (d) IIR of order N = 28, (e) IIR prototype calculated
as in [12] with Nl = 5. FIR of [11] IIR proposed approach

DI ∆t Order DI ∆t Order
the proposed approach and the approaches of [11] and [12], M = 16 0.9985 1.192s 2048 1.0049 0.343s 50
taking into consideration also the delay introduced by the final M = 32 0.9985 2.413s 2048 1.0026 0.457s 30
equalizer. Considering the method proposed in [11], for a L-
taps FIR filter prototype, there are exactly L multiplications M = 64 0.9985 4.950s 2048 1.0009 0.677s 22
and L − 1 additions, giving a complexity of 2L − 1 for
each equalizer band. Taking into consideration the method TABLE IV. D ISTORTION I NDEX FOR UNIFORM M BAND EQUALIZER
WITH A FIR PROTOTYPE OF LENGTH 4096 AND A IIR PROTOTYPE : ∆t
proposed in [12], the IIR prototype length is bounded to the
REPRESENTS THE TIME INTERVAL IN SECOND TO FILTER 4096 SAMPLES .
number of IIR lattice structure used and to the number of
subbands as shown in Table II. This facts allows to have a FIR of [11] IIR proposed approach
good resolution and a sharp transition of the IIR filter, to
DI ∆t Order DI ∆t Order
the detriment of the computational complexity. Considering
the proposed approach, if Ns is the number of fourth order M = 32 0.99978 5.043s 4096 1.0018 0.657s 46
sections for a N length IIR filter, the computational cost M = 64 0.99978 10.025s 4096 1.0034 0.832s 28
is evaluated taking into account Eq.(17) where each section
needs 5 multiplications and 4 additions both for the numerator
and denominator. Furthermore, it is needed to accumulate the M = 8, M = 16, M = 32, and M = 64: incremental gains
output of each second order section in order to obtain the from 1 to 8 have been repeated every 8 bands. Figs. 4ii-7ii
output sample and also N − 1 additions have to be considered. shows a detail of the equalizer curve for each value of M ,
More in details, taking into consideration the example since it is very difficult to distinguish the difference between
reported in Fig. 3, a L = 4096 taps FIR filter prototype for the filter considering all the spectrum as reported in Figs. 4i-7i.
M = 64 subbands equalizer requests a total amount of 8191 It is evident that for the FIR prototype, increasing the number
operations per output bin. If we consider the IIR filter of order of subbands, the length of 1024 is no more sufficient in order to
N = 28 with Ns = 7 fourth order sections, the total amount have a sharp transition band. Therefore, increasing the number
of computational complexity will be equal to 265 operations. of subband, longer FIR prototype are requested. Considering
To obtain good performance considering the approach of [12], the IIR prototype, it has good performance also with a small
a IIR filter with Nl = 5 lattice sections had to be considered number of samples, and it preserve its behaviour increasing
with a computational complexity of 1923 samples. the number of the subbands. Differently from [11] and [12],
there is a computational saving considering the proposed IIR
III. E XPERIMENTAL RESULTS prototype maintaining a good performance level. Taking as
term of comparison the process proposed in [19], it is evident
The validation of the algorithm has been performed through
that the proposed approach is capable to avoid the ripple
a comparison between the FIR Equalizer implemented as
around the transition band while [19] shows an increase of the
described in Section II-A, the proposed approach and the
ripple amount, increasing the number of subband. Furthermore,
method proposed in [19]. This approach has been chosen since
the IIR filter prototype preserve a good phase linearity in the
it tries to solve the problem of frequency response interference,
passband resulting from the least squares approximation of the
considering each band as composed by an high number of
complete FIR filter frequency response, as shown in Fig.(8).
raised-cosine filters, in order to avoid ripple between adjacent
bands. Moreover, a comparison with other simple filter bank To better underline the performances of the proposed
has been already performed in [10], [11]. Then, four cases have approach, an objective performance measurement parameter
been considered with a different number of subbands (i.e., M has been used. Taking into consideration the distortion transfer
= 16, 32, 64, 128): for each case, three FIR filter prototypes of function [16], calculated as follows,
length N = 1024, 2048, 4096 and different IIR filter prototypes
have been considered for the tests. 1 X
M
T (z) = Hi (z)Fi (z) (22)
Figs. 4i-7i show the equalizer frequency response for M i=1

Signal Processing
Speech, Music, and Audio Processing 413
8th International Symposium on Image and Signal Processing and Analysis (ISPA 2013) September 4-6, 2013, Trieste, Italy

10

Magnitude [dB] 8 5

Magnitude [dB]
6
4.5
4

2 a a
b 4 b
c c
0
0 0.1 0.2 0.3 0.4 0.5 0.22 0.23 0.24 0.25 0.26 0.27 0.28
Frequency [Normalized] Frequency [Normalized]
i ii

Fig. 4. Equalizer frequency responses for M = 8 subbands considering (a) 1024 FIR filter, (b) IIR of order N = 50, (c) method of [19]: (i) frequency
behaviour of the first 8 bands, (ii) detail of one band.

10

8 5
Magnitude [dB]

Magnitude [dB]
6
4.5
4

2 a a
b 4 b
c c
0
0 0.05 0.1 0.15 0.2 0.25 0.11 0.115 0.12 0.125 0.13 0.135 0.14
Frequency [Normalized] Frequency [Normalized]
i ii

Fig. 5. Equalizer frequency responses for M = 16 subbands considering (a) 1024 FIR filter, (b) IIR of order N = 34, (c) method of [19]: (i) frequency
behaviour of the first 8 bands, (ii) detail of one band.

8
5
Magnitude [dB]

Magnitude [dB]

4 4.5

2 a a
b 4 b
c c
0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.055 0.06 0.065 0.07
Frequency [Normalized] Frequency [Normalized]
i ii

Fig. 6. Equalizer frequency responses for M = 32 subbands considering (a) 2048 FIR filter, (b) IIR of order N = 30 ,(c) method of [19]: (i) frequency
behaviour of the first 8 bands, (ii) detail of one band.

where Hi and Fi are the frequency responses of the analysis different prototype lengths. Also a representation of the time
and synthesis filter banks, respectively, a distortion index has interval needed to filter 4096 samples is reported. It is evident
been defined allowing direct comparison between different that for all the considered prototype, good performance has
equalizer with different number of bands, i.e., been achieved, obtaining values very close to 1. In particular,
    increasing the number of subbands, the IIR equalizer shows
max T (ejω ) + min T (ejω ) better performance in terms of distortion index, confirming the
DI = . (23)
2 validity of the proposed approach in comparison with the first
FIR implementation. Furthermore, as underlined in Section
This index takes into account the amplitude distortion of II-C, the computational complexity of the IIR implementation
each band: better performance is achieved when it is ap- is lower than FIR implementation as shown in the Table,
proximately 1. Tabs.(II)-(IV) show the results in terms of considering ∆t values.
distortion index for the FIR equalizer considering different
number of subbands and different prototype lengths and for
the IIR equalizer considering different number of subbands and

Signal Processing
Speech, Music, and Audio Processing 414
8th International Symposium on Image and Signal Processing and Analysis (ISPA 2013) September 4-6, 2013, Trieste, Italy

10

Magnitude [dB] 8 5

Magnitude [dB]
6
4.5
4

2 a a
b 4 b
c c
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.028 0.029 0.03 0.031 0.032 0.033 0.034 0.035
Frequency [Normalized] Frequency [Normalized]
i ii

Fig. 7. Equalizer frequency responses for M = 64 subbands considering (a) 4096 FIR filter, (b) IIR of order N = 28, (c) method of [19]: (i) frequency
behaviour of the first 8 bands, (ii) detail of one band.

IV. C ONCLUSION
In this work, a new approach to M -band uniform IIR
equalizer has been presented starting from a previous FIR
implementation based on multirate systems and filter banks
theory. The proposed approach employs IIR filters to improve
the filter banks structure developed to avoid ripple between
adjacent bands. The solution presented here has several advan-
tages increasing the equalization performance in terms of low
computational complexity, low delay, and uniform frequency
response. The effectiveness of the optimized implementation
has been tested taking into comparison the previous FIR
approach and a technique of the literature that try to solve
the same problem. The obtained results in terms of desired
final frequency response and distortion index have confirmed
the validity of the proposed approach, that shows a low compu-
tational complexity and filtering delay especially considering a
great number of subband. Future works will be oriented to the
development of the system on a real time platform, in order
to analyze the obtained performance.
R EFERENCES
[1] S. A. Azizi, “A new concept of interference compensation for parametric
and graphic equalizer banks,” in Proc. 112nd Audio Engineering Society
Convention, Munich, Germany, May 2002.
[2] J. Vieira, “Digital Five Band Equalizer with Linear Phase,” in Proc.
100th Audio Engineering Society Convention, Copenhagen, Denmark,
May 1996.
[3] Y. Lim, “A Digital Filter Bank for Digital Audio Systems,” IEEE Trans.
Circuits Syst., vol. 33, no. 8, pp. 848–849, Aug. 1986.
0
−20
−40
Phase [Deg]
[4] R. A. Greiner and M. Schoessow, “Design aspects of graphic equaliz-
ers,” J. Audio Eng. Soc., vol. 31, no. 6, pp. 394–407, 1983.
[5] D. A. Bohn, “Constant-q graphic equalizers,” J. Audio Eng. Soc.,
vol. 34, no. 9, pp. 611–626, 1986.
[6] J. Henriquez, T. Riemer, and R. Trahan Jr., “A Phase Linear Audio
Equalizer: Design and Implementation,” J. Audio Eng. Soc., vol. 38,
no. 9, pp. 653–666, Sept. 1990.
[7] H. Schopp and H. Hetzel, “A Linear Phase 512 Band Graphic Equalizer
using the Fast Fourier Transform,” in Proc. 96th Audio Engineering
Society Convention, Amsterdam, The Netherlands, Feb. 1994.
[8] A. Gilloire and M. Vetterli, “Adaptive filtering in subbands with
critical sampling: analysis, experiments and application to acoustic echo
cancellation,” IEEE Trans. Signal Process., vol. 40, no. 8, pp. 1862–
1875, Aug. 1992.
[9] M. Petraglia and P. Batalheiro, “Non uniform Subband Adaptive Fil-
tering with Critical Sampling,” in Proc. International Symposium on
Circuits and Systems, vol. 56, no. 2, Island of Kos, Greece, May 2006,
pp. 565–575.
[10] S. Cecchi, P. Peretti, L. Palestini, F. Piazza, F. Bettarelli, and A. Lattanzi,
“Real Time implementation of an Innovative Digital Audio Equalizer,”
in Proc. 123rd Audio Engineering Society Convention, New York, NY,
USA, Oct. 2007.
[11] S. Cecchi, L. Palestini, E. Moretti, and F. Piazza, “A New Approach
to Digital Audio Equalization,” in Proc. Workshop on Applications
of Signal Processing to Audio and Acoustics, New Paltz, NY, USA,
October 2007, pp. 62–65.
[12] M. Virgulti, S. Cecchi, A. Primavera, L. Romoli, E. Ciavattini,
F. Bettarelli, and F. Piazza, “Optimized Implementation of an Innovative
Digital Audio Equalizer,” in Proc. 133rd Audio Engineering Society
Convention, San Francisco, CA, USA, Oct. 2012.
[13] R. D. Koilpillai and P. P. Vaidyanathan, “Cosine-modulated fir filter
banks satisfying perfect reconstruction,” Signal Processing, IEEE Trans-
actions on, vol. 40, no. 4, pp. 770–783, apr 1992.
[14] S. G. Kim and C. D. Yoo, “Highly selective m-channel iir cosine-
modulated filter banks,” Electronics Letters, vol. 39, no. 20, pp. 1478–
1479, oct. 2003.
a [15] M. Petraglia, R. Alves, and P. Diniz, “New Structures for Adaptive
b Filtering in Subbands with Critical Sampling,” IEEE Trans. Signal
Process., vol. 48, no. 12, pp. 3316–3327, Dec. 2000.
[16] P. P. Vaidyanathan, Multirate Systems and Filterbanks. Prentice-Hall,

Inc., 1993.
−60
[17] F. Cruz-Roldan, P. Amo-Lopez, S. Maldonado-Bascon, and S. S. Law-
−80 son, “An Efficient and Simple Method for Designing Prototype Filters
for Cosine-Modulated Pseudo-QMF Banks,” IEEE Signal Process. Lett.,
−100 vol. 9, no. 1, pp. 29–31, Jan. 2002.
[18] N. Wong and C. Lei, “Fir filter approximation by iir filters based
−120 on discrete-time vector fitting,” in Proc. International Symposium on
0 0.05 0.1 0.15
Frequency [Normalized] Circuits and Systems, May 2007.
[19] D. McGrath, J. Baird, and B. Jackson, “Raised cosine equalization
Fig. 8. Phase response for M = 8 subbands considering a FIR filter of 1024 utilizing log scale filter synthesis,” in Proc. 117th Audio Engineering
samples (a) and a IIR filter of order 50 (b). Society Convention, San Francisco, CA, USA, Oct 2004.

Signal Processing
Speech, Music, and Audio Processing 415