2007
DOI: 10.1007/s00034-005-0220-x
|View full text |Cite
|
Sign up to set email alerts
|

Design Equations for Jointly Optimized Frequency-Response Masking Filters

Abstract: The introduction of nonlinear optimization techniques to the design of a frequency-response masking (FRM) filter has changed the way in which an FRM filter is synthesized. It allows all subfilters in an FRM structure to be optimized jointly, resulting in further savings in the number of arithmetic operations. Under the joint optimization, a new set of design equations is necessary, not only for a more computationally efficient filter, but also for the simplification of the design process and the reduction of t… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
16
0

Year Published

2008
2008
2018
2018

Publication Types

Select...
4
1

Relationship

1
4

Authors

Journals

citations
Cited by 7 publications
(16 citation statements)
references
References 26 publications
0
16
0
Order By: Relevance
“…Because there are only few candidates of the L r 's left, and the estimated subfilter orders are very close to the actual orders that minimize the arithmetic complexity, the proposed overall optimization algorithm has to be run very few times. This fact makes the proposed algorithm drastically quicker compared with other existing ones [3,4,7,12,16,19,20], which do not exploit the above-mentioned observations at all.…”
Section: Estimation Of the Design Parameters For The Proposed Synthesmentioning
confidence: 91%
See 1 more Smart Citation
“…Because there are only few candidates of the L r 's left, and the estimated subfilter orders are very close to the actual orders that minimize the arithmetic complexity, the proposed overall optimization algorithm has to be run very few times. This fact makes the proposed algorithm drastically quicker compared with other existing ones [3,4,7,12,16,19,20], which do not exploit the above-mentioned observations at all.…”
Section: Estimation Of the Design Parameters For The Proposed Synthesmentioning
confidence: 91%
“…Consider the following specifications [4,7,11,13,15,17,19]: ω p = 0.4π , ω s = 0.402π , δ p = 0.01, and δ s = 0.001. For the conventional direct-form FIR filter, the minimum order to meet the given criteria is 7 2558, requiring 1290 multipliers and 2558 adders when the coefficient symmetry is exploited.…”
Section: Filter Specificationsmentioning
confidence: 99%
“…However, in the case of a joint optimization of the constituent digital sub-filters, an empirical investigation led to α = 2.25 in [22], while an analytical investigation in [17] led to α = 2.4. In this paper, a better approximation for the interpolation factor M is obtained by tabulating the overall bandpass …”
Section: Design Of Bandpass Frm Fir Digital Filtersmentioning
confidence: 99%
“…Other gradient-based optimization techniques for FRM digital filters include the weighted least-squares approach [23] and the semi-definite [12] and second-order cone programming [13] approaches. Recently, a more attractive technique based on non-linear programming was developed [17] (see also [22]) for the joint optimization of the constituent digital sub-filters, leading to a substantial reduction in the number of addition and multiplication operations in the realization of the overall FRM digital filter. A further reduction was also possible by constructing the corresponding masking digital sub-filters in terms of a common part.…”
Section: Introductionmentioning
confidence: 99%
“…the filter structure, the design method, and the implementation. The introduction of various FRM structures has significantly enhanced the computational efficiency of the FRM technique [19,21,22,25,45,[48][49][50][51][52][53]. Non-linear optimization techniques further reduce the arithmetic operations in the FRM filters, where the coefficients of all subfilters in an FRM structure are optimized simultaneously [6, 15-18, 31, 34, 35].…”
mentioning
confidence: 99%