Implementation of a new paradigm in design of IIR digital filters

Lesnikov, Vladislav; Chastikov, Alexander; Naumovich, Tatiana; Armishev, Sergey V.

doi:10.1109/ewdts.2010.5742051

Cited by 16 publications

(4 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the development of this technique is supposed to take into consideration not only the number of multipliers, but the number of adders. This work is performed in the course of development of a new paradigm of realization of recursive digital filters [13][14]. This technique is most effective at FPGA and ASIC design.…”

Section: Discussionmentioning

confidence: 99%

Estimation of structural complexity of IIR digital filters

Lesnikov

Chastikov

Naumovich

et al. 2013

East-West Design &Amp; Test Symposium (EWDTS 2013)

Self Cite

View full text Add to dashboard Cite

The paper is devoted to technique of comparison of various structures of IIR digital filters on complexity taking into consideration the accuracy of implementation of characteristics. An offered technique is based on dividing of all possible structures into classes of equivalence. All structures belonging to the same equivalence class have the same position of zeros and poles in the z-plane. This location is determined by the algebraic-numerical nature of the poles and zeros of the filters with coefficients of finite word length. Each evaluated structure is assigned a reference canonical form with same zeros and poles. The concept of structural complexity considering complexity of realization of operations of multiplication in estimated and reference structures is introduced.

show abstract

Section: Discussionmentioning

confidence: 99%

Estimation of structural complexity of IIR digital filters

Lesnikov

Chastikov

Naumovich

et al. 2013

East-West Design &Amp; Test Symposium (EWDTS 2013)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In [2] - [4], [6], [7] it is shown, that any structure of digital filter can be described by a topological matrix ( )…”

Section: The Description Of Structures Of Filters By Topological Matrmentioning

confidence: 99%

“…Developed by authors of this paper the new paradigm of synthesis of the IIR digital filters assumes generation of various structures of filters and the subsequent analysis of their noise properties [2] - [4]. Therefore development of a universal technique of estimation of round-off noise for the IIR digital filter with arbitrary structure is an actual problem.…”

Section: Introductionmentioning

confidence: 99%

Universal technique of the analysis of round-off noise in digital filters with arbitrary structure described by topological matrixes

Lesnikov

Naumovich

Chastikov

et al. 2013

East-West Design &Amp; Test Symposium (EWDTS 2013)

Self Cite

View full text Add to dashboard Cite

For the adequate description of digital filters with arbitrary structure it is convenient to use so-called topological matrixes. Elements of these matrixes are transmission coefficients between corresponding nodes of the block diagram. Methods of the analysis of power of round-off noise in the digital filters, based on statespace representation of digital filters are known. In this paper the technique of analysis of round-off noise in the IIR digital filter with any structure described by its topological matrix is offered.

show abstract

“…This filter is characterised in the form of the gain parameter and the transition frequency . Thus, applying the filter architecture as derived in (16)(17), we obtained a first-order shelving filter as both low as well as high frequency. Mathematically, (19) The LIG form can be retrieved in terms of the parameters and the sampling frequency .…”

Section: First-order Shelving Filtersmentioning

confidence: 99%

Parametric Continuously Programmable Infinite Impulse Response Filter for Non-Linear Real-Time DSP Applications

Elavarasi¹,

Manjunath²

2020

IJETAE

View full text Add to dashboard Cite

Considering the exponentially rise in non-linear signal filtering purposes, in this paper a novel parametric continuously programmable infinite impulse response (CPIIR) filter has been developed. Realizing the key upsurging significance of non-linear audio signal filtering, our proposed CPIIR filter model has been realized over parametric equalization system, which can have one or multiple IIR filters in cascade-design to perform continuous filtering. Structurally, our proposed model represents a PE solution containing shelving and second-order peaking CPIIR filters whose design parameters are optimized dynamically by reducing cost-function iteratively. Unlike classical approaches where merely filter coefficients are changed manually to cope up with non-linear signal filtering, our proposed parametric IIR filter employs sum-of-square error (SSE) as the costfunction to update filter design parameters like gain parameter, frequency and bandwidth. Noticeably, the use of SSE as cost-function intends to reduce error (i.e., difference between the target signal and the system frequency response) to optimize global gain parameter that eventually help updating other design parameter adaptively over different frequency-bins or windows. This process retains system response near 0-dB line and thus maintains optimal filtering performance over swiftly varying signal response. Unlike major at hand meta-heuristic based search approaches for parameter tuning, our proposed model applies simple grid search concept that avoided significantly large computation and hardware requirements. MATLAB based simulation with Gaussian white noise embedded non-linear signals reveals that the proposed parametric CPIIR model can achieve optimal filtering performance and suitability towards hardware implementation.

show abstract

Implementation of a new paradigm in design of IIR digital filters

Cited by 16 publications

References 3 publications

Estimation of structural complexity of IIR digital filters

Estimation of structural complexity of IIR digital filters

Universal technique of the analysis of round-off noise in digital filters with arbitrary structure described by topological matrixes

Parametric Continuously Programmable Infinite Impulse Response Filter for Non-Linear Real-Time DSP Applications

Contact Info

Product

Resources

About