Equiripple and minimax (Chebyshev) approximations for recursive digital filters
A. Deczky
Abstract
A. Deczky
Abstract
The problem of designing recursive digital filters whose frequency response approximates an arbitrarily prescribed function in the Chebyshev sense on a single interval is considered. Certain degenerate cases where the best Chebyshev approximation is not equiripple are studied in detail, and an algorithm is given for determining the best Chebyshev as well as the best equiripple approximation. Finally, a number of examples illustrating applications of this algorithm are given.
OpenAlex reports 112 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The problem of designing recursive digital filters whose frequency response approximates an arbitrarily prescribed function in the Chebyshev sense on a single interval is considered. Certain degenerate cases where the best Chebyshev approximation is not equiripple are studied in detail, and an algorithm is given for determining the best Chebyshev as well as the best equiripple approximation. Finally, a number of examples illustrating applications of this algorithm are given.
Key concepts: Chebyshev filter, Equioscillation theorem, Chebyshev iteration, Chebyshev equation, Minimax approximation algorithm, Minimax, Approximation theory, Mathematics