Even-odd structure and parallel structure for implementing variable fractional-delay filters
Tian–Bo Deng
Abstract
Tian–Bo Deng
Abstract
The problem of designing 1D variable fractional-delay (VFD) digital filter can be elegantly simplified as the easier sub-problems that involve 1D constant filter (sub-filter) designs and 1D polynomial approximations by utilizing the singular-value-decomposition (SVD) of the variable design specification. This paper proposes two new structures for efficiently implementing the SVD-based VFD filters, which require less computational complexity than the Farrow structure.
A significance statement is not available in the OpenAlex record.
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 1D variable fractional-delay (VFD) digital filter can be elegantly simplified as the easier sub-problems that involve 1D constant filter (sub-filter) designs and 1D polynomial approximations by utilizing the singular-value-decomposition (SVD) of the variable design specification. This paper proposes two new structures for efficiently implementing the SVD-based VFD filters, which require less computational complexity than the Farrow structure.
Key concepts: Variable (mathematics), Filter (signal processing), Digital filter, Singular value decomposition, Filter design, Computer science, Prototype filter, Control theory (sociology)