Filterbank design for oversampled filterbanks without aliasing in the subbands
M. Harteneck, J.M. Pàez-Borrallo, R.W. Stewart
Abstract
M. Harteneck, J.M. Pàez-Borrallo, R.W. Stewart
Abstract
The authors propose a new strategy for the design of real-valued filterbanks which reduces aliasing in the subbands (‘inband aliasing’) such that applications where two subband signals have to be compared become possible. One example for such an application is subband adaptive filtering, where, with the proposed filterbank structure, the traditionally required crossterm adaptive filters between adjacent subbands can be omitted. The proposed filterbank consists of at least three channels which are subsampled by different subsampling ratios.
OpenAlex reports 21 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 authors propose a new strategy for the design of real-valued filterbanks which reduces aliasing in the subbands (‘inband aliasing’) such that applications where two subband signals have to be compared become possible. One example for such an application is subband adaptive filtering, where, with the proposed filterbank structure, the traditionally required crossterm adaptive filters between adjacent subbands can be omitted. The proposed filterbank consists of at least three channels which are subsampled by different subsampling ratios.
Key concepts: Filter bank, Aliasing, Filter (signal processing), Computer science, Adaptive filter, Anti-aliasing filter, Mathematics, Algorithm