A two-channel hybrid analog and IIR digital filter bank approximating perfect reconstruction
Per Löwenborg, Håkan Johansson, Lars Wanhammar
Abstract
Per Löwenborg, Håkan Johansson, Lars Wanhammar
Abstract
BANKof the filter bank has in this case exact linear phase. The output signal is then simply a sampled version of the input signal shifted by dT, i.e., y(n) = x(nT–dT). In many cases, it is however not necessary to have exact linear phase. It may in such cases be beneficial in terms of filter complexity to instead consider filter banks with perfect (4) where Φ0(ω) and Φ1(ω) denote the phase responses of A0(s) and A1(s), respectively, and where H0R(ω) and H1 jω ( ) je jΦ0 ω ( ) Φ1 ω ()+2-------------------------------------- H1R ω ()=APPROXIMATING PERFECT MAGNITUDE RECONSTRUCTION
OpenAlex reports 1 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.
BANKof the filter bank has in this case exact linear phase. The output signal is then simply a sampled version of the input signal shifted by dT, i.e., y(n) = x(nT–dT). In many cases, it is however not necessary to have exact linear phase. It may in such cases be beneficial in terms of filter complexity to instead consider filter banks with perfect (4) where Φ0(ω) and Φ1(ω) denote the phase responses of A0(s) and A1(s), respectively, and where H0R(ω) and H1 jω ( ) je jΦ0 ω ( ) Φ1 ω ()+2-------------------------------------- H1R ω ()=APPROXIMATING PERFECT MAGNITUDE RECONSTRUCTION
Key concepts: 2D Filters, Infinite impulse response, Digital filter, Computer science, Channel (broadcasting), Filter bank, Analogue filter, Filter (signal processing)