Title :
Optimization of multirate crossover filters
Author :
Hämäläinen, Matti
Author_Institution :
Speech & Audio Syst. Lab., Nokia Res. Center, Tampere, Finland
Abstract :
A digital crossover filter design procedure is developed based on multirate complementary filters. The crossover optimization procedure includes the determination of the optimal multirate filter structure and the overall optimization of the cascaded linear phase multirate FIR filter system. The filter structure can be optimized for several cost criteria e.g. for minimal run-time memory or computational complexity. The following crossover filter optimization is based on frequency sampling method in the weighted least-mean-squared sense. The weighting function combines the weighting for the auditory system and the audio reproduction system
Keywords :
FIR filters; audio signal processing; computational complexity; digital filters; least mean squares methods; loudspeakers; low-pass filters; optimisation; signal sampling; audio reproduction system; auditory system; cascaded linear phase multirate FIR filter system; computational complexity; cost criteria; digital crossover filter design procedure; filter structure; frequency sampling method; loudspeaker system; minimal run-time memory; multirate complementary filters; multirate crossover filters; optimal multirate filter structure; optimization; weighted least-mean-squared sense; weighting function; Auditory system; Computational complexity; Cost function; Digital filters; Finite impulse response filter; Frequency; Nonlinear filters; Optimization methods; Runtime; Sampling methods;
Conference_Titel :
Applications of Signal Processing to Audio and Acoustics, 1999 IEEE Workshop on
Conference_Location :
New Paltz, NY
Print_ISBN :
0-7803-5612-8
DOI :
10.1109/ASPAA.1999.810850
