Title :
Design of a sample-rate converter based on least-square method
Author_Institution :
Mechatron. Inst., Univ. of Electron. Sci. & Technol. of China, Chendu, China
Abstract :
A sample-rate converter in digital domain is regarded as filtering the origin signal, and the precision of resampling is mainly decided by filter designing. To improve the accuracy of signal resampling, a resampling realization method is proposed in this paper. It converts a resampling problem into a time-variable filter designing, so the least-square method can be used to obtain the filter coefficients and obtain better accuracy. To avoid the ill-conditioning problem which may arise in solving the least-square equations, QR decomposition based on householder transformation is used. Finally, the experimental results proved that high accuracy signal resampling can be realized with the proposed method.
Keywords :
filtering theory; least mean squares methods; signal sampling; QR decomposition; digital domain; householder transformation; ill-conditioning problem; least-square equations; least-square method; resampling realization method; sample-rate converter; signal filtering; signal resampling; time-variable filter design; Finite impulse response filter; Noise; least-square method; sample-rate conversion; time-variable filter;
Conference_Titel :
Computer Science and Information Processing (CSIP), 2012 International Conference on
Conference_Location :
Xi´an, Shaanxi
Print_ISBN :
978-1-4673-1410-7
DOI :
10.1109/CSIP.2012.6308862