Title :
On IIR echo cancellation
Author :
Sharma, Piyare Lal
Author_Institution :
Dept. of Electr. Eng., Wisconsin Univ., Platteville, WI, USA
Abstract :
Least mean square (LMS) algorithms using output error criteria suffer from stability and convergence problems when applied to infinite impulse response (IIR) structures. These problems have been overcome to a certain extent using the equation error method. However, the method results in biased estimates in the presence of measurement noise. To obtain consistent estimates other methods like the Steiglitz-McBride method (SMM) and the instrumental variable method (IVM) are used. An echo canceller is implemented on TMS320C25 digital signal processor using a variant of the SMM method
Keywords :
digital filters; digital signal processing chips; echo suppression; stability; IIR echo cancellation; Steiglitz-McBride method; TMS320C25 digital signal processor; convergence problems; echo canceller; infinite impulse response; instrumental variable method; output error criteria; stability; Convergence; Digital signal processors; Echo cancellers; Equations; Instruments; Least squares approximation; Noise cancellation; Noise measurement; Signal processing algorithms; Stability criteria;
Conference_Titel :
Circuits and Systems, 1992., Proceedings of the 35th Midwest Symposium on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-0510-8
DOI :
10.1109/MWSCAS.1992.271211
