Abstract
A signal-processing large-scale integration (LSI) configuration and its advantages with respect to speech communications subscriber circuits on an asynchronous transfer mode (ATM) network are described. The LSI configuration combines the silence span compression/expansion function and the cell loss compensation function into a single group of three LSI chips which are capable of processing media for both pulse code modulation (PCM) and adaptive differential PCM (ADPCM) environments. A near-end echo canceller with a single speech-processing LSI which serves to absorb delays in voice packeting, as well as to absorb the influence of cell propagation delays, is described. A performance evaluation is presented. The results indicate that good speech quality can be realized on an ATM network up to cell loss rate of about 3% for ADPCM speech, and about 8% for PCM speech. >
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
