DocumentCode :
336978
Title :
GSM EFR implementation for TRAU application on DSP16000
Author :
Du, Junchen ; Warner, George ; Vallow, Erik ; Breyer, Penny ; Hollenbach, Tom
Author_Institution :
Wireless & Multimedia Microelectron. Group, Lucent Technol., Allentown, PA, USA
Volume :
4
fYear :
1999
fDate :
15-19 Mar 1999
Firstpage :
2207
Abstract :
An implementation of GSM EFR using traditional single MAC DSPs (DSP1600) takes 24 MIPS to run a single speech channel. Lucent DSP16000 is a dual MAC high performance DSP. Analysis has shown DSP16000 code can run GSM EFR at 10 MIPS per channel with the same program space as DSP1600. The code has been re-structured to minimize the delays and RAM usage for multi-channel transcoder/rate adapter units (TRAU) applications. For a 100 MIPS DSP16210 running 6 EFR speech channels, DSP16210 code takes 15.7 K words of RAM with 3.2 ms maximum delay for the encoder and 0.5 ms for the decoder. Without re-structuring, the numbers are 27 K, 15.4 ms and 1.5 ms respectively. This makes the DSP16000 very attractive for TRAU applications. The initial implementation runs at 12.8 MIPS per channel with 19.3 K words of RAM, 4.7 ms maximum encoding delay and 0.5 ms maximum decoding delay for 6 speech channels
Keywords :
cellular radio; decoding; delays; digital arithmetic; digital signal processing chips; linear predictive coding; random-access storage; speech codecs; speech coding; 0.5 ms; 1.5 ms; 10 MIPS; 100 MIPS; 12.8 MIPS; 15.4 ms; 24 MIPS; 3.2 ms; 4.7 ms; CELP; DSP16000 code; GSM EFR implementation; Lucent DSP16000; RAM; TRAU application; codec; decoder; delay minimisation; dual MAC high performance DSP; maximum decoding delay; maximum encoding delay; multichannel TRAU applications; program space; single MAC DSP; speech channel; transcoder/rate adapter units; Codecs; Costs; Decoding; Delay effects; Digital signal processing; Digital signal processing chips; GSM; Speech coding; Speech processing; Telecommunication standards;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1999. Proceedings., 1999 IEEE International Conference on
Conference_Location :
Phoenix, AZ
ISSN :
1520-6149
Print_ISBN :
0-7803-5041-3
Type :
conf
DOI :
10.1109/ICASSP.1999.758374
Filename :
758374
Link To Document :
بازگشت