DocumentCode :
1722812
Title :
Viterbi decoder design for the IS-95 CDMA forward link
Author :
Lou, H.
Author_Institution :
AT&T Bell Labs., Murray Hill, NJ, USA
Volume :
2
fYear :
1996
Firstpage :
1346
Abstract :
The IS-95 code division multiple access (CDMA) system uses a rate 1/2 256-state convolutional code for error control in the forward link (from the base station to the mobiles). This paper describes the implementation of the Viterbi decoder for the rate-1/2 256-state convolutional code. We first discussed the effects of quantization of the input symbols on the Viterbi decoder performance. We determined that, with the proper placement of coded symbols, 4 bits of uniform quantization will result in about 0.05 dB loss in performance compared to that of using infinite precisions in Gaussian channels. We showed that only negation operations are required to compute the branch metrics, and path metrics can be updated with a straight-forward 2´s complement subtraction without any normalization. We also derived that 5 bits and 8 bits are sufficient to store the branch metrics and the path metrics, respectively, in the decoder. In the control, sync and signaling channels, the convolutional encoder encodes the data continuously. Thus, decoding has to be done continuously. Therefore, we studied the effects of using a different number of latency stages before the survivor path can be traced and a data symbol can be decoded with high confidence
Keywords :
Gaussian channels; Viterbi decoding; cellular radio; code division multiple access; convolutional codes; CDMA forward link; Gaussian channels; IS-95 mobile radio; Viterbi decoder design; branch metrics; code division multiple access; control channel; convolutional code; error control; latency stages; negation operations; path metrics; quantization effects; rate 1/2 256-state code; signaling channel; survivor path; sync channel; Base stations; Convolution; Convolutional codes; Decoding; Error correction; Gaussian channels; Multiaccess communication; Performance loss; Quantization; Viterbi algorithm;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Vehicular Technology Conference, 1996. Mobile Technology for the Human Race., IEEE 46th
Conference_Location :
Atlanta, GA
ISSN :
1090-3038
Print_ISBN :
0-7803-3157-5
Type :
conf
DOI :
10.1109/VETEC.1996.501532
Filename :
501532
Link To Document :
بازگشت