Performance of ML rate determination for an IS-95 downlink SIC receiver

Maximum likelihood (ML) decision rules, such as that presented by Gutierrez, Lee and Mandyam (see. Proc. IEEE 49th Vehicular Technology Conf., Houston, TX, p.417-21, 1999), can be used for rate determination in an IS-95 system to reduce the complexity of the Viterbi decoder. We extend this previous work by determining whether it is beneficial to use rate determination for the successive interference canceller (SIC) in an IS-95 downlink receiver. Assuming certain conditions are met regarding the signal-to-noise ratio (SNR) and the number of received symbols used for rate determination, employing knowledge of bit repetition can decrease the bit error rare (BER) for the corresponding Walsh channels. To further increase the effectiveness of interference cancellation, soft decision decoding can be used instead of a conventional hard decision device. The devices considered here are similar to those of Frey and Reinhardt (see Proc. IEEE 47th Vehicular Technology Conf., Phoenix, AZ, p.155-159, 1997), but they use the knowledge that there can be repeated bits in the IS-95 downlink.