A vector quantizer for additive white and colored Gaussian noise channels

Most of the work on joint source-channel coding deal with discrete channel models. We investigate the design of joint source-channel coding scheme for continuous-amplitude channels. More specifically, we propose a channel-optimized vector quantizer (COVQ) for additive white Gaussian noise (AWGN) and additive colored Gaussian noise (ACGN) channels with soft-decision binary phase-shift keying (BPSK) modulation.