Blind multiuser symbol estimation with iterative decoding

We propose a new iterative receiver structure that jointly optimizes blind multiuser sequence estimation and convolutional decoding. The receiver can be used to decode multiple user signals that are spatially multiplexed over a frequency selective channel. The intersymbol interference (ISI) and multiple access interference (MAI) are jointly corrected and iteratively reduced using the "turbo principle". In addition to equalization and multiuser detection, convolutional coding at the transmitter provides additional coding gain leading to low bit error rates.