Iterative LDPC CDMA receiver with EM estimation

Abstract

This thesis proposed a scheme of obtaining an estimate of channel coefficient and noise power spectral density (PSD), using iterative expectation-maximization (EM) channel estimation, based on a low-density parity-check (LDPC) code-division multiple-access receiver. At the receiver, an initial estimate was obtained with the aid of pilot symbols. Pilot bits were distributed among subframes followed by spreading and binary phase-shift keying. Subsequent values of channel coefficient and noise PSD both were updated iteratively by the soft feedback from the LDPC decoder. The updated channel coefficient and noise PSD were iteratively passed to the LDPC decoder, which resulted in improved decoding accuracy. The algorithm was tested on both a single user for constant noise PSD and a more realistic multiuser environment for a time-varying interference-plus-noise PSD estimation.