Automatic Identification System (AIS) signals contain ship tracking information used in navigation for collision avoidance, by coast guards in search and rescue missions, and for port security.We recently introduced a set of filters matched to AIS tri-bit states and phases and applied them to demodulate simulated AIS signals in additive white Gaussian noise (AWGN) and in co-channel interference. We assumed that the carrier frequency and carrier phase of the target AIS signal were known, and that their contributions were removed prior to demodulation. We also assumed that the filters were properly aligned relative to the baud starts, to avoid phase mismatch. In this article we introduce algorithms for estimating carrier frequency, carrier phase and time alignment offset prior to demodulation. We compare demodulation bit error rate (BER) found with the algorithm estimates to BER obtained when these parameters are known, both in AWGN and in co-channel interference, as a function of signal-to-interference ratio (SIR) and as a function of carrier frequency difference (CFD) between two colliding AIS signals. Simulations showed very mild, if any, BER degradation over a wide range of SNR, SIR and CFD, attesting to the robustness of the algorithms.