Data channel with joint data estimation and timing recovery

Abstract

A data channel includes a data detector that approximates both inter-symbol-interference (ISI) and random timing perturbations using a single finite-state hidden Markov model. The ISI is approximated by a finite impulse response and the timing perturbations are approximated by a first order random walk. The data signal, which is subject to inter-symbol interference and timing perturbations, is sampled periodically over a succession of time epochs without regard to timing perturbations. Timing perturbation values and data states are then assigned for each epoch, and each timing perturbation value is paired with each data state to arrive at a set of composite states. Probabilities are then assigned between composite states in successive epochs to arrive at the most probable composite state sequence corresponding to the sequence of detected data values from the sampled data. A Viterbi algorithm is then applied to find the maximum likelihood sequence of composite states. The resulting composite state sequence then defines both the data sequence and timing trajectory which, jointly, are most likely given the observations.

Dependencies are inferred from the claim text.