Oversampled phase tracking in digital communications with large excess bandwidth

This paper deals with the on-line carrier phase estimation in a digital receiver. We consider a Brownian phase evolution in a data aided scenario. The proposed study uses an oversampled signal model after matched filtering, leading to a coloured reception noise and a non-stationary power signal. The contribution of this paper is twofold. First, we derive the Bayesian Cramér–Rao bound for this estimation problem. Then, based on a state-space model formulation of the problem, we propose an extended Kalman filter to approach this lower bound for a BOC shaping pulse. Our numerical results illustrate the gain resulting from the use of an oversampled version of the received signal to estimate the phase offset, obtaining better performances than using a classical synchronizer.