Decision-directed estimation of phase noise in iterative frequency-domain equalization schemes

Abstract

In this paper we propose a novel joint equalization and phase noise compensation scheme. We consider an iterative receiver that can be regarded as a modified turbo equalizer implemented in the frequency-domain where, within each iteration of the equalizer, a decision-directed estimation of the phase noise and its corresponding compensation is performed. The phase noise is modeled considering both its white noise and flicker noise components, thus describing more accurately the phase noise process. Furthermore, regarding the phase noise estimation operation and since it follows the received signal equalization step, the samples at the output of the equalizer can be described by means of an equivalent additive channel allowing us to combine the use of solutions initially proposed for the additive white Gaussian channel (AWGN) with iterative frequency-domain equalization (IFDE) which supports inter-symbol interference (ISI). Consequently, the maximum likelihood (ML) estimator of the phase noise will be adopted, though other solutions could be of choice. Finally, our simulation results show that the proposed scheme performs well even for highly-dispersive channels and moderate phase noise.