An Extended Kalman Filter Framework for Joint Phase Noise, CFO and Sampling Time Error Estimation

We present a framework for joint estimation and compensation of three major oscillator impairments, namely sampling time error (STE), carrier frequency offset (CFO) and phase noise (PN). In particular, we model these impairments as Wiener processes and introduce a pilot-aided approach which facilitates their joint estimation. The proposed solution is carried out in two steps: first, an initial estimation of the transmitted symbols is acquired by applying an extended Kalman filter (EKF) on the pilot symbols and then, a second EKF is applied on the estimated symbols which yields an accurate tracking of STE, PN and CFO over an additive white Gaussian noise channel. Our numerical results demonstrate the efficacy of the proposed solution.