In this paper, we present a comprehensive foundation for the theory and practice of pilot-based carrier phase synchronization of OFDM symbols in the presence of phase noise. The system covers phase noise added in both the transmitter and receiver together with a frequency-selective channel. A novel, low-complexity, phase noise estimation method is presented, and we show that it is close to optimal over a wide dynamic range of phase noise. Additionally, a method for approximating the phase noise over the OFDM symbol using principal component analysis is proposed, where the number of parameters to estimate can be significantly reduced with minimal impact on performance.

We establish a solid foundation for the theory and practice of pilot-based carrier phase synchronization in line-of-sight (LOS) MIMO systems affected by phase noise. LOS MIMO is a multi-antenna technology used in fixed wireless backhaul systems. As part of this, a realistic and simple oscillator phase-locked loop model is employed. To maximize the rate, a novel pilot-based MMSE estimator is formulated to jointly estimate the received symbols and the unknown phase noise at both the transmitter and receiver. Additionally, we demonstrate that effective power allocation, considering phase noise, is crucial for maximizing the rate in phase noise-limited systems.

We present a solid foundation for the theory and practice of using pilot tones for estimation and compensation of phase noise in a communication link. We analyze several phase estimators and illustrate their pros and cons, and we derive the necessary optimal signal-to-pilot-power relationship when the transmitted power is shared between the communication signal and the pilot tone. We show that the optimal signal-to-pilot-power ratio increases with decreasing phase noise and with increased additive noise.