Fan-Shuo Tseng, Meng-Lin Ku, Tsang-Yi Wang, Bo-Ting Li, Fakhriy Hario Partiansyah
In orthogonal frequency-division multiplexing (OFDM) systems, the received signal is practically impaired by oscillator phase noise (PHN), which varies across OFDM symbols. PHN introduces inter-carrier interference (ICI) and degrades overall detection performance. To address this issue, a novel joint PHN and channel estimation scheme is proposed for single-input single-output (SISO) and multiple-input single-output (MISO) systems under limited pilot conditions. The proposed approach commences by accurately characterizing the statistics of the common phase error (CPE) and deriving an approximate a posteriori probability distribution of the CPE and channel. The overall design is structured into two phases: training and tracking. In the training phase, the CPE and channel are jointly estimated using an approximate maximum a posteriori (MAP) detector, providing a principled approach toward optimality under the assumed model. Since a static channel is assumed and only the PHN varies across OFDM symbols, a small number of comb -type pilots are inserted in the tracking phase to estimate the CPE using the proposed approximate MAP framework, leveraging the channel estimate obtained in the training phase. Subsequently, the updated estimates of the CPE and channel can be used for data symbol detection. Simulation results demonstrate that the proposed scheme outperforms existing methods in terms of normalized mean square error (NMSE), particularly in scenarios with reduced pilot overhead. © 1967-2012 IEEE.
National Sun Yat-sen University, Institute of Communications Engineering, Kaohsiung, Taiwan; National Sun Yat-sen University, Southern Taiwan Industry Promotion Center, Kaohsiung, 804, Taiwan; National Central University, Department of Communications Engineering, Jhongli, Taiwan; Universitas Brawijaya, Department of Electrical Engineering, Malang, Indonesia
