Low-complexity Precoding-aided CFO Estimation for ICA-based MIMO OFDM Systems in URLLC
Zhening Liu,Yufei Jiang,Xu Zhu,Sumei Sun
DOI: https://doi.org/10.1109/tcomm.2024.3462699
IF: 6.166
2024-01-01
IEEE Transactions on Communications
Abstract:Carrier frequency offset (CFO) and channel equalization are two critical problems for multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) wireless communication systems in ultra-reliable and low latency communication (URLLC). In this paper, we propose a semi-blind precoding aided structure that includes two CFO estimation approaches and an independent component analysis (ICA) based equalization scheme for MIMO OFDM systems in URLLC, requiring no pilots. We design a non-redundant balanced precoding strategy, killing two birds with one stone, where reference signals are superimposed into source signals to simultaneously allow CFO estimation and ambiguity elimination in the ICA-equalized signals. The proposed precoding-aided CFO estimation approach performs by maximizing a cost function formulated by the cross-correlations between the reference signal and the received signal. We further propose a low-complexity closed-form CFO estimation approach, by transforming the formulated cost function into a new expression. To maximize bit error rate (BER) performance, particle swarm optimization (PSO) is employed to perform the joint optimization of precoding constant and the number of OFDM blocks for CFO estimation, while avoiding exhaustive search. The proposed semi-blind precoding-aided structure provides a trade-off between performance, complexity and spectral efficiency for MIMO OFDM systems in URLLC.