CFO and Channel Estimation Techniques for GFDM

Abstract

Carrier frequency offset (CFO) caused by the misalignment of the transmitter and receiver local oscillators can adversely affect the performance of any multicarrier system if not accurately estimated and corrected. Thus, in this paper, we propose a CFO and channel estimation technique based on the maximum-likelihood (ML) criterion for generalized frequency division multiplexing (GFDM). Our proposed CFO estimator does not have any limitation on the CFO acquisition range while providing an accurate estimate. We propose a preamble block containing two frequency domain ZC (Zadoff-Chu) sequences for training which leads to a low complexity implementation of the CFO estimator. Compared with the existing solution in the literature with the largest CFO estimation range and precision, our technique brings around two orders of magnitude complexity reduction without any performance penalty. We also evaluate the performance of our proposed technique through numerical simulations while showing its superiority to the existing literature.