Estimation of CFO and STO for an OFDM using general ICI self-cancellation precoding

• Published in: Digital signal processing Vol. 31; pp. 35 - 44
• Main Author: Wang, Yung-Yi
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.08.2014
• Subjects: Algorithms; CFO–STO estimation; Estimates; ICI self cancellation; Mathematical analysis; MUSIC; OFDM; Orthogonal Frequency Division Multiplexing; Samples; Statistical analysis; Statistical methods; Vectors (mathematics); ICI self cancellation; MUSIC; CFO–STO estimation; OFDM
• ISSN: 1051-2004; 1095-4333
• DOI: 10.1016/j.dsp.2014.04.012

This study presents an efficient carrier-frequency-offset (CFO) and symbol-timing-offset (STO) estimation algorithm for time-dispersive orthogonal frequency division multiplexing (OFDM) system using a general inter-carrier interference (ICI) self-cancellation scheme. This study takes advantage of the time shift invariant property in the precoded signal to estimate the CFO. The proposed algorithm first stacks the receive time samples spaced by a pre-determined time interval into sample vectors, which can be expressed in a form having a CFO-directed response vector. This CFO-directed structure enables the proposed approach to estimate the CFO by using the multiple signal classification (MUSIC) algorithm in the time domain. Equalization is required before the MUSIC algorithm to remove the scaling factor generated in the signal stacking. Using the CFO estimate, the proposed approach compensates for the frequency error in the receive signal and then estimates the STO by invoking the MUSIC algorithm in the frequency domain. Unlike conventional algorithms, in addition to having a larger CFO estimation range linearly proportional to the precoding order, the proposed approach can easily handle the case of fractional STO. This study presents some statistical analysis of the undesired equalization residues and the mean square error of the perturbed MUSIC algorithm to provide further insights into the proposed approach.