Zadoff-Chu sequence based time-domain SNR estimation for OFDM systems

Abstract

The optimization of wireless orthogonal frequency division multiplexing (OFDM) systems is based on the knowledge of the signal-to-noise ratio (SNR). Most of the existing methods for the SNR estimation operates in the frequency-domain and assumes perfect carrier frequency offset (CFO) synchronization. However, it has been shown that in real systems in the presence of the residual CFO this estimators are prone to performance degradation. This problem can be avoided by estimating the SNR in the time-domain. In this paper, a novel low-complexity time-domain SNR estimator based on time periodic Zadoff-Chu (ZC) sequence, named TDZCE, is proposed. When applied to the time periodic synchronization preamble, commonly used in packed based OFDM systems, the special property of a ZC sequence preserves the considered comb-type structure both in the frequency and time-domain. This allows for easy adaptation of previously proposed moment-based SNR estimator to be applied to the time-domain samples. The TDZCE outperforms the existing time-domain SNR estimators in frequency-flat channels in the presence of the CFO, and approaches the Cramer-Rao bound (CRB) as the number of periodic parts of the proposed preamble increases. Additionally, it offers considerable computational savings compared to the existing time-domain SNR estimators.