Timing synchronization and frequency offset estimation for ultra-wideband (UWB) multi-band OFDM systems

Abstract

This thesis addresses symbol timing synchronization and carrier frequency offset estimation under multi-band orthogonal frequency division multiplexing ultra-wideband systems in frequency selective fading channels. The proposed algorithm: First significant multipath detection via Threshold comparison of Adjacent samples (FTA) exploits periodic training sequences by comparing the difference between adjacent samples of a proposed metric against a pre-defined threshold. Three correlation algorithms of conventional systems are presented for comparison. The proposed FTA algorithm is at least 75% more accurate than the next best algorithm in terms of mean squared error and at least 53% more likely to achieve timing synchronization under a typical operating signal-to-noise ratio (SNR) of 17 dB. The maximum likelihood (ML) frequency offset (FO) estimator variance and the Cramer Rao Bound (CRB) are derived for comparison. Phase differences between each subcarrier of adjacent identical training symbols are averaged to yield the FTA FO estimator, whose variance suffers a SNR loss of less than 0.5 dB compared against the ML FO estimator variance. Further, the ML estimator variance converges to the CRB towards the high SNR end.