Blind channel and carrier frequency offset estimation using periodic modulation precoders

Abstract

Abstract-Recent results have shown that blind channel estimators, which are resilient to the location of channel zeros, color of additive stationary noise, and channel order overestimation errors, can be developed for communication systems equipped with transmitter-induced cyclostationarity precoders. The present paper extends these blind estimation approaches to the more general problem of estimating the unknown intersymbol interference (ISI) and carrier frequency offset/Doppler effects using such precoders. An all-digital open-loop carrier frequency offset estimator is developed, and its asymptotic (large sample) performance is analyzed and compared to the Cramér-Rao bound (CRB). A subspace-based channel identification approach is proposed for estimating, in closed-form, the unknown channel, regardless of the channel spectral nulls. It is shown that compensating for the carrier frequency offset introduces no penalty in the asymptotic performance of the subspace channel estimator. Simulations are presented to corroborate the performance of the proposed algorithms.

Mixture Kalman filtering for joint carrier recovery and channel estimation in time-selective Rayleigh fading channels

“…The existing non-MKF based CFO estimators either assumes oversampling of the received signal by a factor of 8, [4], or large number of symbols in the transmission frame, (> 500), [4,5] or limited range of frequency offset estimation, ( ) respectively [1,4,5]. We do not assume any oversampling and provide full range acquisition for frequency offset estimation.…”

“…The existing CFO estimators in fading channels can be classified as either data-aided [2,3] or blind [1,4,5]. Data aided schemes use training or pilot symbols and are not bandwidth efficient [5].…”

“…Data aided schemes use training or pilot symbols and are not bandwidth efficient [5]. Blind schemes do not require any initial training sequences.…”

“…Blind schemes do not require any initial training sequences. Existing blind CFO estimators exploit cyclostationary statistics of the received signal [1,4] or use transmitter-induced cyclostationary precoders [5]. These algorithms assume either oversampling of the received signal [4] or large number of symbols in the transmission frame [1,4] or work under limited range of frequency offset estimation [1,4,5].…”

Blind identification with periodic modulation: a time-domain approach

“…Compared with the structured subspace method [2], [14] and the one-cycle subspace method [12], the proposed method requires fewer computations than the former and more computations than the latter. Detailed flop counts for these methods are given in Appendix E.…”