Synchronisation of the superimposed training method for channel estimation in the presence of DC-offset
Enrique Alameda-Hernandez, Des McLernon, S. M. A. Moosvi, M. Lara, Aldo Gustavo Orozco-Lugo, Mounir Ghogho · White Rose Research Online (University of Leeds, The University of Sheffield, University of York) · 2005
The superimposed training method estimates the channel from the induced first-order cyclostationary statistics exhibited by the received signal. In this paper, using vector space decomposition, we show that the information needed for training sequence synchronisation, and for DC-offset estimation, can be extracted from the first-order cyclostationary statistics as well. Necessary and sufficient conditions for channel computation and equalisation are derived, when training sequence synchronisation and DC-offset removal are required. The computational burden of the practical implementation of the method presented here is much lighter than for existing algorithms. At the same time, simulation results show that the performance, in terms of the MSE of the channel estimates and BER, is not diminishedwhen compared to these existing algorithms.