Blind Channel Identi cation Based On Cyclic Statistics

Abstract : Blind channel identi cation and equalization based on second-order statistics by subspace tting and linear prediction have received a lot of attention lately. On the other hand, the use of cyclic statistics in fractionally sampled channels has also raised considerable interest. We propose to use these statistics in subspace tting and linear prediction for (possibly multiuser and multiple antennas) channel identi cation. We base our identi cation schemes on the cyclic statistics, using the stationary multivariate representation introduced by [6] and [9] [10]. This leads to the use of all cyclic statistics. The methods proposed, compared to classic approaches, have equivalent performance for the subspace tting and enhanced performance for linear prediction.