Bayesian estimation of clipped Gaussian processes with application to OFDM

Paolo Banelli, Geert Leus, Georgios B. Giannakis

Research output: Contribution to journalConference articlepeer-review

16 Scopus citations


Several engineering applications ranging from control to communications have to deal with a clipped Gaussian process, observed in the presence of Additive White Gaussian Noise (AWGN). For such a scenario, we derive in this paper a closed form expression of a Bayesian estimator, which recovers the original undistorted Gaussian process by minimizing the mean square estimation error. In addition, we use the obtained closed form expression to show that the Bayesian estimator results in a Bit-Error Rate (BER) improvement compared to existing receivers for an Orthogonal Frequency Division Multiplexing (OFDM) system in an AWGN channel that is impaired by a clipping device at the transmitter.

Original languageEnglish (US)
Article number7071943
JournalEuropean Signal Processing Conference
StatePublished - Mar 27 2002
Event11th European Signal Processing Conference, EUSIPCO 2002 - Toulouse, France
Duration: Sep 3 2002Sep 6 2002


Dive into the research topics of 'Bayesian estimation of clipped Gaussian processes with application to OFDM'. Together they form a unique fingerprint.

Cite this