We consider a free-space optical (FSO) communication system employing on-off keying (OOK) signaling in the presence of background radiation and intersymbol interference (ISI) arising as a consequence of a sampling clock offset. The receiver has no prior information about the instantaneous channel coefficient, and shot noise with signal-dependent power is inserted by the photodetector device. In such a harsh scenario, suitable estimates of the timing error, channel attenuation and noise parameters are needed for reliable data detection. In this work, we employ the Expectation-Maximization (EM) algorithm for the joint estimation of all the aforementioned parameters. In order to obtain a feasible scheme, the maximization step of the EM procedure is split into a sequence of simpler optimization tasks according to the Expectation-Conditional Maximization (ECM) principle. The resulting algorithm operates in a blind fashion using symbol-spaced samples, which is particularly advantageous in high-speed FSO communications where signal oversampling may require sophisticated hardware equipment. Computer simulations indicate that the accuracy of the proposed scheme is close to the relevant modified Cramér-Rao bounds (MCRBs), which amounts to saying that no room is left for a further improvement of the system performance.
Blind Estimation of Timing Error, Channel Attenuation and Noise Parameters in OOK Free-Space Optical Communications
D'Amico A. A.
;Morelli M.
2024-01-01
Abstract
We consider a free-space optical (FSO) communication system employing on-off keying (OOK) signaling in the presence of background radiation and intersymbol interference (ISI) arising as a consequence of a sampling clock offset. The receiver has no prior information about the instantaneous channel coefficient, and shot noise with signal-dependent power is inserted by the photodetector device. In such a harsh scenario, suitable estimates of the timing error, channel attenuation and noise parameters are needed for reliable data detection. In this work, we employ the Expectation-Maximization (EM) algorithm for the joint estimation of all the aforementioned parameters. In order to obtain a feasible scheme, the maximization step of the EM procedure is split into a sequence of simpler optimization tasks according to the Expectation-Conditional Maximization (ECM) principle. The resulting algorithm operates in a blind fashion using symbol-spaced samples, which is particularly advantageous in high-speed FSO communications where signal oversampling may require sophisticated hardware equipment. Computer simulations indicate that the accuracy of the proposed scheme is close to the relevant modified Cramér-Rao bounds (MCRBs), which amounts to saying that no room is left for a further improvement of the system performance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.