Paper abstract bibtex

We investigate carrier estimation (CE) for coherent optical receivers where the received signal is impaired by additive white Gaussian noise, laser phase noise, and frequency offset. Best practical 4-, 8-, and 16-point constellations are identified. A generalized differential encoding rule for signal constellations is presented. Performance of our complex-weighted decision-aided maximum-likelihood (CW-DA-ML) phase noise and frequency offset estimator is analyzed at low signal-to-noise ratio (SNR) and the optimal filter lengths are found. CW-DA-ML CE is put in perspective with respect to two fundamental estimators in the literature: (i) differential frequency estimator followed by block Mth power phase estimator (DiffFE-Mth CE), and (ii) fast Fourier transform based frequency estimator followed by block Mth power phase estimator (FFTbE-Mth CE), in terms of laser linewidth tolerance, frequency estimation range and speed, SNR threshold, and cycle slip probability. CW-DA-ML CE is 2.5 and 10.5 times faster than DiffFE-Mth CE in 4 phase-shift keying and 16 quadrature amplitude modulation signals, respectively, at a 1-dB system penalty for a bit-error rate of 10†. Our CE has lower cycle slip probability and transmission overhead than DiffFE-Mth and FFTbE-Mth CE. Hence, our CE is shown to be favourable in pilot-assisted (PA) systems. A PA CW-DA-ML CE is introduced and shown to be robust against time-varying frequency offset with minimal training overhead. Analog-to-digital convertor quantization error on our CE performance is also addressed.

@article{adaickalavan_meiyappan_decision_2013, title = {On {Decision} {Aided} {Carrier} {Phase} and {Frequency} {Offset} {Estimation} in {Coherent} {Optical} {Receivers}}, volume = {31}, copyright = {\&\#169; 2013 IEEE}, url = {https://www.osapublishing.org/jlt/abstract.cfm?uri=jlt-31-13-2055}, abstract = {We investigate carrier estimation (CE) for coherent optical receivers where the received signal is impaired by additive white Gaussian noise, laser phase noise, and frequency offset. Best practical 4-, 8-, and 16-point constellations are identified. A generalized differential encoding rule for signal constellations is presented. Performance of our complex-weighted decision-aided maximum-likelihood (CW-DA-ML) phase noise and frequency offset estimator is analyzed at low signal-to-noise ratio (SNR) and the optimal filter lengths are found. CW-DA-ML CE is put in perspective with respect to two fundamental estimators in the literature: (i) differential frequency estimator followed by block Mth power phase estimator (DiffFE-Mth CE), and (ii) fast Fourier transform based frequency estimator followed by block Mth power phase estimator (FFTbE-Mth CE), in terms of laser linewidth tolerance, frequency estimation range and speed, SNR threshold, and cycle slip probability. CW-DA-ML CE is 2.5 and 10.5 times faster than DiffFE-Mth CE in 4 phase-shift keying and 16 quadrature amplitude modulation signals, respectively, at a 1-dB system penalty for a bit-error rate of 10\&\#x2020;. Our CE has lower cycle slip probability and transmission overhead than DiffFE-Mth and FFTbE-Mth CE. Hence, our CE is shown to be favourable in pilot-assisted (PA) systems. A PA CW-DA-ML CE is introduced and shown to be robust against time-varying frequency offset with minimal training overhead. Analog-to-digital convertor quantization error on our CE performance is also addressed.}, language = {EN}, number = {13}, urldate = {2018-10-26TZ}, journal = {Journal of Lightwave Technology}, author = {{Adaickalavan Meiyappan} and {Pooi-Yuen Kam} and {Hoon Kim}}, month = jul, year = {2013}, keywords = {Linewidth, Phase noise, Quadrature amplitude modulation}, pages = {2055--2069} }

Downloads: 0