DocumentCode
787580
Title
Detailed noise statistics for an optically preamplified direct detection receiver
Author
Danielsen, Soeren Lykke ; Mikkelsen, Benny ; Durhuus, Terji ; Joergensen, Carsten ; Stubkjaer, Kristian E.
Author_Institution
Electromagn. Inst., Tech. Univ. Denmark, Lyngby, Denmark
Volume
13
Issue
5
fYear
1995
fDate
5/1/1995 12:00:00 AM
Firstpage
977
Lastpage
981
Abstract
We describe the exact statistics of an optically preamplified direct detection receiver by means of the moment generating function. The theory allows an arbitrary shaped electrical filter in the receiver circuit. The moment generating function (MGF) allows for a precise calculation of the error rate by using the inverse Fast Fourier transform (FFT). The exact results are compared with the usual Gaussian approximation (GA), the saddlepoint approximation (SAP) and the modified Chernoff bound (MCB). This comparison shows that the noise is not Gaussian distributed for all values of the optical amplifier gain. In the region from 20-30 dB gain, calculations show that the GA underestimates the receiver sensitivity while the SAP is very close to the results of our exact model. Using the MGF derived in the article we then find the optimal bandwidth of the electrical filter in the receiver circuit and calculate the sensitivity degradation due to inter symbol interference (ISI)
Keywords
fast Fourier transforms; intersymbol interference; inverse problems; optical noise; optical receivers; sensitivity analysis; 20 to 30 dB; Gaussian approximation; electrical filter; error rate; inter symbol interference; inverse Fast Fourier transform; modified Chernoff bound; moment generating function; noise statistics; optically preamplified direct detection receiver; optimal bandwidth; receiver circuit; receiver sensitivity; saddlepoint approximation; Circuit noise; Error analysis; Filtering theory; Noise shaping; Optical detectors; Optical filters; Optical noise; Optical receivers; Optical sensors; Statistics;
fLanguage
English
Journal_Title
Lightwave Technology, Journal of
Publisher
ieee
ISSN
0733-8724
Type
jour
DOI
10.1109/50.387817
Filename
387817
Link To Document