Since the introduction of optical fibers in 1960's in communication systems, researchers have encountered many challenges to improve the signal quality at the receiver as well as transmitting the signal as distant as possible. The former was achieved by employing coherent receivers, which let us use M-array modulation formats, such as QPSK, or QAM, and polarization of the signal. The later is accomplished by the advent of optical amplifiers. Optical amplifiers enable us to compensate for the loss occurred within the fiber optic line, without the need for optical-electrical signal conversion. These amplifiers add noise to the line which interacts with the nonlinearity in the fiber line. This interaction causes phase change in the propagating signal called nonlinear phase noise, which degrades the system performance.
In this study we will derive an analytical expression for the linear and nonlinear phase noise variance in dispersion unmanaged fiber optic systems, using a first-order perturbation theory. We use numerical examples to depict the proposed system performance in terms of nonlinear phase noise variance. We will conclude that the nonlinear phase variance in a dispersion unmanaged system is much lower than the corresponding noise variance in a dispersion managed system. We will use this concept and will introduce more dispersion in the line by adding fiber brag gratings (FBGs) throughout the fiber link. Through numerical simulations, we will illustrate the improvement we get by adding FBG in each span. We will show that employing FBG improves the system performance for systems working at symbol rates 5 GBaud, which we get the best improvement to less than 20 GBaud, and beyond 20 GBaud there will be no improvement.
Nowadays, telecommunication systems based on fiber optics are working at symbol rates around 28 GBaud. We will introduce new models to reduce the nonlinear phase, by splitting digital back propagation (DBP) between transmitter and receiver, and using optical phase conjugation (OPC) in the line. We will prove that the new proposed models lower the phase noise variance significantly, for single pulses. We will also illustrate numerical examples to validate the betterment they provide in terms of Q-factor. / Thesis / Master of Applied Science (MASc)
| Identifer | oai:union.ndltd.org:mcmaster.ca/oai:macsphere.mcmaster.ca:11375/24018 |
| Date | January 2018 |
| Creators | Rahbarfam, Saber |
| Contributors | Kumar, Shiva, Electrical and Computer Engineering |
| Source Sets | McMaster University |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0019 seconds