الفهرس | Only 14 pages are availabe for public view |
Abstract The wave division multiplexing (WDM) is one of the new technologies that have contributed to the increase of subscribers’ solutions very effectively. Dispersion of the optical signal is one of some impairments stand against the subscriber capacity increasing in the WDM. It decreases the WDM network capacity, performance and causes Inter-Symbol Interference (ISI). To reduce the dispersion effect and enhance performance of the WDM system, a communication system model has been proposed and investigated through computer simulation. The model consists of uniform cascaded FBGs act as four stages. Each stage reduces the spectral width of the optical signal. The validation of the model is accomplished by comparing the output for each stage between Matlab and Optisystem7 individually. The FBGs of the model are apodized with many functions and used to enhance the performance of WDM system with bit rate 10 GB/s and optical fiber 100 km. it is found increasing in the performance as increasing in stages. The best apodization functions accomplished the best performance are Cauchy, Sinc and Blackman. After that, the model is used in three proposed dispersion reduction schemes pre, post, and symmetrical to enhance performance of simulated WDM system with Optisystem has a bit rate 10 GB/s and optical fiber 200 km. Each scheme has cases and each case has a certain connection of the mathematical model. Q-factor and BER are noticed before and after implementing the proposed schemes. The obtained results refer enhancing in performance using the proposed schemes than without. Q-factor without the schemes 6.8 where it gives 7.64 for case three in the pre, 9.09 in case two in the post and 9.28 in case two in the symmetrical. Finally the dispersion compensation process of CFBG is performed by using many apodization profiles. Q-factor, BER and eye diagram are noticed for each profile. It is found Hamming, Cauchy and Blackman profiles perform the compensation process and achieve the best Q-factors: 8.27, 7 and 6.78, respectively. Gaussian and uniform represents the smallest value 5.3 and 2.67 respectively where the poorest compensation process. |