Research On Key Technologies In Multi-band Optical Communication Systems

To meet the higher demands on broadband and high-capacity optical fiber communication systems for the emerging technologies such as 5G,artificial intelligence,and autonomous driving,network traffic is growing rapidly,placing higher demands on broadband wide,high-capacity fiber optical communication systems.Currently,the capacity for communication in C+L bands is approaching the Shannon limit and needs to be extended to O,E,S and U bands to provide more available bandwidth.However,As the number of channels systems grows,the interference between channels increase sharply,seriously affecting the quality of signal transmission.Therefore,research on key technologies in multi-band optical comunication systems is of great significance.To improve the transmission performance of multi-band transmission systems,this thesis investigates the theoretical model of nonlinear effects in wavelength division multiplexing(WDM)systems,proposes a Raman effect compensation method based on Raman amplifiers and Partition Particle Swarm Optimization algorithms,and presents a noise compensation method based on regularization and dynamic parameter adjustment of adaptive equalizers.The following are the main research works of this thesis:(1)A channel model for five-band transmission systems(O,E,S,C,L)is established to investigate the impact of nonlinear effects on the transmission system.The model is based on the Gaussian noise(GN)model,considering the frequency-dependent power distribution of the signal,and modeling the nonlinear interference in the system.The influence of channel number,channel spacing,and signal power on the system is analyzed.The simulation results show that the average error of the proposed channel model is 0.33 dB,indicating its reliability.(2)A Raman effect compensation method based on Raman amplifiers and Partition Particle Swarm Optimization algorithms is proposed to address the signal distortion problem caused by power transfer due to the Raman effect in the three-band transmission systems(S,C,L).This method uses Raman amplifiers to amplify the output signal power spectrum and adaptively updates the pump power and pump wavelength of the Raman amplifier using the partition particle swarm optimization algorithm.In the process of the algorithm,particles are divided into preferred and sparse areas,and different update strategies are used to improve the adaptability and convergence speed of the algorithm.The simulation results show that this scheme can reduce the offset of the output signal spectrum from 1.2 mW to 0.29 mW in S,C,and L bands,while improving the generalized mutual information(GMI)of S-L band signals by 2.12 bit/symbol,0.2 bit/symbol,and 0.12 bit/symbol,respectively.Compared to the traditional particle swarm algorithm,this method improves the iteration speed by 10%.(3)A noise copensation scheme based on regularization and dynamic parameter adjustment of adaptive equalizers is proposed to address the comprehensive noise from amplifier spontaneous emission(ASE)and transceivers in five-band transmission systems(O,E,S,C,L).The scheme uses regularization techniques to avoid over-fitting problems and dynamically adjusts the weight vector according to the signal error to accelerate the convergence speed of the algorithm.The simulation results show that compared to the conventional RLS algorithm,the SNR cost of the scheme is reduced by 3.31 dB,1.2 dB,2.4 dB,0.89 dB and 0.6 dB in O,E,S,C and L bands,respectively,when the BER reaches the threshold of 3.8×e-3.