Optimization Of Gain Flatness Properties Of Broadband Fiber Raman Amplifiers

As the internet has come to wide use, modern communications systems are developed ultra-rapidly in the direction of high-capacity, long-distance and high-speed. Fiber Raman amplifiers (FRA) is one of the key devices for improving transmission distance of the dense wavelength division multiplexing (DWDM) optical fiber communication system. FRA uses transmission fiber as a gain medium with properties of broad gain spectrum, high saturation power, high signal to noise ratio and distributed amplification. It is increasingly becoming the research hotspot of optical fiber communications systems. The optimal design algorithms obtaining broadband gain flatness FRA is studied in this paper. The least square method is put forward as a new broadband FRA gain flatness optimal design algorithm for the first time and confirmed feasibility in the study. There are mainly four aspects:Firstly, based on the simplified SRS power transfer equation, a variety of different algorithms which is used to solve the output power of FRA are compared. The variable step Runge-Kutta method and the average power method are used to solve FRA power coupled equations respectively and the variable step-size shooting method is used to solve the two-point boundary problem.Secondly, how the fiber connection loss influence the FRA Characteristics of output power is analyzed for the first time, which permits the research results to access the actual transmission system. Then the actual transmission experimental system can be better optimized. The simulation results show that the impact of fiber connection loss to the FRA output power characteristics is related to the magnitude of depletion and the location of fiber connection and the output power spectrum shape will not be influenced greatly.Thirdly, the least square method is put forward as a new broadband FRA gain flatness optimal design algorithm for the first time. Satisfying results is obtained using this algorithm for three-pumped, four-pumped and five-pumped FRA respectively. Pumped gain flatness of optimized, so the feasibility and effectiveness of the algorithm is confirmed.Fourthly, by comparing the properties of least squares method to other methods, we found that the least square method is more suitable for dynamic optimizing the working FRA. In addition, a dynamic optimization model using least square method is raised to optimize the gain flatness characteristics of FRA. After the characteristics of genetic algorithm and least square method are analyzed and compared, a new optimization method is proposed.