Impact Analysis of the Use of the Power Mask Spectrum Information in the Adaptive Control of Optical Amplifiers

One of the significant challenges in the dynamic optical networks is the adaptive control of operating point (ACOP) of optical amplifiers. The ACOP approaches aim to autonomously define the gain of the amplifier after changes in the amplifier input power. The ACOP approaches proposed so far use the amplifier power mask to define the gains. The full power mask has information about gain and noise figure for each channel considered in characterization. However, most of these approaches consider a simplified version of the power mask that stores just the worst ripple and worst noise figure among all channels for each operating point. In this paper, we analyze the impact of using the full power mask, instead of the simplified power mask, in one of the first proposed ACOP approaches. We also used the nonlinear interference model in all analyzed cases to evaluate the impact of the non-linear effects for each operating point. The results show that it is possible to achieve results for the compact power mask that are close to the ones obtained from the full power mask for a cascade of amplifiers. These results suggest that the more straightforward approach using the worst ripple and the worst noise figure among all channels for each operating point can present adequate results if small optical signal-to-noise ratio and ripple penalties are acceptable.