Local and global approaches for the adaptive control of a cascade of amplifiers

In this work we present and analyze three approaches for the adaptive control of the operating point of a cascade of erbium-doped fiber amplifiers (EDFAs), aimed at optical networks performance enhancement. The first approach is called Annealing Search Heuristic with Backpropagation and flexible output (AsHB flex) and uses machine learning concepts to update the amplifier gains through an iterative process. The second one (Exhaustive Method) uses an exhaustive search to evaluate all possible solutions for the problem and obtain the optimum solution. The last one (MaxGain) is a heuristic method that uses previous knowledge about the problem to obtain the solutions. The amplifier characteristics and specifications were obtained experimentally through measurements of gain, noise figure, gain ripple and power consumption on commercially available EDFAs. We performed comparisons among these approaches and others found in the literature, and the results show that the three proposals outperformed the previous ones in terms of noise figure, gain ripple and BER. For example, in a link with four amplifiers the Exhaustive Method achieved a reduction in the cascade noise figure from 10.05 to 5.18 dB, a reduction in the gain ripple from 24.08 to 18.56 dB and a reduction in the BER in almost two orders of magnitude, when compared with the traditional approach, which defines the gain to compensate the loss of the previous link. However, the computation time of Exhaustive Method becomes prohibitive as the number of amplifiers in the link increases. Both MaxGain and AsHBflex obtained similar solutions, close to the optimum operation point in a reasonable time.