Real-Time Observer and Neuronal Identification of an Erbium-Doped Fiber Laser

This paper presents the implementation of a nonlinear state observer in real-time applied to an erbium-doped fiber laser system. The observer is designed to estimate population inversion, a state variable that cannot be directly measured due to the physical limitations of the sensing devices. Taking advantage of the fact that laser intensity can be acquired in real time, an observer was developed that reconstructs the dynamics of population inversion from this measurable variable. To validate and reinforce the estimate obtained through the observer, a recurrent wavelet first-order neural network (RWFONN) was implemented, trained to identify both state variables: laser intensity and population inversion. This network efficiently captures the nonlinear dynamic characteristics of the system, complementing the observer’s performance. To evaluate the accuracy and reliability of the results, two metrics were applied: the Euclidean distance and the mean squared error (MSE), both of which confirmed the consistency between the estimated and expected values. The ultimate goal of this research is to establish a neural control architecture that combines the estimation capabilities of state observers with the generalization and modeling power of artificial neural networks. This hybrid approach opens the door to the development of more robust and adaptive control schemes for highly dynamic complex laser systems.