As we all know, the change of mode interference caused by the curvature change in multi-mode fiber (MMF) can be well represented as a fiber specklegram recorded by CCD (Charge-coupled Device). However, it's difficult to identify the different bending occurred in several points in short distance. The paper proposes plastic fiber bending sensors can be used to detect the multi-point bending without adding any hardware. The convolutional neural network was used to classify the output speckles under different bending states. Specklegrams from fiber with three sensitization areas can be recognized by the neural network with a bending interval of 15°,10° and 5° with an accuracy rate of 99.2%, 96.1% and 93.5% respectively. Compared with traditional multi-point distributed sensors, this method is lower cost and easier to operate. The method proposed in this paper can find applications in distinguishing the status of certain structures, such as robotic arms and some disabled auxiliary equipment.

There is an increasing demand for data with the development of the world, and various fiber optic multiplexing techniques have become an important research direction to improve transmission capacity. However, the transmitted signals are subject to great interference due to mode coupling and mode dispersion, which require multiple-input multiple-output (MIMO) digital signal processing techniques to restore the quality of the transmitted signals. In this paper, a novel MIMO detector is designed using an adaptive learning recurrent neural network and successfully implemented in a mixed wavelength-division-mode-division-multiplexing (WDM-MDM) optical transmission system, and its performance is compared with that of the forced-zero detector and the minimum-mean-square-error detector. The results show that the introduction of an adaptive machine learning model in MIMO detection for WDM-MDM optical transmission systems can significantly improve the quality of the transmitted signals and achieve better performance than other MIMO detection algorithms while maintaining a faster computational speed and a lower number of parameters.