Learning-enabled recovering scattered data from twisted light transmitted through a long standard multimode fiber

Multiplexing multiple orbital angular momentum (OAM) modes of light has proven to be an effective way to increase data capacity in fiberoptic communications. However, existing techniques for distributing the OAM modes rely on specially designed fibers or couplers. Direct transmission of multiplexed OAM modes through a long standard multimode fiber remains challenging because the strong mode coupling in fibers disables OAM demultiplexing. Here, we propose a deep-learning-based approach to recover the scattered data from multiplexed OAM channels without measuring any phase information. Over a 1-km-long standard multimode fiber, our method is able to identify different OAM modes with an accuracy of more than 99.9% in the parallel demultiplexing of 24 scattered OAM channels. To demonstrate the transmission quality, color images are encoded in multiplexed twisted light and our method achieves decoding the transmitted data with an error rate of 0.13%. Our work shows that the artificial intelligence algorithm could benefit the use of OAM multiplexing in commercial fiber networks and high-performance optical communication in turbulent environments. Published under an exclusive license by AIP Publishing.

Automated summary

This article presents a deep-learning-based approach for recovering scattered data from multiplexed OAM channels. The method achieves high accuracy in identifying OAM modes in a standard multimode fiber and demonstrates low error rates in transmission quality.