Developing a novel algorithm for the automated design and simulation of 3D-printed mode-selective photonic lantern spatial (de)multiplexers, incorporating tolerance analysis for fabrication imperfections

The increasing demand for high-capacity optical communication systems has driven the development of advanced mode-selective photonic devices. This study presents a novel algorithm for the automated design and simulation of 3D-printed mode-selective photonic lantern spatial (de)multiplexers. The proposed algorithm integrates computational optimization techniques to design efficient mode converters while incorporating tolerance analysis to assess fabrication imperfections. By leveraging machine learning-based optimization and finite-difference time-domain (FDTD) simulations, the approach ensures robust device performance despite deviations introduced during the 3D printing process. The results demonstrate enhanced mode selectivity, reduced insertion loss, and improved resilience to structural variations. This work contributes to the advancement of scalable and cost-effective spatial multiplexing solutions for next-generation optical networks