Abstract

The spatial mode demultiplexer is a critical component in the mode division multiplexing communication system, which requires the merits of power-efficiency and low-crosstalk [1]. It has been demonstrated that any unitary transform of spatial modes can be implemented by passing through a succession of phase masks with free-space propagation in between, and the key is to inversely design the phase distribution of each phase mask. One inverse approach is based on wavefront matching method, which is employed in the technique termed multi-plane light conversion (MPLC) [1,2]. This approach enables an efficient optimization of overlap integrals between the obtained modes and the desired modes. However, this method does not have an objective function, meaning that it is incapable of implementing specific optimization. Another approach is based on the gradient descent algorithm [3], which is flexible in choosing objective functions, while the performance in maximizing overlap integrals is less desirable when compared with wavefront matching method.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call