APC 2021 - Programmable Photonics with Silicon Photonic MEMS - Wim Bogaerts

[@meetoptics]

Congratulations for introducing this topic so well in KZbin.This platform let us spread all we know about the field and from MEETOPTICS we are proud to be part of the photonics community and to help engineers and researchers in their search for optical lenses through our site. We celebrate every step forward.

[@chaselewis6534]

I'm curious how long is the time to switch in a MEMS photonic platform? (i'm dumb you said ~10 microseconds) This seems like it would have the ability to create what is essentially a photonic FPGA but likely wouldn't be suitable for live calculations.

[@wbogaert]

Indeed. about 10us. You would not be able to switch the MEMS fast to perform 'live' switching. But the MEMS is more like the 'programming' of an FPGA. It is the light itself, by propagating through the chip, that performs the 'calculations' or 'signal processing'. That being said, there are quite a lot of applications where you would really want very fast (ns) switching.

[@AtlasReburdened]

If you can scale up in terms of number of switches without massive compromises, then there could be little need to have fast switching. My kinda overaching grasp of this is that with all your switches and phase offsets in a given intended state, the switch assembly is essentially simulating a pure equation without the concessions normally associated with handling true analog functions in a digital medium. Inputs go in, and the true solution to that equation/input set comes out. All this to say that with enough switches and phase controllers, the equation being simulated by the switch assembly can be as expansive and convoluted as desired, and theoretically could involve enough series, parallel, and back fed functions that the total propagation and solution time approaches the maximum switching speed. Entropy willing, of course.