x
<<   <   Page 2 / 2
^Eagle^ 12/5/2012 | 5:23:56 PM
re: Photonic Integration Starts Slowly at 100G

Dear SLUT,


I have read ALL of the published papers. &nbsp;And have seen inside a PIC module.


so I ask you a few direct questions:


1) does IFN make all the parts inside a PIC on a single inP die? &nbsp;Or


2) do they make separate die for the transmit and receive functions and co-package them?


3) what dies do they use for the optical mux / demux functions? &nbsp;are those monolithically done in InP and on the SAME die as the TX and RX chips?


4) do they not use any lenses and other bulk optics parts inside the PIC? &nbsp;


&nbsp;


easy way to prove I am wrong. &nbsp;Just show us the inside of a PIC. &nbsp;Note: showing us the inside will not give away much in the way of key intellectual property. &nbsp;If indeed it is monolithically done, it would be virtually impossible to copy any IP based on a photo of the inside of the package. &nbsp;


sailboat

furious_george 12/5/2012 | 5:23:55 PM
re: Photonic Integration Starts Slowly at 100G

Redface, Sailboat,


There are some pictures in slide 4 of this link which illustrates at least what the level of integration was a few years ago.


http://www.ieee802.org/3/hssg/public/jan07/jaeger_01_0107.pdf


Furious George


&nbsp;

redface 12/5/2012 | 5:23:55 PM
re: Photonic Integration Starts Slowly at 100G




Sailboat,


&nbsp;


Now that you posted these interesting questions and seem to know the answer, you have to let us know the answer too. &nbsp;


&nbsp;


"do they make separate die for the transmit and receive functions and co-package them"


&nbsp;


I would think transmit and receive dies have to be separate devices and they do not need to reside in the same package. &nbsp;Am I missing something?


&nbsp;


Thanks for your answer.


&nbsp;


Redface




melao2 12/5/2012 | 5:23:54 PM
re: Photonic Integration Starts Slowly at 100G

This is for&nbsp;10x10G SMF PHY. I suppose now the integration is very different. Even though it is quite impressive for the time the integration.

redface 12/5/2012 | 5:23:53 PM
re: Photonic Integration Starts Slowly at 100G

Furious George,


&nbsp;


Thanks for the link. &nbsp;It is interesting stuff.


&nbsp;


So it looks like INFN has ten transmitters spaced 200 GHz apart on each chip. &nbsp;The outputs from four such chips (with slight wavelength difference) are combined together into one fiber to cover the enire C-band. &nbsp;


&nbsp;


Redface

<<   <   Page 2 / 2
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE