I recently attended a lecture on Optomechanical accelometry for inertial navigation, from NIST (National Institute for Science and Technology in the US).
It's very clever, probably an advance in the art. They are at the noise floor for such devices. One fundamental limitation they mentioned was "laser noise" causing a 1/f noise peak at low frequencies. This rang a bell - didn't Phil Hobbs invent a way to cancel laser relative intensity noise (RIN)?
Yes. And it seems to be a perfect place to use this method, so I told the speakers of US patent 5,134,276 and the companion article in the literature. They had heard of Phil and have his book on electro optics. It would be dead simple to cancel RIN in their application, so I bet they will do it.
The application:
.
Look at the first two papers listed under "Publications", "Broadband thermo mechanically limited sensing with an Optomechanical accelerometer" and "Electrooptic frequency combs for rapid interrogation in cavity optomechanics".
More papers are likely to emerge.
Joe Gwinn