Imagine an object in empty space accelerating rightwards. Now information from beyond a certain distance to its left, limited to light speed, can't catch up so there is an information horizon, and the volume to the left of that horizon can never be seen by the object. A radiation like Hawking radiation comes off this horizon (Unruh radiation).

The first new contribution of MiHsC is that it says Unruh radiation is damped between the object and the horizon. Ernst Mach said, 'what can't be seen, can't exist', so from the point of view of the object all fields beyond the horizon must be zero, so they must be zero on the horizon too. Therefore, only waves with nodes at the horizon can be allowed. This cuts out many Unruh waves on the left side but not on the right, so more Unruh radiation hits the object from the right, slowing its acceleration. This correctly models what we call inertia, which has never before been explained. In general the horizon makes the zero point field, hitherto hidden, suddenly vary in space and able to push on real objects (see reference McCulloch, 2013).

The second part of MiHsC is that the Unruh radiation is also damped slightly by the far off Hubble horizon which cuts out some long wavelength Unruh radiation. This weakens the part of MiHsC described above, and so weakens inertia slightly in a new way. This makes more difference for low acceleration objects (which see long Unruh waves) and in this way MiHsC correctly models galaxy rotation without dark matter, and cosmic acceleration without dark energy (McCulloch, 2012).

Now to the emdrive: the photons in the cavity have inertial mass (photons in a mirrored box do) so MiHsC should apply. If we assume that the cavity walls are like the cosmic horizon in MiHsC then they will damp some Unruh radiation and make inertia weaker, but less so at the wider end. So the photons at the wide end have more inertia, and photons gain mass going from the narrow to the wide end. You'll note that mass-energy is not conserved in the usual way here because the horizon causes the zero point field to become real: just as black hole horizons cause virtual particles to become real (Hawking radiation). To conserve momentum the cavity has to move towards the narrow end. This predicts the emdrive results quite well, see here.

The application of MiHsC to the emdrive is summarised and derived properly (mathematically) in the paper below (McCulloch, 2015) and I'm mentioning it here in the hope of stimulating much needed debate.

References

McCulloch, M.E., 2013. Inertia from an asymmetric Casimir effect, EPL, 101, 59001. Link

McCulloch, M.E., 2012. Testing quantised inertia on galactic scales. Astrophys. and Space Sci., 342, 2, 575- Link

McCulloch, M.E., 2015. Can the emdrive be explained by quantised inertia. Progress in Physics, 11, 1, 78-80. Link