Peter,
You wrote: "You DO need some reference frame to define your emission 'point' and measure from it".
No. As do incremental sensors, I only measure the distance between the emitting point at the moment of emission and the point of arrival at the moment of arrival. Emitter and receiver may have frames of reference that are moving relative to each other.
In particular if their distance doesn't change then it and also the time of flight can be measured, in principle.
You wrote: "1st; Who can properly determine 'Time of Flight'? The answer is nobody! Choose an observer and I'll show you his problem."
I am curious. For the sake of simplicity I consider an emitter E and a receiver/observer R on a line without any reference point. Let the distance between E and R increase during the time of flight from d_0 to d_1.
Notice: There is no third point of reference relative to which E and R could have a velocity of relevance.
The notorious problem is thinking in terms of either the Newtonian ballistic picture where the speed of a bullet depends on the speed of the cannon or of the picture how a wave travels in a medium.
My definition makes an uncommon distinction between empty space without any point of reference and the hypothetical medium aether that was ad hoc attributed to a differently understood space, thought to include a center point or a boundary being equivalent to it. CMBR is not qualified as such boundary.
Eckard