How are Fresnel Zones formed? What phenomena of light allow ellipsoid areas to be in phase? I've tried reading articles, but they more or less introduce me to characteristics of light, and then tell me that Fresnel Zones exist. How does the circular wave coming out of one antenna "bend back" to the other one in the shape of an ellipse?
[Physics] Fresnel Zones-How are they Formed
antennasoptics
Related Solutions
The Wikipedia article that Anna mentioned is an excellent description of holography and I'm not going to try and compete with it, but since I'm guessing that you're not a physicist the following might help make things clearer.
When you "see" things, you see them because they enter your eye, get focussed by the lens and hit the retina. In addition you see 3D because the image recorded by your left and right eyes are slightly different, and the brain can reconstruct a 3D image from the differences.
So if you're looking at a mouse (to use the example from the Wikipedia article) it's the light reflected off the mouse that the eye uses to "see" the mouse. Suppose you could come up with some clever trick to remove the mouse but still send the light to your eyes as if it had come from the mouse. Your brain couldn't tell the difference because your eyes are still receiving the same light as when the mouse was there. This is what a hologram does.
A hologram is a pattern of light and dark areas. When you shine a laser onto a hologram the light and dark areas scatter the light by a process called diffraction. The clever bit is that the light is scattered in exactly the same way as if there were a mouse there, so your brain sees light that looks as if it has come from a mouse, so you see a mouse. It appears in 3D because the hologram scatters light differently depending on the angle you're looking at it, so your left and right eye receive differently scattered light just as they would from a real mouse.
You might think it would be tremndously difficult to make a hologram to scatter light in just the right way to make it appear as a mouse, but actually you make a hologram from a real mouse i.e. it's just a type of photograph.
It's hard to make multicoloured holograms because to "see" the hologram you have to shine a laser on it, and lasers are just a single colour. You could use three lasers, e.g. a red, green and blue laser, but annoyingly the hologram scatters different coloured light in different ways and your multicoloured hologram would be very blurred.
I hope this helps - to get any further you'll need to work through the Wikipedia article, and also understand what diffraction is.
In the figure that you titled as "correct way", indeed everything is correct. You say that you have a problem, "these secondary wavelets are all formed at different times". Yes, the wavelets appear at different times, and this is why the reflected wave-front is inclined. Look at the picture below.
By the time the leftmost ray reaches the point $A$, on the wave-front (dark green) the other two rays are far from the reflecting surface, points $B$, respectively $C$.
By the time the point $B$ reaches the surface, point $B_1$, the surface of the wavelet formed around $A$ touches the point $A_1$.
Finally, by the time the point $C$ reaches the surface, the wavelet around $A$ touches the point $A_2$, and the wavelet around $B_1$ touches the point $B_2$. The reflected wave front is colored red.
Best Answer
When you derive a Huygen-Fresnel Propagator (which is how actual wavefronts propagate according to Maxwell's equations) a Fresnel zone is really the difference (in phase) between surfaces of equal phase on the propagating wavefront and a plane slicing or tangent to that surface of equal phase.
These Fresnel zones are defined when propagating a plane wave incident on some circular aperture.
This picture shows the concept, except the Fresnel zones are shown for a finite conjugate (i.e. phase difference between a propagating wave and a point that we want to use the zone plate act as a "lens").
Basically, the Fresnel zone is the phase difference map between something like a spherical wavefront, and a plane tangent to that wave front. That is exactly what it is physically and conceptually. A zone plate (manufactured on a plane) can then be used to change the propagation properties of the light wave.
I say something like a spherical wavefront above because that is a good approximation, the actual Green's function for the propagator is the derivative of a spherical wavelet, which is required to satisfy the boundary conditions properly.
The figure above shows how canceling the additive and subtractive parts of the wavefront can result in "lens type" behavior.