Ulla,
you can of course 'wiki' Near Field and Far Field as an introduction which is very abbreviated, it is a complex area of investigation but generally technological rather than theoretical.
Briefly, for all that is stated as 'knowns' in the interminable debate of whether EMR is a wave or a particle, the ONLY direct observation of electro-magnetic radiation is at the receiving end of a signal. That is done by measuring the change of intensity of the electric field and magnetic field in the first two wavelengths of any frequency in the immediate proximity of the receiving antennae. That is the Transition Zone in which the electric and magnetic fields settle down to a more or less homogeneously uniform configuration, called the Far Field, wherein the intensity of both fall off at the rate of inverse square of distance which is assumed to carry through the entire distance of transmission. And theoretically, that is assumed to extend the signal to infinity. Here again, that is hypothetical because we only actually observe the changing intensities in near proximity to some sort of antennae. So you see the problem that exists theoretically due to the physical limits of observing systems.
Where it gets interesting is in the distance from the antennae that is approximately equal to one wavelength, and termed the Near Field, which physically displays two distinct regions, one called the 'radiative' and extends across most of the wavelength, and the 'reactive' which is the nearest to the antennae and is roughly 1/2pi lambda; or about 0.159 the distance of the wavelength. More about that region in a moment.
Here is what is intriguing about the Near Field. While the electric field strength falls off at an inverse square rate as would be expected, the magnetic field strength falls of at an inverse CUBE rate! And in the reactive region there is a confusion of fields where the intensity falls off at an inverse exponential rate!
Now... when you consider it from the perspective of Maxwell's determination that the associated electric and magnetic field strengths in a point charge (such as a stationary needle point that is electrified) have intensities that differ by a light velocity proportion, it should immediately become obvious that if an EM signal was stopped instantly by the antennae, that proportional difference would vaporize the antennae! So the Near Field gives us clues to what happens. And that must be that the signal slows exponentially from light velocity when encountering the electromagnetic field of the atomic matter comprising the antennae, and as the energy of (for want of a more definitive word) photon 'stacks up' on its self as the leading edge slows, the magnetic field intensifies by virtue of increasing density to become undifferentiated from the electric field in both intensity (density) and differentiation of characteristic effect. This also suggests that energy density varies in direct inverse proportion to velocity, and may provide a theoretical means of mathematically obtaining an absolute velocity without violating the neutral centrality of SR.
What applies to the modeling of exchange particles, is that given the energies that would need be transferred at inner atomic distances, the wavelengths would have to be at the gamma end of the frequency spectrum, because if 'photons' are to be the exchange particles, their Near Field of that transfer exchange would be equivalent to the distance of ~one wavelength for that frequency.
Sorry to be so long in the tooth with this but it does get rather complicated. jrc