You cannot make a beam of light that never spreads out, I'm afraid. Lasers do pretty well but even laser beams have a width which increases by a predictable amount once they've gone far enough. Moreover, a laser only works because there are many photons - to define a direction of emission. Just one photon has equal probability of 'going in any direction' from an atom (Being detected anywhere around the atom, is a more correct way of saying it because, until it's detected, it could be anywhere).How about little bullets that wave a little bit, snake-like, and can split in two and interfere with itself? Kind of like some of these electrons here:
As far as I know, we can make very narrow beams of light, and it appears the thickness does not variate, so they must have some defined 'cross section' radius, or width and height, which is defined by the peaks of photon amplitude, right? And they also have defined some length since we can emit individual photons with a gap between them, right? So something that has certain cross section radius kind of does look like a bullet, or an arrow, depending on how long they are. Do you know how long photons are?
1. How much did you want them to wave - anything related to the wavelength, perhaps? (careful with your reply to this as it would be difficult to generate a beam of 1MHz Radio waves, only 1mm width).
2. What do you mean by "peak of photon amplitude"? All photons of one frequency have the same energy. If they split, then one photon would have to be sub divided - not just into two (for two slits) but into thousands for a diffraction grating.
3. Do your answers to 1 and 2 extend to photons of LF radio signals with a wavelength of over a km and individual energies which are 10^-10 of the energy of a photon of light? If they don't then you have to think again.
4. Would the (split) bullets also be travelling (and wiggling) through the amplifiers and feeders, too, in a multi-element transmitting antenna?
I realise that, from the perspective of experiments with light, 'little bullets' fit , emotionally and comfortingly, with what we see - or think we see. But the same model absolutely has to fit all cases of EM waves if it can be considered as a candidate for 'the truth'. (And photons are very different from electrons, in many ways)