From the theoretical perspective, light is a wave, and wave tend to spread out following inverse square laws. That is if you are using a standard beam, that doesn't consider the phases of the wave. But if you are using a non-diffracting one, like an Airy or Bessel beam, you can get a wave-packet that doesn't spread out in its working regime.
Airy beam only require FFT level to understand : you spatially phase-modulate a beam with a window function, so that the beam doesn't leak out of a spatial region.
Of course when you need to beam out huge amount of power, this phase-modulation must be done in a very efficient way so as to not destroy your emitter. In optic's lab, for low-power airy beams, you can use lenses sort-of-like fresnel ones, but this won't for higher power. So you have instead to have a large spatially distributed array of sources (antenna) which you can control precisely the phase.
Real Airy beam are only an approximation to the theoretical ones (that require an infinitely spatially-large source). But typically the smaller the wavelength, the more lobes you can match between real and theoretical.
But all your nice calculations and perfect orchestration of signal, then encounter the atmosphere that messes everything in an uncontrollable way.
So you check your wave-length charts for wavelengths that can go through the atmosphere, but would still impact the target. You are now talking X-ray lasers, or gamma-ray lasers. More power but harder to control precisely (phase wise) because the wave-lengths are roughly the size of atoms. At this point you are in high-energy physics can just strap your run of the mill particle accelerator and aim it at the target, and select the particles you want it to beam.
While standard electron-gun won't go through a meter of the atmosphere, because it's electrically charged. So you look at the neutral particles. Neutron beam generator (does seem to have bad radioactive properties so boss won't approve) and they are more difficult to accelerate because they are neutral. What about muonium : One part electron, One part anti-muon, electrically charge so you can accelerate them nicely before combining them and then neutral, they are quite light, they self-disappear after a few micro-seconds (of self-time so you just have to get close enough to c if you want them to last longer).
Let's try that : Muonium-aser, 95 balloons to go...