Still, now you've got me curious and I'd be grateful for some proper boffins to elucidate us further: if you have an transmitter emitting single photons in an isotropic antenna pattern, could they still be received by a receiver even though there is no "wave" to speak of? Even if part of the information is going in another direction? There is some proper double slit-type weirdness going on there.
Optically we make photon counters out of cooled avalanche photodiodes or photomultiplier tubes that will repeatedly trip for each incoming photon (at high probability). But they also randomly trip at a rate because of internal thermal noise.
As to emission, yes, you can make (close to) omnidirectional single photon emitters. But the odds of it interacting with your antenna or detector are low.
Also the quantum limits can still be treated entirely classical, no need to think about "quantum weirdness"