Anyone in this thread familiar with Electronic Warfare? Specifically, the EW arms race between a guided missile's "eyes", and decoy signals created by things that don't want to get hit by said missiles? (Or, better yet, between large-scale MIRV missiles, that don't want their warheads to get hit by interceptor systems, and the sensors on those interceptor counter-missiles. In that case, you'd call these techniques
https://en.wikipedia.org/wiki/Penetration_aid s.)
Consider that we now have reasonably-cheap, high-quality cameras and feature-detection ASICs that we can plop into missiles. Oh, and wide-band software-defined radios, too, to cheaply range-find our targets all over the spectrum. (Remember, SDR radios are expensive for consumers, but that's an artificial supply-limit, precisely because of their military applications.) Consider what sort of decoys would be required to trick all that fancy sensor tech. Now consider what such decoy emissions would look like to human eyes, or any other sensor. If you can trick a Sufficiently-Advanced Missile, you can trick anything.
(For fun, also consider the modern menagerie of cheaply-available parts that you could shove into such a decoy: a laser rangefinder, a short-throw projector, an optical stabilization array, some negative-index metamaterial "scales" that can be flipped on and off—like those glittery pillows!—some more SDRs, a MIMO signal-shaping antenna array. Oh, and an ultrasonic humidifier, because emissions won't diffract off of thin, non-turbulent air.)
Presume you have a bunch of these mesh-networked together (or all slaved to a stealthed aircraft so far above that it's out of range of your instrumentation unless you know exactly where to look), and that they're built with hard real-time links to one-another such that they can execute perfectly-timed hand-offs of the emissions-signature they're working together to project.
In short, picture a bunch of projectors you can't find, working to create a moving multi-spectral hologram in the sky. Picture one of the projector-drones gliding into place "underneath" the projection, just in time to give off a sonar return—and then wandering away again.
None of this is impossible. None of this is even particularly challenging if you spend $100MM/yr on it. And this is just what someone who doesn't know about any of the classified stuff can imagine.