direct is practically useless, because that's point-to-point. lighting up the ionosphere that way seems like the hardest case scenario, requiring a very powerful transmitter, somehow ignored by sensitive high-resolution scanning over a large area of sky. and you'd be disrupted by other occluding objects like water vapor

there is precedent https://en.m.wikipedia.org/wiki/Over-the-horizon_radar but it seems like a limiting factor is suitable frequencies and resolution