My family lives outside of a tier 2 city border, in what used to be farmland in the 90s.
They have Asianet FTTH at 1Gbps, but most of the video/streaming traffic ends at the CDN hosts in the same city.
That CDN push to the edge is why Hotstar is faster to load there - the latency on seeks isn't going around the planet.
You could go active but then your SFP/SFP+ per port cost eats you up.
For less than 1mil fixed wireless is going to cover 2,800km/sq. You are not going to to get anywhere near that cost trying to do the same thing to 2048(or more) subs in that footprint with fiber. That wouldn't even cover your fiber material cost!
LTE is what somebody would do without much telecom experience and more money than sense.
I've built fiber networks and fixed wireless networks. Almost ended up becoming an LTE network as well. It didn't make any sense in any sort of financial modeling, even with spectrum availability.
LTE helps solve "general connectivity". What it does not do is build scalable, reliable, high speed, economical sensitive broadband infrastructure.
One could see India deploying the same density compatible infrastructure in the usual "leapfrog" model of skipping lesser technology implementations in this space.
Mobile data is cheap, but broadband is much cheaper.