Jul 23, 2026
•
1 min read
I tested a "smarter" satellite handover against the classic strongest-signal one, on a live Starlink window. The classic one held up.
Jul 21, 2026
I raced an adaptive refresh schedule against a simple fixed one on a real Starlink pass. The simple one won and that's the useful finding.
Jul 18, 2026
3 min read
Week 6: I built 3GPP's actual answer to the problem Week 5 exposed — and found its real limit.
Jul 12, 2026
2 min read
Week 5: 5G assumes the tower is close and standing still. A satellite is neither.
Jul 9, 2026
Week 4: the prediction works, the obvious way to use it didn't, and the reason why is the point.
Jul 5, 2026
Week 3: I built the smallest honest version of a real satellite hand-off decision.
Jul 2, 2026
Live orbital data, a real pass, and a frequency shift that flips sign mid-flight.
Jun 28, 2026
Build-in-public #1: a link budget, three setups, and whether the signal even arrives.
