@James Rodriguez yeah this is a reasonable fallback strategy, but I'd temper expectations a bit on the RSSI side — at the ranges the MultiNav Pro+ operates at, RSSI-to-distance gets pretty noisy pretty fast. The path loss model works okay in open air with clear LOS, but the moment you're dealing with the environments where GNSS drops (tunnels, indoors, urban clutter), those are exactly the same environments that wreck your RF propagation assumptions, so your RSSI readings are going to have a lot of variance that doesn't map cleanly to distance.
That said, it's not useless — if you're just looking for coarse buckets like "within 100m / 100-500m / 500m+" rather than actual meter-level estimates, you can make it work with some averaging and a decent path loss exponent calibrated to your environment. The key is don't use raw RSSI, do a rolling average over several readings and maybe a Kalman filter pass to smooth the jumps. Also worth calibrating your reference RSSI at 1 meter for your specific hardware and environment rather than trusting a generic value, because the external LNA on the MultiNav Pro+ shifts your baseline compared to stock BLE modules.
The GNSS + RSSI fallback combo you're describing is actually a solid architecture — GNSS for precision when you have sky view, RSSI for "at least tell me roughly where it is" when you don't. Just don't expect RSSI to give you anything better than maybe ±30-50% accuracy at mid-to-long range in degraded environments, and build your application logic around that assumption rather than treating it like a real ranging solution.
