Take a GPS tracker outdoors and it will give you latitude and longitude within tens of seconds; but the moment you walk into a warehouse, a hospital, an underground parking garage, or a steel-frame factory building, the GPS signal immediately fades to nearly unusable. Yet it is precisely these indoor spaces that most need to know the location of forklifts, high-value equipment, medical staff, and visitors. LoRaWAN indoor positioning exists to fill this gap: it does not require every device to carry a high-precision module, but instead uses existing gateways to do zone-level positioning, answering "which zone is the thing in, is the person inside the safe area" at a controllable cost.

1. Why indoor positioning is harder than outdoor

The logic of outdoor positioning is simple: the device receives signals from multiple satellites and calculates its own position. Indoors it is a completely different story:

  • Satellite signals cannot penetrate roofs and walls: reinforced concrete, metal shelving, and floor structures all block GPS signals from outside.
  • Multipath reflection is severe: signals bounce back and forth between walls and equipment, and the signal from the same satellite arrives via different paths, producing huge direct positioning errors.
  • The space is enclosed and dense: a single building may contain hundreds or thousands of people and objects to be located, and it is impossible to deploy an expensive high-precision module on every one of them.

So indoor positioning is usually not "the device calculates its own coordinates," but "the network side infers the device's position from signal characteristics."

2. Several approaches to LoRaWAN indoor positioning

LoRaWAN itself is a communication protocol, not a positioning protocol, but it can leverage gateway signals for positioning. There are three common approaches:

  • RSSI proximity method (fingerprinting): the device reports which gateways heard it and at what signal strength, and the system matches this against a pre-collected signal fingerprint database to determine the location zone. Low cost, easy to deploy, with accuracy typically at room level to corridor level.
  • TDOA (Time Difference of Arrival): the device sends an uplink packet, multiple gateways at known positions record the time of reception, and the position is inferred from the time differences. More accurate than RSSI, reaching several meters, but it requires gateway time synchronization, and both deployment and algorithm costs are higher.
  • AOA (Angle of Arrival): a directional antenna array determines which direction the signal comes from, usually as a supplement.

For most indoor asset management and personnel safety scenarios, the RSSI proximity method is already sufficient; only in high-demand scenarios such as warehouse sorting and hospital operating rooms is it necessary to add Bluetooth beacons or UWB for high-precision coverage filling.

3. Accuracy needs to be sufficient, not as high as possible

This is where indoor positioning most easily goes astray:

  • Zone level (meters to tens of meters): knowing "the device is in Warehouse Zone 3, 2nd floor, Building A" already solves 80% of the find-the-goods and inventory-counting problems.
  • Room level: knowing which room it is in, suitable for personnel safety and visitor management.
  • Sub-meter level: only UWB, vision, and similar technologies can achieve this, and both cost and deployment density are high; the cost-effectiveness may not be appropriate.

The sweet spot of LoRaWAN indoor positioning is precisely to make "zone level" solid using low power and wide coverage, rather than blindly pursuing sub-meter accuracy.

4. Typical application scenarios

  • Large warehouses and logistics centers: zone-level positioning of forklifts, turnover boxes, and high-value goods, shortening the time to find goods.
  • Hospitals: nearby dispatch of mobile medical equipment, infusion pumps, and monitors, reducing the situation where equipment is borrowed and then cannot be found.
  • Factories and underground utility tunnels: personnel badges combined with zone-level positioning, for boundary-crossing alarms and emergency roll calls, in the same vein as the personnel safety solutions mentioned earlier.
  • Underground garages and parking lots: zone-level vehicle guidance, forming a "vacancy detection + reverse car-finding" combination with parking sensors.

5. What to pay attention to during deployment

  • Gateway density determines accuracy: indoor wall obstruction is significant, so gateways cannot simply copy the sparse outdoor deployment; points must be arranged reasonably across floors and corridor corners.
  • Building a fingerprint database has a cost: the RSSI approach requires collecting signal fingerprints on site first, and after shelving is rearranged or equipment changes, it must be updated regularly, otherwise accuracy will degrade.
  • Smooth switching with outdoor positioning: when people or assets go outside, GPS takes over; when they come inside, LoRaWAN indoor positioning takes over, and the platform must handle the seamless handover.
  • Don't confuse communication with positioning: being heard by a gateway ≠ being accurately positioned; positioning performance must be measured on site, not judged solely by the nominal distance.
  • Solve "whether it works" before pursuing "how accurate it is": first build zone alarms and boundary-crossing alerts, then gradually densify gateways to improve accuracy.

Conclusion

The value of LoRaWAN indoor positioning is not in competing with UWB for sub-meter accuracy, but in using an already existing low-power network to answer, at relatively low cost, "roughly where are the people and things indoors, and has anyone crossed a boundary." For manufacturers doing IoT device OEM/ODM, packaging gateway positioning capability, tags, and positioning algorithms together and integrating them with the actual management workflows of warehouse, hospital, and factory customers can find clear deployment space in indoor scenarios where GPS fails.