Validating the deployment
Important: When deploying positioning to a venue, validate the deployment before leaving the site. 1. Fingerprinting/Map Analytics - Beacon environment analytics has good quality on all areas (+link to definition / guide) - Mapping quality analytics shows good quality on all validated areas (+link to definition / guide) 2. Ground Truth (GT) paths from all major areas show acceptable positioning performance - Major corridors - Major floor transition areas - Entrance areas: indoor/outdoor in both directions (if indoor-outdoor transitions are relevant for your app's use cases) 3. App Testing - Test with the app that integrates IndoorAtlas SDK, and make sure you can export IndoorAtlas TraceIds from that app or identify the TraceIDs from IndoorAtlas Heatmap and report to IndoorAtlas support when requested. |
Such paths are an essential check point for you afterwards to go back and review the performance in the venue and if you need advice from IndoorAtlas support team, the Ground Truth paths are by far the best way to communicate the performance you currently have in the venue.
Background and Detailed Instructions
After the whole venue has been fingerprinted and identified deployment issues have been initially fixed (e.g. low fingerprinting quality shown by mapping quality analytics, or missing beacons in some area), you are ready to test the positioning performance throughout the whole venue.
You should test the positioning performance on each floor and check also if all floor transitions work as expected. To get more quantitative data about the positioning performance, you can simply use the MapCreator app's positioning mode to record ground truth paths and replay the trace ids.
Optionally, you can use with Session Viewer to analyze the positioning performance using the stored trace ids. This tool lets you see if all the installed beacons have been detected in your test walk, for example.
Tip: if you've subscribed to a commercial plan, IndoorAtlas support can help you plan the Ground Truth test paths. |
Planning Ground Truth Test Paths
Below are some guidelines for planning the test routes for good coverage:
- Focus on the most probable routes the users might take when they use your indoor positioning-enabled application.
- In single-floor venues, walk through the main areas. Start tests from different initial locations to see how fast and accurate the first positioning fix is: walk for at least 30 seconds to see if the positioning tracking gets accurate.
- In multi-floor venues, plan a test route that goes through all floors. On each floor, walk until the positioning tracking gets accurate – this takes usually 10 to 30 seconds. After getting accurate tracking, take the elevator or stairs to the next floor. Test the route several times to verify that the floor transitions work well.
- Testing indoor-outdoor transitions; plan a test route from outdoors to indoors. Start the testing well outside, at least 20-50 meters away from the entrance. Walk indoors (to a fingerprinted area) and walk at least 50 meters to see if the positioning indoors gets accurate and tracks well.
- Testing multiple devices; use different Android models and iOS devices with your test routes. Use exactly the same test route for each device. Focus on the most common device models you expect your users to use. Repeat the test several times for each device to validate the positioning performance.
Remember: always export or send by email the TraceId of each Ground Truth session for your own records!