Dynamic experiments localize a mobile node moving within the Zone. Zone Operators running dynamic experiments to gather localization data that can be used to improve FOAM is one of the core objectives of the Trust Zone Program. As opposed to static experiments, which localize static Zone Anchors and can be automated, dynamic experiments require Zone Operators to actively run them. This guide details the steps required for Zone Operators to prepare for and run dynamic experiments in their Zones.
Dynamic experiments allow coverage maps to be created of a Zone, showing which areas in the Zone are conducive to producing accurate and consistent localizations. Once enough data is collected from experiments, Foamspace will analyze metrics like signal-to-noise (SNR) ratio and signal transmission quality to better understand the causal factors driving localization performance. Examples of causal factors could include obstructions and interference sources in the Zone. This and other info can then be used to inform refinements to the FOAM technical stack and Zone setup procedures and guidelines for future releases of the network. Running experiments can also reveal issues with a Zone’s hardware (e.g. LTE modem not working), software, or other configurations, and may shed light on the optimal amount of time to run experiments for, saving operators time and money on data plans in the future. Overall, there are many different things can be learned from beta-testing Zones in dynamic experiments.
Keep in mind that you are running scientific experiments, so you need to think and act like a scientist! đź” ⚗️ 🧪 Apply the scientific method by stabilizing all variables - such as the time of day experiments are run, the path, speed, duration of experiments, obstructions in the zone and the hardware used - as much as possible across all experiments! If the Foamspace team wants you to adjust a variable and experiment to see the results, they will tell you to do so. Definitely do not change any hardware (e.g. antennas) without prior approval from the Foamspace team.
âś…Â Each path should be a continuous straight, curved, or circular line.
🚫 Path should not have overlapping or “criss-crossing”. Minimize switch-backs. If doing an out-and-back route, only run the experiment for one of the two directions (i.e. either the “out” or “back”) Not following these directions will make the results difficult to analyze.
đź§Â Consider logical routes normally taken through the area like roads, sidewalks, walking-paths or game-trails.
🗺️ Each individual path will not cover the entire Zone, but it is ideal if each path is unique and contributes to covering as much of the Zone as possible when combined with the other paths (i.e. to minimize having un-tested areas in the Zone).
[ ] It is best to plan a routine and block off a window of time when you can run dynamic experiments each week (e.g. every Friday afternoon, re: keeping variables constant! 🧑‍🔬)
[ ] Check for severe weather in the forecast, and reschedule the testing for another time that week if needed.