Determining the floor where the user is of great importance in any positioning system. Floor detection is a must in all the wayfinding and tracking use cases. Situm is able to perform this task very accuratelly.
Performance in different environments
Situm yields different performance depending on the morphology of the building.
| Type | Description | Performance |
|---|---|---|
| Areas with delimitation between floors | These are areas where there is a total separation in ceilings and walls between floors, such as between two corridors on different floors. This is the simpler case since WiFi/BLE differ a great deal among floor (due to the attenuation of signals in-between floors). | Accuracy is usually >99%, requiring aproximatelly 5-10 seconds to detect a floor change. |
| Areas without delimitation between floors | These are open areas in which the separation between floors is not clearly delimited, as is often the case in many shopping centers. These are the most complicated areas, since as there is no clear physical delimitation between floors, the values of the signals received (WiFi, BLE, etc.) are very similar between floors. | Accuracy is usually >95%, requiring approximately 10-20 seconds to detect a floor change. |
The following Figure contains an example of both types of areas.
Area without delimitation Area with delimitation between floors
between floors
Floor detection modes & floor change accelerators
Situm provides 2 different ways of detecting the floor where the user is.
| Mode | Description | Recommended... |
|---|---|---|
| WiFi/BLE AI detector | An Artificial Intelligence classifier trained with the WiFi/BLE calibration data determines the floor where the user is based on the WiFi/BLE signals received. | This is the default mode and is recommended in most cases. |
| WiFi/BLE basic detector | Determines the floor where the user is by comparing the WiFi/BLE signals received with the WiFi/BLE signals gathered from the calibrations. | This mode is only recommended if you find out that the WiFi/BLE AI detector is not working properly. |
Additionally, Situm provides 2 floor change accelerators that will help you to speed up floor changes when the user goes from one floor to another. They can be used individually or combined to achieve the best of both.
| Accelerators | Description | Recommended... |
|---|---|---|
| Path based (How to apply this mode) | This mode requires that you configure the wayfinding paths in your building indicating the floor change transitions, set up the altitude of each floor and re-train your building. When the user approaches a floor transition and is closer than 20 meters away, the accelerator detects it and helps Situm algorithm to change faster to any of the plausible floors. For instance, when you approach the escalators, Situm will consider that you may go upstairs/downstairs as 2 very plausible options, therefore accelerating the transition detection. | If you want to speed up floor changes, specially in wayfinding apps. |
| Barometer based (How to apply this mode) | This mode requires that you activate the Barometer in Situm Mapping Tool app on the Advanced Settings or at the app SDK level. When the user changes the floor, Situm processes the barometer information in order to determine if the user is going up or down to the respective next floor. This result is obtained from the pressure measures detected by the mobile device through the barometer sensor that provides pressure data at each moment. From these barometer's results the altimeter module calculates the height of the user as well as the direction (Up/Down) of the movement, besides the probability of the action be performed. | If you want to speed up floor changes and at least some of your users have smartphones with a barometer sensor. |