# Mobile Sensing
movisensXS has the ability to do mobile sensing. Mobile sensing is "Using the sensors of a mobile device (i.e. smartphone or tablet computer) to acquire data from the environment." Modern Smartphones are equipped with the sensors to monitor a diverse range of human activities and commonly encountered contexts.
This is particularly useful for the research field of ambulatory assessment.
Mobile sensing data is store in the Unisens format. Read more about the file format here.
The mobile sensing features are only available on request!
# Features Library version 7365
Log App Usage On request
This logs the used applications.
|Limit to Apps On request||Applications which should be logged only. All other applications will not be written in log files. Please split with ';' if you want to specify more than one.||String|
Log Battery Level On request
This logs the battery level to a unisens log.
Log Nearby Devices On request
This logs nearby devices in an unsisens log file. It will send a bluetooth advertisement through IBeacon standard on supported devices and will protocol all other devices which are broadcasting with a beacon standard.
|Broadcast Signal On request||Enable to broadcast nearby signal.||Boolean||true|
|Scanning interval On request||Determines how long the scan should last. Beacon scanning consumes power, a practical value is 10 seconds.||Integer||10|
|Time Between Scans On request||Interval between scans if a beacon is around in seconds. Beacon scanning consumes power, a practical value is 300 seconds.||Integer||300|
Log Device Running On request
This logs device running to a unisens log once per minute.
Log Location On request
This logs the location in an energy efficient way.
It uses GPS, WLAN and Cell to balance the battery usage with precision.
|Accuracy*||ca. 10 m||ca. 40 m||200 - 3000 m|
|Availability in everyday life||80%||In cities > 70%||> 90%|
|Time until first determination||65s||3s||2s|
|Requirement||Clear view to the sky||WLAN nearby, Internet||Mobile cell tower nearby, Internet|
While tracking there are continuous location updates at a maximum rate of every 5s and every 20m. The tracking transitions to a stationary state when the participant remains within 100m of a central position or no location update occurs for 120s. Then more tracking will be done until the participant leaves the 100m radius.
*Location accuracy is the estimated horizontal accuracy of this location, radial, in meters. We define horizontal accuracy as the radius of 68% confidence. In other words, if you draw a circle centered at this location's latitude and longitude, and with a radius equal to the accuracy, then there is a 68% probability that the true location is inside the circle. Furthermore problems can occur with the location tracking using WLAN if the router is moving. This can happen with WLAN in trains for example.
Log Music Listening On request
This logs the currently listened music metadata and the current playback state.
- Tested and working: Google Android Player, Amazon, Spotify
- Not tested but should work: Google Android player, HTC Music, Apollo, Miui, Real, Sonyericsson, Rdio, Samsung Music Player, PowerAmp, Last.fm, Rhapsody, PlayerPro Music Player, Rocket Player, doubleTwist Music Player, Pandora, Winamp, 8tracks playlist radio, jetAudio HD Music Player, Spotify, Soundcloud
- Tested and not working: Apple Music
Log Notifications On request
This Logs all incoming notifications in an eventbased manner.
|Filter On request||Only react to the given packages (Separated with ';')||String|
Log Phone Call On request
This logs the phone activity in an anonymized way (Hashing of phone numbers).
By hashing the phone number, first an individual code will be generated. This code together with the phone number will be processed using the SHA1 hash algorithm. Because of that the hash number can not be assigned to the phone number later on. Furthermore it is not possible to compare the hashs of different participants. How a phone call can be displayed in the results is shown in this example:
Log Physical Activity On request
This logs the physical activity of the user (IN_VEHICLE: 0, ON_BICYCLE: 1, ON_FOOT: 2, STILL: 3, UNKNOWN: 4, TILTING: 5). This implementation is based on Google Play Services which detection algorithm could change even during a running study.
Activity confidence is a value from 0 to 100 indicating the likelihood that the user is performing this activity. The larger the value, the more consistent the data used to perform the classification is with the detected activity. This value will be <= 100. It means that larger values indicate that it's likely that the detected activity is correct, while a value of <= 50 indicates that there may be another activity that is just as or more likely. Multiple activities may have high confidence values. For example, the ON_FOOT may have a confidence of 100 while the RUNNING activity may have a confidence of 1. The sum of the confidences of all detected activities for a classification does not have to be <= 100 since some activities are not mutually exclusive (for example, you can be walking while in a bus) and some activities are hierarchical (ON_FOOT is a generalization of WALKING and RUNNING).
Beginning with Android 5, activities may be received less frequently than minutely if the device is in power save mode and the screen is off. To conserve battery, activity reporting may stop when the device is 'STILL' for an extended period of time. It will resume once the device moves again. This only happens on devices that support the Sensor.TYPE_SIGNIFICANT_MOTION hardware.
Physical Activity cannot be measured accurately with a Smartphone. Smartphone accelerometers are not very accurate, differ between devices and the wearing position of the Smartphone varies. To accurately measure it please use a dedicated activity sensor like the movisens Move4.
|Log interval On request||Minimal interval to log the physical activity in seconds. Activity may be logged more often, if requested by other apps.||Integer||60|
Log SMS On request
This logs the SMS activity in an anonymized way (Hashing of phone numbers).
Log Steps On request
This logs the steps.
Steps cannot be measured accurately with a Smartphone. This is just an estimation based on 5 seconds of measurement of the Smartphone accelerometer in a minute. Smartphone accelerometers are not very accurate, differ between devices and the wearing position of the Smartphone varies. To accurately measure it please use a dedicated activity sensor like the movisens Move4.
Log Traffic On request
This action logs the traffic usage.
The network traffic is measured at the network layer, so they include both TCP and UDP usage. Values increase monotonically since device boot. The bytes received (RX) and transmitted (TX) are stored in the following categories:
- Total: Number of bytes transmitted since device boot.
- App: Number of bytes transmitted by the movisensXS app since device boot.
- Mobile: Number of bytes transmitted across mobile networks since device boot.
Log Display On/Off On request
This condition is true if the display is on.