FenceIt Location Alarms

FenceIt allows you to define location-based alarms and custom triggered actions.

FenceIt is an Android application that allows users to define location-based alarms and custom actions that are executed when the alarms are triggered. The application uses alternative methods to GPS positioning and different data processing algorithms in order to preserve both the battery level and the data connection, while being easy to use and attractively designed.

The idea behind FenceIt is really simple:

1. Define some locations by gathering the required data

2. Define an event that triggers an alarm (e.g. when leaving from or arriving at a location)

3. Define what happens when the alarm is triggered (an action)

FenceIt integrates 4 types of methods to figure out when you are in a specific place:

* the Geographical Coordinates of the particular place (Geographical Coordinates Location)

* the Wi-Fi Network the device is Connected to (Connected WiFi Location)

* the Wi-Fi Networks detectable in range (WiFis Detected Location)

* the Cellular Network Towers the device is Connected to (Cell Network Location)

The application also allows you to define what happens when an alarm is triggered. In the current version, the following options are available:

* set the mobile device to start ringing and to display a message

* change the volume mode of the device (Normal, Vibrate, Silent)

** Advanced Optimizations **

FenceIt has been carefully optimized to make sure it uses as less resources as possible.

Usually it is useful to just determine if a device is or is not in a particular area and not exactly where that area is. For example, for the purpose of a location based alarm application, it would be enough to conclude that the device has reached “Home”, not being necessary to know exactly what address “Home” has or what are its geographical coordinates.

While most of the existing applications in this field use a straightforward approach regarding when to check for updates, acquiring information from the environment at regular time intervals and checking if any conditions have been met, this is not the optimal behavior. In some cases, more information from the context can be used to make the querying algorithm more efficient. An important type of optimization is only applicable to the first type of alarm (using the geographic coordinates) and considers the distance of the target location to the current position. For instance, if there is only one alarm of this particular active in the application and its target location is far from

the current position of the device (for example a few kilometers), querying for GPS location every second is inefficient, as it is not practically possible to arrive at that location in that short amount of time. In such a case, an approximation algorithm that takes into consideration the current speed, direction and distance to target and estimates the arrival moment at the location can be used to dynamically change the frequency of data acquisition from the GPS hardware.

Another optimization considers the inactivity of the device or the lack of movement. If it hasn’t

moved for a considerable amount of time, the frequency at which any type of data acquisition should

be done can decrease, improving the battery consumption. This can prove highly efficient, as it

considers cases when, for instance, the user is at work or at home and the mobile device is stationary

on a table. If this happens for a considerable amount of time, it’s inefficient to acquire location

information every second and an algorithm can be used to compute a dynamic waiting time between

data gatherings.

The third possible type of optimization includes pre-defined inactivity periods, when the user knows

that he/she will most likely not move very often, if at all. As an example, we can consider the case of

time spent at home, during the night, when the user is usually stationary and the frequency of

alarms’ analysis should not be as high as during the day, when the probability of movement is higher.