Performance of geographically restricted flooding for emergency alert applications

For various emergency alert applications like a disaster, accident, or hazard alert, geographically restricted flooding can prove to be extremely vital, especially considering the absence of any fixed infrastructure and the alert having spatio-temporal boundaries. Among various geographically restricted flooding schemes, in this work, a scheme called floating content is considered. Floating content uses geographically restricted flooding to make a piece of information available within a certain geographic area. In this work, an Android-based emergency alert application is deployed and tested in two different environments. One environment spans over a small geographic area (an office), while the other environment is a university campus. The application is used to log the mobility patterns of users for both the considered environments, and these mobility patterns are thoroughly analyzed. Then the performance of the considered emergency alert application in terms of key performance metrics is evaluated for both of the considered settings by considering single and multiple seeders (alert generators). Moreover, for the prediction of performance, two different analytical models are also evaluated, and the results of these models are compared with the ones obtained from real-world experiments, showing the performance limits of geographically restricted flooding.