Sahyog: A middleware for mobile collaborative applications

The ubiquitous nature of mobile devices offering high computing capabilities has attracted a large number of users. The sensors available on these devices provide valuable data about user contexts, which when combined with sensor data from other mobile devices can reveal interesting environmental contexts, which can then be used for applications like real-time traffic monitoring, noise mapping, crowdmapping, etc. Developing applications for inferring environmental context requires participation from multiple users satisfying certain requirements (e.g., location of users etc.); hence, building such applications requires developers to carefully identify potential collaborators satisfying these conditions. This makes the development of such applications cumbersome and time-consuming. In this work, we present Sahyog1 - a publish-subscribe framework that allows developers to specify these requirements and identify users that can actively participate in such applications. Sahyog helps application developers request data from a large number of users without worrying about reaching out to users, communicating data requests and responses, and storing information. We discuss the design and implementation of our system with the underlying research challenges, and evaluate it using several parameters.