Sense-Aid: A Mobile Crowdsensing framework Heng Zhang, Saurabh Bagchi, He Wang, Rajesh K. Panta Purdue University & AT&T Labs Take opportunities of: Traffic packet piggybacking Cellular network radio opportunity Orchestration among all devices Save more energy than any other solutions  Encourage more participants  Every body shares, every body benefits. Good afternoon everyone, I am Heng Zhang from Purdue University. My topic today is Mobile crowdsensing. It is a way to get environmental information from mobile devices. It will benefit users in different applications in a cheap and reliable way. For example, we can build a hyperlocal weather map to show fine-grained weather information by crowdsensing barometers on mobile smartphones at different places. High energy efficiency of mobile crowdsensing apps is a key factor to guarantee enough number of participants. By conducting a survey of people from different backgrounds, we found that 2% of the battery energy is the maximum that most of users are willing to pay. However, all existing mobile crowdsensing solutions are above that maximum but we are not. We had a user study where users ran our framework in their daily activity, through evaluation we found our solution is more energy efficient and fair than two baseline soltutions. user run our framework in their daily activity, through evaluation, we find our solution is more energy efficient and fair than two baseline solutions. Add author --> purdue & att labs  

Edge Internet Sensing data Traffic path Core Network 1 1 2 P-GW MME 1 1 S-GW S-GW 2 Sense-Aid Server eNodeB eNodeB 2 Offload Cloud CAS CAS