We developed a crowd sensing application to estimate road conditions (CRATER). CRATER is a smartphone application that opportunistically measures acceleration when it finds itself on the road in order to map and measure the locations of potholes and speedbumps. It does not require input from users, but can report detected potholes and speedbumps to a cloud-hosted application engine, which stores partially processed data received from smartphones of participating users and jointly processes it to obtain a better estimate of road conditions. The information is published in map form on the web. This map allows both citizens and municipal authorities to localize potholes, road segments in need of repair, and imbalances in infrastructure maintenance efforts across cities. Road tests demonstrate that CRATER succeeds in correctly detecting roughly 90% of potholes and 95% of speedbumps, while generating false alarms only about 10% and 5% of the time, respectively.INDEX TERMS Application of wireless sensor networks, smartphone applications, pattern recognition, mobile systems, crowd sensing. I. INTRODUCTION A. BACKGROUND AND MOTIVATION

No abstract

No abstract

No abstract

Edge computing caters to a wide range of use cases from latency sensitive to bandwidth constrained applications. However, the exact specifications of the edge that give the most benefit for each type of application are still unclear. We investigate the concrete conditions when the edge is feasible, i.e., when users observe performance gains from the edge while costs remain low for the providers, for an application that requires both low latency and high bandwidth: video analytics.