| Cartel: a distributed mobile sensor computing system (2006) | |||||||||||||||||
Abstract | |||||||||||||||||
| This paper describes CarTel, a mobile sensor computing system designed to collect, process, deliver, and visualize data from sensors located on mobile units such as automobiles. A CarTel node is a mobile embedded computer coupled to a set of sensors. Each node gathers and processes sensor readings locally before delivering them to a central portal, where the data is stored in a database for further analysis and visualization. In the automotive context, a variety of on-board and external sensors (GPS, cameras, wireless network monitors, etc.) on cars collect data as users drive around. CarTel’s design meets three sets of challenges: (1) providing a simple programming interface, (2) handling intermittent and variable network connectivity, and (3) handling data heterogeneity (large amounts of data from a variety of sensors). CarTel nodes rely primarily on opportunistic Wi-Fi and Bluetooth connectivity— to the Internet, or to “data mules ” such as other CarTel nodes, mobile phone flash memories, or USB keys—to communicate with the portal. CarTel applications run on the portal, using a delaytolerant continuous query processor, ICEDB, to specify how the mobile nodes should summarize, sub-sample, filter, and dynamically prioritize data. The portal and the mobile nodes use CarTel’s delay-tolerant network stack, CafNet, to communicate with each other. CarTel has been deployed on six cars, running on a small scale in Boston and Seattle for over a year. Three case studies—traffic delay monitoring, metropolitan Wi-Fi network monitoring, and automotive diagnostics—demonstrate the usefulness of the system. | |||||||||||||||||
Publication details | |||||||||||||||||
| |||||||||||||||||