In this project, a group of mobile devices are enabled to perform a common task (involving acquiring content from each other and/or from a remote server) to enjoy higher bandwidth and lower delay than each device would be able to achieve by itself. This is achieved by building a cooperative networking framework that enables several devices to use all their resources together. Each device can use simultaneously two network interfaces: one (cellular or WiFi) to connect to a remote server and another (WiFi or Bluetooth) to connect to the rest of the group. This framework can: (i) provide a connectivity bundle that multiplies the bandwidth at each device by a factor on the order of the number of cooperating phones, (ii) significantly improve the download rate for applications that need to acquire remote content, such as video streaming, and (iii) support highly interactive applications, such as multiplayer games. To demonstrate these capabilities, two prototype applications have been developed; one that supports HTTP video streaming to a group of users, and another which supports local multiplayer games. Key components of the system include the following: (i) a scheduling algorithm for cooperative downloading over all available downlink connections; (ii) an all-to-all local dissemination scheme for sharing content among group members; (iii) a way to efficiently exploit overhearing and network coding on Android phones. The architecture is modular and can be easily optimized for bandwidth or latency, depending on the application.
This project will advance cooperation among mobile devices within proximity of each other. It can not only significantly improve the user experience when using the Internet and the performance of existing applications on mobile devices, but can also enable a new class of collaborative applications. With the proliferation of smartphones and the unprecedented growth of the mobile service segment, this framework has the potential to have a significant social and economic impact. Potential commercialization of these technologies include: standalone apps, providing a development framework and libraries to software developers, and partnerships with telecom providers and device manufacturers.
This ICORPS project explored the commercialization potential of "Microcast" - a research project, at UC Irvine, on video delivery to a group of mobile devices. The intellectual merit of the project lies in the development of a mobile app prototype that can boost the bandwidth available to mobile devices for streaming video faster. In addition to core technology, we collected and processed analytics about performance and user behavior. Furthermore, through the customer discovery process, taught and enabled by ICORPS, we explored several customer segments. In particular, we interviewed potential busineess customers and also we learnt from the use of our prototype app in the wild. Based on this learning, we pivoted our value proposition from (i) delivery of the same video to a group of proximal devices to (ii) boosting the bandwidth for video streaming on a single device, by using multiple interfaces on a single phone and/or multiple proximal devices. At the end of the ICORPS project, we realized that our boosting technology can be deployed in a number of different ways, including: a use-facing app, a software-development-kit incorporated in existing apps, and a service on the mobile device. The broader impact of the project includes the training of the entrepreneurial lead (EL) in both the technical and entrepreneurial aspects of the projec and technology transfer. For example, following the ICORPS training, a company was founded and an NSF SBIR Phase I was awarded.
