|Ph.D Student||Sainz David|
|Subject||Efficient Information Transfer Leveraging Wireless|
|Department||Department of Computer Science||Supervisor||Professor Roy Friedman|
|Full Thesis text|
Mobile phones and tablets are becoming the main devices to access the Internet, and wireless proximity technology has improved significantly over the last years. This creates the right environment for mobile applications that rely on proximity and close peers to improve their performance. One possible use of this kind of applications applies to Opportunistic Networking, leveraging encounters with other devices to efficiently transfer and store information. This generates the need of understanding the patterns and connectivity characteristics of the wireless connections people usually make over time, as well as possible new patterns of file storage in current mobile phones. There is also a need for new algorithms and architectures to transfer and store data.
We first present an analysis of wireless users activity in the Technion campus Wi-Fi network, where users connect through mobile phones, tablets and laptops, over the course of 4 weeks during the semester. The findings shed light on the behavior of users and offer some indications for opportunistic network protocols. The data is analyzed in terms of session characteristics, connection times and contacts with nearby devices as well as connectivity graph metrics that help understand device connectivity.
Next, we report on the analysis of data from Android mobile phones of 38 users, composed of access traces of the users mobile file systems during 30 days. We shed new light on the file usage patterns and present the data in terms of file size distributions, file sessions, file lifetime, file access activity and read / write access patterns. We characterize different distributions and extract conclusions about usage patterns of Android file systems.
Finally, we use the prior analysis to create an implementation of an opportunistic replication system for mobile phones. It enables backing up files in devices that are encountered on a regular basis. The design of the system includes mechanisms for detecting and identifying regular wireless encounters as well as file accesses. Based on these, the system decides which file should be replicated and to which other devices. Results show that a vast majority of the nodes are able to replicate and retrieve their files within a reasonable delay.
Although the analysis is based on connections and files, much more connectivity information can be inferred with more input data. As mobile phones hold a wide array of sensors, they can report on much more than proximity. It is possible to create a Sensor Network out of those devices and retrieve different kinds of information. Software Defined Networks (SDN) have recently gained attention in the research community as an efficient and flexible network architecture. The SDN concept has opened a new research option for Sensor Networks. Thus, we propose our first approach for an architecture that enables SDN on sensor networks, which makes it reprogrammable and efficient in decoupling between hardware and software.