Latest EECS News
Aleksandar Kuzmanovic, Associate Professor gave an invited talk at Dartmouth College on Wednesday, October 30, 2013. He discussed privacy on the Internet, a topic that raised significant attention recently due to alleged large-scale government surveillance.
He presented the Mosaic project (ACM Sigcomm’13) which aims to quantify privacy leakage in mobile networks. The project demonstrated that the popularity of online social networks and smart phones has dramatically increased the amount of identifiable digital footprints in network traffic. These enable an adversary to attribute significant portions of traffic including the ones with no identity leaks to network users' true identities and reveal personal information such as political views, browsing habits, and interests.
He also presented Synthoid, an National Science Foundation (NSF) project that aims to counter such large-scale surveillance attempts. Synthoid enables users to explicitly define and implicitly control their online profiles at all possible trackers and third parties at once by leaving controlled artificial footprints on the Internet. It is capable of completely altering a user’s real profile including features such as age and gender.
Goce Trajcevski received the Best Short Paper Award in the 16th ACM International Conference on Modeling, Analysis and Simulation of Wireless and Mobile Systems (MSWiM 2013) on November 3-8, 2013 in Barcelona, Spain.
The winning paper, entitled "Voronoi Trees for Hierarchical In-Network Data and Space Abstractions in Wireless Sensor Networks," was also co-written by Mohamed Ali (UIC) and Ashfaq Khokhar (IIT).
The work addresses the problem of efficient in-network processing of spatial queries in Wireless Sensor Networks. Specifically, it considers hybrid overlays settings, where the data values may correspond to different physical phenomena measured by distributed sensors, and may be additionally correlated via spatial constraints. Traditional indexing structures such as KD trees, R-Trees, B-Trees, or variations thereof ? along with various clustering and routing trees, have already been used for organizing in-network data aggregation in WSNs.
The awarded paper showed how using a proper "weighted" distance enables Voronoi-based partitions of the sensing field into convex cells. Subsequently, Voronoi Trees are built in a hierarchical manner, mapping both data and space abstractions, for energy-efficient processing of spatial queries. The evaluations demonstrated significant advantages of the Voronoi Trees based indexing structure in terms of accurate field representation at different levels of the tree hierarchy, with acceptable trade-offs in terms of query processing latency.
Over the years, MSWiM has established itself as a major research venue where numerous key results in the area of performance evaluation of wireless and mobile systems have appeared; we are delighted that this year’s event, maintains the tradition of high quality contributions. covering several specializations within mobile and wireless systems.
The 16th ACM MSWiM 2013 Conference featured Symposia and Workshops, specifically on emerging topics related to wireless networking and mobile computing. The four symposia were: MobiWAC, PE-WASUN, DIVANet and Q2SWinet and the three workshops: WMuNeP, PM2HW2N and HP-MOSys. Over the years, these seven events have become successful and quite competitive.