Marco Maier

Vegas -- A Secure and Privacy-Preserving Peer-to-Peer Online Social Network

Although Social Network Service (SNS) providers like Facebook and Google attempt to mitigate security and privacy-related concerns of their users, abuses and misuses of personal data still make the headlines. As centralized storage of personal data is a decisive factor for unintended information disclosure, several architectures for decentralized Online Social Networks (OSNs) have been proposed. System designs range from solutions based on a decentralized client server architecture like Diaspora to P2P systems like PeerSoN. Despite all efforts to accomplish strong decentralization, most proposals cannot achieve sufficient informational self-determination, i.e., users do not have full control over storage and dissemination of their personal data and published content. In this paper we follow a contrary approach and present Vegas, a secure and privacy-preserving P2P OSN which restricts the possibility to browse the social graph to the ego network. We show how Vegas achieves a maximum degree of security and privacy through encryption and decentralization. We present our mobile Vegas prototype and its context-dependent communication channel decision model. Finally we show how Vegas can be extended to support services like social-search and directory services in a secure and privacy-preserving way.

Visual positioning systems — An extension to MoVIPS

Due to the increasing popularity of location-based services, the need for reliable and cost-effective indoor positioning methods is rising. As an alternative to radio-based localization methods, in 2011, we introduced MoVIPS (Mobile Visual Indoor Positioning System), which is based on the idea to extract visual feature points from a query image and compare them to those of previously collected geo-referenced images. The general feasibility of positioning by SURF points on a conventional smartphone was already shown in our previous work. However, the system still faced several shortcomings concerning real-world usage such as request times being too high and distance estimation being unreliable because of the employed estimation method not being rotation invariant. In this paper, three extensions are presented that improve the practical applicability of MoVIPS. To speed up request times, both a dead reckoning approach (based on step counting using the accelerometer) and an orientation estimation (based on the smartphones compass) are introduced to filter relevant images from the database and thus to reduce the amount of images to compare the query image to. Furthermore, the vectors of the SURF points are quantized. For this purpose, clusters are calculated from all SURF points from the database. As a result, each image can be represented by a histogram of cluster frequencies, which can be compared with each other a lot more efficiently. The third extension is an improvement of the distance estimation method, which uses the matched feature points of an image to perform a perspective transformation and to determine the actual position with the aid of the transformation matrix.

Fine-Grained Activity Recognition of Pedestrians Travelling by Subway

With the now widespread usage of increasingly powerful smartphones, pro-active, context-aware, and thereby unobstrusive applications have become possible. A user’s current activity is a primary piece of contextual information, and especially in urban areas, a user’s current mode of transport is an important part of her activity. A lot of research has been conducted on automatically recognizing different means of transport, but up to know, no attempt has been made to perform a fine-grained classification of different activities related to travelling by local public transport.

Visual odometry using motion vectors from visual feature points

In recent years, location-based services and indoor positioning systems gained increasing importance for both, research and industry. Visual localization systems have the advantage of not being dependent on dedicated infrastructure and thus are especially interesting for navigation within buildings. While there are already approaches of using pre-recorded databases of reference images to obtain an absolute position for a given query image, suitable means to estimate the relative movement of pedestrians from an ego perspective video are still missing. This paper presents a novel visual odometry system for pedestrians. The user carries a mobile device while walking - the camera aims into the direction of walking. Using only the video stream as input, the system generates a two-dimensional trajectory, which describes the path traveled by the user. Both, the users current heading as well as the walking direction are estimated based on the movement of visual feature points in successive video frames. In order to assess the accuracy of the system, it is evaluated in three different scenarios (indoors in an university building, in an urban area and in a city park). Not relying on reference points (for instance provided by a database, which references visual feature points with geo-data), the error accumulates with distance traveled. After a walked distance of 100 meters, the average error lies between 4.6 and 13.9 meters (depending on the scenario). Consequently, the system is a promising approach for visual odometry, which can be used in conjunction with existing absolute visual positioning systems or as a core part of a future SLAM (simultaneous localization and mapping) system.

An Analysis of Query Forwarding Strategies for Secure and Privacy-Preserving Social Networks

Decentralized Online Social Networks (OSNs) attempt to improve user privacy and security. One example is Vegas, a Peer-to-Peer (P2P) OSN which attempts to bring its users back into complete control of the data they share. Due to its decentralized characteristics, P2P OSNs cannot support social search functions in the same way users of centralized OSNs like Facebook are familiar with. Well-known and efficient P2P search algorithms cannot always be applied as knowledge about the structure of the social graph can be very limited. In this paper, we present an in-depth analysis of forwarding strategies to enable social search for secure and privacy preserving P2P OSNs. We compare well-known metrics from the field of unstructured P2P networks with metrics from the area of social network analysis and evaluate their applicability for P2P OSNs like Vegas. We simulate all metrics on four distinct datasets which were generated artificially from the ER- and the BA-model and from crawling data of Lastfm and Flickr. Our evaluation shows that prioritization based on knowledge from the ego network often yields the best results.

Preventing Restricted Space Inference in Online Route Planning Services

Online route planning services compute routes from any given location to a desired destination address. Unlike offline implementations, they do so in a traffic-aware fashion by taking into consideration up-to-date map data and real-time traffic information. In return, users have to provide precise location information about a route’s endpoints to a not necessarily trusted service provider. As suchlike leakage of personal information threatens a user’s privacy and anonymity, this paper presents PrOSPR, a comprehensive approach for using current online route planning services in a privacy-preserving way, and introduces the concept of k-immune route requests to avert inference attacks based on restricted space information. Using a map-based approach for creating cloaked regions for the start and destination addresses, our solution queries the online service for routes between subsets of points from these regions. This, however, might result in the returned path deviating from the optimal route. By means of empirical evaluation on a real road network, we demonstrate the feasibility of our approach regarding quality of service and communication overhead.