- PDFSlidesA Self-stabilizing Publish/Subscribe Middleware for Wireless Sensor NetworksThis paper presents a scalable, self-stabilizing middleware for channel-based publish/subscribe systems for wireless sensor networks. The middleware eventually provides safety and liveness properties such as the guaranteed delivery of all published messages to all subscribers of the corresponding channel and the correct handling of subscriptions and unsubscriptions, while no error occurs. We consider transient message and memory corruptions and also respect dynamic network changes such as node and link removals and additions. We assume the message passing model and guarantee delivery of publications to new subscribers after O(n) steps.
Gerry Siegemund (TU Hamburg-Harburg), Khaled Maamra (University of Versailles St-Quentin), Volker Turau (Hamburg University of Technology)
- PDFSlidesOrganizing and Evaluating Publish/Subscribe Systems with ScopesModern applications and services need to process and respond to an ever growing number of internal and external events. In principle, publish/subscribe middleware would be the ideal platform for these event-driven applications, which range from the gathering of simple sensor data in ubiquitous environments up to the correlation and processing of complex events in business workflows. Most publish/subscribe middleware, however, offers no or just little support to adequately structure and organize network, services, and applications making it hard to deploy and adapt them in complex dynamic infrastructures.
In this paper, we refine and concretize a scope concept for publish/subscribe systems to give application developers and system administrators powerful and effective means to organize these systems by controlling the visibility of events, subscriptions, and advertisements. Besides presenting the scope concept and discussing the integration into a middleware, our main contribution is a comprehensive evaluation based on the first publish/subscribe middleware prototype with integrated scoping. The evaluation analyzes the behavior a scoped publish/subscribe system in various settings and quantifies the benefits of using scopes.
Helge Parzyjegla (University of Rostock), Matthias Prellwitz (University of Rostock), Gero Muehl (University of Rostock)
- PDFSlidesInternal Attacks in Anonymous Publish-Subscribe P2P OverlaysPrivacy, in particular anonymity, is desirable in Online Social Networks (OSNs) like Twitter, especially when considering the threat of political repression and censorship. P2P-based publish-subscribe is a well suited paradigm for OSN scenarios as users can publish and follow topics of interest. However, anonymity in P2P-based publish-subscribe (pub-sub) has been hardly analyzed so far. Research on add-on anonymization systems such as Tor mostly focuses on large scale traffic analysis rather than malicious insiders. Therefore, we analyze colluding insider attackers in more detail that operate on the basis of timing information. For that, we model a generic anonymous pub-sub system, present an attacker model, and discuss timing attacks. We analyze these attacks by a realistic simulation model and discuss potential countermeasures. Our findings indicate that even few malicious insiders are capable to disclose a large number of participants, while an attacker using large amounts of colluding nodes achieves only minor additional improvements.
Jörg Daubert (TU Darmstadt), Tim Grube (TU Darmstadt), Max Mühlhäuser (TU Darmstadt), Mathias Fischer (TU Darmstadt)
- PDFSlidesRaLaNS - A Ray Launching Based Propagation Loss Model for ns-3Although a realistic simulation of wireless networks mainly depends on a proper model of radio wave propagation, it is common to simply assume a fixed, circular range of communication. Many protocols and algorithms have been designed based on this assumption, risking that they might not be applicable to real scenarios. We assume that researchers mainly fall back to simple propagation models because more realistic models are rare and difficult to use. Therefore, we present RaLaNS, a propagation loss model for ns-3 that is based on ray launching. The generated signal strength distribution takes into account reflection and diffraction at buildings, and thus, greatly differs from the simple models in ns-3. Therefore, RaLaNS allows for a more realistic evaluation of new algorithms. We demonstrate this by showing a case study of node placement strategies in urban areas in the second part of the paper.
Thomas Hänel (University of Osnabrück), Alexander Bothe (University of Osnabrück), Nils Aschenbruck (University of Osnabrück)
- PDFSlidesAccurate Pollutant Modeling and Mapping: Applying Machine Learning to Participatory Sensing and Urban Topology DataAs sensor networks and mobile and participatory sensing mature, large environmental datasets become available. Environmental scientist are not prepared to use these vast and noisy datasets for environmental modeling. Today, environmental pollutants (e.g., noise) are simulated and the resulting model is verified by a small number of stationary measurements. These models are updated infrequently and provide only limited time resolution.
Recently, people have started to apply regression to train environmental models. This has shown great promise, but the complexity of regression models might not always be needed. Classification, however, has not been investigated as a mean to provide high-resolution environmental models from noisy data. The main contribution of this paper is a thorough investigation on the application of classification to environmental modeling (using noise as example pollutant). We present an end-to-end classification pipeline that predicts six classes of noise pollution with a precision of 80.89% and a recall of 80.90% using 10-fold cross-validation. Furthermore, we show the advantages of our approach regarding robustness to underline the applicability of classification for real-world scenarios.
Axel Schulz (TU Darmstadt), Jakob Karolus (TU Darmstadt), Frederik Janssen (TU Darmstadt), Immanuel Schweizer (TU Darmstadt)
- PDFSlidesA Study of Networking Software Induced LatencyFor long time, high-speed packet processing has been reserved for specialized hardware devices since software based solutions were not able to achieve the required performance. However, off-the-shelf packet processing hardware and software improved over the last years, which is why software based solutions cope with high-speed traffic nowadays. Due to the flexibility of software there is a trend towards doing packet processing in software, e.g. using OpenFlow or virtual switches. Although packet processing in software offers many capabilities, the complexity of such software bases solutions makes it hard to evaluate, optimize, or predict the networking performance of servers, end user hosts, or routers. We present a study that investigates the packet latency caused by the packet processing in the Linux network stack. We develop a simulation model in ns-3 for packet processing via the Linux network stack that helps understanding of its performance implications. We validate our simulation model based on measurements with nanosecond accuracy and software profiling.
Alexander Beifuß (University of Hamburg), Daniel Raumer (TU Munich), Paul Emmerich (TU Munich), Torsten M. Runge (University of Hamburg), Florian Wohlfart (TU Munich), Bernd E. Wolfinger (University of Hamburg), Georg Carle (TU Munich)
- PDFSlidesSmarterPhones: Anticipatory Download Scheduling for Wireless Video StreamingVideo streaming is in high demand by mobile users. In cellular networks, however, the unreliable wireless channel leads to two major problems. Poor channel states degrade video quality and interrupt the playback when a user cannot sufficiently fill its local playout buffer: buffer underruns occur. In contrast, good channel conditions cause common greedy buffering schemes to buffer too much data. Such over-buffering wastes expensive wireless channel capacity.
Assuming that we can anticipate future data rates, we plan the quality and download time of video segments ahead. This anticipatory download scheduling avoids buffer underruns by downloading a large number of segments before a drop in available data rate occurs, without wasting wireless capacity by excessive buffering.
We developed a practical anticipatory scheduling algorithm for segmented video streaming protocols (e.g., HLS or MPEG DASH). Simulation results and testbed measurements show that our solution essentially eliminates playback interruptions without significantly decreasing video quality.
Martin Dräxler (University of Paderborn), Johannes Blobel (University of Paderborn), Philipp Dreimann (University of Paderborn), Stefan Valentin (Bell Labs, Alcatel Lucent Germany), Holger Karl (University of Paderborn)
- PDFSlidesAPP and PHY in Harmony: A Framework Enabling Flexible Physical Layer Processing to Address Application RequirementsMobile data traffic, particularly mobile video, grows at an unprecedented pace. Despite recent advances at the physical layer, today’s wireless network infrastructure cannot keep up with this growth. This is partially due to the missing flexibility to adapt the physical layer continuously to best support both application level as well as network requirements. In this paper we show how to harness the flexibility of advanced physical layers in practice. We designed and implemented a research platform that provides a flexible application-centric physical layer for Android smartphones using software-defined radios (SDRs) as radio interfaces. Our solution allows applications to define lows and apply per-flow settings that are mapped into distinct physical layer settings. As a proof-of-concept and for testbed evaluation, we implemented our system together with a mobile video streaming application. The latter uses a Motion-JPEG based lightweight scalable video codec (SVC) to generate incremental data flows. We show that our system maximizes video quality at the receiver’s side, while keeping the energy consumption at the transmitter at a minimum. Our solution demonstrates that jointly optimizing network traffic and application quality is feasible in practice using a flexible physical layer processing approach.
Matthias Schulz (TU Darmstadt), Denny Stohr (TU Darmstadt), Stefan Wilk (TU Darmstadt), Benedikt Rudolph (TU Darmstadt), Wolfgang Effelsberg (TU Darmstadt), Matthias Hollick (TU Darmstadt)
⇒ Best Paper Award of Netsys 2015
- PDFSlidesThe Potential of Social-aware Multimedia Prefetching on Mobile DevicesThe access to Online Social Networks (OSN) and to media shared over these platforms account for around 20% of today’s mobile Internet traffic. For mobile device users, the access to media content and specifically videos is still challenging and costly. Mobile contracts usually have a data cap and connection qualities can vary greatly, depending on the cellular network coverage. Prefetching mechanisms that fetch content items beforehand, in times when the mobile device is connected to a WiFi network, have a high potential to address these problems. Yet, such a mechanism can only be effective if relevant content can be predicted with a high accuracy. Therefore, in this paper, an analysis of content properties and their potential for prediction are presented. An initial user study with 14 Facebook users running an app on their mobile device was conducted. The results show that video consumption is very diverse across the users. This work discusses the evaluation setup, the data analysis, and their potential to define an effective prefetching algorithm.
Stefan Wilk (TU Darmstadt), Julius Rückert (TU Darmstadt), Timo Thräm (TU Darmstadt), Christian Koch (TU Darmstadt), Wolfgang Effelsberg (TU Darmstadt), David Hausheer (TU Darmstadt)
- PDFSlidesEnabling Resilient Smart Grid Communication over the Information-Centric C-DAX MiddlewareLimited scalability, reliability, and security of todays utility communication infrastructures are main obstacles to the deployment of smart grid applications. The C-DAX project aims at providing and investigating a communication middleware for smart grids to address these problems, applying the information-centric networking and publish/subscribe paradigm. We briefly describe the C-DAX architecture, and extend it with a flexible resilience concept, based on resilient data forwarding and data redundancy. Different levels of resilience support are defined, and their underlying mechanisms are described. Experiments show fast and reliable performance of the resilience mechanism.
Michael Höfling (University of Tuebingen), Florian Heimgärtner (University of Tuebingen), Michael Menth (University of Tuebingen), Konstantinos V. Katsaros (University College London), Paolo Romano (Swiss Federal Institute of Lausanne (EPFL)), Lorenzo Zanni (Swiss Federal Institute of Lausanne (EPFL)), George Kamel (University of Surrey)
- PDFSlidesCCN Forwarding Strategies for Multihomed Mobile TerminalsCurrent IP-based networks are unable to fully exploit the capabilities of the increasing number of multihomed mobile terminals. We argue that Content-Centric Networking (CCN), a novel networking architecture based on named information objects, can fill the gap. In this paper, we elicit requirements for CCN packet forwarding on multihomed mobile terminals. We categorize CCN forwarding strategies according to their ability to fulfill these requirements and provide a real-world performance evaluation in the current CCNx prototype implementation. Moreover, we describe the initial design of an advanced multipath forwarding strategy.
Klaus M. Schneider (Otto-Friedrich-University of Bamberg), Kai Mast (Otto-Friedrich-University of Bamberg), Udo R. Krieger (Otto-Friedrich-University of Bamberg)
- PDFSlidesDynamic Transmission Modes to Support Opportunistic Information-Centric NetworksIn this paper, we describe dynamic unicast to increase communication efficiency in opportunistic Information-centric networks. The approach is based on broadcast requests to quickly find content and dynamically creating unicast links to content sources without the need of neighbor discovery. The links are kept temporarily as long as they deliver content and are quickly removed otherwise. Evaluations in mobile networks show that this approach maintains ICN flexibility to support seamless mobile communication and achieves up to 56.6% shorter transmission times compared to broadcast in case of multiple concurrent requesters. Apart from that, dynamic unicast unburdens listener nodes from processing unwanted content resulting in lower processing overhead and power consumption at these nodes. The approach can be easily included into existing ICN architectures using only available data structures.
Carlos Anastasiades (University of Bern), Torsten Braun (University of Bern)
- PDFSlidesA Novel Middleware for Smart Grid Data Exchange Towards the Energy Efficiency in BuildingsThe objective of this paper is to present a generic middleware conceived to operate as the linking element in platforms hosting smart energy management services to reduce energy consumption in buildings. This kind of solution presents specific requirements related to the need of accessing and managing different sources of information, internal and external to the building, and related to its structure, geometry, or energy consumption. This information is then processed by the system to determine how to improve the energy behavior of the building. In this context, different elements, communicating in a different way and speaking different languages, have to inter-operate with each other to reach the common objective of reducing the energy consumption by executing integrated energy management actions. With the aim of making this system interoperable, coherent, easily expandable, and transparent, the proposed middleware provides a homogeneous level of abstraction in this heterogeneous scenario.
Susana Martin (CARTIF Centro Tecnologico), Jose Hernandez (CARTIF Centro Tecnologico), Cesar Valmaseda (CARTIF Centro Tecnologico)
- PDFSlidesAn Adaptive Middleware Core for a Multi-Agent Coordination LanguageAchieving cooperative behavior in multi-robot systems in dynamic environments with unreliable communication is an active research area. Therefore, several middlewares and protocols have been proposed including different consensus mechanisms. However, these solutions either are tailored for very specific problems and thus lack flexibility or are based on general middleware frameworks that are too heavy-weight and inefficient. Hence, as main contribution we present a comprehensive multi-robot coordination approach, which can adapt its communication middleware core to explicitly address unreliable and range-limited communication for the needs of multi-robot systems. To achieve this, we use a lightweight and fully distributed object space that supports different consistency levels as adaptive middleware core embedded in a multi-robot coordination language.
Andreas Witsch (University of Kassel), Kurt Geihs (University of Kassel)
- PDFSlidesA Cost Efficient Scheduling Strategy to Guarantee Probabilistic Workflow DeadlinesToday, workflows are widely used to model business processes. A recent trend is to use them to model applications in heterogeneous, large-scale distributed systems. In such systems, many, possibly mobile, providers offer independent and interchangeable services that can be used to satisfy the different activities of a workflow. Due to varying server loads, failures, and changing network characteristics, the response time of these services is highly volatile. Thus, it is hard to ensure the timely and reliable execution of workflows depending on such services. A common approach is to invoke several services in parallel to increase the probability of success. This, however, can easily lead to overprovisioning and high cost when needlessly invoked services have to be compensated. In this paper, we investigate the search space between parallel and sequential invocation of services. We propose to invoke independent services staggered over time to ensure timely workflow execution at minimal cost. Evaluations show that our approach reduces the execution cost by up to 85% while it guarantees to fulfill activity deadlines with 99.9% probability.
Thomas Bach (University of Stuttgart), Muhammad Adnan Tariq (University of Stuttgart), Boris Koldehofe (University of Stuttgart), Kurt Rothermel (University of Stuttgart)
- PDFSlidesCOSCAnet-FT: Transparent Network Support for Highly Available Cloud ServicesMore and more applications move to the cloud, even critical systems that need high availability. In current clouds, faults can be handled for stateless HTTP applications. Other protocols and stateful applications cannot be supported. Highly-available stateful services could use active replication, but it typically needs client-side code for supporting complex totally-ordered multicasts. This paper first presents a transport-level router as a service in the network that transparently multicasts TCP traffic to actively replicated service instances. Second, it demonstrates the integration of this concept into a PaaS cloud as a value-added service to customers so that applications can be replicated on demand. Finally, an evaluation of our prototype shows reasonable throughput, latency and recovery time.
Steffen Kächele (University of Ulm), Franz J. Hauck (University of Ulm)
- PDFSlidesPartial Delaunay Triangulation-Based Asynchronous Planarization of Quasi Unit Disk GraphsWe present a distributed and fully asynchronous algorithm for construction of the partial Delaunay triangulation over quasi unit disk graphs. Provided that the ratio of the maximum to the minimum communication range of nodes is bounded from above by square root of two, our algorithm outputs a connected and planar overlay graph of the input graph, which enables the use of localized geographic routing algorithms that guarantee message delivery. Moreover, under the assumption that the input graph is civilized (i.e., any two network nodes have non-zero minimum Euclidean distance), we show that our algorithm is localized. We show by means of simulation that our approach yields output graphs whose Euclidean spanning ratio is on average significantly smaller compared to those constructed by all other asynchronous approaches.
Florentin Neumann (University of Koblenz-Landau), Hannes Frey (University of Koblenz-Landau)
- PDFSlidesIntegrating Churn into the Formal Analysis of Routing AlgorithmsMathematical modeling and analysis of distributed systems, mostly applied with the goal of the correctness or asymptotic behavior of a system, rarely provides concrete results and often disregards or simplifies network dynamics. However, concrete performance bounds on a system under churn are highly useful both as a validation of empirical results and a scalable alternative to simulations. In this paper, we first present an abstract methodology for deriving the success probability of an action, such as routing, in a dynamic system, using the session length distribution as the decisive parameter. We evaluate the developed methodology by giving concrete bounds on the success probability of recursive routing. The results do not only show the adaptability of our model, but also reveal that a considerable fraction of routing attempts fails due to a leaving node on the return path rather than due to not reaching the target.
Stefanie Roos (TU Dresden), Giang T. Nguyen (TU Darmstadt), Thorsten Strufe (TU Dresden)
- PDFSlidesTerrorist Fraud Resistance of Distance Bounding Protocols Employing Physical Unclonable Functions Distance bounding protocols (DBPs) are security protocols that aim to limit the maximum possible distance between two partners in a wireless communication. This enables to ensure locality of interaction between two devices. Despite numerous proposed protocols, recent analyses of DBPs have shown the majority of them to be susceptible to attacks. Most prominent among the unsolved security problems of DBPs is terrorist fraud. This type of attack involves collaboration with a legitimate device, after which the attacker can successfully execute the protocol. We show how terrorist fraud can be prevented by replacing shared secrets – commonly used in classical DBPs – with physical unclonable functions (PUFs). Our new approach can be integrated in all current DBPs with minor modifications. We offer two alternate designs: One utilizing challenge-response PUFs and another using so-called SIMPL systems, a PUF-analogue to public-key cryptography. We use a security model proposed by previous work to demonstrate security of our scheme.
Stephan Kleber (University of Ulm), Rens W. van der Heijden (University of Ulm), Henning Kopp (University of Ulm), Frank Kargl (University of Ulm)
- PDFSlidesInferring Obfuscated Values in FreenetConducting data analysis and system monitoring in a privacy-preserving manner is extremely important for anonymity systems such as the distributed publication system Freenet. The current obfuscation mechanisms for gathering statistics in Freenet are designed to anonymize both the responding node and the response itself. We show that due to the possibility of repeated targeted queries, hidden information, which can be potentially abused to damage both individual users and the system as a whole, about specific nodes can be derived using Bayesian Statistics. Our evaluation, using both an in-depth simulation study and real-world measurements, show that the hidden information can be inferred accurately in more than 86% of all experiments, with a relative error below 0.05 in more than 99.5% of all considered scenarios. Furthermore, we present an initial design for an improved obfuscation method, which is guaranteed to provide k-anonymity.
Stefanie Roos (TU Dresden), Florian Platzer (TU Darmstadt), Jan-Michael Heller (TU Darmstadt), Thorsten Strufe (TU Dresden)
- PDFSlidesOptimized Location Update Protocols for Secure and Efficient Position SharingAlthough location-based applications have seen fast growth in the last decade due to pervasive adoption of GPS enabled mobile devices, their use raises privacy concerns. To mitigate these concerns, a number of approaches have been proposed in literature, many of which rely on a trusted party to regulate user privacy. However, trusted parties are known to be prone to data breaches . Consequently, a novel solution, called Position Sharing, was proposed in  to secure location privacy in fully non-trusted systems. In Position Sharing, obfuscated position shares of the actual user location are distributed among several location servers, each from a different provider, such that there is no single point of failure if the servers get breached. While Position Sharing can exhibit useful properties such as graceful degradation of privacy, it incurs significant communication overhead as position shares are sent to several location servers instead of one.
To this end, we propose a set of location update protocols to minimize the communication overhead of Position Sharing while maintaining the privacy guarantees that it originally provided. As we consider the scenario of frequent location updates, i.e., movement trajectories, our protocols additionally add protection against an attack based on spatio-temporal correlation in published locations. By evaluating on a set of real-world GPS traces, we show that our protocols can reduce the communication overhead by 75% while significantly improving the security guarantees of the original Position Sharing algorithm.
Zohaib Riaz (University of Stuttgart), Frank Dürr (University of Stuttgart), Kurt Rothermel (University of Stuttgart)
- PDFSlidesRetaining Control Over SDN Network ServicesBoth academic researchers and network vendors have started to promote Software-Defined Networking (SDN) as a new network paradigm, in which controller systems play a major role. A modular and extensible design allows network operators to extend the controller’s functionality by so called network services. Unfortunately, in current designs such network services have unlimited access to mandatory SDN resources which enables different kinds of attacks. To retain control over network services (especially third-party ones), we adapt approved security mechanisms and propose a containment mechanism as well as a framework to ease containment configuration. For both proposals, we provide proof-of-concept implementations for an open and industry-supported reference framework and hereby aim to improve security for a wide range of SDN controllers. Finally, our proposals achieve the ability to harden a mandatory SDN component (i.e., the SDN controller) and enable proactive security even against malicious network services.
Christian Röpke (Ruhr-University Bochum), Thorsten Holz (Ruhr-University Bochum)