Felix Juraschek, Mesut Güneş, Matthias Philipp, Bastian Blywis - 2011
Abstract
In this paper we present an experimental evaluation of the distributed greedy algorithm (DGA) for distributed channel assignment in wireless mesh networks. The algorithm has the advantage of preserving the network topology by assigning channels to links instead of interfaces, thus being completely transparent to the routing layer. Our implementation is based on DES-Chan, a framework for the development of distributed channel assignment algorithms. We evaluate the performance in the DES-Testbed, a multi-radio wireless mesh network (WMN) with 100 nodes at the Freie Universitaet Berlin. We present a graph-theoretic analysis of the experiment results and measure the achieved throughput after the channel assignment. We discuss the feasibility of link-based channel assignment and show that the feature of the algorithm of being transparent to the routing layer is not always guaranteed. Additionally, we show the importance of using realistic interference models to fully exploit the performance gain by channel assignment in real network deployments.
Felix Juraschek, Mesut Günes, Matthias Philipp, Bastian Blywis, Oliver Hahm,, Felix Juraschek, Mesut Güneş, Matthias Philipp, Bastian Blywis, Oliver Hahm - 2011
Abstract
In this paper we present the DES-Chan framework for experimentally-driven research on distributed channel assignment algorithms in wireless mesh networks. The implementation process of channel assignment algorithms is a difficult task for the researcher since common operating systems do not support channel assignment algorithms out of the box. DES-Chan provides a set of common services required by distributed channel assignment algorithms. The modular architecture of DES-Chan allows the extensions with further modules or the modifications of existing ones. As a proof of concept, we present a reference implementation of a distributed greedy channel assignment algorithm. We evaluate its performance in the DES-Testbed, a multi-transceiver wireless mesh network (WMN) with 100 nodes at the Freie Universitaet Berlin.
Heiko Will, Felix Juraschek, Mesut Günes, Jochen Schiller,, Heiko Will, Felix Juraschek, Mesut Güneş, Jochen Schiller - 2011
Abstract
The MANIAC Challenge is a competition for cooperation strategies in wireless ad-hoc networks with the focus on experimental evaluation. We present the results of the MANIAC Challenge and discuss characteristics of real networks such as link instability and mobility, which are often simplified in common network simulators. We introduce our strategy Friendly Clustering that won the Performance Award. Friendly Clustering is based on monitoring the neighbors' forwarding behavior and assessing their willingness to relay future data packets. Further on, we describe the challenges in implementing the strategy for real networks, such as the DES-Testbed at the Freie Universitaet Berlin and the ad-hoc network of the MANIAC Challenge.
The MANIAC Challenge is a competition for cooperation strategies in wireless ad-hoc networks with the focus on experimental evaluation. We present the results of the MANIAC Challenge and discuss characteristics of real networks such as link instability and mobility, which are often simplified in common network simulators. We introduce our strategy Friendly Clustering that won the Performance Award. Friendly Clustering is based on monitoring the neighbors' forwarding behavior and assessing their willingness to relay future data packets. Further on, we describe the challenges in implementing the strategy for real networks, such as the DES-Testbed at the Freie Universitaet Berlin and the ad-hoc network of the MANIAC Challenge.
Bastian Blywis, Mesut Günes, Felix Juraschek, Oliver Hahm, Nicolai Schmittberger, Bastian Blywis, Mesut Güneş, Felix Juraschek, Oliver Hahm, Nicolai Schmittberger - 2011
Download
Abstract
The Distributed Embedded Systems Testbed (DES-Testbed) is a hybrid wireless mesh and wireless sensor network that has been deployed at Freie Universität Berlin and was successively extended from November 2007 to December 2010. This technical report gives an overview of the current topology and the properties of the IEEE 802.11 wireless mesh network that is part of the DES-Testbed. The information that was gathered from an experimental study shall enable researchers to optimize their experiment scenarios, to support the evaluation of experiments, and to derive improved models of real world deployments. The differences of testbeds compared with simulation models and how to evaluate and filter the raw data are addressed. The focus of our study is an up-to-date description of the testbed state and to highlight particular issues. We show that the node degree, link ranges, and packet delivery ratios are not normal distributed and that simple means are not sufficient to describe the properties of a real world wireless network. Significant differences of the results from three channels are discussed. As last, the technical report shows that the DES-Testbed is an overall well connected network that is suited for studies of wireless mesh network and wireless mobile ad-hoc network problems.
Thomas C. Schmidt, Gabriel Hege, Matthias Wählisch, Hans L. Cycon, Mark Palkow, Detlev Marpe, Matthias Wählisch - 2011
Download
Abstract
The design of conferencing systems for achieving efficient and flexible communication in a fully distributed, infrastructure-independent fashion is a promising direction, both in terms of research and practical development. In the particular case of video communication, the seamless adaptation to heterogeneous mobile devices poses an additional strong challenge to those seeking for interoperable and easy-to-deploy solutions. In this paper, we make several contributions towards a generic peer-to-peer (P2P) videoconferencing solution that extends into the mobile realm. We describe the essential building blocks for conference management and media distribution that are necessary for a distributed conferencing approach. Establishing a distributed SIP conference focus, participants share the conference according to their individually given capabilities and resources in terms of bandwidth and processing power rather than in a centralized and fixed way. Overall concepts and SIP-primitives for such an autonomous organization are presented. Security issues that derive from this decentralized identity management are resolved by so-called {em Overlay AuthoCast}, a novel use of cryptographically generated identifiers. Furthermore, this work is dedicated to the development of a software-based H.264 video codec implementation and the specific aspects resulting from tuning such a highly resource-intensive software codec to the given target platform of a standard consumer smartphone.
This document describes deployment options for activating multicast listener functions in Proxy Mobile IPv6 domains without modifying mobility and multicast protocol standards. Similar to home agents in Mobile IPv6, Local Mobility Anchors of Proxy Mobile IPv6 serve as multicast subscription anchor points, while Mobile Access Gateways provide Multicast Listener Discovery (MLD) proxy functions. In this scenario, mobile nodes remain agnostic of multicast mobility operations. Support for mobile multicast senders is outside the scope of this document.
This document describes deployment options for activating multicast listener functions in Proxy Mobile IPv6 domains without modifying mobility and multicast protocol standards. Similar to home agents in Mobile IPv6, Local Mobility Anchors of Proxy Mobile IPv6 serve as multicast subscription anchor points, while Mobile Access Gateways provide Multicast Listener Discovery (MLD) proxy functions. In this scenario, mobile nodes remain agnostic of multicast mobility operations. Support for mobile multicast senders is outside the scope of this document.
Thomas C. Schmidt, Matthias Wählisch, Muhamma Omer Farooq, Matthias Wählisch - 2011
Download
Abstract
Multicast communication can be enabled in Proxy Mobile IPv6 domains by deploying MLD Proxy functions at Mobile Access Gateways, and multicast routing functions at Local Mobility Anchors. This document describes the support of mobile multicast senders in Proxy Mobile IPv6 domains that is provided by this base deployment scenario. Mobile sources remain agnostic of multicast mobility operations.
Multicast communication can be enabled in Proxy Mobile IPv6 domains by deploying MLD Proxy functions at Mobile Access Gateways, and multicast routing functions at Local Mobility Anchors. This document describes the support of mobile multicast senders in Proxy Mobile IPv6 domains that is provided by this base deployment scenario. Mobile sources remain agnostic of multicast mobility operations.
Bastian Blywis, Mesut Günes, Felix Juraschek, Oliver Hahm,, Bastian Blywis, Mesut Güneş, Felix Juraschek, Oliver Hahm - 2011
Download
Abstract
The Distributed Embedded Systems Testbed (DES-Testbed) is a hybrid wireless mesh and wireless sensor network that has been deployed at Freie Universität Berlin and was successively extended from November 2007 to December 2010. This technical report gives an overview of the current topology and the properties of the IEEE 802.11 wireless mesh network that is part of the DES-Testbed. The information that was gathered from an experimental study shall enable researchers to optimize their experiment scenarios, to support the evaluation of experiments, and to derive improved models of real world deployments. The differences of testbeds compared with simulation models and how to evaluate and filter the raw data are addressed. The focus of our study is an up-to-date description of the testbed state and to highlight particular issues. We show that the node degree, link ranges, and packet delivery ratios are not normal distributed and that simple means are not sufficient to describe the properties of a real world wireless network. Significant differences of the results from three channels are discussed. As last, the technical report shows that the DES-Testbed is an overall well connected network that is suited for studies of wireless mesh network and wireless mobile ad-hoc network problems.
Matthias Wählisch, Thomas C. Schmidt, Stig Venaas, Matthias Wählisch - 2011
Download
Abstract
Group communication services exist in a large variety of flavors, and
technical implementations at different protocol layers. Multicast
data distribution is most efficiently performed on the lowest
available layer, but a heterogeneous deployment status of multicast
technologies throughout the Internet requires an adaptive service
binding at runtime. Today, it is difficult to write an application
that runs everywhere and at the same time makes use of the most
efficient multicast service available in the network. Facing
robustness requirements, developers are frequently forced to use a
stable, upper layer protocol controlled by the application itself.
This document describes a common multicast API that is suitable for
transparent communication in underlay and overlay, and grants access
to the different multicast flavors. It proposes an abstract naming
by multicast URIs and discusses mapping mechanisms between different
namespaces and distribution technologies. Additionally, it describes
the application of this API for building gateways that interconnect
current multicast domains throughout the Internet.
Channel assignment for Wireless Mesh Networks (WMNs) attempts to increase the network performance by decreasing the interference of simultaneous transmissions. The reduction of interference is achieved by exploiting the availability of fully or partially non-overlapping channels. Although it is still a young research area, many different approaches have already been developed. These approaches can be distinguished into centralized and distributed. Centralized algorithms rely on a central entity, usually called Channel Assignment Server (CAS), which calculates the channel assignment and sends the result to the mesh routers. In distributed approaches, each mesh router calculates its channel assignment decision based on local information. Distributed approaches can react faster to topology changes due to node failures or mobility and usually introduce less protocol overhead since communication with the CAS is not necessary. As a result, distributed approaches are more suitable once the network is operational and running. Distributed approaches can further be classiï¬ed into static and dynamic, in regard to the modus of channel switching. In dynamic approaches, channels can be switched on a per-packet basis, whereas in static approaches radios stay on a speciï¬c channel for a longer period of time. Static assignments have been more in focus, since the channel switching time for current Institute of Electrical and Electronics Engineers (IEEE) 802.11 hardware is in the order of milliseconds which is two orders higher than the packet transmission time. Recently, surveys of channel assignment algorithms have been presented which cover certain aspects of the research ï¬eld. The survey in [1] introduces the problem and presents a couple of distributed algorithms and [2] gives a broad introduction to centralized and distributed approaches. The survey herein is focused on distributed approaches for peer- to-peer network architectures. This report describes the problem formulation for channel assignment in WMNs and the fundamental concepts and challenges of this research area. We present different distributed channel assignment algorithms and characterize them according to a set of classiï¬cation keys. Since channel assignment algorithms may change the connectivity and therefore the network topology, they may have a high impact on routing. Therefore, we present routing metrics that consider channel diversity and adapt better to the multi- radio multi-channel scenario than traditional routing metrics designed for single channel networks. The presented algorithms are discussed and compared focusing on practical evaluations in testbed and network environments. The implementation for real networks is a hard and labor-intensive task because the researcher has to deal with the complexity of the hardware, operating system, and wireless network interface drivers. As a result, frameworks emerged in order to simplify the implementation process. We describe these frameworks and the mechanisms used to help researchers implementing their algorithms and show their limitations and restrictions.
Felix Juraschek, Mesut Günes, Matthias Philipp, Bastian Blywis,, Felix Juraschek, Mesut Güneş, Matthias Philipp, Bastian Blywis - 2011
Abstract
This article presents the DES-Chan framework for experimental research on distributed channel assignment algorithms in wireless mesh testbeds. The implementation process of channel assignment algorithms is a difficult task for the researcher since common operating systems do not support channel assignment algorithms. DES-Chan provides a set of common services required by distributed channel assignment algorithms and eases the implementation effort. The results of experiments to measure the channel characteristics in terms of intra-path and inter-path interference according to the channel distance on the DES-Testbed are also presented. The DES-Testbed is a multi-radio WMN with more than 100 nodes located on the campus of the Freie Universität Berlin. These measurements are an important input to validate common assumptions of WMNs and derive more realistic, measurement-based interference models in contrast to simplified heuristics.
Sebastian Trapp, Matthias Wählisch, Jochen Schiller,, Sebastian Trapp, Matthias Wählisch, Jochen Schiller - 2011
Abstract
In ad hoc communication, data packets are relayed over several hops before reaching their destination. Spontaneous communication requires that nodes trust each other as communication can be intentionally disturbed or privacy compromised by the intermediate nodes. Establishing this trust relationship within a MANET without access to a central authority poses a challenge. In this work, we discuss the problem of ad hoc trust assignment and present an approach that helps to establish trust relationships between smartphones forming a MANET. Inspired by sociological insights we argue that data inherently available at mobiles can be used to define the social relationship of two individuals. Based on a preliminary measurement-based analysis we show that this data can give an initial estimation of trust between two users and their mobiles.
Hans L. Cycon, Gabriel Hege, Detlev Marpe, Mark Palkow, Thomas C. Schmidt, Matthias Wählisch - 2010
Download
Abstract
In this paper, we present a multipoint video conferencing system that adapts to heterogeneous members including mobiles. The system is built upon a low complexity scalable extension of our H.264 codec DAVC, and a congestion-aware dynamic adaptation layer. Our temporally scaled video codec DSVC has the same RD performance as the non-scaled version with comparable configuration. We achieve this by QP cascading, i.e., assigning gradually refining quantization parameters to the declining temporal layers. We present and analyse a mobile-compliant version of DSVC at reduced complexity that still admits comparable performance. Finally, we report on early work of dynamic layer tuning. Derived of delay variation measures, senders exploit scalable video layering to adapt the video transmission to varying network conditions. Initial results indicate that video performance remains close to optimal.
Sebastian Meiling, Dominik Charousset, Thomas C. Schmidt, Matthias Wählisch - 2010
Abstract
The Internet has been successful in providing a uniformly available IPv4 unicast service, accessible via the socket API which eased application deployment. However, this tightly shaped concept failed to promote a heterogeneous network layer (including IPv6) or additional services like multicast. In this paper, we propose methods for diversifying the Internet service capabilities with the help of enhanced end system intelligence and an API abstraction towards the applications. We provide a general architecture for the example of hybrid adaptive mobile multicast (HAMcast) which allows for various realizations and pluralistic deployments. The achievement of our contribution is a universal, robust service access that allows group applications to run everywhere, no matter what the status of regional technological deployment will be.
Matthias Wählisch, Sebastian Meiling, Thomas C. Schmidt, - 2010
Download
Abstract
The Internet has matured to a mission-critical infrastructure, and recently attracted much attention at political and legal levels in many countries. Civil actions regarding the Internet infrastructure require a thorough understanding of the national components of the global Internet to foresee possible impacts of regulations and operations at a country-level. In this paper we report on a methodology, tool chain and results for identifying and classifying a 'national Internet'. We argue for the importance to consider individual IP-blocks and quantify the effects of our proposed approach. The methods have been applied to identify a 'German Internet', but are designed general enough to work for most countries, as well.
Steffen Gliech, Bastian Blywis, Mesut Güneş, Felix Juraschek - 2010
Abstract
Distributed localization has been in increased focus of research since the emergence of wireless multi-hop networks. One class of localization systems uses anchor-free algorithms, where no node in the network has information about its geographic position. None of the nodes acts as anchor and thus the algorithms have very limited requirements for the application in particular scenarios. The Anchor-Free Distributed Localization-Algorithm (AFL) by Priyantha et al. is such an algorithm that has been developed for applications in wireless sensor networks. It creates a relative coordinate system in the first phase and uses a mass-spring approach for the optimization of the nodes' locations in the second phase. In this publication we report our experiences implementing and using AFL in the DES-Testbed, an IEEE 802.11 based wireless mesh network. The Distributed Embedded Systems - Localization Framework for Testbeds (DES-LOFT) is introduced and discussed. It was developed for the implementation of localization systems as there is currently limited support for wireless mesh networks. DES-LOFT enables a holistic approach and interacts with other frameworks because a localization algorithm is never run independent from other services. Based on a first experiment series, we discuss the applicability of AFL in real world networks, propose improvements, and present preliminary results from experiments.
Gossip routing is an approach to reduce the redundancy of flooding in wireless networks. A study by Haas et al. evaluated different gossip routing variants in simulations on regular and random network topologies. Using the DES-Testbed, a wireless mesh network, we tried to replicate their experiments to evaluate whether the findings hold in real world scenarios. Four different gossip routing variants and the experiment setup are elaborated as well as issues regarding the replication of the experiments discussed. With this study we demonstrate that even small wireless network deployments show a bimodal behavior when a certain probability threshold is passed.
Alexander Knauf, Gabriel Hege, Thomas C. Schmidt, Matthias Wählisch - 2010
Abstract
There is an increasing demand to access voice or video group conferences without the burden of a dedicated infrastructure, but at any place and in an ad hoc fashion. Corresponding solutions require a lightweight, fully distributed cooperation among parties that share and manage the conference in an efficient, self-adaptive way. The technology framework of P2PSIP can be seen as a promising starting point to meet these objectives. In this paper, we make several contributions towards such a distributed, virtualized control layer based on P2PSIP that seamlessly scales and adapts to the user needs. We propose a P2P-signaling protocol scheme for a distributed conference control with SIP, that splits the semantic of Identifier and Locator of a SIP conference URI in a standard-compliant manner. This protocol scheme serves as further basis for a virtualization in RELOAD. We further design and evaluate a self-organizing communication layer that provides load sharing and churn resilience with proximity-awareness. Finally, we address key aspects of security and trust, as well as compatibility for conference unaware clients.
Aad van Moorsel, Philipp Reinecke, Katinka Wolter - 2010
Download
Abstract
Although adaptivity, the ability to adapt, is an important property of complex computing systems, so far little thought has been given to its evaluation. In this paper we propose a framework and methodology for the definition of benefit-based adaptivity metrics. The metrics thus defined allow an informed choice between systems based on their adaptivity to be made. We demonstrate application of the framework in a case study of restart strategies for Web Services Reliable Messaging. Additionally, we provide a broad survey of related approaches that may be used in the study of adaptivity (comprising, among others, robustness, performability, and control analysis), and evaluate their respective merits in relation to the proposed adaptivity metric.
Marco Ziegert, Frederik Hermans, Norman Dziengel, Zakaria Kasmi, Stephan Adler, Georg Wittenburg, Jochen Schiller - 2010
Download
Download
Abstract
Distributed event detection in wireless sensor networks is an approach to increase event recognition accuracy by fusing correlated parts of events, recognized by multiple sensor nodes. Events range from simple threshold detection to complex events requiring pattern analysis. Our previous deployment in this area of research investigates the possibility to recognize intrusion events, breaching security of wireless sensor network guarded areas. We present a new platform, consisting of an improved system layer and application layer that advances distributed event detection in wireless sensor networks. We improve the event detection accuracy, by applying an ARM7-based sensor board with more precise acceleration sensors. In addition, the threaded ï¬re-kernel and adjusted routing take the restricted resources on wireless sensor nodes into account. In order to gain a higher reliability in the detection results, we employ a distributed data quality estimator. The architecture’s goal is to carry out large-scale deployments on fences of construction sites that detect events as stated above.
The Fifth International Conference on Mobile Computing and Ubiquitous Networking (ICMU 2010), April 2010, Pardeep Kumar, Mesut Güneş, Qasim Mushtaq, Jochen Schiller - 2010
Abstract
A medium access control (MAC) protocol generally regulates the access of devices to a shared medium. In case of wireless sensor networks (WSNs), it is also responsible to create energy-efficient links between nodes, where messages can be sent to the sink node in a timely manner. This paper evaluates AREA-MAC, a medium access control protocol designed for real-time and energy-efficient WSN applications. AREA-MAC uses the low power listening (LPL) technique with short preamble messages to minimize latency, energy consumption, and control overhead on nodes. Though AREA-MAC is mainly associated with the MAC layer, it exploits simple routing at the network layer and interacts with application as well as PHY layers. In this evaluation a crosslayer design perspective is adopted, where direct interaction between different layers is examined. Through simulation study, we evaluate the timeliness, energy-efficiency, and packet reception ratio for nodes using AREA-MAC with and without cross-layer support. Results show that the cross-layering improves network performance in terms of timeliness and energyefficiency but it comes at the cost of relatively low packet reception ratio at the sink node.
Despite their limited resources, especially energy, wireless sensor networks (WSNs) are being accepted in many application areas. This makes them a challenging yet an appealing research area. Evidently, a lot of research work revolves around energy efficiency in order to prolong WSNs lifetime. A prominent amount of work suggests keeping radios of sensor nodes in low power sleep mode for most of the time. However, this duty-cycling results in increased network latency. Selection of an optimal duty-cycle is always challenging especially for applications where latency, along with energy consumption, plays an important role. This paper evaluates AREA-MAC for delay, energy, and packet delivery ratio factors for several duty-cycle values. AREA-MAC is a medium access control (MAC) protocol designed for realtime and energy-efficient WSN applications. Linear optimization is used to find the optimal value of duty-cycle in order to minimize delay and energy consumption and by keeping packet delivery ratio above a certain threshold.
Thomas C. Schmidt, Godred Fairhurst, Matthias Wählisch - 2010
Download
Abstract
This document discusses current mobility extensions to IP-layer multicast. It describes problems arising from mobile group communication in general, the case of multicast listener mobility, and problems for mobile senders using Any Source Multicast and Source-Specific Multicast. Characteristic aspects of multicast routing and deployment issues for fixed IPv6 networks are summarized. Specific properties and interplays with the underlying network access are surveyed with respect to the relevant technologies in the wireless domain. It outlines the principal approaches to multicast mobility, together with a comprehensive exploration of the mobile multicast problem and solution space. This document concludes with a conceptual road map for initial steps in standardization for use by future mobile multicast protocol designers. This document is a product of the IP Mobility Optimizations (MobOpts) Research Group.
Matthias Wählisch, Thomas C. Schmidt, - 2010
Download
Abstract
Key-based routing has enabled efficient group communication on the application or service middleware layer, stimulated by the need of applications to access multicast. These developments follow a continuous debate about network layer multicast that had lasted for about 30 years history of the Internet. The IP host group model today still faces a strongly divergent state of deployment. In this chapter, we first review the key concepts of multicast and broadcast data distribution on structured overlays. Second, we perform a comprehensive theoretical analysis examining the different distribution trees constructed on top of a key-based routing layer. Characteristic performance measures of the multicast approaches are compared in detail and major structural differences are identified. Overlay multicast overcomes deployment problems on the price of a performance penalty. Hybrid approaches, which dynamically combine multicast in overlay and underlay, adaptively optimize group communication. We discuss current schemes along with its integration in common multicast routing protocols in the third part of this chapter. Finally, we reconsider and enhance approaches to a common API for group communication, which serves the requirements of data distribution and maintenance for multicast and broadcast on a middleware abstraction layer, and in particular facilitates hybrid multicast schemes.
Miklós Telek, Philipp Reinecke, Katinka Wolter - 2010
Abstract
Phase-type (PH) distributions are proven to be very powerful tools in modelling and analysis of a wide range of phenomena in computer systems. The use of these distributions in simulation studies requires efficient methods for generating PH-distributed random numbers. In this work, we discuss algorithms for generating random numbers from PH distributions and propose two algorithms for reducing the cost associated with generating random numbers from Acyclic Phase-Type distributions (APH).
As its rapid growth continues and new applications and requirements arise, the technological advance of the Internet is increasingly limited by the facts and shortcomings of the existing infrastructure. Attempts to address this by incrementally extending or revising the existing standards have met with considerable diffculties, as in the case of IPv6 or Mobile IP. Although this is partly due to economical and practical reasons, the question presents itself if the fundamental technologies underlying the Internet themselves might need an overhaul. Part of the discussion of the future of the Internet thus turns towards clean slate designs, which envision how the Internet would look like if it were re-designed from scratch today. Purposely ignoring practical constraints, clean slate research has the liberty to challenge even the most fundamental concepts of the currentfdesign. TuneInNet is a clean slate design developed at the Freie Universität Berlin and the Universität Zürich. It seeks to banish the intelligence from the network in a more radical manner than the Internet Protocol Suite does, thereby aiming at a backbone that consists of simple, purely optical switches and provides only raw connectivity. In this scenario, data is flooded and ï¬ltered rather than routed based on addresses. This can be construed as a move towards the radio, from which it derives its name: the sender sends data without regard to whom, whereas the receiver filters out - or tunes in to - the data that interests him. This thesis presents and evaluates a prototypical implementation of TuneInNet.
Wireless mesh networks are in the focus of research for more than a decade. After a short simulation based research period, several testbeds have been set up for the study of particular research topics. As wireless mesh networks are entering a phase of wide-spread commercial application, novel approaches for holistic research are required. In this paper we review the trends, advances, and challenges in experimentally driven wireless mesh network research. The evolution of the experimentation facilities is elaborated by distinguishing three generations of testbeds. Based on the review and the current trends of research, open issues are discussed. We introduce the Distributed Embedded Systems Testbed (DES-Testbed) at Freie Universität Berlin as an example of the current third generation. Further on, its features are elaborated and requirements for future holistic research of wireless networks are discussed.
Note: 10.1007/s11036-010-0227-9
Mobile phones have become a basic commodity and smartphone devices claim a continuously increasing market share. Modern smartphone devices are equipped with powerful processing units and high speed mobile internet capabilities. Mobile operators in turn are offering a wide variety of data plans. These developments will lead to an increase of internet access from mobile devices. At the same time the internet is not solely used anymore for displaying data but collaborative editing, evaluation and distribution of data in any form has become widely popular, as can be seen in the so-called ``Web 2.0'' services. Suppliers are offering versions of these services that are especially designed for mobile devices. Today Peer-to-Peer networks are responsible for a considerable amount of the total internet traffic. Although Peer-to-Peer networks mainly gained attention in the media for copyright infringement cases Peer-to-Peer networks offer many positive characteristics. Peer-to-Peer networks are decentralized, self-organizing, scalable and offer no single point of failure. This makes Peer-to-Peer networks an ideal candidate for deploying mobile web services. The MobP2P base system is a foundation for developing Peer-to-Peer based applications for collaborative data processing developed for the Android platform. Applications built on top of the MobP2P base system enable users to create data items and to share these with remote users. Alterations to these data items are automatically synchronized on authorized devices.
Sebastian Hofmann, Bastian Blywis, Mesut Güneş, Felix Juraschek - 2010
Download
Abstract
Gossip routing is an approach to limit the overhead of flooding in wireless networks. Each node, that receives a packet that would normally be flooded, applies a probabilistic approach. The packet is either forwarded with probability p or dropped with 1 − p. This paper is a follow-up to our last study that evaluated the approaches by Haas et al. in the DES-Testbed, a wireless mesh network. Four different gossip routing variants were discussed which used static parameters to determine the value p. In this study, we compare the proposals by four other entities and discuss whether they show an overall improvement regarding the two most important metrics in this domain: reachability and redundancy. We also discuss the assumptions and parameters of the simulation studies in the context of our experiments in a real world deployment.
David J. H. Gutzmann, Bastian Blywis, Mesut Güneş, Felix Juraschek - 2010
Abstract
Dynamic Source Routing (DSR) was one of the first proposed protocols for routing in Mobile Ad-hoc Networks and is among the few that were ratified by the IETF. Despite its source routing overhead, the protocol exhibits particular properties that make it an optimal choice for multi-path routing where multiple paths are discovered for each destination. The alternative paths can either be stored for error recovery situations or used in parallel to improve the throughput. In this paper we present a study of 10 uni- and multi-path DSR variants in a wireless mesh network testbed. A modified ETX metric is proposed for an improved link quality measurement. The experiments are evaluated considering the routing-overhead and packet delivery ratio with focus on the route discovery.
In this paper we investigate the use of stochastic models for analysing service-oriented systems. We propose an iterative hybrid approach using system measurements, testbed observations as well as formal models to derive a quantitative model of service-based systems that allows us to evaluate the effectiveness of the restart method in such systems. In cases where one is fortunate enough as to have access to a real system for measurements the obtained data often is lacking statistical significance or knowledge of the system is not sufficient to explain the data. A testbed may then be preferable as it allows for long experiment series and provides full control of the system's configuration. In order to provide meaningful data the testbed must be equipped with fault-injection using a suitable fault-model and an appropriate load model. We fit phase-type distributions to the data obtained from the testbed in order to represent the observed data in a model that can be used e.g. as a service process in a queueing model of our service-oriented system. The queueing model may be used to analyse different restart policies, buffer size or service disciplines. Results from the model can be fed into the testbed and provide it with better fault and load models thus closing the modelling loop.
Die Erfindung betrifft ein Verfahren zur Datenbereitstellung auf mobilen Endgeräten mit den Schritten: Bereitstellen einer durchgängigen Netzkonnektivität einer Mehrzahl mobiler Endgeräte unterschiedlicher Nutzer, Ausführen einer lokalen Anwendung auf einem der Endgeräte, die zu einem Erstellen oder einer Änderung eines Datensatzes führt, und automatisches Bereitstellen des erstellten oder geänderten Datensatzes bei den anderen Endgeräten. Das automatische Bereitstellen des erstellten oder geänderten Datensatzes bei den anderen Endgeräten erfolgt dabei, indem der erstellte oder geänderte Datensatz mittels eines PUSH-Dienstes an die anderen Endgeräte übertragen und der erstellte oder geänderte Datensatz bei den anderen Endgeräten in die entsprechende lokale Anwendung transparent integriert wird. Die Erfindung betrifft des Weiteren ein mobiles Endgerät zur Durchführung eines solchen Verfahrens.
Service placement optimizes the number of instances of a service and their location within an ad hoc network in light of changes in service demand and network topology. The goal is to reduce bandwidth usage and latencies between clients and servers, while improving scalability and availability of the service. We propose the SPi service placement architecture which is unique in that it addresses the interdependencies between service placement, service discovery, and routing. We give an overview of the system, discuss the service model and algorithms with emphasis on the cost of synchronization between service instances, and present simulation-based results that illustrate the benefits of service placement when compared to traditional client/server architecture.
Christian Wartenburger, Georg Wittenburg, Norman Dziengel, Jochen Schiller - 2010
Download
Download
Abstract
Event detection is a major issue for applications of wireless sensor networks. In order to detect an event, a sensor network has to identify which application-specific incident has occurred based on the raw data gathered by individual sensor nodes. In this context, an event may be anything from a malfunction of monitored machinery to an intrusion into a restricted area. The goal is to provide high-accuracy event detection at minimal energy cost in order to maximize network lifetime. In this paper, we present a system for collaborative event detection directly on the sensor nodes. The system does not require a base station for centralized coordination or processing, and is fully trainable to recognize different classes of application-specific events. Communication overhead is reduced to a minimum by processing raw data directly on the sensor nodes and only reporting which events have been detected. The detection accuracy is evaluated using a 100-node sensor network deployed as a wireless alarm system on the fence of a real-world construction site
Thomas C. Schmidt, Matthias Wählisch - 2009
Download
Abstract
P2P networks enable end users to establish services relying neither on a dedicated infrastructure nor on an ISP deployment of enhanced services at the network layer. Regrettably, overlay traffic is not optimal with respect to native connections and peering agreements, but may decrease network quality at end users at increased transit costs. This issue has been addressed by traffic localization approaches: The general objective is to keep overlay traffic local and to minimize provider crossing. Current efforts foster provider-assisted solutions. Overlays, which approximate network paths from the underlay, promise to significantly limit inter-domain traffic. However, ISPs offering transits rejoice in additional traffic and may provide localization data to de-localize peers. In this paper, we argue that ISP interaction should be provided by neutral authorities, namely the Internet exchange points. We present an architecture which serves unstructured and structured overlay peers with a sloppy generic overlay ID that jointly reflects AS-paths and peering topologies, and is unbiased by unilateral ISP interests.
Philipp Schmidt, Bastian Blywis, Mesut Güneş, Felix Juraschek, Pardeep Kumar - 2009
Abstract
Routing is a general task, yet the implementation of routing protocols requires specific operating system related knowledge. The developer has to deal with particular kernel internals that might have severe side effects. This is especially true for reactive and hybrid protocols where routing and forwarding are heavily intermixed. Furthermore, novel routing protocols require features that are not provided by current operating systems or have to be customized. Thus routing protocols for mobile ad-hoc networks and wireless mesh networks are often studied in simulation environments. However, simulations have limitations that can result in conclusions that do not hold in real networks. A framework for the implementation of routing protocols in operating systems is required to enable real world oriented research. We introduce the DES Simple and Extensible Routing-Framework for Testbeds (DES-SERT). The framework supports the implementation and evaluation of routing protocols in a testbed environment. A structured protocol implementation is advocated by transmitting data in extensions attached to packets and by its pipeline based architecture. Several steps are elaborated how to derive an implementation from a routing protocol specification.
Thomas C. Schmidt, Matthias Wählisch - 2009
Download
Abstract
Hybrid multicast is regarded a promising technology to overcome the inter-domain deployment problem for group communication. Realistic performance estimators are difficult to obtain due to the diversity of overlay concepts and their complex dependence on the global Internet topology that withstands straightforward simulations or measurements. We contribute a simple analytic model for the expected delay distribution. Parametrized by realistic measurement values, this should serve as a first order estimator for quantifying the delay penalties in a global-scale hybrid multicast packet distribution. We diagnose a strong dependence on hop counts and proximity awareness for the overlay multicast approach in use with promising results for most efficient schemes.
Philipp Schmidt, Bastian Blywis, Mesut Güneş, Felix Juraschek - 2009
Abstract
Routing is infact a simple task, yet the implementation of routing protocols requires specific operating system related knowledge. The developer has to deal with particular kernel internals that might have severe side effects. Furthermore, novel routing protocols require features that are not provided by current operating systems. Thus, routing protocols for mobile ad-hoc networks and wireless mesh networks are often studied in simulation environments. However, simulations have limitations that can result in conclusions that do not hold in real networks. A framework for the implementation of routing protocols in operating systems is required to enable real world oriented research. We introduce the DES Simple and Extensible Routing-Framework for Testbeds (DES-SERT). It supports the implementation, experimentation, and evaluation of routing protocols in a testbed environment. A structured protocol implementation is advocated by transmitting data in extensions attached to packets and by its pipeline based architecture. Several steps are elaborated how to derive an implementation from a routing protocol specification. A gossip implementation is presented as a simple example.
Olaf Watteroth, Mesut Güneş, Felix Juraschek, Bastian Blywis - 2009
Abstract
This paper introduces DES-Cript, a domain specific language to define and describe networking experiments which can be executed on a testbed. DES-Cript aims at simplifying the execution of experiments by using a comprehensible and straight-forward design scheme and therefore assuring the repeatability of experiments. By laying out a clear logical structure, DES-Cript provides the user with the possibility to only focus on the crucial matter of experiments. A DES-Cript file serves as the input file for a software tool, which executes the experiment automatically, thus being able to use a reliable timing for all specified actions. DES-Cript is developed for the DES-Testbed at the Freie Universität Berlin, but is not limited to this particular testbed. It is designed as a general approach for experiment descriptions, useful for simulation environments as well as for other testbed networks.
Felix Juraschek, Heiko Will, Mesut Güneş, Jochen Schiller - 2009
Abstract
We intend to demonstrate our strategy Friendly Clustering which was designed and implemented for the MANIAC Challenge. The MANIAC Challenge is a competition to develop cooperation strategies for mobile ad-hoc networks (MANETs). MANETs usually comprise battery-powered nodes which may act selï¬shly by dropping packets destined to other network nodes in order to save energy. The detection of such behavior is important to exclude these nodes from the forwarding decision to minimize packet loss. For the demo, the strategy will compete against different variants of selï¬sh and uncooperative behavior in a MANIAC competition round carried out on the DES-Testbed at the Freie Universitaet Berlin. Friendly Clustering won the Performance Award in the MANIAC Challenge 2009.
Olaf Watteroth, Mesut Güneş, Bastian Blywis, Felix Juraschek - 2009
Download
Abstract
Scientifically sound network studies require the execution of large series of experiments. Researchers usually have to execute experiments manually, a labor-intensive and error-prone task, since there is no automation of the overall experimentation process. This task becomes especially hard on a distributed testbed and the researcher has to deal with additional challenges. In this paper we introduce the Distributed Embedded Systems Testbed Management System (DES-TBMS) for the automation of experimentation in wireless mesh networks and wireless sensor networks. DES-TBMS consists of several components including a domain specific language to describe and define experiments, a scheduler to manage and control experiments, a distributed monitoring system to gather system data, a visualization tool, and an evaluation tool to compute performance metrics from collected experiment data. Thus, DES-TBMS supports the researcher during the design, execution, and evaluation of experiments.
Christian Graff, Bastian Blywis, Felix Juraschek, Mesut Güneş - 2009
Abstract
Mobility is a core feature of future networks, e.g., wireless sensor, wireless mesh, and mobile ad-hoc networks. Thus the ability to generate accurate traces of mobile nodes is an important aspect for wireless network research. Many publications regarding wireless networks rely on simulations. The applied mobility models are often highly abstract and emulate human behavior poorly. Graph-based approaches try to restrict the area of movement to street-like structures yet they do not model real environments. MoNoTrac is a work-in-progress framework to create mobility traces based on real maps provided by emph{OpenStreetMaps}. Its plug-in architecture allows the usage of custom mobility models and provides simplified access for research in the domain of mobile networks.
Thomas C. Schmidt, Matthias Wählisch, Georg Wittenburg - 2009
Download
Abstract
A broadcast mode may augment peer-to-peer overlay networks with an efficient, scalable data replication function, but may also give rise to a virtual link layer in VPNtype solutions. We introduce a generic, simple broadcasting mechanism that operates in the prefix space of distributed hash tables without signaling. This paper concentrates on the performance analysis of the prefix flooding scheme. Starting from simple models of recursive k-ary trees, we analytically derive distributions of hop counts and the replication load. Further on, extensive simulation results are presented based on an implementation within the OverSim framework. Comparisons are drawn to Scribe, taken as a general reference model for group communication according to the shared, rendezvous-point-centered distribution paradigm. The prefix flooding scheme thereby confirmed its widely predictable performance and consistently outperformed Scribe in all metrics. Reverse path selection in overlays is identified as a major cause of performance degradation.
This paper proposes AREA-MAC, a medium access control (MAC) protocol for wireless sensor networks (WSNs). AREA-MAC reduces latency and energy consumption of nodes by using low power listening (LPL) with short preamble messages. Other protocols like B-MAC use long preamble messages that cause higher latency, energy consumption, and control overhead on nodes. AREA-MAC provides a reasonable trade-off between vital parameters, such as system fairness, throughput, scalability, and adaptability to traffic conditions. Additionally, to minimize network latency and maximize its lifetime, two application-based optimization problems are formulated. The gain of AREA-MAC in terms of delay and energy efficiency and an OMNeT++ based simulation framework are also discussed.
Internet services often use e-mail messages to interact with the user. On the other hand, nowadays many mail administrators employ Greylisting cite{Harris03}, i.e. artificial delays of incoming mail, in an attempt to combat spam. While the latter is a noble undertaking, it may also lead to unacceptable delays in the transmission of important mail traffic. Problems of this kind, where the user cannot control the characteristics of the system itself, can be addressed by restart of the failed (or delayed) action. In this extended abstract we consider the application of the restart method to reduce greylisting-related delays in transmitting mails. We perform a small case-study and identify open problems. Furthermore, we point out future research directions in the application of the restart method in service-oriented systems.
Johannes Semmler, Katinka Wolter, Philipp Reinecke - 2009
Abstract
Ensuring Quality of Service (QoS) in wireless networks poses an open problem in many application domains. We propose an automatic on-line QoS monitoring and management infrastructure that can be incorporated into existing network setups. Based on model-based assessment of current and future QoS conditions, our solution will control traffic in the network through a combination of admission control, enforced hand-over, traffic shaping and transmission parameter adjustments. Correctness of the model is evaluated through experimental evaluation and simulations. We implement a prototype of the proposed system using open-source components.
Levente Bodrog, Miklos Telek, Philipp Reinecke, Katinka Wolter - 2009
Abstract
Phase-type (PH) distributions are proven to be very powerful tools in modelling and analysis of a wide range of phenomena in computer systems. The use of these distributions in simulation studies requires efficient methods for generating PH-distributed random numbers. In this work, we consider the cost of PH-distributed random-number generation.
Application development for wireless sensor networks (WSN) demands for exper-tise in distributed as well as embedded programming. To ease the task of applica-tion development and make this area more accessible to non-experts, middleware abstractions are commonly employed. Middleware is defined as software which is located in between software applica-tions. Similar to operating systems, middleware systems provide applications with additional services to implement their functionality in a more abstract manner. Since devices forming a wireless sensor network have only little capabilities in terms of processing power and memory, their corresponding operating systems only provide very basic support for application development. At the same time various kinds of applications do have additional requirements to simplify their implementation. A multitude of middleware approaches are available to fill in this gap, thus pro-vide support for comfortable application development. We will discuss common application building blocks in this domain, discuss a selection of middleware ap-proaches available and provide an evaluation of their applicability by mapping application needs to middleware services.
An ad-hoc network is realized by mobile devices which communicate over radio. Since, experiments with real devices are very difficult simulation is used very often. Among many other important properties which have to be defined for simulative experiments the mobility model and the radio propagation model has to be selected carefully. Both have strong impact on the performance of mobile ad-hoc networks, e.g. the performance of routing protocols changes with these models. There are many mobility and radio propagation models proposed in literature. Each of them was developed with different intentions and is not suited for every scenario. In this chapter we introduce well known models for mobility and radio propagation and discuss their advantages, drawbacks and limitations in respect to the simulation of mobile ad-hoc networks.
Most of medium access control (MAC) protocols proposed for wireless sensor networks (WSN) are targeted only for single main objective, the energy efficiency. Other critical parameters such as low-latency, adaptivity to traffic conditions, scalability, system fairness, and bandwidth utilization are mostly overleaped or dealt as secondary objectives. The demand to address those issues increases with the growing interest in cheap, low-power, low-distance, and embedded WSNs. In this report, along with other vital parameters, we discuss suitability and limitations of different WSN MAC protocols for time critical and energy-efficient applications. As an example, we discuss the working of IEEE 802.15.4 in detail, explore its limitations, and derive efficient application-specific network parameter settings for time, energy, and bandwidth critical applications. Eventually, a new WSN MAC protocol Asynchronous Real-time Energy-efficient and Adaptive MAC (AREA-MAC) is proposed, which is intended to deal efficiently with time critical applications, and at the same time, to provide a better trade-offbetween other vital parameters, such as energy-efficiency, system fairness, throughput, scalability, and adaptivity to traffic conditions. On the other hand, two different optimization problems have been formulated using application-based traffic generating scenario to minimize network latency and maximize its lifetime.
Thomas C. Schmidt, Olaf Christ, Gabriel Hege, Matthias Wählisch - 2008
Download
Abstract
Mobility is considered a key technology of the next generation Internet and has been standardized within the IETF. Rapidly emerging multimedia group applications such as IPTV, MMORPGs and video conferencing increase the demand for mobile group communication, but a standard design of mobile multicast is still awaited. The open problem poses significant operational and security challenges to the Internet infrastructure. This paper introduces a protocol framework for authenticating multicast sources and securing their mobility handovers. Its contribution is twofold: At first, the current mobile multicast problem and solution spaces are summarized from the security perspective. At second, a solution to the mobile source authentication problem is presented that complies to IPv6 mobility signaling standards. Using an autonomous, one-way authentication based on cryptographically generated addresses, a common design is derived to jointly comply with the mobile any source and source specific multicast protocols that are currently proposed. This light-weight scheme smoothly extends the unicast enhanced route optimization for mobile IPv6 and adds only little overhead to multicast packets and protocol operations.
Note: Special issue Secure Multimedia Communication
Amine Kchiche, Farouk Kamoun, Sadeq Ali Makram, Mesut Güneş - 2008
Abstract
Communication and content sharing between cars has become one of the most interesting services for mobile users. Nevertheless, most efforts in this research field still is focused on direct opportunistic meetings between cars, which may result in poor performance. In this paper, we present a global architecture for vehicular ad hoc networks (VANET) that overcomes the intermittent nature of connectivity on VANETs and provides a consistent global peer-to-peer sharing service among participating vehicles. We propose a traffic aware organization of access points. Besides enabling the connection of distant vehicles and though extending the file-sharing environment, we study through simulation, the influence of incorporating vehicles traffic in the organization of access points on the performance of the sharing service.
Pierrick Seite (Ed.), Peny Yang (Ed.), Dirk {von Hugo}, Hitoshi Asaeda, Thomas C. Schmidt (Ed.), Suresh Krishnan, John Zhao, Hui Deng (Ed.), Matthias Wählisch - 2008
Download
Abd Al Basset Almamou, Intesab Hussain, Pardeep Kumar, Mesut Güneş - 2008
Abstract
The recently standardized IEEE 802.15.4 protocol is emerging as a basic building block for wireless sensor network applications by providing many appealing features, such as low power, low cost, low data-rate, and simplicity. However, when treating applications with strict timeliness, energy, and bandwidth requirements several problems arise, triggering the need of a solution for an efficient twinning of IEEE 802.15.4 and wireless sensor networks. In this paper, after discussing the working of IEEE 802.15.4, we point out its limitations for such applications and propose a solution to overcome those limitations. We also derive an efficient IEEE 802.15.4-based network parameters setting to improve the network performance in terms of latency, energy, and bandwidth utilization.
Abd Al Basset Almamou, Pardeep Kumar, Mesut Güneş, Jochen Schiller - 2008
Abstract
The recently standardized IEEE 802.15.4 protocol provides many appealing features, such as low power, low cost, low data-rate, simplicity, and timeliness guarantee for delay-bound, energy, and bandwidth critical applications. However, when treating applications with such requirements as in the medical field, several problems arise, triggering the need of an efficient enhancement of the IEEE 802.15.4 protocol including the fine tuning of its parameters. In this paper, we discuss the suitability of the protocol and explore its limitations for health care applications. We propose a solution to overcome those limitations and derive the most efficient IEEE 802.15.4-based network parameters setting for urgent medical services.
Adam Dunkels, Michael Baar, Heiko Will, Jochen Schiller - 2008
Download
Abstract
Sensor network hardware designs consist of a central micro-controller, to which sensors and communication peripherals are connected. Resource arbitration and concurrency man-agement must be implemented in software. Existing hardware arbitration mechanisms use explicit locking to protect against resource conflicts. Explicit locking may lead to deadlock, which must be avoided for long-term sensor network deploy-ments. We present a power-saving resource arbitration archi-tecture that is deadlock-free, portable, and resource-efficient. The architecture explicitly manages inter-device dependencies to know what devices to power down.
Michael Baar, Heiko Will, Bastian Blywis, Thomas Hillebrandt, Achim Liers, Georg Wittenburg, Jochen Schiller - 2008
Download
Download
Abstract
Wireless sensor networks (WSN) are in a transition from research to real world applications. Robust and efficient hardware platforms are needed. These have to offer sufficient processing power and memory while retaining energy efficiency. In this technical report we present the newest ScatterWeb hardware platform that fits the needs of research and prototyping applications of the near future.
Efficient group communication within the Internet has been implemented by multicast. Unfortunately, its global deployment is missing. Nevertheless, emerging and progressively establishing popular applications, like IPTV or large-scale social video chats, require an economical data distribution throughout the Internet. To overcome the limitations of multicast deployment, we introduce and analyze BIDIR-SAM, the first structured overlay multicast scheme based on bi-directional shared prefix trees. BIDIR-SAM admits predictable costs growing logarithmically with increasing group size. We also present a broadcast approach for DHT-enabled P2P networks. Both schemes are integrated in a standard compliant hybrid group communication architecture, bridging the gap between overlay and underlay as well as between inter- and intra-domain multicast.
Abd al basset Almamou, Houda Labiod, Jochen Schiller, Mesut Güneş - 2008
Abstract
Wireless sensor nodes can organize themselves after deployment in a large scale ad hoc network topology. Some traditional Ad hoc routing protocols do not take into account that a sensor node has limited capacities, e.g. energy. Therefore, they try to perform the sensing task regardless of the maximum hops and duration it take. In wireless sensor networks (WSNs), this routing solution is unrealistic, so we derive from the Distributed Hash Table (DHT) an overlay routing for WSNs called ScatterDHT as a routing solution, develop a prototype implementation and apply it over ScatterWeb sensor nodes. This paper describes the ScatterDHT overlay routing protocol and figures out the performance evaluation of the implemented prototype over real-world sensor nodes regarding to two metrics (a) memory utilization and (b) power consumption.
Chunlei An, Koojana Kuladinithi, Andreas Timm-Giel, Carmelita Görg, Qasim Mushtaq - 2008
Abstract
Routing protocols designed for ad hoc networks enable a node to communicate over a long distance through multi-hops. The available routing protocols always strive to make multi-hop route between the source and the destination through the lowest possible hop count. However, the lower hop count cannot always guarantee the higher throughput if the link quality among the nodes is degrading. In such situations, routes with higher hop count may give better throughput. In this work, standard ad hoc routing has been improved by discovering the paths with better link quality to provide better application performance. The mostly used reactive protocol of AODV ( Ad hoc On-Demand Distance Vector) routing is modified to incorporate link quality when discovering routes as a deciding parameter in addition to a lower hop count. The Quality of Service (QoS) requirements are considered in order to satisfy the user requirements in terms of throughput, packet loss rate, end-to-end delay etc. The proposed protocol, called as QoS AODV (QAODV), is implemented and tested for performance against AODV in a real test-bed setup with mobility. Experiments show improved performance compared to original AODV in terms of throughput and delay.
Mesut Güneş, Bastian Blywis, Felix Juraschek - 2008
Download
Abstract
Wireless mesh networks are an emerging and versatile communication technology. The most common application of these networks is to provide access of any number of users to the world wide Internet. They can be set up by Internet service providers or even individuals joined in communities. Due to the wireless medium that is shared by all participants, effects like short-time fading, or the multi-hop property of the network topology many issues are still in the focus of research. Testbeds are a powerful tool to study wireless mesh networks as close as possible to real world application scenarios. In this technical report we describe the design, architecture, and implementation of our work-in-progress wireless testbed at emph{Freie Universität Berlin} consisting of 100 mesh routers that span multiple buildings. The testbed is hybrid as it combines wireless mesh network routers with a wireless sensor network.
Philipp Schmidt, Mesut Güneş, Bastian Blywis, Felix Juraschek - 2008
Download
Abstract
Testbeds are a powerful tool to study wireless mesh and sensor networks as close as possible to real world application scenarios. In contrast to simulation or analytical approaches these installations face various kinds of environment parameters. Challenges related to the shared physical medium, operating system, and used hardware components do arise. In this technical report about the work-in-progress emph{Distributed Embedded Systems} testbed of 100 routers deployed at the emph{Freie Universität Berlin} we focus on the software architecture and give an introduction to the network protocol stack of the Linux kernel. Furthermore, we discuss our first experiences with a pilot network setup, the encountered problems and the achieved solutions. This writing continues our first publication and builds upon the discussed overall testbed architecture, our experiment methodology, and aspired research objectives.
Efficient group communication within the Internet has been implemented by multicast. Unfortunately, its global deployment is missing. Nevertheless, emerging and progressively establishing popular applications, like IPTV or large-scale social video chats, require an economical data distribution throughout the Internet. To overcome the limitations of multicast deployment, we introduce and analyze BIDIR-SAM, the first structured overlay multicast scheme based on bi-directional shared prefix trees. BIDIR-SAM admits predictable costs growing logarithmically with increasing group size. We also present a broadcast approach for DHT-enabled P2P networks. Both schemes are integrated in a standard compliant hybrid group communication architecture, bridging the gap between overlay and underlay as well as between inter- and intra-domain multicast.
The research area of wireless sensor networks is in a mature phase. Many fundamental problems have been identified and various approaches proposed as solutions. However, there still is a lack of long-term studies about performance, reliability, and maintenance of wireless sensor networks. Industrial style products used for safety critical application scenarios need extended and reproducible test runs. Simulations alone cannot suffice the evaluation requirements for real world deployments in health care, disaster management, or fire detection. In this paper we address the lack of long-term experiments of large-scale wireless sensor networks. We introduce our work in progress testbed for such trials in reference to current and future projects of our research group.
Thomas C. Schmidt, Waldemar Spät, Matthias Wählisch - 2008
Download
Abstract
The Internet topology has evolved over the past decades in an evolutionary process and continues to grow. Recently, it has attracted much attention from the networking and physics communities, as it forms a unique operational instance of a planetary-scale network environment. Several measurement projects observing the Internet have been undertaken over the past years, out of which Skitter and Dimes have established as continuous recordings of the vivid process of network formation. In this paper we compare Internet measurement data obtained from Skitter and Dimes by analyzing the Internet evolution, its degree distributions and correlations at IP node level. This comparative analysis was enabled by a data conversion and processing tool-chain implemented as an extension to the BRITE topology generator which we introduce, as well. Our results show significant differences in higher nodal degrees. Correlation analysis indicates that DIMES scans discover Internet links to a fairly uniform degree, while parts remain invisible within Skitter data. Mid-range, oscillating autocorrelations are discovered as a signature of memory effects in Internet topology. Further on we analyze implications of the Internet structure as attained in both, its core and edge vicinities, for mobility management and mobile multicast routing performance.
The utilization of domain-specific programming abstractions for wireless sensor actor networks can greatly ease applica- tion development. A high level of abstraction from under- lying system complexity fosters fast prototyping and less erroneous source code. However, especially in this domain access to low-level parameters can be mandatory to allow for sophisticated application fine-tuning. In this paper, we present a layered system access interface we developed for our middleware platform FACTS to address this challenge. Context-based configuration and control of system-level resources is facilitated by relying on lean ad- ditions to the FACTS framework. These will be discussed with the help of a common use case to illustrate the simple, yet powerful abstraction provided.
The aggregate capacity of wireless mesh networks can be improved significantly by equipping each node with multiple interfaces and by using multiple channels in order to reduce the effect of interference. Since the number of available channels is limited, it is desired to allocate and reallocate channels on-demand. In this paper, a Cluster Channel Assignment (CCA) approach is proposed, to maximize the aggregate throughput by exploiting spatial reuse and local dynamic switching of the channels. A clustering approach is employed in order to maximize the network capacity while minimizing the interference and taking advantage of the possibility of reuse of channels among clusters.
Within the past decade, wireless technologies have become a crucial part of our day to day life, generally allow-ing us to overcome inconveniences caused by spatial dependencies of services or devices. Especially in the do-main of health monitoring, patient supervision and care for the elderly the possibility to physically decouple health parameter acquisition from data analysis by substituting wires with wireless links has a great impact on the quality of life of patients. This paper targets to supply an overview of available wireless technologies that may be used in a health monitor-ing context. Since this domain includes a rich set of different applications, essential parameters shared by a ma-jority of current or envisioned applications are pinpointed and used as a basis for an evaluation of the decision on the usage of a particular wireless standard.
Service placement deals with the problem of selecting which node in a network is most suitable for hosting a service that responds to queries from other nodes. Optimally placing services reduces network traffic and improves connectivity between clients and servers. Service placement algorithms may thus be regarded as an interesting building block for research into service-oriented middleware. Recently, new approaches to address the service placement problem in the field of ad-hoc networking have been proposed. This paper surveys, classifies and evaluates ten representative approaches, thereby providing a summary of the state of the art in service placement.
Die verteilte Anordnung von Sensorknoten in einem drahtlosen Sensornetz ermöglicht es, Eigenschaften und Veränderungen der Umwelt an unterschiedlichen Stellen zu beobachten. Durch ein Zusammenwirken der Sensorknoten können in der Umwelt auftretende Ereignisse verteilt erkannt werden. Es existiert ein verteiltes Erkennungssystem, welches Ereignisdaten mehrerer Sensorknoten im Sensornetz fusioniert und anhand einer lokal durchgeführten Mustererkennung klassifiziert. Dieses System wurde unter Laborbedingungen entwickelt und getestet. Mit dieser Arbeit wird eine Weiterentwicklung des verteilen Erkennungssystems vorgestellt, welches den Anforderungen einer praktischen Anwendbarkeit gerecht wird. Während des Mustertrainings können Ereignisdaten von einer dynamischen Anzahl von Sensorknoten verarbeitet werden. Die Daten werden dabei auf die für die Erkennung geeignetsten und aussagekräftigsten Werte reduziert. Die verteilte Ereigniserkennung ist nicht auf den Ort des Trainings beschränkt. Ein unbekanntes Ereignis kann an jedem Punkt des Sensornetzes von den das Ereignis wahrnehmenden Sensorknoten unter Austausch von Merkmalswerten klassifiziert werden. Ort und Typ des Ereignisses werden auf einer Basisstation angezeigt. Das in der vorliegenden Arbeit entwickelte System wird anhand eines praktischen Anwendungsfalls getestet. Dabei handelt es sich um die Erkennung sicherheitsrelevanter Ereignisse an einem Bauzaun. Die Evaluation erfolgt für die Erkennung der Ereignisse am Ort gleich und ungleich des Trainings. Es wird untersucht, wie sich die Änderung des Ortes auf die Erkennungsraten auswirkt und wie viel Einfluss die Bedingungen der realen Welt auf die Ergebnisse im Vergleich zu denen unter Laborbedingungen besitzen. Am Ort des Trainings wird eine Korrektklassifikationsrate von 74,8% erzielt. Im Vergleich zum Projekt Fence Monitoring, welches sich ebenfalls mit der verteilten Erkennung von Ereignissen an Bauzäunen beschäftigt, wird eine Steigerung von 15,9 Prozentpunkten erzielt. In Bezug auf das dieser Arbeit zugrunde liegenden Erkennungssystem, dessen Ergebnisse durch Versuche unter Laborbedingungen entstanden, ist ein Rückgang von 21,5 Prozentpunkten zu verzeichnen. Für den Fall der Erkennung am Ort ungleich des Trainings ergibt sich eine Korrektklassifikationsrate von 57,7%.
Norman Dziengel, Georg Wittenburg, Jochen Schiller - 2008
Download
Abstract
Distributed event detection in wireless sensor networks (WSNs) is the process of observing and evaluating an event using multiple sensor nodes without the help of a base station or other means of central coordination and processing. Current approaches to event detection in WSNs transmit raw data to an external entity for evaluation or rely on simplistic pattern recognition schemes. This implies either high communication overhead or low event detection accuracy, especially for complex events. In this paper, we present our currently on-going work on a system for distributed event detection that particularly suits the specific characteristics of WSNs. Adapting traditional pattern recognition algorithms to highly embedded devices, it uses the distributed sampling of sensor nodes to optimize the accuracy of the event detection process. Four different algorithms for distributing, classifying and fusing ``fingerprints'' of the raw data sampled on each sensor are proposed and quantitatively evaluated in a small-scale experiment.
In this paper we introduce our work-in-progress wireless testbed. The testbed consists of hybrid nodes. Each node contains a sensor node and a mesh router. We present concepts which exploit the features of an existing wireless mesh network for the wireless sensor network. The concepts include the emph{wireless sensor testbed integration} (WSTI) which simplifies the management and provides a basic infrastructure for the sensor network and experiments. The integration into the Linux kernel is realized by the emph{wireless sensor kernel interface} (WSKI). emph{Network check points} (NCPs) enable long-term experiments that may span several weeks although they cannot be done in one continuous session.
Thomas C. Schmidt, Hans L. Cycon, Mark Palkow, Matthias Wählisch - 2008
Abstract
Mobile Geräte und multimediale Kommunikation bilden seit einigen Jahren einen der wesentlichen Innovations- und Wachstumsmotoren überhaupt. Gruppenkommunikation auf mobilen Geräten, etwa in Sprach-- und Videokonferenzen oder komplexen kollaborativen Umgebungen, wächst hierbei an Bedeutung, als das Internet skalierbare, dynamische adaptierende Dienste zur Kommunikation in einer Gruppe anbietet, welche die in ihren Kapazitäten eingeschränkten mobilen Endgeräte wesentlich unterstützen. Das Projekt MOVIECAST bündelt die auf eine Gesamtlösung mobiler, multimedialer Gruppenkommunikation abgestimmten Entwicklungskomponenten. In diesem Beitrag stellen wir zunächst das gerade begonnene Forschungsprojekt vor und präsentieren erste Lösungskonzepte. Wir führen eine Erweiterung von SIP ein, geeignet für die Aushandlung von Gruppensitzungen basierend auf Source Specific Multicast. Die vorgeschlagene SIP--Protollspezifikation und --semantik kommt ohne das Hinzufügen neuer Methoden aus. Als Implementierungsreferenz werden wir eine Multimedia--Kommunikationssoftware mit einer verteilten, leichtgewichtigen Architektur vorstellen. Weiterhin diskutieren wir ein für mobiles SSM geeignetes Routing--Protokoll und evaluieren dieses auf der Basis realer Internettopologien.
Note: Contribution WCI 2006, on invitation
In order to avoid transmitting raw data to a base station, sensor nodes are trained to cooperatively recognize deployment-specific events based on the data sampled by their sensors. As both training and event detection are performed without the need for central coordination or processing, only information about the detected event needs to be reported.
Note: (engl.: Specification and implementation of a modular and visual development environment with embedded firmware for the wireless sensor node MSB430)
Otto Spaniol, Martin Wenig, Alexander Zimmermann, Sadeq Ali Makram, Ulrich Meis, Mesut Güneş - 2007
Abstract
Drahtlose multi-hop Mesh-Netzwerke bestehen aus einer drahtlosen Infrastruktur. Dadurch erlauben sie eine günstige Abdeckung großer Flächen und bieten gleichzeitig eine hohe Flexibilität. In dieser Arbeit werden die Architektur und die Methodik beschrieben, die im Rahmen des UMIC-Mesh-Testbeds an der RWTH Aachen verfolgt werden. Auf der Basis erster Experimente werden die Vorzüge, M"oglichkeiten und Grenzen von Mesh-Netzwerken diskutiert.
Martin Wenig, Alexander Zimmermann, Mesut Güneş - 2007
Abstract
Recent research has shown the poor accuracy of widely used simulators in the area of wireless networks. Beside many other parameters two are of particular interest: i) the mobility model, and ii) the radio wave propagation model. The first is responsible for the network topology and the latter for the perceiption of transmitted data. Both have strong impact on the performance of mobile ad-hoc networks, e.g. the performance of routing protocols changes with these models. We developed a framework combining a realistic mobility model and radio wave propagation model. To generate mobility multiple well understood random mobility models are combined in a scenario graph. The graph also includes obstacles which restrict the movement and the radio wave propagation. The radio wave propagation is calculated using a ray-tracing approach. We present an extensive simulation study on the effect of the mobility and radio wave propagation on simulation results.
Autonomous networks of small embedded computers equipped with modern, miniaturized sensors and wireless communication system open new ways for informationâ€gathering. Placed in numbers of hundreds or thousands they collect data and delivered it to their owner. Use case are manifold. They cover extensive, yet detailed, mesh networks for environment studies as well as monitoring of large droves or body functions of a single person. Digital trunked radio is on it’s way to replace existing analogue professional mobile radio systems. New features like direct†and groupâ€calls as well dataâ€services become possible. The TETRA digital trunked radio system has been standardised by the European Telecommunication Standards Institute (ETSI). Unlike existing digital radio solutions TETRA terminals and infrastructure of different manufacturers can be used in combination. A large group of potential users in the commercial sector are industry and conveyances. The largest and most demanding user group are the public authorities. The integration of sensor networks into digital trunked radio enables new sensornetworkâ€based applications for these (potential) users. The common task of all these applications is the collection of data in a distant environment. In this master thesis a number of possible use cases for sensor networks in the sphere of trunked radio users are presented. Besides buildingâ€automation and industrial usage the main case is the use in security critical facilities and the use by public relief services. Main applications are early hazard detection and support for disaster site exploration. After an introduction of the technical basics the core requirements for this kind of application are analysed. In combination with the technical restrictions a system model is developed and different ways of realisation are discussed. The abstract application is then refined into a prototype which shows the feasibility of a configurable system for information transport between a sensor network and an operator station connected to a digital trunked radio network. For the prototype the ScatterWeb sensor network of Freie Universität Berlin and the TETRA trunked radio implementation by Motorola GmbH are used. To connect both networks a gateway is designed and implemented. For the overall application a concept for configurable information transport in sensor networks is designed and realised. The main concepts and structures are explained along diagrams.
Alexander Zimmermann, Martin Wenig, Ulrich Meis, Jan Ritzerfeld, Mesut Güneş - 2007
Abstract
Simulation is the most famous way to study wireless an mobile networks since they offer a convenient combination of flexibility and controllability. However, their largest disadvantage is that the gained results are difficult to transfer into reality since not only the abstraction of the upper network layer are typically high, but also the environment of mobile and wireless networks is very complex. This is due to two reasons. First there are typically many simplifications in the models of the upper networking layers, and second the environment of mobile and wireless networks is in particular complicated and thus difficult to be considered in all details. In this paper we introduce UMIC-Mesh, a hybrid testbed approach, that consists of real mesh nodes and a virtualization environment. On the one hand the virtualization allows the development and testing of software as if it was executed on real mesh routers, but in a more repeatable and controllable way. On the other hand the results and conclusions gained by a software evaluation in the testbed can be easily transferred into reality, since the testbed represents a high degree of realism.
With the emergence of wireless sensor networks, the issues of event recognition and processing have been partially shifted into the embedded domain. New processing capabilities on small devices allow for physically close event monitoring and fast filtering without having to set up a wired infrastructure beforehand. This opportunity for flexible deployments, local data storage and demand-driven event forwarding opens up new application areas for event-centric architectures. However, the convenience of localized event processing comes at a cost, such as sparse resources or medium contention when relying on wireless communication. Several parameters have to be evaluated to decide whether pushing the application logic into a sensor network is worthwhile, or whether a conventional server-centered deployment is to be preferred. In this paper, we discuss parameters influencing an architectural decision and their interdependencies, illustrate our contribution with the help of an example and provide a generic cost model for estimating this decision.
Martin Wenig, Alexander Zimmermann, Mesut Güneş - 2007
Abstract
Two main steps on the way to more realistic simulations of mobile ad-hoc networks are the introduction of realistic mobility and sophisticated radio wave propagation models. Both have strong impact on the performance of mobile ad-hoc networks, e.g. the performance of routing protocols changes with these models. In this paper we introduce a framework which combines realistic mobility and radio wave propagation models. Our approach consists of a zone-based mobility generator and a high accuracy radio wave propagation model. For the mobility generation a wide variety of well understood random mobility models is combined with a graph based zone model, where each zone has its own mobility model. To achieve a realistic radio wave propagation model a ray tracing approach is used. The integration of these two techniques allows to create simulation setups that closely model reality.
Thomas C. Schmidt, Ying Zhang, Matthias Wählisch - 2007
Download
Abstract
Significant effort has been invested recently to accelerate handover operations in a next generation mobile Internet. Corresponding works for developing efficient mobile multicast management are emergent. Both problems simultaneously expose routing complexity between subsequent points of attachment as a characteristic parameter for handover performance in access networks. As continuous mobility handovers necessarily occur between access routers located in geographic vicinity, this paper investigates on the hypothesis that geographically adjacent edge networks attain a reduced network distances as compared to arbitrary Internet nodes. We therefore evaluate and analyze edge distance distributions in various regions for clustered IP ranges on their geographic location such as a city. We use traceroute to collect packet forwarding path and round-trip-time of each intermediate node to scan-wise derive an upper bound of the node distances. Results of different scanning origins are compared to obtain the best estimation of network distance of each pair. Our results are compared with corresponding analysis of CAIDA Skitter data, overall leading to fairly stable, reproducible edge distance distributions. First conclusions on expected impact on handover performance measures are drawn.
Research in wireless sensor networks needs a robust hardware platform that allows access to hardware parameters to achieve optimum solutions. In this poster we present the new ScatterWeb MSB hardware platform which we believe soundly fits the needs of research and prototyping applications of the near future.
Drahtlose Sensornetze haben aufgrund ihrer verteilten Struktur die Möglichkeit, differenziert Informationen an unterschiedlichen Stellen der Umwelt zu erfassen. Die Erkennung von Ereignissen, die in ihrer Umgebung mehrere Parameter und Messpunkte verschiedentlich beeinflussen, wird mittels drahtlosen Sensornetzen möglich. Das vorgestellte System nutzt die Vorteile drahtloser Sensornetze, indem mehrere Sensorknoten das Problem der verteilten Ereigniserkennung gemeinsam lösen. Mit dieser Arbeit wird ein Erkennungssystem vorgestellt, das verteilt zu betrachtende Ereignisdaten auf Basis einer lokal auf den Sensorknoten durchgeführten Mustererkennung im Netz fusioniert und klassifiziert. Eine dynamische Kalibrierung des Beschleunigungssensors und ein Mustertraining mit den Sensorknoten ermöglichen eine nachfolgende differenzierte Erkennung der gelernten Bewegungsmuster. Die erfassten Beschleunigungsdaten werden lokal auf den jeweiligen Sensorknoten komprimiert und mit den Daten benachbarter Sensorknoten fusioniert, um im Bezug auf das globale Ereignis ausgewertet zu werden. Verschiedene Fusionstechniken auf Ebene der Klassifizierung sowie der Mustermerkmale werden getrennt und miteinander verknüpft untersucht und bewertet. Ein qualitativer Vergleich zwischen der hier erarbeiteten lokalen und verteilten Ereigniserkennung sowie einem bestehenden verteilten Erkennungsansatz wird gezogen. Die lokale Ereigniserkennung erreicht eine Korrektklassifikationsrate von 89,4 %. Basierend auf der Komponente der lokalen Erkennung erzielt das verteilte Erkennungssystem in den untersuchten Fusionsmethoden eine Korrektklassifikationsrate von 93,8 % bis 96,3 %. Die Erkennungsratensteigerungen werden auf den Einsatz der verteilten Erkennungsstrategie zurückgeführt. Die energietechnisch kostenintensivste Operation in drahtlosen Sensornetzen ist das Versenden von Datenpaketen. Aufgrund dieser Problematik wird die Rentabilität der Datenübertragung innerhalb des Sensornetzes in Bezug auf die erbrachte Korrektklassifikationsrate in den Fusionsmethoden analysiert. Au{ss}erdem werden während der lokalen und verteilten Mustererkennung die Einflussstärken der Probanden auf die ermittelten Kennwerte untersucht und evaluiert.
Various types of multi-hop wireless networks have been subject to research for more than two decades. Famous representatives are mobile ad-hoc networks, wireless sensor networks, and wireless mesh networks. These networks have been studied mainly based on simulations, but there is currently a trend towards testbed-based research, since some characteristics of these networks are hard to model in simulation environments. In this paper we review architectures for next generation wireless networks including the aforementioned ones according to future applications. We introduce the DES-Testbed at the emph{Freie Universität Berlin}, which we believe is suitable to study these architectures or emph{network configurations}, as we call them. Furthermore, we discuss an approach on how to set-up network configurations and perform long-term experiments.
Energy efficient, scalable and reliable data dissemination is of critical importance in wireless sensor networks. Reliability comes at the cost of control overhead, which in such networks directly translates into additional energy consumption. This document analyzes existing data dissemination solutions under the premise of increased reliability requirements. The analysis targets a) dissemination of code updates, an application with increased reliability requirements and b) reliable data dissemination in general.
lexander Zimmermann, Daniel Schaffrath, Michael Faber, Martin Wenig, Mesut Güneş - 2007
Abstract
The TCP was designed with fixed, wired networks in mind. As a result TCP performs suboptimal in networks with noisy links and changing paths, e.g., wireless multi-hop networks. The main reason is that TCP assumes packet loss indicates congestion. However, such fluttering networks drop a non negligible amount of packets because of corruption, route failures and disconnections. In this paper we introduce and evaluate the ECRFN algorithm. ECRFN improves TCP performance in such environments. It treats all possible loss types and features a smart timing scheme in case of path disconnections. It employs an optional router enhancement and exploits common router messages.
Thomas C. Schmidt, Matthias Wählisch - 2007
Download
Abstract
Multicast communication services are among the longest debated areas in the 30 years history of the Internet. Innumerous solutions and controversies rank around the IP host group model and led to a strongly divergent state of deployment. Stimulated by application needs alternative multicast mechanisms have been developed. P2P technologies enabled group distributions on the application or service middleware layer. A significantly simplified routing approach gave rise to the lean source specific multicast in IP. Henceforth the debate elaborated to which approach proves suitable for providing the best benefit of a scalable, efficient and deployable group communication service. This paper discusses problems, requirements and current trends for deploying group communication in real-world scenarios from an integrative perspective. We introduce Hybrid Shared Tree, a new architecture and routing approach to combine network and subnetwork layer multicast services in end system domains with transparent, structured overlays on the inter-domain level. This hybrid solution is highly scalable, robust and offers provider-oriented features to stimulate deployment. Furthermore, a straight forward perspective is identified towards a mobility agnostic routing layer in future use.
Note: Selected papers from the TERENA networking conference 2007
Sadeq Ali Makram, Martin Wenig, Alexander Zimmermann, Mesut Güneş - 2007
Abstract
One of the most promising wireless technologies are Wireless Mesh Networks (WMN). The aggregate capacity of wireless mesh networks can be improved significantly be equipping each node with several Wireless Network Interfaces (WNICs) and by using multiple channels in order to minimize the interference and to provide high performance in such networks. To achieve these benefits it requires a good channel assignment planning. The channels have to be assigned in such a way, that interference is decreased and performance is increased at the same time. Since the number of available channels is limited, it is desired to allocate and reallocate channels dynamically on-demand. In this paper, a dynamic channel assignment, is proposed to the aforementioned problem which is adaptive to the load in the wireless mesh network. The algorithm add or select a channel for those heavy loaded nodes, based on the local informations of the neighbor nodes. The selected or the added channel that minimizes the interference and insure the network connectivity. The simulation results show that our algorithm improve the network capacity by approximately 50% in comparison to a static channel assignment.
Wireless mesh networks (WMNs) are an emerging technology which provides broadband wireless access services. One of the challenges that still faces high performance WMNs is the capacity reduction due to interference of wireless links. In this paper, we address the problem of assigning channels to nodes in WMNs. For this, we introduce the Cluster Channel Assignment (CCA) approach with the objective of reducing network interference to increase the overall network performance. A clustering approach is employed in order to reduce the complexity of channel assignment into local problems within clusters and taking the advantage of the possibility to reuse the channels in different clusters.
In the field of wireless sensor networks, network simulators are commonly used to evaluate properties of software components or the network as a whole. Their advantages in reduced experimental overhead, flexibility, and repeatability come at the expense of questionable credibility of the results. In order to quantify the simulation accuracy of wireless sensor networks, we have conducted a field test measuring the packet loss rate and compared the data with the results obtained from a carefully configured simulation of the same scenario. Our evaluation gives insight into how much trust can be put into the results of simulations of comparable scenarios.
A wireless sensor network consists of a large quantity of small, low-cost sensor nodes that are limited in terms of memory, available energy and processing capacity. Generally, these sensor nodes are distributed in space to obtain physical parameters such as temperature, humidity, vibration or light conditions, and transmit the measured values to a central entity. The measurements are tagged with the corresponding location of the nodes in the network and the time of sampling, to enable a view on the value distribution in space and time later on. Positioning of wireless sensor nodes without dedicated hardware is an open research question. Especially in the domain of embedded networked sensors, many applications rely on spatial information to relate collected data to the location of its origin. As a first step towards localization, an estimation of the distance between two nodes is often carried out to determine their positions. So far, the majority of approaches therefore explore physical properties of radio signals such as the strength of a received signal or its trip time. However, this is problematic since either the complexity on the software or on the hardware side is not adequate for embedded systems, or the approaches lack the required accuracy. In this paper we present the WDNI algorithm (Weighted Density of Node Intersection) to determine the distance between two nodes, relying solely on the investigation of local node densities. To evaluate the accuracy of this algorithm, we ran extensive simulations and experimented with different testbed setups using real sensor nodes, and finally compared WDNI to a range-free distance estimation algorithm based the analysis of RSSI values.
A mobile ad-hoc network (MANET) is a collection of mobile nodes which communicate over radio. These kind of networks are very flexible, thus they do not require any existing infrastructure or central administration. Therefore, mobile ad-hoc networks are suitable for temporary communication links. The biggest challenge in this kind of networks is to find a path between the communication end points, what is aggravated through the node mobility. In this paper we present a new on-demand routing algorithm for mobile, multi-hop ad-hoc networks. The protocol is based on swarm intelligence and especially on the ant colony based meta heuristic. These approaches try to map the solution capability of swarms to mathematical and engineering problems. The introduced routing protocol is highly adaptive, efficient and scalable. The main goal in the design of the protocol was to reduce the overhead for routing. We refer to the protocol as the Ant-Colony-Based Routing Algorithm (ARA)
Wireless mesh networks are a special kind of ad hoc networks in which most nodes are static. The application purpose of wireless mesh networks is also different than that of ad hoc networks and is focused on broadband access services to the Internet. Recent studies have shown that nodes in a wireless mesh network have to be equipped with several radio interfaces for high performance. However, one of the challenges that still faces high performance wireless mesh networks is the capacity reduction due to interference of wireless links. In this paper, we address the problem of assigning channels to nodes in wireless mesh networks. For this, we propose and study the Cluster Channel Assignment (CCA) approach with the goal of reducing the network interference to increase the overall network performance.
Sadeq Ali Makram, Martin Wenig, Alexander Zimmermann, Ulrich Meis, Mesut Güneş - 2007
Abstract
Wireless Mesh Networks (WMN) promise to provide high bandwidth and access to the Internet. Recent studies have shown that multi-channel WMN supported with several Wireless Network Interfaces (WNIC) per node can provide high performance. To achieve these benefits it requires a good channel assignment planning. Channel assignment is an important factor to support QoS and load balancing in the WMN. The channels have to be assigned according to the distribution of traffic load in the WMN. Since the number of available channels is limited, it is desired to allocate and reallocate channels dynamically on-demand. This paper proposes a dynamic channel assignment approach to balance the load aiming to maximize the throughput and utilize the channels efficiently.
Georg Wittenburg, Kirsten Terfloth, Freddy Lopez Villafuerte, Tomasz Naumowicz, Hartmut Ritter, Jochen Schiller - 2007
Download
Download
Abstract
In-network data processing and event detection on resource-constrained devices are widely regarded as distinctive and novel features of wireless sensor networks. The vision is that through cooperation of many sensor nodes the accuracy of event detection can be greatly improved. On the practical side however, little real-world experience exists in how far these goals can be achieved.In this paper, we present the results of a small deployment of sensor nodes attached to a fence with the goal of collaboratively detecting and reporting security relevant incidents, such as a person climbing over the fence. Based on experimental data we discuss in detail the process of in-network event detection both from the conceptual side and by evaluating the results obtained. Reusing the same traces in a simulated network, we also look into the impact of multi-hop event reporting.
Alexander Zimmermann, Sadeq Ali Makram, Ulrich Meis, Michael Faber, Mesut Güneş - 2007
Abstract
WMN are supposed to provide flexible and high-performance wireless network access for large areas. However, these claims are mostly substantiated by simulation studies only as real testbeds are inflexible and associated with high maintenance costs. In this work we present a hybrid, i.e., partly real and partly virtualized, WMN testbed. This provides a high degree of realism while still allowing the flexibility known from simulations. In addition to the architectural discussion we present measurement results from our testbed highlighting the optimisation potential of small protocol parameter changes.
Thomas C. Schmidt, Matthias Wählisch - 2006
Download
Abstract
Source Specific Multicast (SSM) promises a wider dissemination of group distribution services than Any Source Multicast, as it relies on simpler routing strategies with reduced demands on the infrastructure. However, SSM is designed for 'a priori known and changeless addresses of multicast sources and thus withstands any easy extension to mobility. Up until now only few approaches arose from the Internet research community, leaving SSM source mobility as a major open problem. The purpose of this paper is twofold. At first we analyze characteristic properties of multicast shortest path trees evolving under source mobility. Analytically and by stochastic simulations we derive measures on the complexity of SSM routing under source mobility. At second we introduce a straightforward extension to multicast routing for transforming (morphing) source specific delivery trees into optimal trees rooted at a relocated source. All packet forwarding is done free of tunneling. Multicast service disruption and signaling overhead for the algorithms remain close to minimal. Further on we evaluate the proposed scheme using both, analytical estimates and stochastic simulations based on a variety of real-world Internet topology data. Detailed comparisons are drawn to bi-directional tunneling, as well as to proposals on concurrent distribution trees.
Alexander Zimmermann, Martin Wenig, Jan Ritzerfeld, Ulrich Meis, Mesut Güneş - 2006
Abstract
The study of wireless and mobile networks is mainly based on simulations. Although simulation environments offer a convenient combination of flexibility and controllability, their largest disadvantage is that the results gained by using them are difficult to transfer into reality. This is due to the complex environment of mobile and wireless networks. In this paper we introduce a hybrid testbed approach, which consists of real mesh nodes and a virtualization environment. This combination provides on the one hand a flexible development environment for distributed network protocols and applications, and on the other hand a high degree in realism. Therefore, it allows the design and conduction of large scale networks where the results are easily transferred to the real world.
Alexander Zimmermann, Martin Wenig, Jan Ritzerfeld, Ulrich Meis, Mesut Güneş - 2006
Abstract
The study of wireless and mobile networks is mainly based on simulations. Although simulation environments offer a convenient combination of flexibility and controllability, their largest disadvantage is that the results gained by using them are difficult to transfer into reality. This is due to the complex environment of mobile and wireless networks. In this paper we introduce a hybrid testbed approach, which consists of real mesh nodes and a virtualization environment. This combination provides on the one hand a flexible development environment for distributed network protocols and applications, and on the other hand a high degree in realism. Therefore, it allows the design and conduction of large scale networks where the results are easily transferred to the real world.
Alexander Zimmermann, Martin Wenig, Jan Ritzerfeld, Ulrich Meis, Mesut Güneş - 2006
Download
Abstract
The study of wireless and mobile networks is mainly based on simulations. According to recent publications 76% of the studies in this area are based on simulations. Although simulation environments offer a convenient combination of flexibility and controllability, their largest disadvantage is that the results gained by using them are difficult to transfer into reality. This is due to the complex environment of mobile and wireless networks. In this paper we introduce a hybrid testbed approach, which consists of real mesh nodes and a virtualization environment. This combination provides on the one hand a flexible development environment for distributed network protocols and applications, and on the other hand a high degree of realism. Therefore, it allows the design and evaluation of large scale networks where the results are easily transferred to the real world.
Thomas C. Schmidt, Matthias Wählisch - 2006
Download
Abstract
Source Specific Multicast (SSM) promises a wider dissemination of group distribution services than Any Source Multicast, as it relies on simpler routing strategies with reduced demands on the infrastructure. However, SSM is designed for 'a priori known and changeless addresses of multicast sources and thus withstands any easy extension to mobility. Up until now only few approaches arose from the Internet research community, leaving SSM source mobility as a major open problem. This paper introduces a straightforward extension to multicast routing for transforming (morphing) source specific delivery trees into optimal trees rooted at a relocated source. All packet forwarding is done free of tunneling. Multicast service disruption and signaling overhead for the algorithms remain close to minimal. Furtheron we evaluate the proposed scheme using both, analytical estimates and stochastic simulations based on a variety of real-world Internet topology data. Detailed comparisons are drawn to bi-directional tunneling, as well as to proposals on concurrent distribution trees.
Zinaida Benenson, Martin Wenig, Mesut Güneş - 2006
Abstract
The deployment of sensor networks in the area of social relationships is very attractive, since by assisting people it may improve cooperation and prevent conflicts. We propose the Personality Sensors system (PerSens), which consists of a sensor network embedded into the clothes and other accessories of the person. PerSens will determine the personality type of the owner. Besides, it will also notify the owner of his behaviour in the current context and how it may appear to his counterparts.
Thomas C. Schmidt, Hans L. Cycon, Mark Palkow, Henrik Regensburg, Matthias Wählisch - 2006
Download
Abstract
Mobile multimedia communication will particularly rely on IP multicasting, as users commonly share frequency bands of limited capacities. This paper is dedicated to the problem of mobile Source Specific Multicast (SSM) senders. We propose extensions to multicast routing for transforming (morphing) previous delivery trees into optimal trees rooted at a relocated source. This extension scheme only requires basic signaling mechanisms, explicit joins and prunes. First evaluations grounded on real-world Internet topologies indicate network performance superior to traditional distribution schemes. As corresponding application we furtheron introduce a videoconferencing and communication software of distributed architecture. It is built as a simple, ready-to-use scheme for distributed presenting, recording and streaming multimedia content over next generation unicast or multicast networks.
Location-based services offer a powerful approach to provide highly relevant information and functionality to a mobile user. Addressing the problem of location inference, we expand the design space by proposing to employ rule-based programming techniques. Based on position coordinates as well as data gathered by environmental sensors, either a precise location or an abstract location class can be deduced. Both types of location may then be utilized to trigger services in an event-centric manner.
Thomas C. Schmidt, Hans L. Cycon, Mark Palkow, Henrik Regensburg, Matthias Wählisch - 2006
Download
Abstract
Mobile multimedia communication will particularly rely on IP multicasting, as users commonly share frequency bands of limited capacities. This paper is dedicated to the problem of mobile Source Specific Multicast (SSM) senders. We propose extensions to multicast routing for transforming (morphing) previous delivery trees into optimal trees rooted at a relocated source. This extension scheme only requires basic signaling mechanisms, explicit joins and prunes. First evaluations grounded on real-world Internet topologies indicate network performance superior to traditional distribution schemes. As corresponding application we furtheron introduce a videoconferencing and communication software of distributed architecture. It is built as a simple, ready-to-use scheme for distributed presenting, recording and streaming multimedia content over next generation unicast or multicast networks.
Recently, Distributed Hash Tables (DHT) explicitly designed for the use in MANETs have been proposed. Thus, many DHT-based distributed network applications from the domain of the Internet can be expected to be efficiently ported to MANETs. While the exact key lookups provided by such DHTs might be sufficient for many applications, range queries are often a desirable feature in wireless ad hoc networks (e.g. in sensor networks). However, the implementation of range queries using DHTs is a non-trivial task.In this paper we present a straight-forward implementation of Distributed Segment Trees as proposed in [4] on top of MADPastry [3] to provide DHT-based range queries for MANETs. The main goal of this work is to gain a first insight into the question whether DHT-based approaches for range queries are feasible in MANETs. First experimental results indicate that DHTs can indeed enable efficient range queries in MANETs.
In the domain of wireless sensor networks, simulation is the preferred way of evaluating new algorithms. One commonly found drawback of simulation tools is that they provide a programming environment that does not match the one present on real-world platforms. In this paper, we address this problem by presenting the steps required to port an existing software stack to a popular network simulator. This novel procedure allows for applications to run both in the simulated environment and on real-world sensor nodes without any changes to the source code. We verify our results by means of performance analyses of several simulated applications.
Kirsten Terfloth, Georg Wittenburg, Jochen Schiller - 2006
Download
Abstract
Introducing a middleware layer into wireless sensor networks is a widely accepted solution to facilitate application programming and to allow network organization. In this paper we introduce FACTS, a highly flexible middleware architecture able to provide support for a wide range of different applications. Instead of developing middleware and application apart from one another, we seek to combine them at programming level. Our rule-based language, tailored to this concept and the domain of networked sensors, enables high-level software development. The objective is to combine advantages of event-centric processing and rule-based execution while preserving low resource usage.
Kirsten Terfloth, Georg Wittenburg, Jochen Schiller - 2006
Download
Abstract
Data-centric, distributed programming for embedded systems with harsh resource constraints poses a heavy burden upon a developer. In this paper, we describe how rule-based programming can alleviate these problems by combining middleware and application at the programming level. We describe in detail the programming primitives and the implementation of the FACTS middleware architecture. Based on statistics derived from three representative tasks specific to wireless sensor networks, we illustrate how our approach allows for aggressive optimization as well as writing expressive application-level code. We summarize our experience by proposing several rule-oriented programming patterns.
Two main steps on the way to more realistic simulations of mobile ad-hoc networks are the introduction of realistic mobility and sophisticated radio wave propagation models. Both have strong impact on the performance of mobile ad-hoc networks, e.g. the performance of routing protocols changes with these models. In this paper we introduce a framework which combines realistic mobility and radio wave propagation models. Our approach consists of a mobility generator and an obstacle model for the radio wave propagation. It enables researchers to create realistic simulation setups and thus helps to correctly evaluate new algorithms and protocols. For the mobility generation a wide variety of well understood random mobility models is combined with a graph based zone model, where each zone has its own mobility model. To achieve a realistic radio wave propagation model a ray-tracing approach is used. The integration of these two techniques allows to create simulation setups that closely model reality.
This document extends the Hierarchical Mobile IPv6 Internet Draft to include the reception and transmission of Any Source Multicast traffic at the Mobile Node. It introduces handover mechanisms for IPv6 mobile multicast listeners and mobile multicast senders. Operations are based on a Mobile IPv6 environment with local mobility anchor points. These local anchor points are conformal with a Hierarchical Mobile IPv6 proxy infrastructure. Handover latencies in the proposed scheme remain bound to link switching delays with respect to these local proxy points. Thus the M-HMIPv6 achieves seamless mobility, even though no bicasting of multicast streams is used. Multicast listeners in addition encounter the option to optimize multicast routing by turning to a direct data reception. The mechanisms described in this document do not rely on assumptions of any specific multicast routing protocol in use. The M-HMIPv6 protocol operations utilize the existing HMIPv6, MIPv6 and MLDv2 messages under minor extensions.
Thomas C. Schmidt, Matthias Wählisch - 2005
Download
Abstract
Handovers in mobile packet networks commonly produce packet loss, delay and jitter, thereby significantly degrading network performance. Mobile IPv6 handover performance is strongly topology dependent and results in inferior service quality in wide area scenarios. To approach seamless mobility in IPv6 networks predictive, reactive and proxy schemes have been proposed for improvement. In this article we analyse and compare handover performance and frequencies for the corresponding protocols, as they are an immediate measure on service quality. Using analytical methods as well as stochastic simulations, we calculate the performance decreases originating from different handover schemes, the expected number of handovers as functions of mobility and proxy ratios, as well as the mean correctness of predictions. In detail we treat the more delicate case of these rates in mobile multicast communication. It is obtained that performance benefits, expected from simple analysis of predictive schemes, do not hold in practice. Reactive and predictive handovers rather admit comparable performance. Hierarchical proxy environments --- foremost in regions of high mobility --- can significantly reduce the processing of inter--network changes. Reliability of handover predictions is found on average at about 50%.
One of the most important components of mobile ad-hoc network simulations is the mobility model, since it defines the movement of mobile nodes and thus indirectly the network topology. The network topology at a given time in turn influences the performance of an ad-hoc network, for example the performance of routing algorithms changes with the mobility model. In this paper we introduce a communication and mobility scenario generator for mobile multi-hop ad-hoc networks. The goal is to aid researchers in the design of 'realistic' simulation scenarios which emulate real cities. Our approach combines a wide variety of well understood random mobility models with a graph-based zone model, where each zone has its own mobility model and parameters. The combination of directed, weighted graphs where the weights correspond to the flow of mobile nodes between neighboring zones and zones with different mobility models, allows the researcher design more realistic simulation scenarios.
In the first days of the emergence of Wireless Sensor Networks (WSN) programming software involved expertise in both hard-ware and networking. Since then, various middleware architectures have been proposed to achieve a suitable abstraction from distribution and management tasks of sensor devices. This allows users to focus on the application development. The approaches suggested so far differ in concept, functionality and envisioned abstraction. This paper surveys some representative approaches and states that only three different flavors of middleware exist, each of them addressing some characteristic abstraction middleware architectures are eager to provide. For future middleware implementations it will be beneficial to carefully balance the components of each class to obtain a mature design.
There are several deployment scenarios for mobile ad-hoc networks discussed in the literature. However, the most results have been made in artificial environments. In this paper we study the performance of state of the art routing protocols for mobile multi-hop ad-hoc networks in an environment which emulates a city downtown. The studied simulation environment differs in three aspects from that of well known: i) The used mobility model emulates a city downtown with several zones and different mobility models. ii) The number of mobile nodes and the number of connections is inspired from real traces. iii) We use duplex-connections.
Multicast data delivery carries distinct importance in mobile wireless environments, where bandwidth is limited and transmission channels are shared between users. The development of mobile multicast protocols consequently exhibits emerging interest. As the common approach of Any Source Multicast distribution is burdened with intricate routing procedures, the spread of multicast--enabled network infrastructure remains hesitant. It is widely believed that simpler mechanisms for group distribution in Source Specific Multicast (SSM) will lead to a pervasive dissemination of multicast infrastructure and services. However, SSM is designed for the {'a} priori known and changeless addresses of multicast sources and thus withstands any easy extension to mobility. This paper reviews the state--of--the--art in current work to extend SSM to Mobile IPv6 networks. The principle conceptual problems are discussed and analysed. Propositions for improvement and possible directions to proceed further in SSM source mobility are presented.
Note: Selected papers from the TERENA networking conference 2005
Wireless Sensor Networks (WSN) are an emerging technology with applications in the fields of environmental monitoring, facility management, and high-precision data gathering in general. They are deployed in the form of a large set of inexpensive sensing units that communicate over an ad-hoc wireless network. Developing applications for WSNs is challenging due to very limited resources available on each sensor node and the distributed nature of the algorithms used.In order to ease the development effort, we propose the FACTS middleware architecture for wireless sensor networks. Built around a rule-based programming paradigm, FACTS supports event-driven and data-centric applications to be written in the ruleset definition language. Components of the middleware are implemented in rulesets, compiled into bytecode and interpreted on the sensor nodes. We illustrate the capabilities of the FACTS middleware architecture by simulating it using the ns-2 network simulator and the ScatterWeb WSN platform.
Hans L. Cycon, Mark Palkow, Thomas C. Schmidt, Detlev Marpe, Matthias Wählisch - 2004
Download
Abstract
The purpose of this paper is twofold: On the one hand, we propose a fast wavelet- based video codec which is implemented into a real-time video conferencing tool. The proposed codec uses temporal frame difference coding, a computationally low-complex 5/3 tap wavelet transform, and a fast entropy coding scheme based on Golomb-Rice codes. On the other hand, we present an application of the video conferencing tool in a serverless peer-to-peer IP-based communication framework. For mobile communication we propose a simple, ready-to-use location scheme for video conference users in a global network.
Note: Special issue, selected papers from the Third International Conference on Wavelet Analysis and Its Applications (ICWAA '03)
Michael Friedrich, Gerd Nusser, Wolfgang Köchlin, Kirsten Terfloth - 2004
Download
Abstract
Mobile Agents are a well-known programming paradigm nowadays. There is a multitude of research concerning Mobile Agent Systems with emphasize on agent coordination, agent languages and agent migration technology. On the one hand, it is often argued, that Mobile Agents are well-suited for the use in the Internet and especially with Mobile Devices and roaming users. On the other hand, there are only few publications describing the actual implementation of a Mobile Agent System for Mobile Devices. In this paper we present our implementation of the Mobile Agent System Okeanos for the use in mobile environments with emphasize in agent routing and forwarding.
In this work, we present a distortion-based erasure code for realizing unequal error protection in video delivery services. Prior to transmission, the resulting distortion caused by packet erasure is estimated for each of the video packets. This Information Is used to protect the video packets with higher distortion values strongly, whereas video packets with very low distortion values are not worth being protected. We present a local search algorithm that delivers near-optimal parity codes which minimize the expected overall distortion at the receiver. Furthermore, we suggest the deployment of our FEC erasure code at video proxies located at the edge to access networks with high packet erasure rates (e.g. mobile networks). This helps detecting and correcting packet losses efficiently, by tailoring the error protection code characteristics to the observed packet erasure rate of the receiver. Our erasure code is a parity code and relics solely on XOR-operations for encoding and decoding; hence, it shows very high efficiency compared to other more sophisticated codes. Simulation results show that our introduced erasure code outperforms the classical Reed-Solomon erasure codes in terms of perceived video quality, while running considerably faster.
A mobile ad-hoc network (MANET) is a collection of mobile nodes which communicate over radio. These networks have an important advantage, they do not require any existing infrastructure or central administration. Therefore, mobile ad-hoc networks are suitable for temporary communication links. This flexibility, however, comes at a price: communication is dif cult to organize due to frequent topology changes. The Ant-Colony-Based Routing Algorithm (ARA) is highly adaptive, efficient and scalable. It is based on ant algorithms which are a class of swarm intelligence. Ant algorithms try to map the solution capability of ant colonies to mathematical problems. In this paper we present some extensions to the basic idea and show through simulation results the performance gain and compare it with AODV and DSR. Furthermore, we discuss the extensibility of the approach.
A mobile ad-hoc network (MANET) is a collection of mobile nodes, which communicate over radio. These networks have an important advantage; they do not require any existing infrastructure or central administration. Two nodes can communicate together as soon as they are in communication range. Two nodes apart from each other need the help of intermediate nodes relay their data. Mobile ad-hoc networks are suitable for temporary communication links. This flexibility, however, comes at a price. Communication is difficult to organize due to frequent topology changes. In this thesis, fundamental questions in regard to mobile ad-hoc networks, which are needed to realize and develop such networks are considered. While the difference between mobile ad-hoc networks and classical networks are minor, ad-hoc networks have inherently some properties, which make them much more difficult to handle. Especially, two aspects of ad-hoc networks make more intricate their realization: The first aspect is the used communication medium, which, compared to wired communication, has higher bit error rates. The other aspect is the dynamic network topology, which is caused by node mobility. These two problems of ad-hoc networks have effects on all communication layers. This thesis deals with two main issues in ad-hoc networking. Namely routing and automatic configuration. Routing Routing in ad-hoc networks is aggravated by the dynamic network topology. In the worst case the topology is changing continuously. To guarantee efficient data communication, it is necessary to transmit data packets over a shortest path. This requires high adaptability from routing algorithms. At the same time it is important to take into consideration, that the routing must be realized by all nodes in the ad-hoc network. This thesis presents a new on-demand routing algorithm based on ant algorithms named 'ant routing algorithm (ARA)' and evaluates its performance. Address configuration The second topic of this thesis is automatic configuration of ad-hoc networks, since zero configuration networking is also associated with ad-hoc networks. Many investigations assume, that each node in an ad-hoc network has an unique IP-address, but it is an open question how this can be performed. This thesis presents a distributed algorithm for the assignment of IP-addresses in ad-hoc networks and evaluates the protocol.
Ein Ad-hoc-Netzwerk besteht aus einer Menge von Knoten, die über Funk miteinander kommunizieren und dafür keinerlei Infrastruktur ben"otigen. Zwei Knoten, die sich in ihrer gegenseitigen Reichweite befinden, k"onnen direkt miteinander kommunizieren. Knoten, die voneinander entfernt sind, ben"otigen die Hilfe von weiteren Knoten, die sich zwischen ihnen befinden. Ein Ad-hoc-Netzwerk ist flexibel, sowohl hinsichtlich der Selbstkonfiguration als auch bezüglich der Anpassung an die Netzwerkstruktur, d.h. an die vorhandenen Netzwerkteilnehmer und die Netzwerktopologie. In dieser Arbeit werden grundlegende Verfahren betrachtet, die für die Entwicklung und Realisierung von zukünftigen Ad-hoc-Netzen erforderlich sind. Obwohl sich Ad-hoc-Netze von klassischen leitungsgebundenen Netzen in nur wenigen Punkten unterscheiden, besitzen sie von Haus aus bestimmte Eigenschaften, wodurch sie sehr schwer handhabbar sind. Vor allem zwei Aspekte von Ad-hoc-Netzen erschweren ihre effiziente Realisierung. Der erste Aspekt ist das verwendete "Ubertragungsmedium, nämlich die Luftschnittstelle, die, verglichen mit anderen Medien, schlechte Eigenschaften für die Kommunikation besitzt. Der zweite Aspekt ist die sich ändernde Netzwerktopologie, welche durch die Knotenmobilität verursacht wird. Diese beiden Problemursachen haben Auswirkungen auf alle Schichten des Kommunika-tions-protokolls. Im Rahmen dieser Arbeit werden zwei aktuelle Fragestellungen von Ad-hoc-Netzen behandelt. Die erste Fragestellung betrifft das Routing und die zweite Fragestellung die Autokonfiguration in Ad-hoc-Netzen. Routing: Das Routing wird in Ad-hoc-Netzen durch die Knotenmobilität erschwert, da im ungünstigsten Fall die Netzwerktopologie sich ständig ändert. Um die Kommunikation effizient gestalten zu k"onnen, müssen die Datenpakete auf einem günstigsten (bzw. kürzesten) Pfad zwischen dem Quell- und Zielknoten übertragen werden. Dies erfordert von den eingesetzten Routingalgorithmen eine hohe Adaptionsfähigkeit an die Netzwerktopologie. Dabei ist darauf zu achten, dass in einem Ad-hoc-Netz, wegen des Fehlens einer Infrastruktur, die Funktionalität von den teilnehmenden Knoten erbracht werden muss. In dieser Arbeit wird ein neuartiger Routingalgorithmus auf der Basis von Ameisenalgorithmen, die ein Teilgebiet der Schwarmintelligenz sind, vorgestellt und mit bekannten Routingalgorithmen für Ad-hoc-Netze verglichen. Adresskonfiguration: Die zweite Frage, mit der sich diese Arbeit beschäftigt, ist die automatische Konfiguration von Ad-hoc-Netzen. Da in Zusammenhang mit Ad-hoc-Netzen implizit auch über Zero-Konfiguration-Netzwerke gesprochen wird, spielt diese Fragestellung eine wichtige Rolle. Es wird also angenommen, dass ein Ad-hoc-Netz sich automatisch selbstkonfiguriert, ohne dass Benutzereingriffe erforderlich sind. Viele Untersuchungen im Bereich der Ad-hoc-Netze gehen davon aus, dass alle Knoten in einem Ad-hoc-Netz eine eindeutige Adresse haben. Bevor jedoch z.B. ein Routingalgorithmus einen Pfad zwischen Quell- und Zielknoten finden kann, müssen die Knoten auf der Netzwerkschicht identifiziert werden. Diese Arbeit stellt einen verteilten Algorithmus für die automatische Konfiguration von Ad-hoc-Netzen und das zugeh"orige Protokoll vor und bewertet seine Leistung.
Mobile multi-hop Ad-hoc-Netze (MANET) zeichnen sich durch ihre hohe Flexibilität aus, da sie keine Infrastruktur ben"otigen. Durch ihre inhärente schwierige Umgebung, die durch die Knotenmobilität bedingt ist, erfordern diese Netze adaptive Verfahren. Eine der wichtigsten Herausforderungen von Ad-hoc-Netzen stellt das Routing dar, an dem seit Jahrzehnten gearbeitet wird. In dieser Arbeit wird ein neuartiger Routingalgorithmus für mobile multi-hop Ad-hoc-Netze vorgestellt, der auf Ameisenalgorithmen basiert. Durch das Nachahmen des Futtersuchverhaltens von Ameisenkolonien ist der Ameisenroutingalgorithmus in der Lage, sich der ständig ändernden Netztopologie von mobilen multi-hop Ad-hoc-Netzen anzupassen und dabei den ben"otigten Overhead zu reduzieren.
A mobile ad-hoc network (MANET) is a collection of mobile nodes which communicate over radio. These networks have an important advantage, they do not require any existing infrastructure or central administration. Therefore, mobile ad-hoc networks are suitable for temporary communication links. This exibility, however, comes at a price: communication is dif cult to organize due to frequent topology changes. The Ant-Colony-Based Routing Algorithm (ARA) is highly adaptive, efficient and scalable. It is based on ant algorithms which are a class of swarm intelligence. Ant algorithms try to map the solution capability of ant colonies to mathematical and engineering problems. In this paper we present some extensions to the basic idea and show through simulation results the performance gain and compare it with AODV and DSR.
In this work, we propose an approach for the deployment of video proxy caching for robust video transmission over lossy networks. Video packets are classified by the server according to their valuableness for the decoding process and the perceived video quality. This can be easily realized using a framework for objective video quality measurement along with the frame prediction type. When deployed within the network, e.g. in intermediate multicast routers or proxy caches, this technique helps achieving great quality improvements. By identifying the most important video packets and caching them for short time, retransmission request can be processed quickly, while saving precious time and bandwidth over retransmissions issued by the server. An analytical model as well as experiments have been developed and carried out to show the gain in video quality at the receiver side, that can be achieved through selective video packet caching and retransmission.
The trend of wireless networking is growing. This is due to the success of IEEE 802.11 WLAN technology which allows the extension of local area networks by mobile stations and establishing of ad-hoc networks. The latter does not need any pre-installed infrastructure. Nodes communicate directly or the traffic is relayed over intermediate wireless nodes. While there are only few nodes the performance of the network is acceptable, but with increasing number of nodes the performance of the network decreases. If TCP is used additional performance degradation is expected. However, TCP is necessary since most applications and application level protocols are based on it. The poor performance of TCP in wireless networks has several reasons, especially the communication medium with high bit error rates (BER) and TCP's optimization on fixed networks. In this paper, we present an analysis of TCP's performance in wireless and ad-hoc networks. We emphasize the role of MAC layer retransmissions without breaking the end-to-end semantic of TCP. For this, we introduce the dynamic short retry limit which allows the adaptation of RTS/CTS retransmissions.
A mobile ad-hoc network (MANET) is a collection of mobile nodes which communicate over radio. These networks have an important advantage, they do not require any existing infrastructure or central administration. Therefore, mobile ad-hoc networks are suitable for temporary communication links. This flexibility, however, comes at a price: communication is difficult to organize due to frequent topology changes. Ant algorithms are a class of swarm intelligence and try to map the solution capability of ant colonies to mathematical and engineering problems. The Ant-Colony-Based Routing Algorithm (ARA) is based on ant algorithms. The main properties of the algorithm are high adaptive, efficiency. In this paper we present some extensions to the basic idea and show through simulation results the performance gain and compare it with AODV and DSR.
Video streaming applications are getting more and more popular among Internet users. Video conferencing, Video-on-Demand and video surveillance are just examples of video streaming applications. An emerging need to make such applications available in mobile multi-hop ad-hoc networks can be observed clearly through the last research effort made in this direction. Several challenges, related to the nature of mobile wireless networks have to be dealt with, in order to make the deployment of video streaming applications over mobile multi-hop adh-hoc networks feasible with an acceptable quality. In this paper, we present a framework for streaming video over mobile multi-hop wireless networks. The framework is composed of four components handling several problems of video streaming over multi-hop ad-hoc networks. The focus in this paper is nevertheless set on the loss control component, realized through augmented functionality of intermediate nodes. We also present some experimental results which identify the reasons for the quality deterioration and show the improvements achieved through the deployment of our framework.
Streaming video over the best-effort Internet raises some challenging issues. The available bandwidth and packet transmission delays are often subject to severe changes. As a consequence, a significant ratio of video packets is dropped by network routers because of lack of capacity on congested links. The impacts of video packet drops on the observed quality at the receiver may vary from unnoticeable to unacceptable. This is due to the different values of each video packet for the decoding process at the receiver. Classical queueing disciplines do not have the ability to distinguish between different packet values. It is not unusual that a packet of higher value is dropped, in favor of a lower valued packet. The value of a packet is measured in terms of the induced distortion, when the packet is not received correctly. In this paper, we introduce an application-specific queueing system, which drops packets in a size-distortion optimized manner. Out of all packets present in the queue, the system selects the packets to be dropped, if necessary, in order to meet a queue size constraint while minimizing the overall distortion. Great improvements in video quality for the receiver, and an efficient utilization of the available bandwidth have been observed through different simulations.
Thomas C. Schmidt, Hans L. Cycon, Mark Palkow, Matthias Wählisch - 2003
Download
Abstract
In recent years the capabilities of the common Internet infrastructure have increased to an extent where data intensive communication services may mature to become popular, reliable applications. Videoconferencing over IP can be seen as such a highly prominent candidate. However, heavy infrastructure and complicated call handling hinder acceptance of standard solutions. This paper presents a more lightweight framework - both communication scheme and conferencing software - to overcome these deficiencies. A simple, ready-to-use global location scheme for conference users is proposed. First practical experiences are reported.
Note: Selected papers from the TERENA networking conference 2002
This paper describes the replay attack against traditional and channel-based Chaumian mixes and puts it into context with other attacks on anonymizing services. It then evaluates different possible approaches to a defense against said attack, concentrating on the question of efficiently recognizing potentially compromised messages within a channel-based cascade of mixes. Finally, an exemplary implementation of a solution is presented for the anonymizing cascade of mixes of the AN.ON project.
A mobile ad-hoc network (MANET) is a collection of mobile nodes which communicate over radio. These networks have an important advantage, they do not require any existing infrastructure or central administration. Therefore, mobile ad-hoc networks are suitable for temporary communication links. This exibility, however, comes at a price: communication is dif cult to organize due to frequent topology changes. In this paper we present a new on-demand routing algorithm for mobile, multi-hop ad-hoc networks. The algorithm is based on ant algorithms which are a class of swarm intelligence. Ant algorithms try to map the solution capability of ant colonies to mathematical and engineering problems. The Ant-Colony-Based Routing Algorithm (ARA) is highly adaptive, ef cient and scalable. The main goal in the design of the algorithm was to reduce the overhead for routing. Furthermore, we compare the performance of ARA with other routing protocols, including DSDV, AODV, and DSR through simulation results.
Mobile multi-hop ad-hoc networks promise an efficient and very exible way of establishing a communication connection. One aspect of these kind of networks is their convenience, which is some times neglected in research. Especially for civil applications, this aspect of ad-hoc networks need more attention. It is expected that users neither need to know much of networks nor perform complex configuration tasks to establish a connection. Therefore this kinds of networks have to not only handle a very difficult environment, but also have to be very easy to use. One of the basic requirement of networking is that each node in the network has a unique identification, also called an address. In the internet and in local area networks IP addresses are common. Mobile ad-hoc networks should also use IP addresses thus the devices will be same and also to allow the interconnection of ad-hoc networks to the internet. It is an open question how to provide IP addresses to nodes in mobile multi-hop ad-hoc networks. In this work we introduce a new dynamic IP address configuration algorithm for mobile multi-hop ad-hoc networks. Further we show through simulation results, made with ns-2, the performance of the proposed addressing algorithm and compare it with other approaches.
Mobile wireless networks are gaining increasing interest as an alternative to infrastructure-based wired networks. For this to take place, applications running over traditional networks should also be applicable over wireless networks with a reasonable quality. One of the popular applications of networking are streaming multimedia applications. A better support of streaming multimedia applications over wireless networks will speed up the spreading of such networks. In this paper, we present a framework for streaming video over mobile multi-hop wireless networks. The framework is composed of three components handling all problems of video streaming. However, we focus in this paper on the congestion control component, realized through augmented functionality of intermediate nodes. We also present some experimental results with which we tried to identify the causes of the deterioration in quality, when streaming video in a wireless network.
Mobile ad-hoc networks (MANET) enable the communication of mobile wireless nodes without any pre-defined infrastructure. Therefore, they are predestined for future mobile applications. To use existing applications and to inter-operate with the world wide internet, ad-hoc networks have to use TCP/IP, which were not designed for mobile adhoc networks and yield poor performance when deployed in this kind of networks. The poor performance of TCP in adhoc networks has several reasons, especially the communication medium with high bit error rates (BER) and TCP's optimization on fixed networks. In this paper, we present an enhancement, the restricted congestion window enlargement (TCP/RCWE), and show its ability to handle TCP's problems in mobile ad-hoc networks. We also illustrate the performance gain of TCP by the enhancement.
Mobile Ad-Hoc Netze (MANET) bestehen aus einer Menge von mobilen Knoten, die drahtlos miteinander kommunizieren. Durch die Knotenmobilität ist die Netztopologie ständigen Veränderungen unterworfen und es existieren keine Einheiten, die jederzeit verfugbar sind. Hauptproblem bei mobilen Ad-Hoc Netzen ist die schnelle Pfad ndung, welche durch die Knotenmobilitat erschwert wird. Bevor jedoch ein Pfad zwischen zwei Kommunikationsendpunkten ermittelt werden kann, mussen die Knoten eindeutig identi ziert werden. Dabei werden zwei Komponenten ben otigt: i) einheitliche Adressen, ii) ein Verfahren, welches den Knoten eindeutig Adressen zuweist. Als Adressen bieten sich IP-Adressen an. In dieser Arbeit wird die Agentenbasierte Adressierung, ein dynamisches Adressierungsverfahren fur mobile Ad-Hoc Netze, vorgestellt und bewertet. Weiterhin wird ein bekanntes Verfahren aus der Praxis und ein neuer Vorschlag aus der aktuellen Diskussion untersucht und bewertet.
Thomas C. Schmidt, Hans L. Cycon, Mark Palkow, Matthias Wählisch - 2002
Download
Abstract
Die Leistungsmerkmale der Internetinfrastruktur in ihrem gegenw�rtigen Ausbaustand lassen uns darauf vertrauen, da� auch synchrone Telekommunikationsdienste �ber IP alltagstauglich werden. W�hrend Voice over IP vornehmlich unter Kostengesichtspunkten adressiert wird, bildet Videoconferencing over IP einen neuen, bisher in keinem Medium verbreiteten Kommunikationsdienst. Videokonferenzanwendungen haben deshalb beste Aussichten, als k�nftige Standardapplikation die Popularit�tskurve des Internets pr�gend anzuf�hren. Diese Entwicklungspotentiale werden gegenw�rtig jedoch von einer wenig anwenderfreundlichen Kommunikationsvermittlung sowie schweren Infrastrukturb�rden gehemmt. Wir stellen nachfolgend einen leicht zug�nglichen L�sungsansatz vor, welcher sowohl in einem einfachen Kommunikationsschema als auch in einer fertigen Videokonferenzsoftware f�r Endanwender besteht. Bestandteile dieser L�sung sind eine einfache Partneradressierung und ein infrastrukturneutrales, benutzungsfertiges globales Teilnehmerlokalisierungsschema. Erste Erfahrungen der L�sungseinf�hrung in unserem Hause werden berichtet.
The book deals with modeling and analysis of computer and communication systems. It is mainly an introduction to queueing systems. Language is german.
Note: ISBN: 3-86073-729-5
Arndt Tochatschek, Thomas C. Schmidt, Matthias Wählisch - 2001
Download
Abstract
Die Untervermittlung von Netzdienstleistungen an rechtlich unabhägige Abnehmer erfordert Technologien, die sowohl Sicherheitsanforderungen als auch dem Anspruch an Verbrauchsabrechenbarkeit genügen. Auf der Basis von Standard Ethernet, 802.1Q VLAN-Techniken und RMON präsentieren wir eine einfache, kostenminimierte Lösung für dieses Problem. Auf der Grundlage dieses Ansatzes stellen wir weiterhin ein Betriebsszenario zur vollautomatischen Konfiguration und Abrechnung vor.
This book deals with security issues in communication systems. Introduces basic problems of secure and private communication, shows commong pitfalls in communication protocol design, and discusses solutions. Language is german.
Note: ISBN: 3-86073-727-9
Using hypertext to support work, learning and research, has have increasingly popular. The user can arbitrarily navigate in the information space. Yet many approaches do not consider the aims, interests and the knowledge of the user. This paper introduces a method for the intelligent and dynamic generation of hypertexts from an information space, and for the support of navigation in large, complex hyper documents, which is based on the modelling of the application domain. Intelligent generation of hypertext is based on different inference strategies, for horn clause resolution. The generated hypertext is problem-oriented and presented in a context sensitive manner. Furthermore, the structure of the hypertext is presented graphically and serves as a navigation support for the user. All tools related to this work are implemented in Java.
The VIVE project is validating applications that reliably detect events with cooperative working wireless sensor nodes. Please, visit us at the Hannover Messe:BMBF hall 2, both C24. BMBF-Flyer.
The new BMBF project VIVE validates the capability of innovation of distributed event detection in wireles sensor networks. Cooperative and autonomous event detecting sensor nodes in wireless sensor networks share and fuse information of a common event within the network. VIVE started: February 2012.
The SKIMS project is working on developing a cooperative immune system for mobile phones. Please, visit us at the CeBIT:BMBF hall 26, both E50. Overview.
Diese Grafiken werden nur in der Druckvorschau verwendet:
Download
