Publikationen 2011

The future wireless telecommunication environment is characterized by the co-existence of a multitude of wireless networks such as LTE, LTE-Advanced, UMTS, WiMAX, WiFi etc. In order to be able to offer the best connectivity, according to the requirements of the users and to the preferences of the operator, a novel functionality was introduced in the core network and in the mobile devices for access network discovery and selection. This paper introduces this functionality as standardized in the 3GPP Evolved Packet Core (EPC), highlighting its main concepts and technical scenarios. Further, a set of novel optimizations are presented and evaluated, followed by the description of their implementation as part of the Fraunhofer FOKUS OpenEPC toolkit.

Telecommunication and Internet services are constantly subject to changes, seeking the customer's full satisfaction. Enriching these services with innovative approaches such as context-aware, social, mobile, adaptable and interactive mechanisms enable users to experience a variety of personalized services seamlessly across different platforms and technologies. In this sense, Service Oriented Architectures play a central role in allowing component reuse and low cost service creation. Together with IP Multimedia Subsystem enable the convergence of telecommunications and web services, allowing the network transport technologies to be abstracted from the services above. By integrating these technologies, a number of synergies can be explored. Existing services can be easily enriched with context information, made available on a variety of networks and new services can be composed using previously existing building blocks. This paper explains how this integration can be achieved, and demonstrates the potentialities of this architectural paradigm with a prototype service.

Networks, services, and business models need to be piloted in a large-scale environment that mimics as far as possible the Internet and its future version, although we certainly do not know what this will look like. In this highly demanding and dynamic context, new frameworks and (meta-) architectures for supporting experimentation are proposed, aiming at discovering how existing and emerging testbeds and experimental resources can be put together in such a way that testing and experimentation may be carried out according to specific requirements (industry, academia, etc.). The work presented in this article addresses a number of the fundamental principles and their corresponding technology implementations that enable the provisioning of large-scale testbeds for testing and experimentation as well as deploying future Internet platforms for piloting novel applications. The proposed concepts and architecture of Panlab, a pan-European testbed federation, aspires to facilitate and support user needs in these new areas.

As the Internet is developing over the next decade, context-aware computing will be one of the driving technologies. User generated context and machine generated context (e.g. by sensors or RFID chips) about the location, presence, or environment status etc. of persons or devices will be used for many different kinds of application services like telecommunication services, logistics, weather forecasts or for environment monitoring in general. To facilitate the development of such context-aware applications, effective and efficient systems that support the development of such services are needed. Otherwise the development of context-aware applications will lead to custom solutions that decrease development time and limit reusability. To avoid this, generic tools are needed that assist the developer to produce context-aware services, that are able to support different kind of context information and do not limit the way how context information can be conveyed by different signaling approaches. The aim of this thesis is to design and implement a generic framework for the development of context-aware server applications based on the event notification mechanism. The approach is to have a universal basis that handles the signaling and to have a flexible way to add business logic to process the content of the messages. By design the framework will not be limited to a certain messaging protocol, but will be extendible for implementing other messaging protocols. Developer of context-aware server applications will be able to decrease the development time by focusing on the imp lementation of the business logic and not considering about the message protocol handling. Applications will benefit from considering different context sources which will facilitate to create new context-aware services that bridge the gap between telecommunication and Internet services, machine-to-machine communication and the Internet of Things.

This doctorate thesis introduces a framework for information and communication technology resource federation. Resource federation in the context of this work is a concept for sharing and allowing access to heterogeneous resources across multiple administrative domains. Three design artifacts are delivered as the main research result: a federation model, a federation methodology, and a system instantiation. The artifacts abstract from the types of resources to be federated. Examples of resources include general-purpose physical and virtual machines, software packages and systems, as well as specialized proprietary devices. Resource capabilities are exposed in a service oriented manner and can be assembled in meaningful combinations and configurations to satisfy the service consumer's requirements. One of the potential user segments is the scientific community where resources are requested for running large scale research experiments that require a certain setup of heterogeneous resources. The benefit provided to users of the federation is the seamless access to a large collection of resources via unified interfaces. Resource providers benefit from an additional sales channel, increasing their average resource utilization, some sort of compensation for the provided services, and the access to additional user groups. A federation organization maintains relationships with all stakeholders and offers central services to ensure the overall operation.

Das Dokument gibt eine erste Übersicht über den Ansatz zum Aufbau des BEST-SELECT Demonstrators. Der Demonstrator wird auf dem Information-Management-System CCDS basieren und im FUSECO Playground implementiert, einem Berliner Testbed für drahtlose Breitband-Kommunikationssysteme, das auch einen Zugang zum OpenEPC bietet. Damit können die BEST-SELECT-Konzepte in einem realistischen Demonstrator gezeigt und evaluiert werden. Dieses Dokument beschreibt die einzelnen Grundlagen, auf denen der BEST-SELECT-Demonstrator entwickelt wird, die Demonstrationsszenarien und erste Ansätze zur Implementierung des Demonstrators.

This Deliverable, D4.2, on 'Distributed Self-Growing Architecture and Interface Description' provides a proposal for a CONSERN self-growing architecture and describes its components and interfaces on a high-level. The architecture builds on the requirements that were derived in the earlier deliverables D1.1 and D4.1 and will be used and refined in future CONSERN research.

This Deliverable, D4.3, on "Formalisation of self-growing strategies, policies and algorithms" specifies the different mechanisms that constitute self-growing, with emphasis on specific mechanisms that are necessary for proof-of-concept of self-growing. It identifies the requirements towards self-growing from the full list of technical requirements of CONSERN, and prioritizes those that are necessary for a proof-of-concept. It elaborates on the relevant mechanisms that are implemented and/or studied within all the Work-Packages of CONSERN. And it verifies that all the requirements for proof-of-concept of self-growing are covered within CONSERN.

A framework for lowering the borders of service delivery and service composition in telecommunication networks is proposed in this work. In our opinion such an agile framework is needed for two main reasons. At first the cycle of introducing new services in the telecommunication industry usually takes several months, while the general cycles in the Web domain are much faster. Besides, customers want to use a flexible amount of services from different kinds of sources (e.g. Google, Facebook). The proposed agile service environment is intended to interact with services of the traditional telecommunication industry and is enabling the third party development process to support services coming from all kind of domains. Service platforms for Next Generation Networks adapting the principles of Service Oriented Architectures and the initial steps in the standardization towards service delivery frameworks are done. Special requirements that have to be considered are aspects of servicelife-cycle management and service composition that are critical regarding high performing real-time communication processes as needed in a telecommunications environment. Beyond the requirements of dynamic service deployment, management, evolution and service termination there is the intention of a well performing agile service composition mechanism. In order to fulfill the discussed requirements some related work in this area is presented and the promising IT / Network standards are enlightened. Furthermore, the architectural design of the developed framework is presented. Finally the paper demonstrates a practical use case scenario that shows the dynamic integration and combination of telecommunication services with services of the Web domain.

In this work a new Universal Communications Framework (UCF) is defined, based on IMS concepts and on EPC technologies. The UCF extends the IMS concepts to a generic middleware solution that intermediates between underlying network infrastructure and various application functions by offering generalized service APIs. One use case scenario to demonstrate the overall system is the integration of non human-interacting services like mobile Machine-to-Machine (M2M) communication in the emerging Internet of Things (IoT). Furthermore, the framework is focused on non human-interacting services like mobile Machine-to-Machine communication enabled with dedicated application-driven Quality of Service (QoS).

With the increasing deployment of autonomous machine type devices which highly surpass in number the devices directly controlled by humans, the wireless environment should support efficiently and in a scalable manner the Machine-2-Machine (M2M) type of communication enabling new services and business opportunities. However, the current mobile network architecture is designed for human interaction, missing the optimization which can be derived from the basic characteristics of the machine type communication. This article introduces the concept of device connectivity management as a core network means for indicating governing policies to the wireless connected devices on when and how to communicate in order to be able to deliver the machine services in an efficient manner in regard to core network available capabilities and to service requirements. Also it includes the execution means in the wireless device for handling these policies. Furthermore, the concept is exemplif ied asan addition to the current 3GPP Evolved Packet Core architecture and to the Fraunhofer OpenEPC testbed.

This paper presents initial work towards an architecture suitable for purpose-driven, self-growing networking as realized by distributed cognitive decision engines within the network. Starting from describing basic modules enabling for self-growing, the paper derives a logical architecture for realizing the concept at various network levels. Mapping results to the UMTS stratum model shows potential for immediate applicability of the concept to deployed networks.

With the deployment of novel wireless broadband technologies such as LTE and LTE-Advanced, new types of devices will be able to communicate remotely such as sensors and actuators, enabling the deployment of new services and applications generically named Machine-to-Machine (M2M) communication. However, the current core network architectures are designed only for Human-to-Human communication, lacking the fundamental support for the machine type communication patterns for functions like mobility and resource management, charging control etc. This article introduces a novel solution for mitigating with the basic support for resource management for M2M through wireless device based customization and grouping and introduction of cached based resource reservation and event notification mechanisms. Furthermore, the concept is exemplified on the 3GPP Evolved Packet Core architecture and evaluated as a testbed realization based on the Fraunhofer OpenEPC testbed.

Abstract: This half day workshop will provide an overview of the Packet Core Network evolution starting with the introduction of global fixed and mobile next generationnetwork standards and outlining its evolution in regard to the emerging Future Internet, which is driven by international research programmes, such as GENI in the US, FIRE in Europe and Akari in Japan. The main focus will be on the network and service control options and optimization within the 3GPP Evolved Packet Core (EPC) forming today the common packet core architecture for various broadband access networks, including LTE and WiMAX. The tutorial will outline the EPC architecture and address potential EPC application domains, including IP Multimedia Subsystem (IMS) based VoIP as well as internet over the top service architectures with their own service layer control capabilities. Starting from here the tutorial looks at the emerging Future Internet, comprising different research views, such as the internet of things, the internet of services, and the network of the future. Here we introduce the current visions and research topics related to cross-layer functional composition and network virtualisation. The tutorial terminates with outlining how FI concepts could be exploited in NGN/EPC evolution and introducing related software toolkits and experimental platforms of TU Berlin / Fraunhofer FOKUS, namely the OpenEPC (www.openepc.net) and the FUSECO-Playground (www.fuseco-playground.org) enabling comprehensive prototyping in the context of academic and industry research.

Spectrum resources unused in the spatio-temporal domain, so-called whitespace, can be utilised by opportunistic devices during the absence of their incumbent users. The possibility to opportunistically use whitespace implies the knowledge of diverse constraints: knowledge about whitespace utilisation, which can be obtained by spectrum sensing or from a data repository and constraints internal to the opportunistic system, such as quality of service (QoS) levels and mobility targets. The use of whitespace requires therefore a multi-faced management of the above constraints. In this paper, an architecture for use of whitespaces under QoS and mobility constraints is proposed. The proposed architecture flexibly adapts to different operating scenarios also described in this paper. Examples show how different realisations of the same architecture are derived. The interactions of constituent blocks are illustrated also with the help of charts, showing the management of contextinformation and its use.

Resource federation is a concept for sharing information and communication technology (ICT) resources beyond administrative domains. The aim is to implement some common service drawing upon the resources committed by participating organizations. This concept is currently applied in a number of projects to provide large scale experimental facilities serving Future Internet research and development. In this paper we show how federated distributed resources can be described in terms of a common information model to support distributed resource management and Future Internet experimentation.

Currently, a number of research activities worldwide focus on resource federations to enable remote resource access and the setup of large scale experiments. With the global efforts in sharing experimental facility resources across organizational boundaries, interoperability issues are becoming increasingly important. To this end, this paper describes a model that defines the necessary entities for inter resource federation scenarios. The model is then discussed by means of a concrete use case federating Panlab and PlanetLab resources. In this respect Panlab and PlanetLab are considered to be federations themselves. Our findings show that the two approaches can be matched following the general model. This allows for federating resources not only across the boundaries of administrative (organizational) domains but even across the boundaries of federations. Those federations in themselves provide collections of resources offered by several administrative domains and use different control frameworks to enable resource access and management.

The challenging and ever increasing requirements of future applications demand new concepts for better control and management of network resources. The Third Generation Partnership Project (3GPP) introduced in their latest specifications the Evolved Packet Core (EPC) architecture for transparently unifying the parameters of different technologies, like the UMTS, WLAN, non-3GPP access technologies and a future Evolved Radio Access Network, called Long Term Evolution (LTE), with the use of multiple application platforms such as IP Multimedia Subsystem (IMS) and the Internet. This paper describes a testbed implementation of the Evolved Packet Core named OpenEPC which provides a reference implementation of 3GPPs EPC developed by the Fraunhofer Institute FOKUS. OpenEPC is a set of software com ponents offering advanced IP mobility schemes, policy-based QoS control, and integration with different application platforms in converging network environments. This initiative, in addition to fostering research and development, enables academic and industry researchers to rapidly realize state-of-the-art NGMN infrastructures and application testbeds.

Mobility in a wireless heterogeneous scenario in which the mobile devices are able to connect to more than one access technology available in their vicinity requires a re-consideration of the access network reselection mechanisms as to ensure seamless handovers for real deployments. This paper describes and evaluates a new proactive vertical handover optimization which enables a fast reselection, independent and in addition to the classic proactive procedures, using as a central concept that in specific conditions, the access network selection functionality located in the network should trigger a network internal proactive procedure which does not imply a high usage of the target access network resources according to the probability that a handover will be necessary in the near future as t o reduce the actual handover delay. This concept is exemplified on the 3GPP Evolved Packet Core and evaluated on a minimal prototype implementation.

Handling the complexity of future network infrastructure presents great challenges to network management. Introduction of self-management features will allow for the automation of common management tasks. Based on an architecture for measurement and self-management we analyze in this paper a use case for achieving energy-efficient adaptations in access networks comparing various options of node cooperation and information exchange.

Service-Oriented Network Architecture, Multimedia, Quality of Service, Internet housekeeping

In diesem Projekt werden die in den AV Vorlesungen NGN und FIT erworbenen Kenntnisse durch praktische Programmieraufgaben vertieft. Dabei werden mit Blick auf Exzellenz und Nachhaltigkeit in kleinen Gruppen neue Komponenten, Schnittstellen, Anwendungen und Werkzeuge fuer die AV/FOKUS OSTP, Next Generation Mobile Networks und FI Testbeds entwickelt. Die genauen Themen sind der AV Homepage zu entnehmen unter www.av.tu-berlin.de.

In diesem Projekt werden die in den AV Vorlesungen NGN und FIT erworbenen Kenntnisse durch praktische Programmieraufgaben vertieft. Dabei werden mit Blick auf Exzellenz und Nachhaltigkeit in kleinen Gruppen neue Komponenten, Schnittstellen, Anwendungen und Werkzeuge für NGN und FI Testbeds entwickelt die bei AV/Fraunhofer FOKUS zum Einsatz kommen. Die Technologien kommen aus den Bereichen Service Composition und Web Mash Ups, Identity and Policy Management, Converged Instant Messaging, IP-TV, Videostreaming, Infrastructure as a Service, Virtualisation.

In diesem Projekt werden die in den AV Vorlesungen NGN und FIT erworbenen Kenntnisse durch praktische Programmieraufgaben vertieft. Dabei werden mit Blick auf Exzellenz und Nachhaltigkeit in kleinen Gruppen neue Komponenten, Schnittstellen, Anwendungen und Werkzeuge fuer die AV/FOKUS OSTP, Next Generation Mobile Networks und FI Testbeds entwickelt. Die genauen Themen sind der AV Homepage zu entnehmen unter www.av.tu-berlin.de.

In diesem Projekt werden die in den AV Vorlesungen NGN und FIT erworbenen Kenntnisse durch praktische Programmieraufgaben vertieft. Dabei werden mit Blick auf Exzellenz und Nachhaltigkeit in kleinen Gruppen neue Komponenten, Schnittstellen, Anwendungen und Werkzeuge fuer die AV/FOKUS OSTP, Next Generation Mobile Networks und FI Testbeds entwickelt. Die genauen Themen sind der AV Homepage zu entnehmen unter www.av.tu-berlin.de.

In dieser VL werden grundlegende Konzepte, Architekturen, Protokolle, und Schnittstellen von aktuellen und zukünftigen Offenen Dienstplattformen auf der Basis konvergierender Netzwerktechnologien wie drahtlose und drahtgebundene Telekommunikationsnetzwerke, Next Generation Networks (NGNs) und Next Generation Mobile Networks vorgestellt. Insbesondere werden vorgestellt: VoIP, das IP Multimedia Subsystem (IMS) , aktuelle Entwicklungen des Evolved Packet Core und das Zusammenspiel der grundlegenden Technologien: SIP, Diameter, SIP Server, Media Server, Application Server, Service Delivery Platforms wie Telco JSON APIs, Telco JSON APIs, IN/CAMEL, OSA/Parlay, JAIN Web2.0 Mash ups, etc.