"This site requires JavaScript to work correctly"
minimize Header

Prof. Dr. Andreas Fischer

  • Applied Artificial Intelligence
  • Generative Artificial Intelligence
  • Software-defined Networking and Network Function Virtualization
  • Network Resilience
  • Experimental Algorithmics

Faculty of Computer Science

Professors

Professor

  • Chief Information Officer (CIO)
  • Scientific Manager of the IT Center
  • Assistant member of the doctoral committee DigiTech

DEGG's 2.12

0991/3615-653

consulting time

By request per e-mail

Sortierung:
Contribution
  • Andreas Fischer
  • Andreas Berl
  • H. Meer

Virtual Network Management with XEN.

In: Tagungsband Herbsttreffen 2008 der GI/VTG Fachgruppe Betriebssysteme und KuVS (Oktober 2008; Garching).

  • (2008)
Contribution
  • Andreas Fischer
  • Andreas Berl
  • H. Meer

Virtualized Networks based on System Virtualization.

In: Proceedings of the 2nd GI/ITG KuVS Workshop on The Future Internet (November 2008; Karlsruhe).

  • (2008)
Contribution
  • S. Davy
  • C. Fahy
  • L. Griffin
  • Z. Boudjemil
  • Andreas Berl
  • Andreas Fischer
  • H. Meer
  • J. Strassner

Towards a Policy-Based Autonomic Virtual Network to Support Differentiated Security Services.

In: Proceedings of the International Conference on Telecommunications and Multimedia (TEMU 2008) [July 16-18, 2008; Ierapetra, Crete, Greece].

  • (2008)
Contribution
  • Andreas Berl
  • Andreas Fischer
  • H. Meer
  • A. Galis
  • J. Rubio-Loyola

Management of Virtual Networks.

In: Middleware technologies for enabling next generation network services and applications (Samos Island, Greece, 22nd - 26th September 2008; Proceedings of the 5th International Workshop on Next Generation Networking Middleware (NGNM 2008) and 4th IEEE/IFIP International Workshop on End-to-end Virtualization and Grid Management (EVGM 2008)). (Multicon lecture note series) pg. 197-202

  • Eds.:
  • T. Magedanz

Multicon Multimedia Consulting Schöneiche bei Berlin

  • (2008)
Journal article
  • Andreas Berl
  • Andreas Fischer
  • H. Meer

Using System Virtualization to Create Virtualized Networks.

In: Electronic Communications of the EASST (Workshops der Wissenschaftlichen Konferenz Kommunikation in Verteilten Systemen (WowKiVS2009) vol. 17 pg. 1-12

  • (2009)

show abstract

The method of system virtualization is very popular for the use in data centers and desktop virtualization today. In this work, system virtualization is applied to core network elements (routers and links) in order to create a virtualized network. The selection of this virtualization method crucially determines the emerging network model. The network model consists of virtual networks, virtual routers, and virtual links that form overlays on top of the physical network. The properties, features, and limitations of this network model are analyzed and described in this paper. Additionally, a proof of concept implementation using currently available technology and infrastructure is presented. Finally the dynamic configurability of virtual resources in such a system virtualization based virtualized network is evaluated.
Lecture
  • G. Lovász
  • Andreas Fischer
  • H. Meer

Network Virtualization and Energy Efficiency.

In: 9th Würzburg Workshop on IP: Joint ITG and Euro-NF Workshop on Visions of Future Generations Networks (EuroView 2009)

Universität Würzburg Würzburg, Germany University of Würzburg

  • 2009 (2009)
Contribution
  • J. Rubio-Loyola
  • J. Serrat
  • A. Astorga
  • Andreas Fischer
  • Andreas Berl
  • H. Meer
  • G. Koumoutsos

A Viewpoint of the Network Management Paradigm for Future Internet Networks.

In: Proceedings of the 1st IFIP/IEEE International Workshop on Management of the Future Internet (ManFI 2009). pg. 93-100

IEEE Press

  • (2009)
Contribution
  • A. Galis
  • S. Denazis
  • A. Bassi
  • P. Giacomin
  • Andreas Berl
  • Andreas Fischer
  • H. Meer
  • J. Strassner
  • S. Davy
  • D. Macedo
  • G. Pujolle
  • J. Rubio-Loyola
  • J. Serrat
  • L. Lefevre
  • A. Cheniour

Management Architecture and Systems for Future Internet Networks.

In: Towards the Future Internet. A European Research Perspective pg. 112-122

  • Eds.:
  • S. Krco
  • A. Gavras
  • J. Domingue
  • D. Hausheer
  • V. Lotz
  • A. Galis
  • T. Zahariadis
  • G. Tselentis

IOS Press Washington, DC

  • (2009)
ConferenceProceedings

Demonstrating Distributed Virtual Networks.

In: Lecture Notes in Computer Science (LNCS) vol. 6481

  • Eds.:
  • H. Meer
  • Andreas Fischer
  • A. Paler
  • R. Yahyapour
  • E. Di Nitto

Springer-Verlag

  • (2010)

show abstract

The current internet is commonly perceived as being too inflexible. Network Virtualization has been recognized lately as a method to overcome these limitations. Previous implementations of Virtual Network (VN) deployment software either provided limited access to the network layer (like PlanetLab), or focused only on specific application scenarios (like VNUML or VLAN). A first step toward the expected flexibility, with a solution that is both universal and thorough, is the implementation of network virtualization using system virtualization approaches [2].
Contribution
  • J. Rubio-Loyola
  • A. Astorga
  • J. Serrat
  • W. Chai
  • L. Mamatas
  • A. Galis
  • S. Clayman
  • A. Cheniour
  • L. Lefevre
  • O. Mornard
  • Andreas Fischer
  • A. Paler
  • H. Meer

Platforms and Software Systems for an Autonomic Internet.

In: Proceedings of the IEEE Global Communications Conference (IEEE GLOBECOM 2010) [Miami, FL, USA; December 6-10, 2010]. pg. 1-6

  • (2010)

DOI: 10.1109/GLOCOM.2010.5683380

show abstract

The current Internet does not enable easy introduction and deployment of new network technologies and services. This paper aims to progress the Future Internet (FI) by introduction of a service composition and execution environment that re-use existing components of access and core networks. This paper presents essential service-centric platforms and software systems that have been developed with the aim to create a flexible environment for an Autonomic Internet.
Journal article
  • Andreas Berl
  • Andreas Fischer
  • H. Meer

Virtualisierung im Future Internet - Virtualisierungsmethoden und Anwendungen.

In: Informatik-Spektrum vol. 33 pg. 186-194

  • (2010)

DOI: 10.1007/s00287-010-0420-z

show abstract

Das Future Internet stellt sich der Herausforderung, die Verknöcherung und die Unzulänglichkeiten des heutigen Internets zu überwinden. Es soll beispielsweise robust, zuverlässig und fehlertolerant sein und dabei Dienste auf energieeffiziente Weise erbringen. Zusätzlich müssen auch neue Anforderungen antizipiert werden, die zukünftige Dienste und Netze stellen, um der Entwicklung von neuen Diensten und Protokollen nicht im Wege zu stehen. Erforderlich ist hierbei die Entwicklung einer äußerst variablen und rekonfigurierbaren Netzwerkarchitektur, die eine möglichst einfache und autonome Netzwerkverwaltung mit einschließt. Die Virtualisierung von Host- und Netzwerkressourcen stellt in diesem Zusammenhang eine der Schlüsseltechnologien dar, mit der diese notwendige Flexibilität erreicht werden kann. Virtualisierung verbirgt die Komplexität von physikalischen Netzwerkinfrastrukturen und stellt homogene, flexible und dynamisch rekonfigurierbare virtuelle Ressourcen zur Verfügung. Dieser Artikel gibt einen Überblick über verschiedene Host- und Netzwerkvirtualisierungsmethoden. An mehreren Beispielen wird aufgezeigt, wie verschiedene Herausforderungen angegangen werden können, die sowohl im heutigen Internet als auch im Internet der Zukunft eine Rolle spielen.
Contribution
  • J. Rubio-Loyola
  • A. Astorga
  • J. Serrat
  • L. Lefevre
  • A. Cheniour
  • D. Muldowney
  • S. Davy
  • A. Galis
  • L. Mamatas
  • S. Clayman
  • D. Macedo
  • Z. Movahedi
  • G. Pujolle
  • Andreas Fischer
  • H. Meer

Managability of Future Internet Networks from a Practical Viewpoint.

In: Towards the Future Internet - Emerging Trends from European Research. pg. 105-114

  • Eds.:
  • S. Krco
  • E. Simperl
  • A. Gavras
  • V. Lotz
  • A. Galis
  • T. Zahariadis
  • B. Stiller
  • G. Tselentis

IOS Press

  • (2010)

DOI: 10.3233/978-1-60750-539-6-105

Contribution
  • J. Botero
  • X. Hesselbach
  • M. Duelli
  • D. Schlosser
  • Andreas Fischer
  • H. Meer

Flexible VNE Algorithms Analysis using ALEVIN.

In: Proceedings of the 11th Würzburg Workshop on IP: Joint ITG and Euro-NF Workshop "Visions of Future Generation Networks" (EuroView2011).

  • (2011)

show abstract

Network virtualization is recognized as an enabling technology for the Future Internet that overcomes network ossification. However, it introduces a set of challenges. In any network virtualization environment, the problem of optimally mapping virtual demands to physical resources, known as virtual network embedding (VNE), is a crucial challenge. This paper analyses the behaviour of the main algorithms proposed to solve VNE by means of the ALEVIN framework. The VNE algorithms are evaluated with regard to appropriate metrics such as: cost, revenue, and virtual network acceptance ratio. We also analyse the impact of the recently introduced hidden hop demand concept in the performance of the VNE algorithms.
Contribution
  • M. Duelli
  • D. Schlosser
  • J. Botero
  • X. Hesselbach
  • Andreas Fischer
  • H. Meer

VNREAL: Virtual Network Resource Embedding Algorithms in the Framework ALEVIN.

In: Proceedings of the 7th Euro-NF Conference on Next-Generation Internet (NGI 2011) [Kaiserslautern; June 27-29, 2011]. pg. 1-2

IEEE

  • (2011)

DOI: 10.1109/NGI.2011.5985874

show abstract

Network virtualization is recognized as an enabling technology for the Future Internet that overcomes network ossification. However, it introduces a set of challenges. In any network virtualization environment, the problem of optimally mapping virtual resources to physical resources, known as virtual network embedding (VNE), is a critical challenge. Several algorithms attempting to solve this problem have been proposed in literature, so far. However, comparison of existing and new VNE algorithms is hard, as each algorithm focuses on different criteria. To that end, the VNREAL project introduces ALEVIN, a framework to compare different algorithms according to a set of metrics, easily incorporate new VNE algorithms, and evaluate these algorithms on a given scenario for arbitrary parameters.
Contribution
  • A. Galis
  • S. Clayman
  • L. Lefevre
  • Andreas Fischer
  • H. Meer
  • J. Rubio-Loyola
  • J. Serrat
  • S. Davy

Towards In-Network Clouds in Future Internet.

In: The Future Internet. (Lecture Notes of Computer Science (LNCS)) vol. 6656 pg. 19-33

  • Eds.:
  • H. Müller
  • S. Krco
  • A. Gavras
  • J. Domingue
  • F. Cleary
  • M. Nilsson
  • V. Lotz
  • A. Galis
  • D. Lambert
  • T. Zahariadis
  • S. Avessta
  • B. Stiller
  • S. Karnouskos
  • H. Schaffers
  • P. Daras
  • F. Alvarez
  • M.-S. Li

Springer

  • (2011)

DOI: 10.1007/978-3-642-20898-0_2

show abstract

One of the key aspect fundamentally missing from the current Internet infrastructure is an advanced service networking platform and facilities, which take advantage of flexible sharing of available connectivity, computation, and storage resources. This paper aims to explore the architectural co-existence of new and legacy services and networks, via virtualisation of connectivity and computation resources and self-management capabilities, by fully integrating networking with cloud computing in order to create In-Network Clouds. It also presents the designs and experiments with a number of In-Network Clouds platforms, which have the aim to create a flexible environment for autonomic deployment and management of virtual networks and services as experimented with and validated on large-scale testbeds.
Journal article
  • Andreas Fischer
  • H. Meer

Position Paper: Secure Virtual Network Embedding.

In: Praxis der Informationsverarbeitung und Kommunikation vol. 34 pg. 190-193

  • (2011)

DOI: 10.1515/piko.2011.040

show abstract

Network virtualization has been recognized as an important technique to overcome the perceived ossification of the current Internet. Several variations of network virtualization have already been discussed in the literature. These approaches use virtualization to partition and/or combine physical network resources into virtual network resources. An actual deployment of virtual networks then requires the network operator to perform a mapping of virtual resources onto physical resources. The question of how this mapping can be performed in an optimal way is commonly known as the Virtual Network Embedding (VNE) problem. Several algorithms to solve this problem have been proposed already. These algorithms, however, focus on optimizing the use of resources with regard to performance. Security constraints to the VNE problem have not been investigated in depth, so far.
Journal article
  • J. Rubio-Loyola
  • A. Galis
  • A. Astorga
  • J. Serrat
  • L. Lefevre
  • Andreas Fischer
  • A. Paler
  • H. Meer

Scalable Service Deployment on Software Defined Networks.

In: IEEE Communications Magazine vol. 49 pg. 84-93

  • (2011)

DOI: 10.1109/MCOM.2011.6094010

show abstract

The network of the future will require a greater degree of service-awareness, and an optimal use of network resources. This paper presents the architectural design developed in the AutoI project for an open software-defined network infrastructure that enables the composition of fast and guaranteed services in an efficient manner and the execution of these services in an adaptive way taking into account better shared network resources provided by network virtualisation. Validation results are provided with special emphasis on service deployment scalability over virtualized network infrastructures.
Journal article
  • Andreas Fischer
  • J. Botero
  • M. Duelli
  • D. Schlosser
  • X. Hesselbach
  • H. Meer

ALEVIN - A Framework to Develop, Compare, and Analyze Virtual Network Embedding Algorithms.

In: Electronic Communications of the EASST vol. 37 pg. 1-12

EASST

  • (2011)

show abstract

Network virtualization is recognized as an enabling technology for the Future Internet. Applying virtualization of network resources leads to the problem of mapping virtual resources to physical resources, known as “Virtual Network Embedding” (VNE). Several algorithms attempting to solve this problem have been discussed in the literature, so far. However, comparison of VNE algorithms is hard, as each algorithm focuses on different criteria. To that end, we introduce a framework to compare different algorithms according to a set of metrics, which allow to evaluate the algorithms and compute their results on a given scenario for arbitrary parameters.
Contribution
  • Andreas Fischer
  • A. Fessi
  • G. Carle
  • H. Meer

Wide-Area Virtual Machine Migration as Resilience Mechanism.

In: Proceedings of the 1st International Workshop on Network Resilience: From Research to Practice (WNR 2011) [Madrid, Spain; October 4, 2011]. pg. 72-77

IEEE

  • (2011)

DOI: 10.1109/SRDSW.2011.16

show abstract

The resilience of services in the Internet has become an important issue and is expected to become even more important in the future. Virtualization is one of the means which can be deployed for resilience purposes. In this paper we follow a systematic approach to the use of virtualization to increase the resilience of network services. First, we provide an analysis of the potential failures of services running within Virtual Machines (VM) and how VM migration or replication can be used to address these failures. Then, we address the problem of re-establishing connectivity between a service and its clients upon successful migration, by leveraging results from mobility research. A special focus is given to wide-area VM migration, since it is considered as the solution for some difficult failures, e.g., large-scale failures due to natural disasters.
Contribution
  • G. Polyzos
  • G. Marias
  • S. Arkoulis
  • P. Frangoudis
  • M. Fiedler
  • A. Popescu
  • H. Meer
  • R. Herkenhoener
  • Andreas Fischer
  • J. Oberender

ASPECTS: Agile Spectrum Security.

In: Proceedings of the 7th Euro-NF Conference on Next-Generation Internet (NGI 2011) [Kaiserslautern; June 27-29, 2011]. pg. 1-2

IEEE

  • (2011)

DOI: 10.1109/NGI.2011.5985872

show abstract

The ASPECTS project considered an underlying dynamic spectrum access radio network, spectrum sensing technology and techniques, and spectrum sharing policies and rules and addressed vulnerabilities in all spectrum sharing phases, i.e. spectrum monitoring, negotiation, dissemination of access rules and sharing implementation. In the following sections, we describe our research objectives and our approach towards attaining them, present the main outcome of ASPECTS and conclude with a discussion on the relevance of our project to the Euro-NF vision towards the Future Internet.
Journal article
  • A. Paler
  • Andreas Fischer

vCPI – virtual Component Programming Interface.

In: Praxis der Informationsverarbeitung und Kommunikation vol. 34 pg. 142-143

  • (2011)

DOI: 10.1515/piko.2011.028

Journal article
  • J. Botero
  • X. Hesselbach
  • M. Duelli
  • D. Schlosser
  • Andreas Fischer
  • H. Meer

Energy Efficient Virtual Network Embedding.

In: IEEE Communications Letters vol. 16 pg. 756-759

  • (2012)

DOI: 10.1109/LCOMM.2012.030912.120082

show abstract

Waste of energy due to over-provisioning and overdimensioning of network infrastructures has recently stimulated the interest on energy consumption reduction by Internet Service Providers (ISPs). By means of resource consolidation, network virtualization based architectures will enable energy saving. In this letter, we extend the well-known virtual network embedding problem (VNE) to energy awareness and propose a mixed integer program (MIP) which provides optimal energy efficient embeddings. Simulation results show the energy gains of the proposed MIP over the existing cost-based VNE approach.
Contribution
  • M. Beck
  • Andreas Fischer
  • H. Meer

Distributed Virtual Network Embedding.

In: Proceedings of the 7th GI/ITG KuVS Workshop on Future Internet.

  • (2012)

show abstract

Network virtualization applies virtualization concepts to network resources in order to provide an abstraction from specific hardware. The added flexibility manifests itself in the possibility to create arbitrarily structured virtual networks that do not necessarily have to reflect the topology or the properties of the underlying physical network. The problem of mapping virtual resources to physical resources in such an environment is commonly known as ,,Virtual Network Embedding”, and almost all existing embedding algorithms do this in a centralized manner. The exploitation of distributed approaches aims to improve performance and to increase the level of reliability: stability and scalability of the whole system should be ensured. In this work, we present our ideas how to build such a new approach.
Journal article
  • G. Marias
  • J. Barros
  • M. Fiedler
  • Andreas Fischer
  • H. Hauff
  • R. Herkenhoener
  • A. Grillo
  • A. Lentini
  • L. Lima
  • C. Lorentzen
  • W. Mazurczyk
  • H. Meer
  • P. Oliveira
  • G. Polyzos
  • E. Pujol
  • K. Szczypiorski
  • J. Vilela
  • T.T.V. Vinhoza

Security and Privacy issues for the Network of the Future.

In: Security and Communication Networks vol. 5 pg. 987-1005

  • (2012)

DOI: 10.1002/sec.384

show abstract

The vision towards the Network of the Future cannot be separated from the fact that today’s networks, and networking services are subject to sophisticated and very effective attacks. When these attacks first appeared, spoofing and distributed denial-of-service attacks were treated as apocalypse for networking. Now, they are considered moderate damage, whereas more sophisticated and inconspicuous attacks, such as botnets activities, might have greater and far reaching impact. As the Internet is expanding to mobile phones and ‘smart dust’ and as its social coverage is liberalized towards the realization of ubiquitous computing (with communication), the concerns on security and privacy have become deeper and the problems more challenging than ever. Re-designing the Internet as the Network of the Future is self-motivating for researchers, and security and privacy cannot be provided again as separate, external, add-on, solutions. In this paper, we discuss the security and privacy challenges of the Network of the Future and try to delimit the solutions space on the basis of emerging techniques. We also review methods that help the quantification of security and privacy in an effort to provide a more systematic and quantitative treatment of the area in the future.
Journal article
  • J. Botero
  • X. Hesselbach
  • Andreas Fischer
  • H. Meer

Optimal mapping of virtual networks with hidden hops.

In: Telecommunications Systems: Special Issue on Future Internet Services and Architectures (Trends and Visions) vol. 51 pg. 273-282

  • (2012)

DOI: 10.1007/s11235-011-9437-0

show abstract

Network Virtualization has emerged as a so-lution for the Internet inability to address the requiredchallenges caused by the lack of coordination amongInternet service providers for the deployment of newservices. The allocation of resources is one of the mainproblems in network virtualization, mainly in the map-ping of virtual nodes and links to specific substratenodes and paths, also known as the virtual networkembedding problem. This paper proposes an algorithmbased on optimization theory, to map the virtual linksand nodes requiring a specific demand, looking for themaximization of the spare bandwidth and spare CPUin the substrate network, taking into account the band-width demanded by thehidden hopswhen a virtuallink is mapped. The components of the virtual networks(nodes and links) that do not ask for an specific demandare then allocated following a fairness criteria.
Contribution
  • M. Beck
  • Andreas Fischer
  • H. Meer
  • J. Botero
  • X. Hesselbach

A distributed, parallel, and generic virtual network embedding framework.

In: IEEE International Conference on Communications (ICC 2013) [Budapest, Hungary; June 9-13, 2013]. pg. 3471-3475

  • (2013)

DOI: 10.1109/ICC.2013.6655087

Contribution
  • Andreas Fischer
  • M. Beck
  • H. Meer

An Approach to Energy-efficient Virtual Network Embeddings.

In: 2013 IFIP/IEEE International Symposium on Integrated Network Management (IM 2013) [Ghent, Belgium; May 27-31, 2013]. pg. 1142-1147

IEEE

  • (2013)

show abstract

Network Virtualization is recognized as a key technology for the Future Internet. Energy-efficiency is one of the main challenges in future networking environments. Most algorithms for mapping virtual resources to substrate resources however do not consider energy as a factor for the mapping. In order to evaluate the energy-efficiency of such a mapping, an energy model and appropriate energy-aware metrics are needed. This paper discusses how an algorithm can be modified to take energy-efficiency into account. The modified algorithm is then evaluated, showing that energy-efficiency can be increased with only a minor impact on embedding quality regarding other metrics.
Journal article
  • Andreas Berl
  • M. Niedermeier
  • Andreas Fischer
  • D. Hutchinson
  • H. Meer

Virtual Energy Information Network: A Resilience Perspective.

In: e&i Elektrotechnik & Informationstechnik (Special Issue on ’Smart Grid and Security’) vol. 130 pg. 121-126

  • (2013)

show abstract

Increasing demand in energy consumption, missed modernisations, and the increasing difficulties in predicting power production due to volatile renewable energy sources (e.g., based on wind or sun) impose major challenges to the power grid. Power supply and power demand are closely interconnected with the need to maintain the power grid in a stable state with a sufficient quality of power. This requires energy-relevant information to be exchanged through the so called Energy Information Network. Communication, however, is challenging within the Energy Information Network due to privacy, security, resiliency, and quality-of-service requirements. Particularly, the resilience of communication within the Energy Information Network needs to be considered to maintain the power grid in a stable and controlled state. This paper suggests a Virtualised Energy Information Network (VEIN), where the Energy Information Network is divided into multiple virtual networks that run over a common substrate network. Furthermore, this paper discusses benefits of this approach in terms of privacy, security, and resilience and points out open research questions.
Journal article
  • Andreas Fischer
  • J. Botero
  • M. Beck
  • H. Meer
  • X. Hesselbach

Virtual Network Embedding: A Survey.

In: IEEE Communications Surveys and Tutorials vol. 15 pg. 1888-1906

  • (2013)

DOI: 10.1109/SURV.2013.013013.00155

show abstract

Network virtualization is recognized as an enabling technology for the future Internet. It aims to overcome the resistance of the current Internet to architectural change. Application of this technology relies on algorithms that can instantiate virtualized networks on a substrate infrastructure, optimizing the layout for service-relevant metrics. This class of algorithms is commonly known as “Virtual Network Embedding (VNE)” algorithms. This paper presents a survey of current research in the VNE area. Based upon a novel classification scheme for VNE algorithms a taxonomy of current approaches to the VNE problem is provided and opportunities for further research are discussed.
Contribution
  • M. Beck
  • Andreas Fischer
  • F. Kokot
  • C. Linnhoff-Popien
  • H. Meer

A Simulation Framework for Virtual Network Embedding Algorithms.

In: Proceedings of the 16th International Telecommunications Network Strategy and Planning Symposium (Networks 2014) [Funchal, Madeira Island, Portugal; September 17-19, 2014]. pg. 1-6

IEEE

  • (2014)

DOI: 10.1109/NETWKS.2014.6959238

show abstract

Network virtualization is seen as an enabling technology for the Future Internet. In this context, many Virtual Network Embedding algorithms have been introduced in literature so far. This paper discusses an open source framework for the evaluation of such algorithms. The paper describes features provided by the framework, how to use the framework for evaluating these algorithms, and how to extend the software with respect to novel algorithms and simulation scenarios. Lessons learned are presented, describing how the software evolved towards a mature and highly extensible simulation framework.
Contribution
  • J. Benze
  • Andreas Berl
  • K. Daniel
  • G. Eibl
  • D. Engel
  • Andreas Fischer
  • U. Hofmann
  • A. Kießling
  • S. Köpsell
  • L. Langer
  • H. Meer
  • C. Neureiter
  • M. Niedermeier
  • T. Pfeiffenberger
  • M. Pietsch
  • A. Veichtlbauer

VDE-Positionspapier Energieinformationsnetze und -systeme (Smart Grid Security).

In: VDE-Kongress 2014 Smart Cities (Intelligente Lösungen für das Leben in der Zukunft, Kongressbeiträge 20./21.10.2014, Frankfurt/Main, Messe).

  • Eds.:
  • VDE Verband der Elektrotechnik Elektronik Informationstechnik e.V.

VDE VERLAG Berlin

  • (2014)
Contribution
  • B. Taubmann
  • H. Reiser
  • T. Kittel
  • Andreas Fischer
  • W. Mandarawi
  • H. Meer

CloudIDEA: Cloud Intrusion Detection, Evidence Preservation and Analysis.

In: Proceedings of the Tenth European Conference on Computer Systems (EuroSys 2015) [Bordeaux, France; April 21-24, 2015].

  • (2015)

Contribution
  • A. Cervero
  • M. Chincoli
  • L. Dittmann
  • Andreas Fischer
  • A. Garcia
  • J. Galan-Jimenez
  • L. Lefevre
  • H. Meer
  • T. Monteil
  • P. Monti
  • A.-C. Orgerie
  • L.-F. Pau
  • C. Phillips
  • S. Ricciardi
  • R. Sharrock
  • P. Stolf
  • T. Trinh
  • L. Valcarenghi

Green Wired Networks.

In: Large-Scale Distributed Systems and Energy Efficiency: A Holistic View. (Wiley Series on Parallel and Distributed Computing) pg. 39-77

  • Eds.:
  • J.-M. Pierson

Wiley

  • (2015)
Journal article
  • M. Beck
  • Andreas Fischer
  • J. Botero
  • C. Linnhoff-Popien
  • H. Meer

Distributed and scalable embedding of virtual networks.

In: Journal of Network and Computer Applications vol. 56 pg. 124-136

  • (2015)

DOI: 10.1016/j.jnca.2015.06.012

show abstract

Abstract Network virtualization is widely regarded as a key technology for the Future Internet, enabling the deployment of new network protocols without changing dissimilar hardware devices. This leads to the problem of mapping virtual demands to physical resources, known as Virtual Network Embedding (VNE). Current VNE algorithms do not scale with respect to the substrate network size. Therefore, these algorithms are not applicable in large-scale scenarios where virtual networks have to be embedded in a timely manner. This paper discusses DPVNE, a Distributed and Generic VNE framework: It runs cost-oriented centralized embedding algorithms in a distributed way, spreading workload across the substrate network instead of concentrating it on one single node (as centralized algorithms do). Several state-of-the-art algorithms were evaluated running inside the DPVNE framework. Results show that DPVNE leads to runtime improvements in large-scale scenarios and embedding results are kept comparable.
Contribution
  • Andreas Fischer
  • T. Kittel
  • B. Kolosnjaji
  • T. Lengyel
  • W. Mandarawi
  • H. Reiser
  • B. Taubmann
  • E. Weishäupl
  • H. Meer
  • T. Müller
  • M. Protsenko

CloudIDEA: A Malware Defense Architecture for Cloud Data Centers.

In: Proceedings of the 5th International Symposium on Cloud Computing, Trusted Computing and Secure Virtual Infrastructures (C&TC 2015) [Rhodes, Greece; October 26-28, 2015] .

  • (2015)
Contribution
  • W. Mandarawi
  • Andreas Fischer
  • H. Meer
  • E. Weishäupl

QoS-Aware Secure Live Migration of Virtual Machines.

In: Proceedings of the 2nd Workshop on Security in Highly Connected IT Systems (SHCIS 2015).

  • (2015)
Patent
  • A. Houyou
  • Andreas Fischer
  • W. Mandarawi
  • H. Meer
  • H.-P. Huth

Device and method for allocating communication resources in a system employing network slicing.

  • 29.09.2016 (2016)

Contribution
  • W. Mandarawi
  • Andreas Fischer
  • A. Houyou
  • H.-P. Huth
  • H. Meer

Constraint-Based Virtualization of Industrial Networks.

In: Principles of Performance and Reliability Modeling and Evaluation: Essays in Honor of Kishor Trivedi on his 70th Birthday. (Springer Series in Reliability Engineering) pg. 567-586

  • Eds.:
  • A. Puliafito
  • L. Fiondella

Springer Cham

  • (2016)

DOI: 10.1007/978-3-319-30599-8_22

Contribution
  • Andreas Fischer
  • R. Kühn
  • W. Mandarawi
  • H. Meer

Modeling Security Requirements for VNE algorithms.

In: Proceedings of VALUETOOLS 2016/10th EAI International Conference on Performance Evaluation Methodologies and Tools (Taormina, Italy; October 25-28, 2016).

  • (2016)

show abstract

Public and private Infrastructure as a Service (IaaS) clouds are widely used by individuals and organizations to provision flexible virtual computing resources on demand. Virtual Network Embedding (VNE) algorithms are employed in this context to provide an automated resource assignment. With multiple involved parties security-aware Virtual Machine (VM) placement becomes highly relevant for production environments. Moreover, VNE algorithms should also consider the security requirements of the interconnections between VMs, thereby extending the problem to networks. This paper discusses security requirements of Virtual Networks (VNs) and shows how they can be modeled in VNE to map them to the provided security mechanisms in the physical network. The paper also presents an implementation of this security-aware VNE model in the public simulation platform ALEVIN, demonstrating the applicability with a realistic use case of such a model.
Journal article
  • Andreas Fischer
  • H. Meer

Generating Virtual Network Embedding Problems with Guaranteed Solutions.

In: IEEE Transactions on Network and Service Management vol. 13 pg. 1-14

IEEE

  • (2016)

DOI: 10.1109/TNSM.2016.2596802

show abstract

The efficiency of network virtualization depends on the appropriate assignment of resources. The underlying problem, called Virtual Network Embedding, has been much discussed in the literature, and many algorithms have been proposed, attempting to optimize the resource assignment in various respects. Evaluation of those algorithms requires a large number of randomly generated embedding scenarios. This paper presents a novel scenario generation approach and demonstrates how to produce scenarios with a guaranteed exact solution, thereby facilitating better evaluation of embedding algorithms.
Thesis
  • Andreas Fischer

An Evaluation Methodology for Virtual Network Embedding.

Universität Passau Passau Fakultät für Informatik und Mathematik

  • 2017 (2017)

show abstract

The increasing scale and complexity of computer networks imposes a need for highly flexible management mechanisms. The concept of network virtualization promises to provide this flexibility. Multiple arbitrary virtual networks can be constructed on top of a single substrate network. This allows network operators and service providers to tailor their network topologies to the specific needs of any offered service. However, the assignment of resources proves to be a problem. Each newly defined virtual network must be realized by assigning appropriate physical resources. For a given set of virtual networks, two questions arise: Can all virtual networks be accommodated in the given substrate network? And how should the respective resources be assigned? The underlying problem is commonly known as the Virtual Network Embedding problem. A multitude of algorithms has already been proposed, aiming to provide solutions to that problem under various constraints. For the evaluation of these algorithms typically an empirical approach is adopted, using artificially created random problem instances. However, due to complex effects of random problem generation the obtained results can be hard to interpret correctly. A structured evaluation methodology that can avoid these effects is currently missing. This thesis aims to fill that gap. Based on a thorough understanding of the problem itself, the effects of random problem generation are highlighted. A new simulation architecture is defined, increasing the flexibility for experimentation with embedding algorithms. A novel way of generating embedding problems is presented which migitates the effects of conventional problem generation approaches. An evaluation using these newly defined concepts demonstrates how new insights on algorithm behavior can be gained. The proposed concepts support experimenters in obtaining more precise and tangible evaluation data for embedding algorithms.
Contribution
  • Andreas Fischer

Performance Evaluation for Service Function Chains through Automated Model Building.

In: Proceedings of the 11th EAI International Conference on Performance Evaluation Methodologies and Tools (ValueTools2017) [Venice, Italy; December 5-7, 2017].

  • (2017)

Contribution
  • Andreas Fischer

Generating Random Service Function Chain Embedding Problems.

In: IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN’17) [Berlin; November 6-8, 2017].

  • (2017)

DOI: 10.1109/NFV-SDN.2017.8169834

show abstract

The combination of Software Defined Networking (SDN) and Network Function Virtualization (NFV) promises to provide highly flexible and configurable network infrastructures. This relies, however, on an efficient assignment of the respective Service Function Chain (SFC). This is related to Virtual Network Embedding (VNE), where algorithms are devised to provide such an assignment. To evaluate and compare the efficiency of such algorithms, well-designed embedding problems have to be generated. This paper presents a new mechanism for generating embedding problems: Problems are generated from a given set of SFCs such that each generated problem is known in advance to have an optimal solution. Experimenters can use this approach to investigate specific properties of embedding algorithms. The approach, thereby, facilitates more detailed evaluation.
Contribution
  • Andreas Fischer
  • A. Paler

On the analogy between quantum circuit design automation and virtual network embedding.

In: Proceedings of 34th ACM/SIGAPP Symposium on Applied Computing (SAC'19) [April 8-12, 2019; Limassol, Cyprus]. pg. 1378-1383

ACM New York, NY, USA

  • (2019)

DOI: 10.1145/3297280.3297419

show abstract

Quantum computing and network virtualisation share more similarities than it would seem on first thought: both reach their potential by using the underlying hardware as efficiently as possible - a classical computer science problem. Quantum computing research has already been focusing on key problems related to the scarcity of the quantum hardware, hoping that once scalable quantum computers will be available, quantum circuit design automation methods (QCDA) are mature. Consequently, a significant methodological apparatus exists for a technology envisioned to scale in the future. Likewise, the optimisation of resource assignment for virtual networks has received much attention in the literature. As such, any scalability improvement of virtual network embedding (VNE) would be of high practical importance in the present. There is an interesting and promising relationship between VNE and QCDA. This work introduces the possibility to use QCDA methodology for VNE, and vice versa. VNE problem instances can be modeled using the quantum circuit formalism, and we offer some basic examples. To the best of our knowledge, this is the first work that highlights this analogy.
Contribution
  • Andreas Fischer
  • D. Bhamare
  • Andreas Kassler

On the Construction of Optimal Embedding Problems for Delay-Sensitive Service Function Chains.

In: Proceedings of the 28th International Conference on Computer Communications and Networks (ICCCN 2019) [July 29-August 1, 2019; Valencia, Spain]. pg. 1-10

  • (2019)
Contribution
  • R. Kühn
  • Andreas Fischer
  • H. Meer

Modeling Security Requirements for VNE Algorithms: A Practical Approach.

In: Systems Modeling: Methodologies and Tools. (EAI/Springer Innovations in Communication and Computing) pg. 165-179

  • Eds.:
  • A. Puliafito
  • K. Trivedi

Springer International Publishing, Imprint: Springer Cham

  • (2019)
Journal article
  • Marco Kretschmann
  • Andreas Fischer
  • Benedikt Elser

Extracting Keywords from Publication Abstracts for an Automated Researcher Recommendation System.

In: Digitale Welt (Proceedings of the First International Symposium on Applied Artificial Intelligence in Conjunction with DIGICON) vol. 4 pg. 20-25

  • (2020)

DOI: 10.1007/s42354-019-0227-2

show abstract

This paper presents an automated keyword assignment system for scientific abstracts. That system is applied to paper abstracts collected in a local publication database and used to drive a researcher recommendation system. Problems like low data volume and missing keywords are discussed. For remediation, training is performed on an extended data set based on large online publication databases. Additionally a closer look at label imbalance in the dataset is taken. Ten multi-label classification algorithms for assigning keywords from a given catalogue to a scientific abstract are compared. The usage of binary relevance as transformation method with LightGBM as classifier yields the best results. Random oversampling before the training phase additionally increases the F1-Score by around 5-6%.
Lecture
  • Andreas Fischer

Reflections on Network Resource Assignment. Invited Talk.

In: Colloquium Series/Workshop at the Computer Science Department

Universität Karlstad Karlstad

  • 06.02.2020 (2020)

show abstract

The efficient assignment of network resources has been discussed for several years, but continues to be an interesting topic. Initially sparked by the option to virtualize entire communication networks, the problem of assigning dedicated node and link resources has quickly been found to be multi-faceted and computationally hard to solve. The rise of cloud computing and energy-efficiency concerns strengthened the need for efficient resource assignment algorithms. The current drive for SDN and NFV infrastructures brought a new dimension to the problem. This talk will discuss the current state of the art and future research directions in this area.
Contribution
  • Andreas Fischer
  • M. Amesberger

Improving image tracing with artificial intelligence.

In: Proceedings of the 2021 11th International Conference on Advanced Computer Information Technologies (ACIT) [September 15-17, 2021; Deggendorf]. pg. 714-717

  • Eds.:
  • Institute of Electrical and Electronics Engineers Inc.

  • (2021)

DOI: 10.1109/ACIT52158.2021.9548635

show abstract

Image tracing describes the task to convert a raster image into a vector format. This paper investigates whether dimensionality reduction techniques can improve the results of image tracing by producing simpler vector graphic files while keeping image quality reasonably high. In particular, an Autoencoding Neural Network and Principal Component Analysis are investigated as pre-processing steps before performing the actual tracing. Results indicate that using an Autoencoder as pre-processing step before image tracing can reduce SVG file complexity by over 70% with acceptable impact on image quality.
Journal article
  • M. Welzl
  • S. Islam
  • M. Gundersen
  • Andreas Fischer

Transport Services: A Modern API for an Adaptive Internet Transport Layer.

In: IEEE Communications Magazine vol. 59 pg. 16-22

  • (2021)

DOI: 10.1109/MCOM.001.2000870

show abstract

Transport services (TAPS) is a working group of the the Internet Engineering Task Force (IETF). TAPS defines a new recommended API for the Internet's transport layer. This API gives access to a wide variety of services from various protocols, and it is protocol-independent: the transport layer becomes adaptive, and applications are no longer statically bound to a particular protocol and/or network interface. We give an overview of the TAPS API, and we demonstrate its flexibility and ease of use with an example using a Python-based open source implementation.
Lecture
  • Andreas Fischer

Was ist eigentlich KI?.

In: Innovationssymposium „KI in der Glasindustrie"

Technische Hochschule Deggendorf (THD)/TAZ Spiegelau Zoom

  • 08.07.2021 (2021)
Lecture
  • Andreas Fischer

Tinder für Forscher und Cat Content: Zwei interne Forschungsprojekte.

In: 12. DACHS-Symposium für Lehre und angewandte Forschung in Informatik und Wirtschaftsinformatik

Deggendorf

  • 13.09.2021 (2021)
Lecture
  • Andreas Fischer

Was ist eigentlich KI?.

In: 1. GlasTAOO4.0 Symposium der Glas-Technologie-Allianz Oberfranken-Ostbayern

Universität Bayreuth Bayreuth

  • 30.09.2021 (2021)
Contribution
  • Zineddine Bettouche
  • Andreas Fischer

Mapping Researcher Activity based on Publication Data by means of Transformers.

In: Proceedings of the II Interdisciplinary Conference on Mechanics, Computers and Electrics (ICMECE 2022).

  • (2022)
Contribution
  • Andreas Fischer
  • Zineddine Bettouche

High-pass Filters Preprocessing in Image Tracing with Convolutional Autoencoders.

In: COMPUTATION TOOLS 2022 : Proceedings of The Thirteenth International Conference on Computational Logics, Algebras, Programming, Tools, and Benchmarking. pg. 7-12

  • (2022)

show abstract

Image tracing describes the task of converting a raster image into a vector format. This paper investigates different processing pipelines that can extract an abstract representation of an image by means of high-pass filtering, autoencoding, and vectorization. Results indicate that reconstructing an image using Autoencoders, then filtering it with high-pass filters, and finally vectorizing it, can represent the image more abstractly while improving the efficiency of the vectorization process.
Contribution
  • M. Hollmer
  • Andreas Fischer

Hands-on detection for steering wheels with neural networks.

In: Proceedings of the II Interdisciplinary Conference on Mechanics, Computers and Electrics (ICMECE 2022).

  • (2022)
Journal article
  • Zineddine Bettouche
  • Andreas Fischer

Improving Image Tracing with Convolutional Autoencoders by High-Pass Filter Preprocessing.

In: International Journal on Advances in Software (IARIA Journals) vol. 15 pg. 141 - 151

  • (2022)

DOI: 10.48550/arXiv.2306.09039

show abstract

The process of transforming a raster image into a vector representation is known as image tracing. This study looks into several processing methods that include high-pass filtering, autoencoding, and vectorization to extract an abstract representation of an image. According to the findings, rebuilding an image with autoencoders, high-pass filtering it, and then vectorizing it can represent the image more abstractly while increasing the effectiveness of the vectorization process.
Lecture
  • Andreas Fischer

Pitfalls in Empirical Evaluation of Algorithms. Keynote.

In: ComputationWorld 2022 Congress

Barcelona, Spain

  • 24.04.2022 (2022)
Lecture
  • Andreas Fischer

Erstellen einer Forschungslandkarte der TH Deggendorf.

In: 13. DACHS-Symposium

Berner Fachhochschule Biel, Schweiz

  • 12.09.2022 (2022)
Contribution
  • Zineddine Bettouche
  • Johannes Reisinger
  • Andreas Fischer

Analysis of Features-Brightness Bias in Computer Vision.

In: 2023 13th International Conference on Advanced Computer Information Technologies (ACIT). pg. 556-561

IEEE

  • (2023)

DOI: 10.1109/ACIT58437.2023.10275561

show abstract

This paper explores the impact of mask brightness on various computer vision techniques. We conducted experiments on multiple applications, including reconstruction, generation, object recognition, and vectorization. Darker features relative to the background resulted in improved performance for autoencoders, Potrace vectorization, and GAN training stability. However, this relationship did not hold for Mask R-CNN and YOLO. Nonetheless, the brightness difference between masks and backgrounds influenced object recognition confidence. Our findings highlight the importance of considering brightness variations to avoid biases and provide insights for further research
Journal article
  • Zineddine Bettouche
  • Andreas Fischer

Topical Clustering of Unlabeled Transformer-Encoded Researcher Activity.

In: Bavarian Journal of Applied Sciences pg. 504-525

  • (2023)

DOI: 10.25929/1rjp-d197

show abstract

Transformer models have the ability to understand the meaning of text efficiently through the use of self-attention mechanisms. We investigate the bundled meanings in clusters of transformer-generated embeddings by evaluating the topical clustering accuracy of the unlabeled scientific papers of the DIT publications database. After experimenting with SciBERT and German-BERT, we focus on mBERT as we work with multilingual papers. We create a landscape representation of the scientific fields with active research through the encoding and clustering of research publications. With the absence of topic labels in the data (no ground truth), the clustering metrics cannot evaluate the accuracy of the topical clustering. Therefore, we make use of the coauthorship aspect in the papers to perform a coauthorship analysis in two parts: the investigation of the authors’ uniqueness in each cluster and the construction of coauthorship-based social networks. The calculated high uniqueness of authors in the formed clusters and the found homogeneity of topics across the connected components (in social networks) imply an accurate topical clustering of our encodings. Moreover, the constructed social networks indicate the existence of a set of connecting internal authors, whose collaborations with each other formed a large network, holding 74% of all papers in the database.
Lecture
  • Andreas Fischer

Using Transformers Beyond Natural Language Processing.

In: KI Campus Ostbayern: Grenzgänger und Brückenbauer

Ostbayerische Technische Hochschule (OTH) Regensburg Regensburg

  • 28.06.2023 (2023)
Lecture
  • Andreas Fischer

Erstellen einer Forschungslandkarte an der THD - eine praktische Anwendung moderner Large Language Models.

In: IT-Forum Niederbayern

Landshut

  • 13.07.2023 (2023)
Lecture
  • Andreas Fischer

ChatGPT: Probleme und Chancen.

In: Rotary Club Deggendorf

Deggendorf

  • 15.11.2023 (2023)
Lecture
  • Andreas Fischer

AI4Leaders: Wie kann KI den Arbeitsalltag von Führungskräften unterstützen?.

In: Führungskräfteseminar Webasto

Hengersberg

  • 16.11.2023 (2023)
Lecture
  • Andreas Fischer

Erstellen von Forschungslandkarten auf Basis der Publikationen eines Teams.

In: Teambuilding des Lehrstuhls für Rechnernetze und Rechnerkommunikation der Universität Passau

Passau

  • 24.11.2023 (2023)
NewspaperArticle
  • Andreas Fischer

Die schöne neue KI-Welt.

In: Alt und Jung Metten vol. 90 pg. 282-286

Metten

  • 2024 (2024)
Contribution
  • Zineddine Bettouche
  • A. Safi
  • Andreas Fischer

Contextual Categorization Enhancement through LLMs Latent-Space.

In: ThinkMind COMPUTATION TOOLS 2024: The Fifteenth International Conference on Computational Logics, Algebras, Programming, Tools, and Benchmarking. pg. 6-11

  • (2024)

show abstract

Managing the semantic quality of the categorization in large textual datasets, such as Wikipedia, presents significant challenges in terms of complexity and cost. In this paper, we propose leveraging transformer models to distill semantic information from texts in the Wikipedia dataset and its associated categories into a latent space. We then explore different approaches based on these encodings to assess and enhance the semantic identity of the categories. Our graphical approach is powered by Convex Hull, while we utilize Hierarchical Navigable Small Worlds (HNSWs) for the hierarchical approach. As a solution to the information loss caused by the dimensionality reduction, we modulate the following mathematical solution: an exponential decay function driven by the Euclidean distances between the high-dimensional encodings of the textual categories. This function represents a filter built around a contextual category and retrieves items with a certain Reconsideration Probability (RP). Retrieving high-RP items serves as a tool for database administrators to improve data groupings by providing recommendations and identifying outliers within a contextual framework.
Contribution
  • Johannes Reisinger
  • Andreas Fischer
  • Christoph Goller

The Importance of Clustering in Word-Sense Induction.

In: 2024 14th International Conference on Advanced Computer Information Technologies (ACIT). pg. 735-738

IEEE

  • (2024)

DOI: 10.1109/ACIT62333.2024.10712629

show abstract

This electronic document is a “live” template and already We propose a novel spatial clustering score to evaluate word-sense clusters in for word-sense induction. The score is then evaluated using a basic word-sense induction-pipeline, containing a generally trained Word2Vec model and few selected adjustments, focused on cluster evaluation. We overhauled the former parameter score and propose the novel spatial clustering score to evaluate density-based spatial clujstering of applications with noise clusters and provide automatic parameter selection. Our method was evaluated on the SemEval 2010 Task 14 dataset, achieving results en-par with the state-of-the-art
Contribution
  • Anjali Singh
  • Zineddine Bettouche
  • Andreas Fischer

Synthetic Training-Data Generation for ML-based Process Mining Tools.

In: 2024 14th International Conference on Advanced Computer Information Technologies (ACIT). pg. 705-709

IEEE

  • (2024)

DOI: 10.1109/ACIT62333.2024.10712516

show abstract

This work addresses the challenge of data scarcity in process mining by proposing the creation of synthetic training data using generative models. A comparative analysis is conducted between a Long Short-Term Memory (LSTM) model and the Generative Adversarial Network (GAN) model, using two distinct datasets. Multiple evaluation methods are employed to compare the results from the two models based on: precision, fidelity, diversity, and novelty. Results indicate that while LSTM accurately reproduces the initial data structure, GAN introduces more variability, offering a wider range of training scenarios. This highlights the potential of GAN-generated data to enhance the effectiveness and reliability of machine learning-based process mining tools
Lecture
  • Andreas Fischer

Generative KI und ihr Einsatz im Unterricht.

In: Treffen des Arbeitskreises benediktinischer Schulen in Bayern

St.-Michaels-Gymnasium Metten Metten

  • 24.01.2024 (2024)
Lecture
  • Andreas Fischer

Open Data and data management infrastructure in Bavaria. Expert training (I).

In: Workshop on Open Science and Open Science Infrastructure Development (ERASMUS+-Projekt "Towards Open Science Communitites development in Sub-Saharan Africa Region" [OSCAR 2.0])

Technische Hochschule Deggendorf Deggendorf

  • 04.11.2024 (2024)
Journal article
  • Zineddine Bettouche
  • Khalid Ali
  • Andreas Fischer
  • Andreas Kassler

Enhancing Spatiotemporal Networks with xLSTM: A Scalar LSTM Approach for 5G Traffic Forecasting.

In: Online-Publikationssystem der Universität Tübingen

Eberhard Karls Universität Tübingen

  • (2025)

DOI: 10.15496/publikation-105099

show abstract

Accurate spatiotemporal traffic forecasting is vital for optimizing 5G networks. Traditional LSTM models struggle with capturing complex spatiotemporal dependencies, limiting predictive performance. To address this, we propose an enhanced Spatiotemporal Network (STN) integrating Scalar LSTM (sLSTM), a more efficient variant designed to improve temporal modeling while reducing computational complexity. Our dualpath STN processes the input through an sLSTM for sequential feature extraction and a three-layer Conv3D path for spatial feature learning, with both outputs fused in a dedicated fusion layer for enhanced spatiotemporal representation. By incorporating sLSTM, our model stabilizes gradients, accelerates convergence, and enhances accuracy. Experiments on real-world mobile traffic datasets show a 23% MAE reduction over ConvLSTM, with a 30% improvement on unseen data, demonstrating superior generalization for 5G traffic prediction.
Book
  • Özgür Kaynak
  • Andreas Kassler
  • Andreas Fischer
  • O. Dobrijevic
  • H. Chahed

TSN Scheduling Robust to Wireless Performance Uncertainties: A Problem and Model Definition.

Universität Würzburg Würzburg

  • (2025)

show abstract

Time-sensitive Networking (TSN) is a set of extensions to the Ethernet standard for providing deterministic communication services over a converged network infrastructure. A key element thereby is meeting end-to-end latency requirements, which can be achieved for wired links by properly configuring TSN endpoints and switches using IEEE 802.1AS time synchronization and an IEEE 802.1Qbv time-aware shaper. However, wireless links commonly exhibit performance uncertainties, which introduce additional challenges for deterministic communication. This paper proposes a new method for configuring TSN networks with wireless links, focusing on cyclically scheduled traffic. We design a linear program that synthesizes TSN configurations robust to performance uncertainties of wireless links, by adjusting the transmission schedule at the first wired network node. Moreover, we develop an opportunistic version of the scheduling when the senders transmit multiple frames within their sending interval.

projects

SEMIARID: Natural Language Semantic Search in Big Data; KIGA: AI-based analysis of business processes

labs

Network Lab DEGGs 1.4

Vita

  • since 2020: CIO of the Deggendorf Institute of Technology
  • since 2017: Professor in “Computer Science for Engineers”, Deggendorf Institute of Technology
  • 2017: Postdoctoral researcher, Karlstad University
  • 2017: PhD graduation (Dr. rer. nat.), Universität Passau
  • 2008-2017: Research Associate, Universität Passau
  • 2008: Graduate as “Diplom Informatiker” (Master-equivalent), Universität Passau

Other

Please find further information on my personal homepage: https://andreas.fischer-family.online/