Panagiotis Gouvas

Ubitech, Greece

Talk title: Deploying and orchestrating modern cloud-native applications in the programmable 5G infrastructure   watch presentation

Jaafar Elmirghani

University of Leeds, UK

Talk title: Network Intelligence and Energy Efficient Resources Provisioning in 6G   watch presentation

Talk abstract:

Network intelligence is starting to enable unprecedented levels of flexibility, adaptation and efficient resource utilization in communication networks. It can also enable optimized user centric services by learning not only about the network, but also about the user. The added network intelligence has implications however in terms of power consumption in future 6G networks and beyond. This talk draws on the work carried out in the EPSRC Intelligent Energy Aware Networks (INTERNET) project and the EPSRC Terabit Optical Wireless Systems for 6G (TOWS) project. In particular, it casts the problem of energy efficient resource provisioning in 6G into an optimization framework considering wireless and access networks, IoT networks, core networks and data centre networks. For example, attention is given to the use of machine learning algorithms and their use to understand the user instantaneous requirements as well as the network conditions, thus matching the two in a healthcare application. Attention is also given to energy efficient network slicing and virtual machine placement in cloud - mobile edge computing architectures. Energy efficiency in a service oriented IoT framework is presented together with energy efficient future data centre disaggregation. Test-bed implementations are outlined and Future directions are proposed.

Ingrid Moerman

imec - Ghent University

Talk title: How ORCA solutions meet beyond-5G evolution   watch presentation

Talk abstract:

The H2020 ORCA project (https://www.orca-project.eu) offers experimentation facilities to promote wireless innovation especially focusing on SDR. The goal of the ORCA project is to enable wireless experimenters to unlock the potential of reconfigurable radio technology, by setting up complex experiments that require control of multiple novel technologies or cooperation between multiple networked SDR platforms with 5G or beyond 5G performance constraints on latency or throughput, well before new radio technologies become available on the market in commercial off-the-shelf products. The ORCA project further aims to bridge Software-Defined Radio (SDR) with Software-Defined Networking (SDN) technology, enabling the creation of parallel on demand wireless network slices, each slice optimized for a specific set of requirements.
Some recent trends in software defined radio research that will be highlighted are:
- Solving the dilemma between software, associated with high flexibility and short development cycle, and hardware, associated with high performance and long development cycle, through softwarization;
- Offering QoS-aware end-to-end wireless services with very diverse requirements through softwarization and virtualisation techniques, not only suitable for powerful and centralised core/cloud/edge infrastructures, but also in end devices;
- Simplifying standardisation through hierarchical orchestration, a technology neutral end-to-end approach for managing different networks across heterogeneous networks.

Antonio Manzalini

TIM (Chief Innovation & Partnership Office / Technology Innovation)

Talk title: Pervasive Artificial Intelligence in 5G and beyond scenarios   watch presentation

Talk abstract:

MEC/Edge Computing aims at allocating part of the service-specific processing and data storage from the Cloud computing to the edge nodes. Among the expected benefits of EC deployment in 5G, there are: performance improvements, traffic optimization and new ultra-low-latency services. At the same time, in 5G we are witnessing a growing development of AI (Artificial Intelligence) for a wide spectrum of applications, such as: intelligent personal assistants, video/audio surveillance, smart cities’ applications, self-driving, Industry 4.0. The requirements of these applications seem calling an AI’s resources-hungry model, whose cloud-centric execution appears in the opposite direction with a migration of computing, storage and networking resources at the edge. In reality, the two technology trends are crossing in the EI (Edge intelligence): an emerging paradigm meeting the challenging requirements of future pervasive services scenarios, also beyond 5G, where optical-radio networks requires automatic real-time joint optimization of heterogeneous resources, high dimensional fast configurations (e.g., selecting and combining optimum network functions and inference techniques) and autonomic/cognitive operations.

Luka Koršič

Internet Institute Ltd., Ljubljana, Slovenia

Talk title: Next Generation Public Protection and Disaster Relief Systems   watch presentation

Talk abstract:

Public Protection and Disaster Relief Systems (PPDR) require applications and services to operate reliably and with high survivability and availability characteristics. For such features, support of high-performance communications with added service provisioning intelligence is required in both normal day-to-day conditions and in extreme situations since the provided emergency applications are executed in a distributed system with distributed service intelligence. This presents complex requirements for service provisioning on top of a next generation telecom and cloud infrastructure. With MATILDA-based framework that allows distributed cloud and network services deployment, the use case for next generation PPDR systems can show, how cloud-based PPDR services providing real time emergency intervention monitoring and critical infrastructure protection can benefit with orchestrated deployment. Furthermore, such critical services would require powerful SLS/SLA monitoring, which can be enabled with qMON performance monitoring engine deployed along with application services in NFV OSM-compliant environment.

Sławomir Kukliński

R&D Expert, Orange Labs Polska

Talk title: Carrier-grade scalable architecture for network slicing   watch presentation

Talk abstract:

In the era of virtualization, the network and service management has to be redefined. A new decomposition introduced by MANO introduces the separation of orchestration from classical management functions. Moreover, due to the functional separation of the infrastructure and virtualization, the orchestration and management of slices or services do not have to cope with hardware. The network slicing approach (based on MANO) enables dynamic creation of multiple network slice instances (i.e. logical networks) that can be created for the needs of specific services or verticals. The services may have a different lifetime than network slice instances. Moreover, in some cases, the verticals should see their network slices as their ‘private’ networks and therefore, they need a management interface in order to customize the slices and services attached to them. The new situation calls for a specific approach to network/service management that can provide appropriate functionalities, performance and scalability. The proposed by ETSI MANO approach is centralized, and its operations are driven by single OSS/BSS. The approach raises scalability issues, moreover using a single OSS/BSS for multiple, by definition ‘isolated’ network slices seems to be not appropriate. In the talk, a concept of embedded (in opposite to external) slice/service management that is based on the autonomic/cognitive network management paradigm will be outlined. In the concept there have been introduced multiple orchestration domains, each virtual function has its embedded management component, each slice (or service) has its embedded manager that provides a simple, intent-based management interface to slice tenant and the inter-domain management and operations are handled by dedicated entities.  Moreover, management functions can be dynamically upgraded.  The described framework is ready for federated AI operations.

Kay Burow

BIBA, Germany

Talk title: Towards Flexible Production Systems – Chances and Limits by using Cellular Networks   watch presentation

Talk abstract:

During the last years, a change occurred in the industrial environment. Topics like Industry 4.0, Digital Twins, 5G, Cloud Services and as an incremental part of it, Everything-as-a-Service are phrases which can be heard very often in this context. However, with new ideas, new challenges arise. Concerning the Industry 4.0 by reducing it here to logistics and production, an overview of existing limits in actual industrial scenarios will be shown. Also, the demands that are coming from industry stakeholders and their clients for future applications and products. As part of the MATILDA project, the two use cases will be introduced towards 5G-ready applications; one has its focus on logistics, the other one in production. Both use cases in MATILDA were designed to consider future industrial problems/interests and were scaled to the very base to examine the essential issues and differ in their requirements. In the logistics scenario, the main assets are collecting data in almost real-time and store them online in a database. The challenge here is to provide automated services of interconnected facilities and factories with high reliability and availability to deliver the data at all time for higher flexibility and transparency. Further, low power consumption as a critical aspect to extend the lifetime of simple Internet of Things (IoT) components. In the production scenario, the main asset lies in human-robot-collaboration (HRC). For the interplay between robots and humans, the reaction time (e.g. low latency) is of high value to avoid any clashes, but also a running system needs ultra-high reliability and availability to guarantee the workers safety. By considering the most critical KPIs for further/future applications, we will show a set-up of a combined and autonomous logistics-production scenario which can be of high interest towards developing 6G.

Rui Aguiar

University of Aveiro, Portugal

Talk title: On the network of control   watch presentation

Talk abstract:

The presentation will highlight the changes in paradigm required for properly exploring the multiple use cases forecast for the future digital society. The highlight will be on the needs of the network for control aspects.

Artur Hecker

Director of Future Network Technologies at Huawei MRC, Munich – Associate Professor, Telecom ParisTech, France.

Talk title: From 5G Slicing to fine-granular, autonomic resource management   watch presentation

Talk abstract:

In this talk, we are trying to show, how the current 5G development and deployment, if successful, will naturally lead to a fine-grained resource management as a new common grounds for the future, beyond 5G, systems. By analyzing the current situation and stretching it out into the future along the most probable vectors of development (e.g. 3GPP next releases, technological development, whitepapers from operators and IT, from the European Commission, from vertical industry associations and other key stakeholders), we identify and formally define distributed resource management as the core problem, which we then, under typical environmental and user profile assumptions, analytically break down to a set of required, novel mechanisms and protocols, which we, as community, currently do not have at our disposal. Therefore, this talk tries to lay out a concrete set of proposals for the future work towards B5G.

Miloud Bagaa

Aalto University, Finland

Talk title: High Resolution Media on Demand Vertical   watch presentation

Talk abstract:

Internet video traffic is already overwhelming, current network's traffic is mainly carried out over 4G and is expected to reach 82% by 2022. The massive proliferation of different kinds of mobile devices and the faster and ubiquitous connectivity offered by 4G networks have played a key role in video traffic dominance. For this reason, the use-case “High-Resolution Media on Demand Vertical” is one of the most important verticals targeted by MATILDA project for efficiently handling video traffic in the 5G system and beyond. For this purpose, we have leveraged the MATILDA framework for designing the high-resolution media on demand platform that has a maximum level of flexibility for its integration with different public and private infrastructure as a service (IaaS) providers such as Amazon AWS service, Microsoft Azure, Rackspace, and OpenStack. The platform offers Vertical Application Orchestrator (VAO) that gives the possibility of customizing and updating the blueprint of different slices before its instantiation. The platform harnesses the flexibility and programmability of Software Defined Networks (SDN) to securely interconnect a slice's components and give the full control over the network to the slice owner, GeoDNS-based streamer load balancing is leveraged to ensure delivery of content from the closest network node, and hence enhance the QoE perceived by different users. The platform also provides a real-time monitoring system, stats and analytics engine from which a slice owner can monitor and analyze content consumption, Slice performance, end-users geographical distribution and many more. The platform can be accessed from its VAO interface at: https://www.sdn-enabled-cdnaas.com.