The dynamics of Low Earth Orbit (LEO) constellations introduce frequent and unpredictable Line-of-Sight (LoS) interruptions that challenge conventional handover procedures in Non-Terrestrial Networks (NTNs). Previous analytical work has established that, once the residual LoS interval becomes shorter than the end-to-end control-plane delay, network-assisted preparation loses feasibility and the decision logic must shift toward the terminal. Building on this premise, this paper introduces a learning-based reactive handover model representing a UE-centric continuity mechanism under LoS-limited conditions. The proposed framework employs a lightweight Q-learning agent that selects target satellites based on locally observable indicators such as relative elevation and received signal level. Within a deterministic geometric environment, the agent converges toward stable switching behaviour and reproduces the qualitative trade-offs predicted by the analytical model of reactive continuity. The achieved 25-30% alignment improvement over a geometry-only baseline demonstrates that even simplified reinforcement learning can capture timing dependencies essential to NTN handover robustness. While not intended as a deployment result, the study provides a reproducible benchmark and a methodological foundation for future deep- or federated-learning extensions operating within realistic NTN channel and protocol constraints.

366

529

Traditional rule-based or static management approaches struggle to cope with the dynamic, multi-layered nature of 5G/6G networks, creating a strong motivation for AI-native solutions – management systems built from the ground up with artificial intelligence – to enable autonomous, real-time network control. In this work, we introduce GNN-ATIVE, an AI-native orchestration framework that leverages Graph Neural Networks (GNNs) and knowledge graphs (KGs) in a unified graph-based paradigm for network management. GNN-ATIVE uses a semantic knowledge graph to represent the network’s state and context, employing standard ontologies to ensure consistency and interoperability. Building on this foundation, we design Knowledge Graph enabled Generative Pretrained Transformer (KG-GPT), a novel graph-to-graph Transformer model that performs knowledge-driven reasoning on the KG. KG ingests the structured network state (nodes, links, and attributes) and infers optimal configurations or management actions, serving as a high-level decision engine for the orchestrator. We implement and evaluate GNN-ATIVE on an Optical Transport Network (OTN) testbed using real network components. The results demonstrate that GNN-ATIVE can effectively manage OTN resources and adapt to network changes while achieving low-latency inference for decision making.

Improved RAN scheduling algorithms and the integration of edge functionality tackle low-latency requirements by reducing end-to-end communication delays. Despite these innovations, key architectural limitations remain, specifically the requirement to route user plane traffic through the core network's User Plane Functions (UPFs). This constraint is relevant in the case of traffic between two User Equipment devices (UEs) served by the same gNodeB (gNB), which is a common scenario in factory floor or campus network deployments where a large amount of the traffic would be between UEs in a small area. To address this limitation, an intra-gNB user plane shortcut mechanism is proposed. This shortcut mechanism involves a gNB directly routing user plane traffic between UEs connected to it, bypassing the UPF and some upper-layer user plane processing on the gNB. Considering the current 5G specifications, the proposed mechanism is investigated, and the necessary extensions are discussed. A proof of concept of this mechanism is also introduced and evaluated, built within the 5G RAN implementation of srsRAN. Our theoretical analysis demonstrates latency reduction, supported by prototype measurements conducted in a simulated environment, illustrating promising results for the shortcut concept in reducing end-to-end latency.

Fraunhofer FOKUS' Open6GCore is the first practical implementation of a 6G core network. It challenges the traditional view of network functions as rigidly connected components, by embracing a highly modular software-driven approach that offers significant adaptability and reduced complexity. The flexibility of Open6GCore allows dynamic adaptation to requirements, whether as a 5G core network, simplified microservices, microcore, or a distributed large-scale core network. By implementing 3GPP standards with modern IT methodologies, Open6GCore achieves scalability and resilience and simplifies the addition of experimental features. Thus, it is particularly well-suited for research environments, facilitating rapid prototyping and seamless integration of new functionalities. Initial functional validation demonstrates successful deployment, interoperability with 5G User Equipment and Radio Access Networks, and efficient core network operations. These results demonstrate Open6GCore's readiness as a platform for 5G and 6G testbeds, trials, and innovation. This paper highlights the architecture, development, validation, and integration techniques of Open6GCore, offering insight into its role as a foundational platform for next-generation networks.

As the requirement for flexibility in networks grows, new mechanisms are needed to enable extreme adaptability to diverse environments and the ability to react to unknown situations. The rise of AI provides a significant opportunity to enhance network efficiency, scalability, and resilience beyond what is possible with traditional policy-based approaches. However, current network architectures rely on predefined rules and event-triggered responses, insufficient to dynamically address unpredictable and diverse conditions. This paper introduces a new perspective on AI-driven autonomous network agents, capable of continuous perception, proactive decision-making, and adaptive optimization across multiple network layers. These agents automate deployment-phase tasks, optimize runtime operations, and enhance cross-layer coordination, addressing critical challenges like mobility management, resource scheduling, and dynamic service adaptation. Unlike traditional static policies, these agents learn from operational feedback, enabling networks to adjust and refine their behavior over time. With their self-learning and intent-driven automation, AI-driven agents integrated into future network architectures, including beyond 5G and 6G systems, can enable greater autonomy, adaptive optimization, and improved robustness, effectively responding to evolving communication requirements.

With software-only core networks, a new level of flexibility has been achieved in how network services can be distributed and where they can be deployed. The 6 G network is foreseen to become increasingly heterogeneous regarding compute resources and their interconnectivity by incorporating new deployments such as fixed and nomadic campus networks, onpremise edge networks, and non-terrestrial networks. In this new environment, service placement is one of the key mechanisms to improve the system's response time, performance, resource utilization, scalability, and robustness. This paper presents a study on network service placement for 6G networks, employing MixedInteger Programming (MIP) and Multi-Criteria Optimization (MCO) for efficient resource allocation. We propose two MIP models addressing minimizing latency, installation costs, and maximizing success rate. Our models evaluate network service placement considering constraints such as node capacities, link bandwidth, and failure scenarios. Numerical results demonstrate the scalability and effectiveness of our approaches, providing insights into optimal network service deployment strategies for future 6G-ready networks.

The growth of telecommunication networks towards 6 G, owing to the complexity and heterogeneity, mandates a paradigm shift in network management. Due to the nature of the traditional Policy-based Management and Control (PBMC), aspects of network management have to be well thought out during the design phase making the system inflexible. Artificial Intelligence (AI) is expected to augment PBMC by introducing run-time flexibility and adaptability. The use of AI becomes more pressing, when 6 G systems are considered to host more diverse technologies catering to specific use cases thus increasing the burden on operators. This article presents “GraphGPT”, a transformer-based Knowledge Graph (KG) reasoning model prepared using light-weight and tailored datasets for 6 G. The implementation and evaluation details present the first look of the model.

Network management systems for beyond 5G (B5G) and 6G networks today require efficient approaches for handling increased heterogeneity, network-function dis-aggregation, performance requirements, and optimizing networks to support highly diverse use cases. While the Open-Radio Access Network (O-RAN) Service Management and Orchestration (SMO) frameworks efficiently manage RAN and cloud infrastructure, the fragmentation of management platforms across RAN, Core Network (CN), and Transport Network (TN) introduces operational inefficiencies, particularly in Non-Public Networks (NPNs) deployments. This paper proposes a novel extension to the O-RAN SMO architecture, integrating CN and TN management functionalities into a unified control framework. By exploiting AI/ML-driven$\mathrm{x} / \text{rApps}$and a converged data analytics pipeline, the proposed architecture enhances SMO's fault management, resource optimization, and service continuity capabilities. Our implementation validates the feasibility of the proposed SMO extension, demonstrating subscriber-specific QoS assurance through O-RAN-based mobility management mechanisms. The proposed approach successfully shows how RAN-/Core-converged SMOs enable significant enhancements to O-RAN's x/rApps, allowing for subscriber-specific as well as application-specific differentiated QoS assurance.

To enable seamless 5G service continuity for mobile ground users traversing infrastructure-limited or obstructed regions, the integration of Non-Terrestrial Networks based on Low Earth Orbit satellite constellations is being actively pursued. However, the mobility dynamics of ground terminals frequently result in sudden Line-of-Sight loss, which undermines the feasibility of conventional handover preparation involving context transfer, synchronization, and identity continuity. Without pre-coordination, such transitions lead to uplink blindness, session interruption, and degraded performance. This paper addresses the handover problem under Non-Terrestrial Network-specific constraints by (i) defining a structured taxonomy of service continuity mechanisms suited for impaired Line-of-Sight conditions, (ii) introducing an analytical model that captures handover delay, authentication feasibility, and buffer dimensioning under varying visibility regimes, and (iii) quantitatively evaluating these mechanisms in a high-mobility highway scenario. The results underline that while context mirroring ensures superior continuity under full coordination, handover with reactive preparation emerges as the most practical and resilient fallback under constrained Non-Terrestrial-Networks conditions.

In the context of healthcare for elderly patients, especially in areas located far from rehabilitation clinics, pro-viding remotely assisted prevention and treatment enabled by reliable and secure connectivity through micro-networks based on technologies like 5G and beyond, could be a viable alternative to regularly transporting the elderly to the clinics, potentially improving their quality of life. This paper analyses the needs of such a system featuring fall detection and risk assessment as well as rehabilitation at home and presents the design of the End-to- End interoperable and secure software system and the planned validation.

In the context of healthcare for patients suffering from cronic wounds, especially in areas located far from clinics, providing remotely assisted treatment enabled by reliable and secure connectivity through nomadic micro-networks based on technologies like 5G and beyond, could be a viable alternative to transporting the patients regularly to the clinics or even have them hospitalized for a long time in order to treat the wound. This paper analyses the needs of such a system featuring high precision 3D scanning of wound and bioprinting of a matching hydrogel patch, present the planned topology as well as the ecosystem of the solution under development together with the current evaluation framework.

The arrival of mega-constellations with inter-satellite links marks a major shift in space communication, evolving toward large-scale networking systems. This spurred significant interest in creating new routing mechanisms designed to tackle the unique challenges of space-based networks, characterized by predictable but high-latency and predictably changing links. This paper proposes a Software Defined Networks (SDN)-based routing solution for mega-constellations, designed to significantly cut routing convergence time and reduce processing demands on space nodes. It relies on a centralized terrestrial controller that asynchronously computes routing information and deploys it to space nodes prior to its usage. These nodes are equipped with lightweight semantic routing capabilities that allows them to autonomously determine optimal routing based on their current network status. As any SDN solution, the design is flexible enough to accommodate various routing optimizations available in the literature, facilitating their practical implementation. The paper provides a detailed description of the proposed solution, assesses its feasibility using an example routing algorithm, and compares it with existing routing approaches, highlighting its strong potential for real-world commercial system application.

In 5th Generation networks, the User Plane Function forwards the data of the different users in the Radio Access Networks to the destination networks. To improve the throughput of this component, we used the software acceleration technique eXpress Data Path (XDP) in previous work to shift the processing of GTP-U tunnel connections in the kernel. The developed solution for an XDP Receive Side Scaling (RSS) in the uplink direction distributes the encapsulated traffic across the CPUs, but the measured throughput was significantly lower than in the downlink direction. In this work, we investigate the impact of multiple connections in the downlink direction with RSS by Network Interface Card (NIC). For leveraging this NIC feature in the uplink direction, we propose a sender-side source port variation in combination with XDP enhancements. In RFC2544 frame loss tests, we evaluate the performance with an increasing number of GTP-U tunnels.

This article presents a new interpolation algorithm for received signal strength in mobile networks. The algorithm aims to fill in the missing values in a network environment by creating a full signal strength map. It is based on the deep learning Generative Adversary Network (GAN), which has previously been used successfully only for fixing pixels in image repairs. The algorithm was trained and tested using publicly available data on outdoor mobile network signal strength under 6GHz, and the results were evaluated using two performance metrics: accuracy of measurement and size of predicted values. The results show that the accuracy of the algorithm depends heavily on the characteristics of the input data. When the interpolating points are evenly distributed, the accuracy is high and the number of predicted values is satisfactory. Our algorithm reduces the number of manual measurements required to obtain a complete signal strength map, simplifying network planning and quality assurance processes.

With the accelerating development and deployment of 5G Non-Public Networks (NPNs), many customized and private networks, providing services to local users and devices, are being installed across the world. To fully use the capacity of these networks, their services should also be reachable from other locations outside their limited coverage area. The only solution provided by 5G and beyond-5G 3GPP standardization at the current moment is based on roaming interfaces, which presumes that the NPN operator makes costly and complex peering roaming agreements with many different large scale and NPN operators. This article proposes an alternative to the roaming mechanism based on an Over-The-Top (OTT) peering approach. Similar to the non-3GPP integration, the connectivity in the visited network is transparently used to establish a binding with the home domain. Furthermore, the proposed architecture is implemented and evaluated in a basic form using Fraunhofer 5G Playground, a comprehensive 5G testbed network using commercial base stations and the Fraunhofer Open5GCore in conjunction with standard Android OS devices. Showing that such a solution represents a cost-effective, reduced functionality and relatively easy to implement solution outperforming the roaming approach in both dynamicity and ease of management.

The tunneling protocol WireGuard outperforms its main competitors OpenVPN and IPsec in terms of throughput and latencies. These improvements are due to WireGuard's use of faster crypto primitives, as well as to the implementation of WireGuard as a Linux kernel module that uses multithreading and advanced locking strategies. Independently of the WireGuard project, Lukaszewski et al. demonstrated improvements in end-to-end goodput when tunneling protocols exploit alternative communication paths. In this poster, we combine these two research directions by proposing multipath extensions for WireGuard. Our extensions involve additions to the WireGuard header, which enable obtaining real-time statistics on the performance of each path. Further, these real-time path performance statistics enable a self-adaptive selection of paths. As a proof of concept, we adapted the WireGuard Linux kernel module accordingly and prototyped four example path schedulers, two of which adopt multi-armed bandit algorithms.

Standardisation, research and development efforts for the fifth generation (5G) of mobile telecommunication networks are well under way. Software Defined Networking (SDN) and Network Function Virtualisation (NFV) are two of the key enabling technologies, considered in these efforts. The need for a flexible, high performant and efficient architecture is well established. Network slicing, which combines SDN and NFV, can contribute to such an architecture. It allows the parallel deployment of differing network stacks on top of any physical infrastructure. SDN's separation of control and data plane components allows for flexible deployments. How can SDN's flexibility be leveraged in a sliced, 5G network infrastructure? There needs to be an efficient way to integrate SDN into 5G networks. In this paper, we posit a way of integration, which allows decoupling the data plane components from any particular control plane. This can improve flexibility, utilisation and extensibility. We envision utilising an SDN switch implementation as the User Plane Function (UPF), and introducing an SDN controller between Session Management Function (SMF) and UPF, to effectively decouple the two. Based on this decoupling, control and data plane components can be deployed in separatelyand new slice orchestration opportunities can be developed. Furthermore, we can leverage deep data plane programmability, to enhance the system in terms of function and flexibility.

To pave the way towards using adapted wide scale telco technologies within dedicated networks environment, there is a dire need to increase the reliability of the end-to-end communication beyond what is currently offered to individual subscribers towards the levels previously available in operations technology. This can be achieved by adding a new dynamic Software Defined Networks (SDN) routing overlay on top of the existing point-to-point forwarding technologies. This article introduces an SDN-based solution for dedicated industrial environments, aiming at providing a new level of flexibility and adaptability in case of network or service failure. A gap analysis of the existing infrastructures gives the basis for the network evolution, especially addressing the control of the routing within the given deployment. Furthermore, a new set of functional features are proposed to be added to the existing wired and to the emerging wireless infrastructure in order to further increase the network flexibility and reliability from network availability and optimal data paths perspective. The different features are then exemplified as existing and further to be implemented extensions of the Fraunhofer FOKUS OpenSDNCore toolkit, assessing the effectiveness of the approach for different deployments and use cases.

5G represents the next generation of communication networks and services, and will bring a new set of use cases and scenarios. These in turn will address a new set of challenges from the network and service management perspective, such as network traffic and resource management, big data management and energy efficiency. Consequently, novel techniques and strategies are required to address these challenges in a smarter way. In this paper, we present the limitations of the current network and service management and describe in detail the challenges that 5G is expected to face from a management perspective. The main contribution of this paper is presenting a set of use cases and scenarios of 5G in which machine learning can aid in addressing their management challenges. It is expected that machine learning can provide a higher and more intelligent level of monitoring and management of networks and applications, improve operational efficiencies and facilitate the requirements of the future 5G network.

Cloud computing enables the on-demand delivery of resources for a multitude of services and gives the opportunity for small agile companies to compete with large industries. In the telco world, cloud computing is currently mostly used by mobile network operators (MNO) for hosting non-critical support services and selling cloud services such as applications and data storage. MNOs are investigating the use of cloud computing to deliver key telecommunication services in the access and core networks. Without this, MNOs lose the opportunities of both combining this with over-the-top (OTT) and value-added services to their fundamental service offerings and leveraging cost-effective commodity hardware. Being able to leverage cloud computing technology effectively for the telco world is the focus of mobile cloud networking (MCN). This paper presents the key results of MCN integrated project that includes its architecture advancements, prototype implementation, and evaluation. Results show the efficiency and the simplicity that a MNO can deploy and manage the complete service lifecycle of fully cloudified, composed services that combine OTT/IT- and mobile-network-based services running on commodity hardware. The extensive performance evaluation of MCN using two key proof-of-concept scenarios that compose together many services to deliver novel converged elastic, on-demand mobile-based but innovative OTT services proves the feasibility of such fully virtualized deployments. Results show that it is beneficial to extend cloud computing to telco usage and run fully cloudified mobile-network-based systems with clear advantages and new service opportunities for MNOs and end-users.

Service Function Chaining (SFC) supports services/applications through linking an ordered list of service functions (such as firewalls, deep packet inspections and load balancers) in the network. Using Software Defined Networking (SDN) and Network Function Virtualization (NFV) provides a simpler and shorter SFC process. In order to provide high-availability on a multi-site SFC, it is required a flexible algorithm able to properly select Service Function instances while creating the Service Function Path (SFP). This paper presents a new service function selection algorithm providing a flexible tradeoff between the SFP distance and the loads per service function instances at the deployment stage. The flexible tradeoff provided in the algorithm can be adjusted by a parameter. This parameter value can differ among different types of SFCs based on the service/application (real time or non-real time) dedicated for its SFC. Matlab has been utilized for validating it with different scenarios. It is also compared to other existing algorithms such as, load balancing and shortest path. The results showed that the proposed algorithm could provide the best SFP end-to-end distance performance as the shortest path algorithm for the real time services/applications, and that will result a good Quality of Service (QoS) performance for real time services/applications. Furthermore, the proposed algorithm provides an acceptable load distribution over the available service functions instances better than shortest path algorithm and not far from and load balancing algorithms.

Technical and economic opportunities of cloud computing become the focus for Internet applications and at the same time also for telco support infrastructures and network services. In fact, many telco providers are consolidating their service infrastructures towards converged and all-IP next generation networks providing typical telco services within LTE (and soon 5G) and also fixed network environments, e.g., often still adopting IP Multimedia Subsystem (IMS) architecture solutions. This telco service infrastructure evolution requires a significant upfront investment in the necessary hardware and software, thereby slowing down the adoption process significantly more than any other Internet application. Cloud computing applied to telco infrastructures can allow pay-per-use business models and significantly lower investment risks by providing telco infrastructure functionality as Virtual Network Functions (VNFs) on top of a Network Functions Virtualization (NFV) platform. For this purpose we propose a quality audit and resource brokering framework that is fully NFV-compliant. Its current and reported implementation specifically targets IMS services because of the still relevant role played by IMS in converged provisioning and the wide availability of IMS deployment testbeds to validate the proposal. In particular, the proposed solution can monitor the quality offered by VNFs and scale in/out depending on dynamic requirements; it is fully based on industrial standards and open-source reference implementations, thus enabling rapid adoption in real industrial environments.

The maturity reached by virtualisation technology enabled great innovation for efficient applications and services development and delivery, independent of the underlying hardware equipment, especially with the large deployment of off-the-shelf hardware based cloud infrastructures. In order to take advantage of this technology, the existing network functions have to be developed and adapted to the new paradigm. However, traditional telecom services are still implemented on dedicated hardware resulting in high deployment and maintenance costs compared to the other already cloudified services. ETSI Network Functions Virtualisation (NFV) aims to fill this gap by applying to telecom the virtualisation technologies. This paper introduces a set of three software architectures for efficient virtualisation of IP Multimedia Subsystem (IMS) in different operator environments responding to the high level requirements of the ETSI NFV use case for virtualizing operator core network functions. Additionally, a management architecture for simplifying the deployment and runtime orchestration of such a virtual service on top of a cloud infrastructure is presented. Furthermore, one of the IMS software architectures was implemented based on the Fraunhofer FOKUS Open IMS Core, measured and evaluated on top of an OpenStack cloud.

Self-management will be a distinctive feature in future mobile network infrastructures. Decision points and control loops in the network depend on having the right information at the right place at the right time. For this task we present an Information Gathering and Processing System (IGPS) and show how meta-information (e.g. information costs) can be exploited as an organizing principle in the involved processes. Standardization in this area should at least provide suitable high-level measurement parameters that allow for a technology independent characterization of the network situation.

With the development of novel access technologies e.g. LTE, WiMAX etc., the mobile devices are able to establish services over multiple wireless connections with different momentary capabilities. Being imported from the fixed line communication, the current services deployed are lacking an efficient mechanism of adaptation to the momentary connectivity conditions, which deters the efficiency of the service delivery. This paper introduces a novel solution for the adaptation of the services to the momentary network conditions of the mobile device which does not require a complete end-to-end re-signaling, by this reducing the communication over the wireless link and enabling efficient provisioning of the services and the development of new ones customized for the wireless environment. Furthermore, the concept is exemplified on the 3GPP Evolved Packet Core architecture and evaluated using the Fraunhofer OpenEPC testbed implementation.

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 on the access network reselection mechanisms. The present paper describes and evaluates the opportunity of introducing novel access reselection procedures which enable a network provider to optimize the allocation of the users on the different access networks available using as central concept that handovers can and should be triggered by the modifications on the resources required by the mobile devices in order to optimize the overall usage of the wireless environment. This concept is exemplified on the 3GPP Evolved Packet Core and evaluated on a minimal prototype implementation.

Mobility in a wireless heterogeneous scenario in which the mobile devices are able to connect to more than one access technology requires a new architectural development for the network connectivity management. The present paper describes and evaluates a set of architectural options for integrating the novel network discovery and selection functionality in the 3GPP evolved packet core which enables an operator to manage the connections of the mobile devices. As the design options here presented consider different levels of network control on the mobility at the cost of new interfaces and new procedures, an evaluation based on these two criteria is given.

Optimized processing and cost efficient solutions for providing and controlling the quality of service (QoS) are important issues for the 3GPP operators. Although the main frame of the IMS architecture in the area was defined, i.e. the policy and charging control architecture (PCC), it misses a clear definition and standardized processing of the interchanged messages. This allows for a set of different perspectives on how the main information unit of the architecture, the PCC rule, is deployable. Based on the incomplete set of requirements already existent, this paper proposes two solutions on handling QoS provisioning architectural aspects of the IMS. It also considers a comparative evaluation of the prototypes developed in this work for proving the concept of the two approaches.

In this paper we provide the description of a new mobility service for an IP Multimedia Subsystem (IMS) platform. It optimizes the operation of multimedia services in a wireless heterogeneous environment containing multiple access networks with different cost per bandwidth ratio. It extends the SIP Presence Service with mobility related information, i.e. Media Independent Handover (MIH). The terminals are able to exchange specific handover knowledge with the remote service provisioning infrastructure and to receive commands adapted to the profile of all the parties involved in multimedia sessions, transparent to the actual multimedia service. The system was designed to enable the adaptation of multimedia sessions to the profile of the users and to the momentary network conditions on an end-to-end basis.