6G networks are expected to deliver ultra-low latency, high reliability, and real-time intelligence for emerging services such as interactive Augmented Reality (AR), autonomous robotics, and digital twins. Achieving these requirements in practice demands tight coordination between networking, computing, and control domains, spanning RAN, transport, edge, and cloud. However, current 5G deployments lack pervasive telemetry, fine-grained observability, and automated control mechanisms capable of reacting at the time scales required by latency-sensitive applications. This paper presents a full integrated demonstration of DESIRE6G, a cloud-native 6G-ready architecture that leverages programmable data planes with P4 for flexible routing and telemetry using an implementation of novel data plane protocols, achieves distributed optimization of service deployment and runtime monitoring and reconfiguration via secure multi-agent systems (MAS), combined with intent-based orchestration layer for end-to-end service assurance. The system is validated on the federated ARNO testbed using a real distributed AR application involving a remotely-controlled drone as a User-Equipment that is equipped with a camera streaming a live video through the DESIRE6G network to a Kubernetes edge cluster that executes serverless inference functions for object detection and recognition, the video is then augmented with object information and shown on a Quest 3 AR headset. The MAS monitors the end-to-end latency in real time through P4 Telemetry and responds to changes in network conditions by reconfiguring the affected segments, while the Kubernetes monitoring provides real-time visibility and scalability across different segments. Overall, three hierarchical service assurance loops are demonstrated: (i) In-Network Control (INC) executing microsecond-scale congestion recovery in the P4 data plane, (ii) Infrastructure Management Layer (IML) performing millisecond-scale function migration and scaling, and (iii) MAS-driven cross-domain optimization operating at sub-second time scales to resolve RAN latency anomalies. Evaluation results show stable end-to-end latency in the 15–25 ms range in steady-state conditions, with fast recovery during induced congestion ≤ 1 . 5 ms data plane reroute via P4 INC.

Traditional 5G user plane function (UPF) architectures are software-based implementations that struggle to maintain performance. To meet the stringent quality of service (QoS) and scalability requirements of 5G under high session and data plane loads, a highly efficient next-generation UPF is required. In this paper, we present a next-generation 6GUPF that fully leverages the hardware acceleration of SmartNICs/DPUs. The 6GUPF is developed using DOCA Flow API, which enables fast, programmable packet processing directly in the data path, specifically for the N3 gNB network and N6 interfaces of datanet. The architecture integrates flow-based acceleration for stateful flow tracking and symmetric Receive Side Scaling (RSS) to utilize cores and manage non-blocking states efficiently. This will help minimize latency and avoid contention in multi-core environments. Our experiments show that hardware-offloaded UPF achieves an aggregate line rate of 400 Gbit/s, supports 1 million concurrent data streams, and scales up to 500,000 active User Equipment (UE) instances while maintaining QoS and session isolation. It also enables 40% lower latency compared to traditional software UPFs. Unlike Tofino-based P4 switch UPF designs, which are limited by SRAM and TCAM constraints when storing large-scale data streams, our DPU-based solution is ideal for high-density Protocol Data Unit (PDU) deployments without sacrificing programmability and performance. It creates a path for cloud-native, 6G UPF deployments that can scale to a wide range of workloads, from ultra-low-latency applications to large-scale IoT backhaul.

The growing complexity of network attacks has outpaced the capabilities of traditional intrusion detection systems (IDS), which often rely on flat data structures that fail to capture complex relationships within networks. To address this limitation, we propose IDS-NGNN, a novel IDS that integrates hardware-offload SmartNIC preprocessing with a nested graph neural network (NGNN) architecture. Unlike standard Graph Neural Networks (GNN), IDS-NGNN jointly captures local and global dependencies using a three-layer design: an internal GNN for host-level activity, a nested graph module for hierarchical aggregation, and an external GNN for inter-host communication. SmartNIC acceleration enables efficient real-time processing of large-scale graph-structured network data at the edge. We evaluate IDS-NGNN on six public IDS datasets, including CIC-IDS-2017, CSE-CIC-IDS-2018, and ToN-IoT. Experimental results demonstrate that IDS-NGNN achieves up to 95% accuracy and 92% F1-score, while maintaining efficiency suitable for real-time 100 Gbps deployments.

Modern network infrastructures face increasing demands for scalability, performance, and security, particularly in multi-tenant environments. Ensuring efficient traffic isolation and secure communication across multiple virtualized systems is a critical challenge, especially when dealing with high-throughput applications. In this paper, we address these challenges by leveraging SR-IOV-based virtualization to create isolated network slices and establishing secure communication channels, using Kyber and Dilithium (ML-KEM/ML-DSA) and IPsec tunnels. We connect 16 Virtual Machines (VMs) and another host system on a direct link. We reach a total of 71.79Gbit/s on the 100Gbit/s NVIDIA BlueField-2 Data Processing Unit (DPU), by configuring the OpenvSwitch (OvS) for steering the traffic on its ARM cores, and testing using multiple instances of iPerf3 on all the (v)hosts. This result represents a 95.36% utilization of the baseline network performance.

A multi-objective optimization problem involves optimizing two or more conflicting objectives simultaneously. This type of problem arises in many scientific and industrial areas and it is classified as NP-Hard. Network routing optimization with multiple objectives falls into this category. In the context of 6G networks, solving this problem will become even more challenging due to the exponential growth of Internet of Things devices and the high quality of service requirements. Finding good quality solutions for large-scale networks will be increasingly difficult. In this paper, we introduce a quantum-inspired routing optimization scheme in which noisy-intermediate scale quantum computers (NISQ) can be used to solve the Multi-Objective Routing Problem (MORP). We evaluate the application of the proposed scheme in detail by first developing the mathematical formulas for both single-objective and multi-objective routing and mapping the problem onto gate-based models by using the quadratic unconstrained binary optimization (QUBO) approach. To validate the proposed scheme, we use the quantum approximate optimization algorithm (QAOA), the go-to approach for solving combinatorial optimization problems that are classically intractable. For the simulation, we use the IBM-Qasm simulator and Qiskit framework. Additionally, we use the Chernoff Bound as a standard technique to estimate the sample complexity of QAOA. Finally, we provide a detailed numerical and theoretical analysis of the proposed scheme, including its time complexity, resource requirements, and the challenges associated with it. Our results demonstrate that the proposed approach operates with a time complexity of O ( E 2 ) per iteration in both single and multi-objective scenarios, with an overall runtime of ( n iteration + n CB ) ⋅ O ( E 2 ) influenced by the sampling overhead, significantly outperforming Dijkstra’s algorithm in the multi-objective case, where the complexity increases to O ( 2 k ( N ( k + log N ) + 2 k E ) ) .

The emergence of quantum computing poses a threat to classical cryptography algorithms, necessitating a shift to quantum secure cryptography. Hybrid protocols combining at least one classical and one quantum-resistant cryptographic algorithm are becoming the standard for securing communications. In this work, we present our novel solution for integrating three different cryptographic assumptions (two of them quantumresistant) into hybrid network security protocols, ensuring that three different cryptographic assumptions must be broken before the protocol becomes vulnerable. Our solution allows for a seamless integration of classical and post-quantum (PQ) cryptography, and quantum key distribution (QKD) into existing network security protocols (e.g., TLS, IPsec) without any major modifications to the protocols themselves. This crypto-agility ensures the mitigation of some of the most well known challenges of both PQ cryptography and QKD. Our findings demonstrate the feasibility of such triple-hybrid network security protocols, showing non-substantial decrease in performance and almost no added packet overhead compared to state of the art protocols. In exchange, we pave the way towards next generation networks where the potential of new quantum-resistant cryptographic schemes can be leveraged in a dynamic and agile fashion, thus fostering a new era of unbreakable communication systems.

Cloud forecasting plays a crucial role in various fields such as agriculture, energy systems, and air travel. An accurate forecasting system can offer significant benefits by improving decision-making efficiency in these areas. This study investigates the use of Echo State Network (ESN)-based architectures for weather forecasting, focusing on cloud prediction across Central Europe using the CloudCast benchmark, which integrates data from Meteosat satellites and the European Centre for Medium-Range Weather Forecasts (ECMWF) model. Two novel techniques are included in this study, evaluated in two different phases. First, the Multi-Reservoir Weighted ESN (MWESN) architecture is proposed, featuring optimized inter-reservoir connections that enhance both the effectiveness and adaptability of the model. This model is evaluated along with advanced ESN architectures, including Multi-Reservoir ESN, Deep ESN among others. Second, the Error-Guided Regional Training (ERT) method is introduced to minimize the computational resources required for forecasting at the pixel level while maintaining high accuracy. Combined, MWESN and ERT demonstrate a 1.41% improvement in accuracy, effectively capturing complex spatio-temporal dynamics while significantly reducing computational demands compared to existing state-of-the-art methods. Additionally, models are tested on low-resource devices such as Raspberry Pi units, illustrating their feasibility for real-world meteorological applications.

Quantum key distribution (QKD) is a symmetric key exchange mechanism designed to enhance the security of current communication systems. Although QKD provides unconditional security, it also comes with deployment challenges. One challenge is the distance limitation between the transmitter and receiver, which restricts the large-scale adoption of QKD. To overcome this, trusted relays are used as intermediate nodes, allowing QKD to evolve from distance-limited point-to-point connections to unlimited QKD networks (QKDN). With an increasing deployment of QKD and a growing number of applications requesting keys, quality of service (QoS) constraints must be set for efficient key resource management to alleviate the risk of rejecting application requests. In this work, we integrate key delivery delay and maximum requested key size as QoS constraints in a novel key allocation algorithm built on three QKDN layers – quantum, key management and service. Moreover, we design key relaying using quantum key pools and virtual quantum key pools as storage mechanisms. We use static and dynamic weights for relay path computation and evaluate their impact on the proposed key allocation strategies with QoS constraints and find that static weights show a good overall performance for QKDN, while performance with dynamic weights varies based on historical key consumption data.

The fast-paced growth of 5G networks, along with the emergence of 6G technology, has emphasized the crucial importance of strong security measures to safeguard communication infrastructures. A key security issue in 5G data networks is Distributed Denial-of-Service (DDoS) at tacks, which specifically target the GTP-based protocol which is a significant threat. However, network telemetry data provides a rich source of information about the nature of network traffic, which can be used to detect and predict DDoS attacks. We propose a novel framework for collecting and processing large amounts of telemetry data in 5G networks leveraging state-of-the-art technologies, including data-plane programmability in P4-based User-Plane Function (UPF) and Data Processing Unit (DPU). Furthermore, we propose an anomaly-detection method for performing live deep learning analysis on network traffic using a Convolutional Neural Network (CNN) to detect DDoS attacks. Our results demonstrate the effectiveness of our framework, achieving an impressive 98.6% accuracy and 98% F1-score.

The imminent advent of Quantum Computers poses a significant threat to the cryptographic algorithms supporting the public key infrastructure (PKI) of widely used communication protocols. High Performance Computing (HPC) data centers among other interested parties are well aware of the catastrophic consequences quantum attacks could have on their PKI and are consequently transitioning to Post-Quantum Cryptographic (PQC) methods, despite the substantial overhead this introduces for handling incoming network packets. This work addresses the transition to PQC within the context of the Cloud-Edge Continuum by integrating the Open Quantum Safe (OQS) library into the accelerated strongSwan developed by Mellanox for Data Processing Units (DPUs). This integration offloads cryptographic operations from central servers to data DPUs distributed across the cloud-edge continuum. Our solution ensures quantum security by providing PQ authentication through CRYSTALS-Dilithium or CRYSTALS-FALCON, PQ key exchanges via CRYSTALS-Kyber, and confidential data transmission using AES-256. Additionally, the deployment of this implementation on DPUs helps reduce the computational load on both HPC data centers and edge devices, promoting more efficient and secure operations across the entire cloud-edge continuum.

Data Processing Units (DPUs) are becoming increasingly popular, especially for use in conjunction with Warehouse-Scale Computers (WSCs) due to their ability to handle networking functions and data-centric workloads. Cost-performance, energy efficiency, network 1/0, and batch processing workloads are important design factors for WSCs. Recent developments in AI and the never-ending increase in demand for data processing, cloud computing, and HPC set the optimisation of all those design factors as a high priority. DPUs can be utilised to achieve significant improvements in all those areas. This includes in-line network processing and upcoming enhanced security paradigms such as post-quantum cryptography (PQC) for quantum re-silient communications or software-defined perimeters (SDP) for confidential computing implementations. Being P4-enabled and dRMT-based, DPUs allow for the reconfigurability of the network traffic without the need to change the hardware. However, the network performance on such devices is only sometimes deter-ministic since the actual traffic and the rules, both of which have to do with packet processing, are not known during compile time. In this paper, we envision how the network performance on DPUs can be accelerated. We describe the challenges that negatively impact the bandwidth and latency: the complex steering pipeline and the massive runtime needed to optimise. These challenges arise from the lack of information during compile time that is only known during runtime. Thus, we envision optimising during runtime by leveraging DPUs' reconfigurability on the network 110. For this, we discuss the significant factors that must be considered to accelerate the network performance on such devices and we propose a solution for them.

The reliance on asymmetric public key cryptography (PKC) and symmetric encryption for cyber-security in current telecommunication networks is threatened by the emergence of powerful quantum computing technology. This is due to the ability of quantum computers to efficiently solve problems such as factorization or discrete logarithms, which are the basis for classical PKC schemes. Thus, the assumption that communications networks are secure no longer holds true. Quantum Key Distribution (QKD) and post-quantum cryptography (PQC) are the first cyber-security technologies that allow communications to resist the attacks of a quantum computer. To achieve quantum-resistant communications, the aforementioned technologies need to be incorporated into a network security protocol such as Transport Layer Security (TLS). In this paper, we describe and implement two novel, hybrid solutions in which QKD and PQC are combined inside TLS for achieving quantum-resistant authenticated key exchange: Concatenation and Exclusively-OR (XOR). We present the results, in terms of complexity and security enhancement, of integrating state-of-the-art QKD and PQC technologies into a practical, industry-ready TLS implementation. Our findings demonstrate that the adoption of a PQC-only approach enhances the TLS handshake performance by approximately 9 % compared to classical methods. Furthermore, our hybrid PQC-QKD quantum-resistant TLS comes at a performance cost of approximately 117 % during the key establishment process. In return, we substantially augment the security of the handshake, paving the road for the development of future-proof quantum-resistant communication systems based on QKD and PQC.

Abstract Echo state networks (ESNs) belong to the class of recurrent neural networks and have demonstrated robust performance in time series prediction tasks. In this study, we investigate the capability of different ESN architectures to capture spatial relationships in images without transforming them into temporal sequences. We begin with three pre-existing ESN-based architectures and enhance their design by incorporating multiple output layers, customising them for a classification task. Our investigation involves an examination of the behaviour of these modified networks, coupled with a comprehensive performance comparison against the baseline vanilla ESN architecture. Our experiments on the MNIST data set reveal that a network with multiple independent reservoirs working in parallel outperforms other ESN-based architectures for this task, achieving a classification accuracy of 98.43%. This improvement on the classical ESN architecture is accompanied by reduced training times. While the accuracy of ESN-based architectures lags behind that of convolutional neural network-based architectures, the significantly lower training times of ESNs with multiple reservoirs operating in parallel make them a compelling choice for learning spatial relationships in scenarios prioritising energy efficiency and rapid training. This multi-reservoir ESN architecture overcomes standard ESN limitations regarding memory requirements and training times for large networks, providing more accurate predictions than other ESN-based models. These findings contribute to a deeper understanding of the potential of ESNs as a tool for image classification.

The need for real-time analytics and faster decision-making mechanisms has led to the adoption of hardware accelerators such as GPUs and FPGAs within the edge cloud computing continuum. However, their programmability and lack of orchestration mechanisms for seamless deployment make them difficult to use efficiently. We address these challenges by presenting SERRANO, a project for transparent application deployment in a secure, accelerated, and cognitive cloud continuum. In this work, we introduce the SERRANO platform and its software, orchestration, and deployment services, focusing on its methods for automated GPU/FPGA acceleration and efficient, isolated, and secure deployments. By evaluating these services against representative use cases, we highlight SERRANO 's ability to simplify the development and deployment process without sacrificing performance.

We present an end-to-end solution to facilitate the seamless execution of hardware-accelerated compute-intensive tasks on heterogeneous hardware platforms spanning the Cloud-Edge continuum. Our approach includes a programming interface, orchestration, application management components, the vAccel framework, and a library of hardware-accelerated kernels. These components enable a Function-as-a-Service (FaaS) based operational flow that supports numerous diverse use cases while minimizing the time required for the developer to integrate their code and for the vendor to provide hardware acceleration capabilities to end users. Experimental results showcase the merits of our approach.

The recent availability of smart network interface cards (smart NICs) and Data Processing units (DPUs) providing hardware-accelerated networking and computing functionalities is opening the way towards new applications and use cases beyond the traditional data center scenarios. In this paper, three different use cases for edge scenarios that leverage on the innovative programmability enabled by DPUs are presented and discussed. The first use case focuses on a pervasive monitoring infrastructure to support accurate and decentralized network awareness for low-latency 5G services. The second one focuses on the implementation of power-efficient edge-to-cloud continuum. The third use case refers to effective network security functions at the DPU.

Flexible spectrum assignment in Elastic Optical Networks (EON) has emerged as a potential solution for allowing dynamic and elastic management of available bandwidth resources. In this paper, we demonstrate and evaluate our developed flexi-grid optical domain controller based on NETCONF/YANG. Our proposed modular architecture, based on Finite State Machines (FSMs), allows the flexibility to deploy the controller either in a centralized or in a distributed state for on the fly encrypted device management connections. A testbed composed of two physical Sliceable Bandwidth Variable Transponders (SBVTs) and an emulated flexi-grid optical network was used for our software evaluation. Controller startup and synchronization time, as well as media channel setup time are evaluated to compare the two deployment options and assess network scaling effects. Results demonstrate that our software is scalable by maintaining a relatively constant startup time on the networks tested (i.e., 1 to 64 nodes) in both deployment options. Software scalability is also supported by the media channel setup time, which presents a modest log scale growth when increasing the number of nodes from one to 64.

Two pulse shapes for IR-UWB transmission under the Russian spectral emission mask are proposed and their potential experimentally demonstrated. Pulses based on the hyperbolic secant square function and the frequency B-spline wavelet are shown to enable transmission of 1.25Gbit/s signals, reaching a maximum transmission distance of 6.5 m.

For the first time Ultra-Wideband record data transmission rates up to 35.1 Gbps and 21.6 Gbps are achieved, compliant with the restrictions on the effective radiated power established by both the United States Federal Communications Commission and the European Electronic Communications Committee, respectively. To achieve these record bit rates, the multi-band approach of Carrierless Amplitude Phase modulation scheme was employed. Wireless transmissions were achieved with a BER below the 7% overhead FEC threshold of 3.810-3.

In this work, we evaluate the performance of Wavelet-Coding into offering robustness for OFDM signals against the combined effects of varying fading and noise bursts. Wavelet-Code enables high diversity gains with a low complex receiver, and, most notably, without compromising the system's spectral efficiency. The results show that the Wavelet-Coded OFDM system achieves a BER of 1E-3 with nearly 6 dB less SNR than the convolutional coded OFDM system in frequency selective channels with a normalized channel response variation rate of ζ=1E-4.The proposed system fits as a key enabler for the use of mm-wave frequencies in future generation mobile communication due to its robustness against multipath fading.

Methods to upgrade the network infrastructure to cope with current traffic demands has attracted increasing research efforts. A promising alternative is signal slicing. Signal slicing aims at re-using low bandwidth equipment to satisfy high bandwidth traffic demands. This technique has been used also for implementing full signal path symmetry in real-time oscilloscopes to provide performance and signal fidelity (i.e. lower noise and jitter). In this paper the key digital signal processing (DSP) subsystems required to achieve signal slicing are surveyed. It also presents, for the first time, a comprehensive DSP power consumption analysis for both WDM and TDM systems at 1 Gbps and 10 Gbps, discussing latency penalties for each approach. For 1 Gbps WDM system 278 pJ per information bit for 4 slices is reported at 105 ns latency penalties, whereas 3898.4 pJ per information bit at 183.5 μs latency penalty is reported for 10 Gbps. Power savings of the order of hundreds of Watts can be obtained when using signal slicing as an alternative to 10 Gbps implemented access networks.