NVS2 – Elements for 6G Vision
Wednesday, 5 June 2024, 11:00-13:00, room Gorilla 5
Session Chair: Seppo Yrjölä (Nokia & Centre for Wireless Communications, University of Oulu, FI)
Energy Efficiency in AI for 5G and Beyond: A DeepRx Case Study
Amine Lbath (France); Ibtissam Labriji (Nokia Bell Labs, France)
This study addresses the challenge of balancing energy efficiency with performance in AI/ML models, focusing on DeepRX, a deep learning receiver based on a fully convolutional ResNet architecture. We evaluate the energy consumption of DeepRX, considering factors including FLOPs/Watt and FLOPs/clock, and find consistency between estimated and actual energy usage, influenced by memory access patterns. The research extends to comparing energy dynamics during training and inference phases. A key contribution is the application of knowledge distillation (KD) to train a compact DeepRX ‘student’ model that emulates the ‘teacher’ model’s performance but with reduced energy consumption. We experiment with different student model sizes, optimal teacher sizes, and KD hyperparameters. Performance is measured by comparing the Bit Error Rate (BER) performance versus Signal-to-Interference & Noise Ratio (SINR) values of the distilled model and a model trained from scratch. The distilled models demonstrate a lower error floor across SINR levels, highlighting the effectiveness of KD in achieving energy-efficient AI solutions.
On the Potential of Radio Adaptations for 6G Network Energy Saving
Abolfazl Amiri (Nokia, Denmark); Saeed Hakimi (Aalborg University, Denmark); Mads Lauridsen and Bahram Khan (Nokia, Denmark); Dileep Kumar (Nokia, Finland); Preben Mogensen (Aalborg University, Denmark)
This paper addresses the challenge of energy consumption in wireless mobile networks. These networks typically operate at an average radio resource usage of around 30%, as they are designed to handle peak capacity and coverage, while traffic demand is unevenly distributed in time and space. To enhance network efficiency and minimize energy consumption, it is crucial to optimize radio adaptations that deactivate unnecessary resources during periods of low traffic demand. Our objective is to determine the most efficient radio adaptations across time, frequency, antenna, and power domains. The study evaluates and compares adaptations in these four domains using the base station power consumption model defined by the 3GPP. The primary goal is to maximize energy saving while maintaining network performance, which is evaluated based on downlink achievable rate and excess packet latency. The analysis shows that reducing the number of active antenna ports and adjusting transmit power levels are the most effective approaches for conserving network energy. These strategies enable a more energy-efficient utilization of radio resources while still meeting the required performance criteria, which makes them attractive for the 6G design.
Towards Efficient Confluent Edge Networks
Rishu Raj and Devika Dass (Trinity College Dublin, Ireland); Kaida Kaeval (Tallinn University of Technology, Estonia); Sai Kireet Patri (Adtran Networks SE, Germany & Technical University of Munich, Germany); Vincent A.J.M. Sleiffer (Adtran, Sweden); Benedikt Baeuerle (Polariton, Ireland); Wolfgang Heni (Polariton Technologies, Switzerland); Marco Ruffini (CONNECT, Trinity College Dublin, Ireland); Colm Browning (MBRYONICS Ltd., Ireland & Dublin City University, Ireland); Alan Naughton (mBryonics, Ireland); Ruth Mackey (mBryonics Ltd, Ireland); David Mackey (mBryonics, Ireland); Boris Vukovic (ETH Zurich, Ireland); Jasmin Smajic (ETH Zürich, Switzerland); Juerg Leuthold (ETH Zurich, Switzerland); Carlos Natalino, Lena Wosinska and Tommy Svensson (Chalmers University of Technology, Sweden); Aleksandra M Kaszubowska-Anandarajah (Trinity College Dublin, Ireland & Connect Centre, Ireland); Ioanna Mesogiti (COSMOTE Mobile Telecommunications S.A., Greece); Simon Pryor (Accelleran NV, Belgium); Anna Tzanakaki (National and Kapodistrian University of Athens, Greece); Reza Nejabati (University of Bristol, United Kingdom (Great Britain)); Paolo Monti (Chalmers University of Technology, Sweden); Daniel Kilper (Trinity College Dublin & CONNECT Centre, Ireland)
We provide a vision for 6G fixed networks based on flexible and scalable high-capacity transmission technologies that form mesh edge networks to achieve ultra energy-efficient highly available networks with low latency. These networks will be controlled by AI-native orchestration across mobile, fixed, and compute domains. Mesh networking at the edge will be enabled by a seamless ‘confluence’ of radio fixed wireless (RFW), free space optical (FSO), and switched flex grid wavelength division multiplexed (Flex-WDM) transport using optical-spectrum-as-a-service (OSaaS) and integrated sensing and communication capabilities. In addition, in the frame of the European Smart Network Services Joint Undertaking, the ECO-eNET project will investigate key technologies and concepts to determine the full potential of confluent networks as a viable and scalable platform for 6G.
Advancing Security for 6G Smart Networks and Services
Madhusanka Liyanage (University College Dublin, Ireland); Pawani Porambage (VTT Technical Research Centre of Finland, Finland); Engin Zeydan (Centre Tecnològic de Telecomunicacions de Catalunya (CTTC), Spain); Thulitha Senevirathna (University College Dublin, Ireland); Yushan Siriwardhana (University of Oulu, Finland); Awaneesh Kumar Yadav (Indian Institute of Roorkee, Roorkee, Uttarakhand, India); Bartlomiej Siniarski (University College Dublin, Ireland)
6G Smart Networks and Services are poised to shape civilization’s development in 2030’s world, supporting the convergence of digital and physical worlds. The advent of 6G networks brings unprecedented challenges and opportunities, requiring robust security measures to safeguard against emerging threats. Thus, several complementary issues should be addressed to advance the security of 6G smart networks and services This research paper explores a multi-faceted approach to 6G security, addressing key areas of security of 6G smart networks and services such as distributed trusted AI/ML, zero-touch holistic end-to-end (E2E) security, energy efficient security and privacy enablers, real-time resilience for timing-sensitive 6G software technologies and quantum-safe 6G communications.
Decentralized Learning for 6G Security: Open Issues and Future Directions
Janani Kehelwala, Yushan Siriwardhana and Tharaka Mawanane Hewa (University of Oulu, Finland); Madhusanka Liyanage (University College Dublin, Ireland); Mika E Ylianttila (University of Oulu, Finland)
6G is envisioned with stringent performance requirements served using an open and hyper-dynamic architecture where intelligence is embedded across multiple logical layers. Artificial Intelligence (AI), a key enabling technology in implementing this vision, is often proposed in a centralized mode of operation that does not serve the scalability and fault tolerance required for the self-sustainability objectives of 6G. This article conceptualizes decentralized learning as an alternative enabling technology for 6G capable of serving fault-tolerance and scalability objectives alongside additional security and privacy prospects. It further discusses the altered threat landscape decentralization may cause and presents open issues and future research directions that must be addressed for a robust 6G infrastructure.
Unlocking Spectrum Potential: A Blockchain-Powered Paradigm for Dynamic Spectrum Management
Lavan Oshada Perera, Pasika Ranaweera, Shen Wang and Madhusanka Liyanage (University College Dublin, Ireland); Sachitha Kusaladharma (Concordia University, Canada)
The rise of mobile users, IoT devices, and data-intensive applications has led to an unprecedented surge in spectrum demand. However, current Spectrum Sharing (SS) methods, characterized by centralized control and inflexible architectures, fall short of meeting this escalating challenge. The solutions of Dynamic Spectrum Access (DSA) and Dynamic Spectrum Management (DSM) have emerged to enhance spectral efficiency and facilitate novel services in the realm of 5G networks. The successful implementation of DSM requires rapid sensing, coordination, and management, all the while upholding stringent standards for security and privacy. Unfortunately, existing DSM approaches relying on spectrum databases and Cognitive Radio (CR) techniques face issues related to reliability, security, and privacy. Blockchain (BC) emerges as a promising solution for decentralized DSM, offering superior security and privacy capabilities. Distinct features of BC, such as Smart Contracts (SCs), empower the establishment of complex Service Level Agreements (SLA) among operators. Additionally, the utilization of tokens within BC ensures a reliable and trustworthy environment for spectrum trading. Furthermore, BC’s seamless integration with artificial intelligence (AI) and related Machine Learning (ML) techniques presents an opportunity to automate and enhance the adaptability of DSM frameworks. Despite the potential, there exist research gaps that warrant attention. This paper aims to comprehensively analyze BC-based DSM as the primary solution to DSM challenges and offers clear future directions for addressing BC deployment challenges.
GANs for EVT Based Model Parameter Estimation in Real-Time Ultra-Reliable Communication
Parmida Valiahdi and Sinem Coleri (Koc University, Turkey)
The Ultra-Reliable Low-Latency Communications (URLLC) paradigm in sixth-generation (6G) systems heavily relies on precise channel modeling, especially when dealing with rare and extreme events within wireless communication channels. This paper explores a novel methodology integrating Extreme Value Theory (EVT) and Generative Adversarial Networks (GANs) to achieve the precise channel modeling in real-time. The proposed approach harnesses EVT by employing the Generalized Pareto Distribution (GPD) to model the distribution of extreme events. Subsequently, Generative Adversarial Networks (GANs) are employed to estimate the parameters of the GPD. In contrast to conventional GAN configurations that focus on estimating the overall distribution, the proposed approach involves the incorporation of an additional block within the GAN structure. This specific augmentation is designed with the explicit purpose of directly estimating the parameters of the Generalized Pareto Distribution (GPD). Through extensive simulations across different sample sizes, the proposed GAN based approach consistently demonstrates superior adaptability, surpassing Maximum Likelihood Estimation (MLE), particularly in scenarios with limited sample sizes.
