AIU1 – IoT Service Management
Wednesday, 7 June 2023, 11:00-12:30, Room R22-R23
Session Chair: Benoît-Marie Robaglia (LTCI, Télécom Paris, Institut Polytechnique de Paris, France)
Dynamic and Quality-Aware Network Slice Management in 5G Testbeds
Vincent Charpentier (University of Antwerp – imec, Belgium); Nina Slamnik-Krijestorac (University of Antwerp-IMEC, Belgium); Juan Brenes (Nextworks, Italy); Andreas Gavrielides (eBOS Technologies Limited, Cyprus); Marius Iordache (Orange, Romania); Georgios Tsiouris (National Technical University of Athens, Greece); Xiangyu Lian (Telenet, Belgium); Johann M. Marquez-Barja (University of Antwerpen & imec, Belgium)
The proliferation of 5G technology is enabling vertical industries to improve their day-to-day operations by leveraging enhanced Quality of Service (QoS). One of the key enablers for such 5G performance is network slicing, which allows telco operators to logically split the network into various virtualized networks, whose configuration and thus performance can be tailored to verticals and their low-latency and high throughput requirements. However, given the end-to-end perspective of 5G ecosystems where slicing needs to be applied on all network segments, including radio, edge, transport, and core, managing the deployment of slices is becoming excessively demanding. There are also various verticals with strict requirements that need to be fulfilled. Thus, in this paper, we focus on the solution for dynamic and qualityaware network slice management and orchestration, which is simultaneously orchestrating network slices that are deployed on top of the three 5G testbeds built for transport and logistics use cases. The slice orchestration system is dynamically interacting with the testbeds, while at the same time monitoring the real-time performance of allocated slices, which is triggering decisions to either allocate new slices or reconfigure the existing ones. In this paper, we illustrate the scenarios where dynamic provisioning of slices is required in one of the testbeds while taking into account specific latency/throughput/location requirements coming from the verticals and their end users.
DESK: Distributed Observability Framework for Edge-Based Containerized Microservices
Muhammad Usman, Anna Brunstrom and Javid Taheri (Karlstad University, Sweden); Simone Ferlin (Red Hat and Karlstad University, Sweden)
Modern information technology (IT) infrastructures are becoming more complex to meet the diverse demands of emerging technology paradigms such as 5G/6G networks, edge, and internet of things (IoT). The intricacy of these infrastructures grows further when hosting containerized workloads as microservices, resulting in the challenge to detect and troubleshoot performance issues, incidents or even outages of critical use cases like industrial automation processes. Thus, fine-grained measurements and associated visualization are essential for operation observability of these IT infrastructures. However, most existing observability tools operate independently without systematically covering the entire data workflow. This paper presents an integrated design for multi-stage observability workflows, denoted as DistributEd obServability frameworK (DESK). The proposed framework aims to improve observability workflows for measurement, collection, fusion, storage, visualization, and notification. As a proof of concept, we deployed the framework in a Kubernetesbased testbed to demonstrate the successful integration of various components and usability of collected observability data. We also conducted a comprehensive study to determine the caused overhead by DESK agents at the reasonably powerful edge node hardware, which shows on average a CPU and memory overhead of around 2.5% of total available hardware resource.
SeqDQN: Multi-Agent Deep Reinforcement Learning for Uplink URLLC with Strict Deadlines
Benoît-Marie Robaglia and Marceau Coupechoux (LTCI, Télécom Paris, Institut Polytechnique de Paris, France); Dimitrios Tsilimantos (Paris Research Center, Huawei Technologies Co. Ltd., France); Apostolos Destounis (Ericsson France, France)
Recent studies suggest that Multi-Agent Reinforcement Learning (MARL) can be a promising approach to tackle wireless telecommunication problems and Multiple Access (MA) in particular. The most relevant MARL algorithms for distributed MA are those with “decentralized execution”, where an agent’s actions are only functions of their own local observation history and agents cannot exchange any information. Centralized-Training-Decentralized-Execution (CTDE) and Independent Learning (IL) are the two main families in this category. However, while the former suffers from high communication overhead during the centralized training, the latter suffers from various theoretical shortcomings. In this paper, we first study the performance of these two MARL frameworks in the context of Ultra Reliable Low Latency Communication (URLLC), where MA is constrained by strict deadlines. Second, we propose a new distributed MARL framework, namely SeqDQN, leveraging the constraints of our URLLC problem to train agents in a more efficient way. We demonstrate that not only does our solution outperform the traditional random access baselines, but it also outperforms state-of-the-art MARL algorithms in terms of performance and convergence time.
Research Challenges in Trustworthy Artificial Intelligence and Computing for Health: The Case of the PRE-ACT Project
Foivos Charalampakos, Thomas Tsouparopoulos and Yiannis Papageorgiou (Athens University of Economics and Business, Greece); Guido Bologna (University of Applied Sciences and Arts of Western Switzerland, Switzerland); André Panisson (CENTAI, Italy); Alan Perotti (CENTAI Institute, Italy); Iordanis Koutsopoulos (Athens University of Economics and Business, Greece)
The PRE-ACT project is a newly launched Horizon Europe project that aims to use Artificial Intelligence (AI) towards predicting the risk of side effects of radiotherapy treatment for breast cancer patients. In this paper, we outline four main threads pertaining to AI and computing that are part of the project’s research agenda, namely: (i) Explainable AI techniques to make the risk prediction interpretable for the patient and the clinician; (ii) Fair AI techniques to identify and explain potential biases in clinical decision support systems; (iii) Training of AI models from distributed data through Federated Learning algorithms to ensure data privacy; (iv) Mobile applications to provide the patients and clinicians with an interface for the side effect risk prediction. For each of these directions, we provide an overview of the state-of-the-art, with emphasis on techniques that are more relevant for the project. Collectively, these four threads can be seen as enforcing Trustworthy AI and pave the way to transparent and responsible AI systems that are adopted by end-users and may thus unleash the full potential of AI.
Toward Privacy-Preserving Localization and Mapping in eXtended Reality: A Privacy Threat Model
Martina Brachmann (Ericsson Research, Sweden); Gregoire Phillips (Ericsson Research, USA); Utku Gülen (Ericsson Research, Turkey); Valentin Tudor (Ericsson Research, Sweden)
This study focuses on modelling privacy threats in the eXtended Reality (XR) domain with an emphasis on generic localization and mapping processes that run on or are offloaded by widely used equipment such as XR glasses. With the help of a privacy threat assessment framework, we model privacy threats that can affect the different XR processes and equipment. In addition to the threat analysis results, we provide a brief overview of threat mitigation techniques for each threat category. This strengthens privacy in XR processes and equipment and allows for privacy-preserving localization and mapping in XR scenario.