WOS22024-05-17T13:38:27+00:00

WOS2 – Emerging Technologies and Strategies for Enhanced Connectivity and Efficient Resource Utilization in Future Telecommunication Networks

Wednesday, 5 June 2024, 16:00-17:30, room Toucan 1

Session Chair: Dries Naudts (Ghent University & Imec, BE)

An Efficient Access Point Assignment for Optical-Radio Networks by Multi-Attribute Decision-Making
Mohammad Khalili, Guanghui Ma, Konstantin Mikhaylov and Marcos Katz (University of Oulu, Finland)
In the rapidly evolving domain of optical wireless communication and radio frequency heterogeneous network (OWC/RF HetNet), the Access Point (AP) assignment is an important and multi-attribute problem. The challenges of this problem arise from the distinct characteristics of OWC and RF networks, combined with the difference between uplink and downlinks to and from optical and RF access points (APs). Traditional AP assignment methods, with their limited focus on a few attributes, often overlook a wider array of essential factors. Addressing this, we introduce a Fuzzy Technique for Order Preference by Similarity to the Ideal Solution algorithm as a multi-criteria approach to concurrently consider different attributes for AP assignment. In particular, this study considers some attributes like achievable data rates, node-APs distances, and cost-effectiveness for the AP assignment in the OWC/RF HetNet. Simulation results show that the proposed approach can provide a trade-off between various attributes like data rate, coverage range, and costs. Furthermore, this study paves the way by providing a robust dataset, primed for machine learning approaches in subsequent AP assignments for OWC/RF HetNet.

Age of Information and Value of Information Aware Optimal Packet Transmission for an AGV System
Shreya Tayade (German Research Center for Artificial Intelligence, Germany); Hans Dieter Schotten (Deutsches Forschungszentrum für Künstliche Intelligenz GmbH, Germany); Hans D. Schotten (University of Kaiserslautern, Germany)
In this paper we propose control aware packet transmissions for an industrial control use case. We consider an edge cloud based Automated Guided Vehicle (AGV) control system operating over a correlated wireless fading channel. The AGV follows a reference track and the control packets are sent periodically in the downlink by an edge cloud controller. Transmitting the control packets frequently with high reliability ensures control system stability but increases the resource usage. On contrary, the resource efficiency can be increased by transmitting the control packet less frequently. However, the control system may become unstable due to higher Age of Information (AoI). Therefore, we derive an optimal control packet transmit time that minimizes the control error and ensures the stability considering the AoI, Value of Information (VoI), control error and channel conditions. The results show that the channel coherence time influences the optimal transmit time and the AGV error performance significantly. The control packets can be transmitted less frequently in case of a higher error tolerance and lower VoI. Moreover, we evaluate the maximum permissible time to transmit control packets to ensure that the AGV’s control error is within a given threshold.

Decentralized RL-Based Data Transmission Scheme for Energy Efficient Harvesting
Rafaela Scaciota (University of Oulu, Finland); Glauber Brante (Federal University of Technology – Paraná (UTFPR), Brazil); Richard Demo Souza (Federal University of Santa Catarina, Brazil); Onel L. A. López and Septimia Sarbu (University of Oulu, Finland); Mehdi Bennis (Centre of Wireless Communications, University of Oulu, Finland); Sumudu Samarakoon (University of Oulu, Finland)
The evolving landscape of the Internet of Things (IoT) has given rise to a pressing need for an efficient communication scheme. As the IoT user ecosystem continues to expand, traditional communication protocols grapple with substantial challenges in meeting its burgeoning demands, including energy consumption, scalability, data management, and interference. In response to this, the integration of wireless power transfer and data transmission has emerged as a promising solution. This paper considers an energy harvesting (EH)-oriented data transmission scheme, where a set of users are charged by their own multi-antenna power beacon (PB) and subsequently transmits data to a base station (BS) using an irregular slotted aloha (IRSA) channel access protocol. We propose a closed-form expression to model energy consumption for the present scheme, employing average channel state information (A-CSI) beamforming in the wireless power channel. Subsequently, we employ the reinforcement learning (RL) methodology, wherein every user functions as an agent tasked with the goal of uncovering their most effective strategy for replicating transmissions. This strategy is devised while factoring in their energy constraints and the maximum number of packets they need to transmit. Our results underscore the viability of this solution, particularly when the PB can be strategically positioned to ensure a strong line-of-sight connection with the user, highlighting the potential benefits of optimal deployment.

DNN Partitioning and Inference Task Offloading in 6G Resource-Constrained Networks
Dimitrios Kafetzis and Iordanis Koutsopoulos (Athens University of Economics and Business, Greece)
In the emerging 6G network landscape, edge computing applications are tasked with performing Machine-Learning (ML) inference using Deep Neural Networks (DNNs) on resource-constrained devices in terms of computational power, memory, and energy. This paradigm shift, brought forth by 6G’s promise of ultra-reliable low-latency communication necessitates novel approaches for managing DNN tasks. This paper presents an optimization-driven methodology for DNN partitioning and selective offloading of ML inference tasks to a Base Station (BS) or an edge gateway node equipped with substantial computational resources. We consider a set of resource-constrained devices, each of which has an inference task (abstracted as a DNN) to execute. DNN partitioning determines the part of the neural network to run locally, and the part to offload to the BS. We are interested to determine a suitable task partition for each device, namely determine the neural network layer up to which computation will take place locally on the device. A certain task partition on a device affects the inference delays of tasks of other devices due to task coupling, when task scheduling for processing is decided at the BS. We formulate the optimization problem of DNN partitioning for minimizing total execution delays of tasks or for providing fair treatment to DNN inference tasks, by ensuring balanced task execution delays. We propose a greedy heuristic algorithm to solve the problem, and we evaluate it through numerical simulations, using input from experiment measurements on Raspberry Pi devices. The proposed approach is shown to perform much better than the baseline approach where each task is entirely executed locally on the device.

Network Intelligence in Action: The DAEMON Perspective
Livia Elena Chatzieleftheriou (IMDEA Networks Institute, Spain); Marco Gramaglia (Universidad Carlos III de Madrid, Spain); Marco Fiore (IMDEA Networks Institute, Spain); Nina Slamnik-Krijestorac (University of Antwerp-IMEC, Belgium); Miguel Camelo (University of Antwerp – imec, Belgium); Paola Soto (IMEC, Belgium & Universidad de Antioquia, Colombia); Evangelos Kosmatos (WINGS ICT Solutions, Greece); Andres Garcia-Saavedra (NEC Labs Europe, Germany); Michele Gucciardo (IMDEA Networks Institute, Spain)
The NIO plays a pivotal role in facilitating the lifecycle management of network intelligence within the DAEMON project. Working in tandem with the NIF Managers and the NIF Component Manager, which handle lower-level factors, such as container management for specific intelligence com- ponents, the NIO encompasses several modules that are of paramount importance within the scope of WP3. The NIO plays a pivotal role in facilitating the lifecycle management of network intelligence within the DAEMON project. Working in tandem with the NIF Managers and the NIF Component Manager, which handle lower-level factors, such as container management for specific intelligence com- ponents, the NIO encompasses several modules that are of paramount importance within the scope of WP3. The NIO plays a pivotal role in facilitating the lifecycle management of network intelligence within the DAEMON project. Working in tandem with the NIF Managers and the NIF Component Manager, which handle lower-level factors, such as container management for specific intelligence com- ponents, the NIO encompasses several modules that are of paramount importance within the scope of WP3.

Go to Top