kth.sePublications
1231 of 3
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Predictive Quality of Service for Enhanced Wireless Vehicular Applications
KTH, School of Electrical Engineering and Computer Science (EECS), Computer Science, Network and Systems Engineering. Ericsson Research, Sweden.ORCID iD: 0000-0001-8499-9162
2024 (English)Licentiate thesis, comprehensive summary (Other academic)
Abstract [en]

In recent years, the rapid advancement of emerging technologies has significantly fueled the expansion of the Internet of Things (IoT) and the increase of wireless connected devices in society. IoT devices with a safety-critical nature, such as remotely operated vehicles, demand a high quality of the communication service to function reliably. However, consistently achieving high quality of service (QoS) can be challenging for vehicles with high mobility due to changes in interference and the propagation environment, which may cause fluctuations in the statistical distributions that govern the channel and that model the wireless communication performance. This makes it very difficult to predict future communication conditions and performance, such as the wireless channel conditions and QoS metrics. Yet, by forecasting these factors, cellular networks can transition from a reactive approach and become more proactive. By anticipating performance degradations and allocating resources accordingly, safety-critical applications can operate without disruptions.

Unfortunately, the rapid fluctuations of the statistical parameters make it very difficult to predict them by model-based methods. Recently, research has highlighted the potential of machine learning (ML) to constitute models for predictive QoS (pQoS). ML algorithms can learn from vast amounts of data and identify complex patterns that may not be apparent from traditional methods. Its capability to adapt to new data by dynamical model updates makes ML particularly suitable for environments where the QoS is constantly changing. By leveraging ML for predictive purposes, network operators can ensure more efficient resource allocation and a robust network infrastructure. 

The first part of this thesis provides an essential overview of the dynamics in wireless communication systems, focusing on the wireless channel and QoS. The foundations of how ML learns from datasets along with an overview of popular deep neural networks (DNNs) are presented. We summarize the course of our research including a survey on wireless channel prediction, a novel pQoS model, and a QoS prediction framework along with a network digital twin (NDT). A summary of the principal contributions from our research concludes the overview of the thesis.

In the second part of this thesis, we report our major results. We introduce the innovative pQoS model, specifically for connected vehicles, which creates geographical segments, clusters the segments, optimizes the number of clusters, and trains a pQoS model for each cluster using federated learning (FL). We show how this predictive framework improves approaches commonly implemented in previous research, only considering one global predictive model. Moreover, an overview of wireless channel prediction is provided together with an extensive numerical evaluation of DNNs for the purpose of channel prediction, addressing the gap in previous research. Finally, a proof of concept of a real-time NDT based on experimental data is presented to predict the QoS in an enterprise process.

Abstract [sv]

Den hastiga utvecklingen av nya teknologier under de senaste åren har främjat tillväxten av Internet of Things (IoT) och därmed kraftigt ökat antalet trådlöst anslutna enheter i samhället. IoT-enheter med säkerhetskritiska applikationer, såsom autonoma fordon, kräver en hög quality of service (QoS) av det trådlösa nätverket för att fungera pålitligt. Att konsekvent uppnå en hög QoS kan dock vara utmanande för fordon med hög mobilitet på grund av förändringar i den omgivande miljön. Dessa förändringar kan orsaka kraftiga fluktuationer, samt variationer över längre tid, av de underliggande statistiska egenskaperna för trådlösa kommunikationsförhållanden. En innovativ strategi för att hantera de utmaningar som sådana variationer medför är att förutsäga framtida kommunikationsförhållanden och prestanda, såsom de trådlösa kanalförhållandena och den realiserbara QoS-nivån. Genom att förutsäga dessa faktorer kan operatörer av trådlösa nätverk övergå från ett reaktivt tillvägagångssätt och bli mer proaktiva. Med en proaktiv strategi kan mobiloperatörer fördela resurser utifrån förutsedda QoS-svackor och därmed tillgodose säkerhetskritiska applikationer med den prestanda de kräver.

Ny forskning betonar maskininlärnings potential samt dess kapacitet för att utgöra modeller för prediktiv QoS (pQoS). Maskininlärning lär sig från stora datamängder och identifierar komplexa mönster som inte är uppenbara med traditionella metoder. Dess förmåga att anpassa sig till ny data genom dynamiska modelluppdateringar gör maskininlärning särskilt lämplig för att förutsäga QoS där kommunikationsförhållandena ständigt förändras. Med hjälp av maskininlärning kan nätverksoperatörer säkerställa en mer effektiv resursallokering och en robust infrastruktur i trådlösa nätverk.

Den första delen av denna avhandling presenterar en överskådlig bakgrund över dynamiken i trådlös kommunikation, med fokus på den trådlösa kanalen och QoS. Grunderna för maskininlärning presenteras tillsammans med flera framstående neurala nätverk inom forskningsfältet. Kärnan i vår forskning sammanfattas, där vi föreslår en ny pQoS-modell, ett ramverk för pQoS med en digital tvilling av nätverket, och en översikt över trådlös kanalprediktion. Denna sammanfattning av de huvudsakliga bidragen från vår forskning avslutar överblicken av avhandlingen.

I den andra delen av denna avhandling introducerar vi den föreslagna pQoS-modellen, specifikt skapad för anslutna fordon med hög mobilitet, som konsturerar geografiska segment, delar in segmenten kluster, optimerar antalet kluster och tränar en pQoS-modell för varje kluster med hjälp av federated learning (FL). Vi visar hur detta prediktiva ramverk förbättrar tillvägagångssätt som vanligtvis implementerats i tidigare forskning, som endast överväger en globalt prediktiv modell. Dessutom presenteras en översikt för trådlös kanalprediktion tillsammans med en omfattande numerisk utvärdering av djupa neurala nätverk för ändamålet att prediktera den trådlösa kanalen, vilket adresserar den vetenskapliga luckan från föregående forskning. Slutligen presenteras en koncepttest utav en digital tvilling av ett trådlöst nätverk i realtid, baserat på experimentella data, för att förutsäga QoS i en fabriksprocess.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology, 2024. , p. viii, 35
Series
TRITA-EECS-AVL ; 2024:70
National Category
Communication Systems Telecommunications
Research subject
Electrical Engineering
Identifiers
URN: urn:nbn:se:kth:diva-354816ISBN: 978-91-8106-047-8 (print)OAI: oai:DiVA.org:kth-354816DiVA, id: diva2:1905403
Presentation
2024-11-07, https://kth-se.zoom.us/j/68286726367, D3, Lindstedtsvägen 9, Stockholm, 13:00 (English)
Opponent
Supervisors
Funder
Wallenberg AI, Autonomous Systems and Software Program (WASP)
Note

QC 20241014

Available from: 2024-10-14 Created: 2024-10-14 Last updated: 2024-10-21Bibliographically approved
List of papers
1. Clustered Predictive Quality of Service for Connected Vehicles Using Federated Learning
Open this publication in new window or tab >>Clustered Predictive Quality of Service for Connected Vehicles Using Federated Learning
2024 (English)In: 2024 IEEE International Conference on Communications Workshops, ICC Workshops 2024, Institute of Electrical and Electronics Engineers (IEEE) , 2024, p. 626-631Conference paper, Published paper (Refereed)
Abstract [en]

To enable the development of multi-functional cellular networks that aim to satisfy increasing expectations of connectivity and trustworthiness, it is crucial to provide reliable quality of service (QoS) guarantees for end users. With predictive QoS (pQoS), cellular networks become proactive to meet QoS requirements for a wide variety of new use cases, including advanced driver assistance applications. This work introduces a novel predictive framework to improve the availability and performance of pQoS in cellular networks, especially for advanced road transport applications. We show that by dividing the road into geographical segments, clustering segments with similar data, and assigning each cluster a predictive model, the adversary effects of the propagation environment and interference on QoS become manageable. To this end, each predictive cluster model is trained locally on vehicles within the cluster boundaries by data driven Federated Learning, resulting in personalized predictive models for each cluster. Our numerical results show that the clustered predictive model outperforms the more common predictive approach proposed by previous works that train a single global predictive model for an entire dataset.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2024
Keywords
6G, Advanced driver assistance systems, Deep Learning, Federated Learning, Quality of Service, Vehicular and wireless technologies
National Category
Computer Sciences Telecommunications
Identifiers
urn:nbn:se:kth:diva-353548 (URN)10.1109/ICCWorkshops59551.2024.10615431 (DOI)2-s2.0-85202452496 (Scopus ID)
Conference
59th Annual IEEE International Conference on Communications Workshops, ICC Workshops 2024, Denver, United States of America, Jun 9 2024 - Jun 13 2024
Note

Part of ISBN 9798350304053

QC 20240924

Available from: 2024-09-19 Created: 2024-09-19 Last updated: 2024-10-14Bibliographically approved
2. AI-aided Channel Prediction
Open this publication in new window or tab >>AI-aided Channel Prediction
2025 (English)In: Artificial Intelligence for Future Networks / [ed] Mohammad A. Matin, Sotirios K. Goudos, George K. Karagiannidis, Wiley , 2025Chapter in book (Other academic)
Abstract [en]

The wireless communication systems of today rely to a large extent on the condition of the accessible channel state information (CSI) at the transmitter and receiver. Channel aging, denoting the temporal and spatial evolution of wireless communication channels, is influenced by obstructions, interference, traffic load, and user mobility. Accurate CSI estimation and prediction empower the network to proactively counteract performance degradation resulting from channel dynamics, such as channel aging, by employing network management strategies such as power allocation. Prior studies have introduced approaches aimed at preserving high-quality CSI such as temporal prediction schemes, particularly in scenarios involving high mobility and channel aging. Conventional model-based estimators and predictors have historically been considered state-of-the-art. Recently, the development of artificial intelligence (AI) has increased the interest in developing models based on AI. Previous works have shown high potential of AI-aided channel estimation and prediction, which inclines the state-of-the-art title from model-based methods to be confiscated. However, there are many aspects to consider in channel estimation and prediction employed by AI in terms of prediction quality, training complexity, and practical feasibility. To investigate these aspects, this chapter provides an overview of state-of-the-art neural networks, applicable to channel estimation and channel prediction. The principal neural networks from the overview of channel prediction are empirically compared in terms of prediction quality. An innovative comparative analysis is conducted for five prospective neural networks characterized by distinct prediction horizons. The widely acknowledged tapped delay line (TDL) channel model, as endorsed by the Third Generation Partnership Project (3GPP), is employed to ensure a standardized evaluation of the neural networks. This comparative assessment enables a comprehensive examination of the merits and demerits inherent in each neural network. Subsequent to this analysis, insights are offered to provide guidelines for the selection of the most appropriate neural network in channel prediction applications.

Place, publisher, year, edition, pages
Wiley, 2025
National Category
Communication Systems Telecommunications
Identifiers
urn:nbn:se:kth:diva-354801 (URN)
Note

Part of book ISBN 978-1-394-22792-1

QC 20241015

Available from: 2024-10-14 Created: 2024-10-14 Last updated: 2024-10-15Bibliographically approved
3. AI-based Real-Time Network Digital Twin: Proof of Concept
Open this publication in new window or tab >>AI-based Real-Time Network Digital Twin: Proof of Concept
Show others...
(English)Manuscript (preprint) (Other academic)
Abstract [en]

Network digital twins (NDTs) are virtual representations of network assets and processes, such as in 5G or 6G networks, synchronized to the physical properties. NDTs can be leveraged for network monitoring, automation, and optimization tasks, and may even interact with other external digital twins, such as an enterprise digital twin (EDT). The interaction between an NDT and an EDT can be beneficial for either optimizing enterprise processes based on network insights or optimizing network services based on inputs from the enterprise processes. \os{For example, predictions from the NDT can be provided upon generated critical scenarios requested from an EDT, to scout the feasibility of certain industrial processes from a network connectivity quality perspective. This article presents insights and results from an NDT proof of concept, realized by processing real-time data obtained from commercially available communication equipment in an indoor industrial-like environment.} Neural networks, assisted by comprehensive feature selection and extraction, are integrated into the NDT to predict the key performance indicator (KPI), namely the downlink user throughput. KPI predictions from the NDT are provided based on requests from the generated demanding scenarios. The results illustrate the significance of having access to an NDT and an EDT in symbiosis, under varying network conditions.

National Category
Communication Systems
Identifiers
urn:nbn:se:kth:diva-354803 (URN)
Note

QC 20241015

Available from: 2024-10-14 Created: 2024-10-14 Last updated: 2024-10-15Bibliographically approved
4. Clustering of Geographical Segments for Predictive Quality of Service of Connected Vehicles
Open this publication in new window or tab >>Clustering of Geographical Segments for Predictive Quality of Service of Connected Vehicles
(English)Manuscript (preprint) (Other academic)
Abstract [en]

To meet the growing demands for connectivity and reliability in cellular networks, it is essential to ensure reliable quality of service (QoS) guarantees for end users. The integration of predictive QoS (pQoS) in cellular networks enables proactive fulfillment of QoS requirements for a diverse range of applications, including intelligent transportation systems. This study presents a pQoS framework in cellular networks, particularly for connected vehicles. By segmenting the road into segments, clustering them, and assigning a pQoS model to each cluster, we mitigate concept drift of the pQoS model caused by the propagation environment and interference. Each predictive cluster model is locally trained on vehicles traveling within the cluster boundaries using federated learning (FL). A significant challenge is balancing the trade-off between the number of clusters, prediction accuracy, and communication overhead for updating local models. This trade-off suggests the novel problem of performing a joint optimization of the training and number of clusters. To address such difficult optimization, we propose an iterative approximate solution using proximal alternative minimization for which we provide convergence guarantees. Ultimately, our numerical findings indicate that the clustered predictive model achieves higher prediction accuracy than conventional predictive approaches proposed by prior studies.

National Category
Communication Systems
Identifiers
urn:nbn:se:kth:diva-354802 (URN)
Note

QC 20241015

Available from: 2024-10-14 Created: 2024-10-14 Last updated: 2024-10-15Bibliographically approved

Open Access in DiVA

summary(12185 kB)57 downloads
File information
File name SUMMARY01.pdfFile size 12185 kBChecksum SHA-512
0d2057a23c0ad9a713fed36779d92be492072f9faea7ace2f3148c9aa149e5369d08d071f07f9713145a7abf447611957be11668eab0c34ca8fc5e4cd3e50f3d
Type summaryMimetype application/pdf

Authority records

Stenhammar, Oscar

Search in DiVA

By author/editor
Stenhammar, Oscar
By organisation
Network and Systems Engineering
Communication SystemsTelecommunications

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 144 hits
1231 of 3
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf