Wi-XR
Wi-XR
Wi-XR
The massive adoption of eXtended Reality (XR) in most of our daily activities regarding education, work, health, entertainment and personal life will heavily transform our society in the following years. Those XR applications will mostly reach the final user - to both consume and create XR contents - through wireless networks, and Wi-Fi in particular, due to their flexibility, mobility support and ease of use. The success of such an XR-enabled society will depend to a great extent on the ability of the wireless networks to support their stringent performance requirements. Real-time, and interactive XR applications - which mainly consist of exchanging high quality video along with metadata - require both high throughput (hundreds of Mbps per user) and low-latency (ms, or even sub-ms), which generally are opposed performance metrics.
Current wireless networks are not yet ready to cope with a massive use of XR applications in an efficient way, and so both academia and industry are pushing to find innovative solutions. The target is to have them ready by 2030. By then, Wi-Fi will likely continue being the preferred wireless access solution in unlicensed bands, and so the most used technology to consume XR experiences. However, although current Wi-Fi technologies are extremely advanced, reaching speeds of several dozens of Gbps, they are far from being ready to efficiently offer low-latency guarantees.
To support XR needs in future Wi-Fi networks a paradigm change is required from throughput-centered to latency-aware networking. This paradigm change can be supported by allowing Wi-Fi to opportunistically use different spectrum bands in a coordinated way, enable cooperation between different Access Points, implement sub-frame resource allocation schemes, exchange control information with other layers, and support the use of Machine Learning for a seamless adaptation to diverse and rapidly evolving scenarios.
This project aims to build on the aforementioned points to offer innovative and practical solutions for next-generation Wi-Fi networks, so they can successfully offer both high-throughput and low-latency guarantees to interactive streaming applications including video contents, cloud gaming, and immersive XR applications, among others. By considering new principles such as multi Access Point cooperation and opportunistic multi-channel/band access, we will be able to design particular solutions able to satisfy the needs of XR applications in dense and dynamic wireless network deployments. Moreover, we will go beyond traditional protocol/mechanism design: motivated by the need to find/learn specific functionalities that suit each unique scenario, we will focus on the design of intelligent protocols/mechanisms by embedding Machine Learning techniques in the different Wi-Fi functionalities.
The generated results will extend our knowledge in wireless networking, Wi-Fi technologies, real-time and interactive video applications (cloud gaming, and XR), and in the design of intelligent protocols. It will also contribute to developing the required new mathematical, simulation and experimental frameworks to be able to characterize the performance gains of the upcoming latency-aware Wi-Fi networks. All in all, this project will help to build a better future society by contributing to enable the use of interactive and real-time XR applications and services.
Journal Publications
2024
Wojnar, Maksymilian, Wojciech Ciezobka, Katarzyna Kosek-Szott, Krzysztof Rusek, Szymon Szott, David Nunez, and Boris Bellalta. "IEEE 802.11 bn Multi-AP Coordinated Spatial Reuse with Hierarchical Multi-Armed Bandits". IEEE Communications Letters 2024. [Article]
Wilhelmi, Francesc, Szymon Szott, Katarzyna Kosek-Szott, and Boris Bellalta. "Machine Learning & Wi-Fi: Unveiling the Path Towards AI/ML-Native IEEE 802.11 Networks." IEEE Communications Magazine 2024. [Article]
Seyedeh Soheila Shaabanzadeh, Marc Carrascosa-Zamacois, Juan Sánchez-González, Costas Michaelides, Boris Bellalta; "Virtual reality traffic prioritization for Wi-Fi quality of service improvement using machine learning classification techniques". Elsevier J. Netw. Comput. Appl. 2024. [Article]
Miguel Casasnovas, Costas Michaelides, Marc Carrascosa-Zamacois, Boris Bellalta; "Experimental Evaluation of Interactive Edge/Cloud Virtual Reality Gaming over Wi-Fi using Unity Render Streaming". Elsevier Computer Communications 2024. [https://arxiv.org/pdf/2402.00540]
Giordano, Lorenzo Galati, Giovanni Geraci, Marc Carrascosa, and Boris Bellalta; "What will Wi-Fi 8 be? A primer on IEEE 802.11 bn ultra high reliability." IEEE Communications Magazine 2024 [https://arxiv.org/pdf/2303.10442]
Marc Carrascosa-Zamacois, Giovanni Geraci, Edward W. Knightly, Boris Bellalta; "Wi-Fi Multi-Link Operation: An Experimental Study of Latency and Throughput." IEEE/ACM Trans. Netw. 32(1): 308-322 (2024) [https://arxiv.org/pdf/2305.02052]
2023
C Michaelides, Boris Bellalta; "Buffer Resets: A Packet-Discarding Policy for Timely Physiological Data Collection in Virtual Reality Gaming Systems" IEEE Sensors Letters 7 (12), 1-4 [TechArxiv link]
Boris Bellalta, Marc Carrascosa, Lorenzo Galati Giordano, Giovanni Geraci; "Delay Analysis of IEEE 802.11be Multi-Link Operation Under Finite Load". IEEE Wireless Commun. Lett. 12(4): 595-599 (2023) [https://arxiv.org/pdf/2212.12420]
Conference/Workshop Publications
Submitted
F. Wilhelmi, B. Bellalta, S. Szott, K. Kosek-Szott, S. Barrachina-Muñoz; "Coordinated Multi-Armed Bandits for Improved Spatial Reuse in Wi-Fi" [Article]
2024
Ferran Maura, Miguel Casasnovas, Boris Bellalta; "Experimenting with Adaptive Bitrate Algorithms for Virtual Reality Streaming over Wi-Fi". ACM Wintech@MobiCom 2024. [Article] [GitHub]
Carrascosa-Zamacois, Marc, Lorenzo Galati-Giordano, Francesc Wilhelmi, Gianluca Fontanesi, Anders Jonsson, Giovanni Geraci, and Boris Bellalta. "Performance Evaluation of MLO for XR Streaming: Can Wi-Fi 7 Meet the Expectations?." IEEE CAMAD 2024. [Article]
D Nunez, F Wilhelmi, L Galati-Giordano, G Geraci, B Bellalta; "Spatial Reuse in IEEE 802.11 bn Coordinated Multi-AP WLANs: A Throughput Analysis" IEEE CSCN 2024. [Article]
C. Michaelides, Boris Bellalta; "Responsive and Timely Virtual Reality Gaming: Which Frame Rate Should One Choose?" IEEE Conference on Games, 2024.
Boris Bellalta, Katarzyna Kosek-Szott, Szymon Szott, Francesc Wilhelmi; "Towards an AI/ML-defined Radio for Wi-Fi: Overview, Challenges, and Roadmap" IEEE Future Networks Technical Community (FNTC)'s "International Network Generations Roadmap" (INGR), AI/ML Chapter, 2024 Edition. [Invited Paper]
2023
C Michaelides, M Casasnovas, D Marchitelli, Boris Bellalta; "Is Wi-Fi 6 Ready for Virtual Reality Mayhem? A Case Study Using One AP and Three HMDs" IEEE Future Networks 2023. [Best Paper Award] [TechArxiv Link]
Rashid Ali, Boris Bellalta; “A Federated Reinforcement Learning Framework for Link Activation in Multi-Link Wi-Fi Networks”. IEEE BlackSeaCom 2023: 360-365 [https://arxiv.org/pdf/2304.14720]
Francesc Wilhelmi, Lorenzo Galati Giordano, Giovanni Geraci, Boris Bellalta, Gianluca Fontanesi, David Nunez; “Throughput Analysis of IEEE 802.11bn Coordinated Spatial Reuse”. IEEE CSCN 2023: 401-407 [https://arxiv.org/pdf/2309.09169]
David Nunez, Malcolm Smith, Boris Bellalta; “Multi-AP Coordinated Spatial Reuse for Wi-Fi 8: Group Creation and Scheduling”. MedComNet 2023: 203-208 [https://arxiv.org/pdf/2305.04846]
Marc Carrascosa-Zamacois, Giovanni Geraci, Lorenzo Galati Giordano, Anders Jonsson, Boris Bellalta: “Understanding Multi-link Operation in Wi-Fi 7: Performance, Anomalies, and Solutions”. IEEE PIMRC 2023: 1-6 [https://arxiv.org/pdf/2210.07695]
Marc Carrascosa-Zamacois, Lorenzo Galati Giordano, Anders Jonsson, Giovanni Geraci, Boris Bellalta: “Performance and Coexistence Evaluation of IEEE 802.11be Multi-link Operation”. IEEE WCNC 2023: 1-6 [Joint work with Nokia Bell-Labs, Collaboration with AI/ML group from UPF] [Link to IEEE Xplore]
Rooms and wgpuEngine
Rooms and wgpuEngine are open-source under the MIT license and is being actively maintained at GitHub:
Tutorials & Talks
HiPEAC 2025: "Towards Highly Reliable Wi-Fi". January 2025.
IEEE PIMRC 2024: “Where Wi-Fi Meets Ultra-High Reliability” (tutorial), with L. Galati. and F. Wilhelmi; and ”How Wi-Fi 8 Will Look Like in 2028?” (panel), with L. Galati, I. Val, F. Meneghello Sept. 2024.
IEEE MedComNet 2024: “On the Way to Wi-Fi 8: From Extremely High Throughput to Ultra High Reliability”. With G. Geraci, L. Galati. and F. Wilhelmi. Sept. 2023.
IEEE ICMLCN 2024: “Will ML Disrupt Wi-Fi? Status and Future of ML Adoption in Wi-Fi”. With K. Kosek-Szott, S. Szott, F. Wilhelmi. May 2024
”What will Wi-Fi 8 be? From Extremely High Throughput to Ultra High Reliability”. Resilent Wordls. April 2024.
IEEE PIMRC 2023: “On the Way to Wi-Fi 8: From Extremely High Throughput to Ultra High Reliability”. With G. Geraci, L. Galati. and F. Wilhelmi. Sept. 2023.
ACM Mobicom 2023: “Machine Learning and Wi-Fi: Confluences, Ongoing Activities, and Ways Forward”. With S. Szott, K. Kosek-Szott, and F. Wilhelmi. Madrid. October 2023.
IEEE ICC 2023: “On the Way to Wi-Fi 8: From Extremely High Throughput to Ultra High Reliability”. With G. Geraci and L. Galati. June 2023.
IEEE WCNC 2023: “Towards Wi-Fi 8: From Extremely High Throughput to Ultra High Reliability”. With G. Geraci and L. Galati. March 2023.
IEEE CCNC 2023: “IEEE 802.11be and Beyond: All You Need to Know about Next-generation Wi-Fi”. With G. Geraci and L. Galati. Jan. 2023.
IEEE Globecom 2022: “IEEE 802.11be and Beyond: All You Need to Know about Next-generation Wi-Fi”. With G. Geraci and L. Galati. Sept. 2022.
IEEE 802.11 standardization
Co-authors of the "IEEE 802.11 AIML TIG Technical Report." IEEE 802.11-22/0987r925, 2023. https://mentor.ieee.org/802.11/dcn/22/11-22-0987-02-aiml-aiml-tig-technical-report-draft.doc
Others
Special Session in the 21st International IEEE Conference on Factory Communication Systems (WFCS 2025): SS04 - Unlicensed Spectrum Technologies: The Next Frontier for Reliable Industrial Wireless. ORGANIZED BY: Lorenzo Galati Giordano, Nokia Bell Labs, Germany, Stefano Avallone, University of Naples, Italy, Boris Bellalta, Universitat Pompeu Fabra, Spain, Volker Jungnickel, Fraunhofer Heinrich Hertz Institute, Germany, Rainer Strobel, Maxlinear, Germany; [link]
MCIN/FEDER