Skip to main content

MACHINE LEARNING FOR WIRELESS NETWORKING IN HIGHLY DYNAMIC SCENARIOS

PGC2018-099959-B-I00

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal de Generación de Conocimiento y Fortalecimiento Científico y Tecnológico del Sistema de I+D+i
Subprogram Subprograma Estatal de Generación de Conocimiento
Call Proyectos de I+D (Excelencia / Generación de Conocimiento)
Call year 2018
State Plan 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSITAT POMPEU FABRA CCT

Resultados relacionados

Resultados totales (Incluyendo duplicados): 10
Encontrada(s) 1 página(s)

Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

TXOP sharing with coordinated spatial reuse in multi-AP cooperative IEEE 802.11be WLANs

  • Núñez Cuadrado, David
  • Wilhelmi Roca, Francesc
  • Avallone, Stefano
  • Smith, M.
  • Bellalta, Boris
Comunicació presentada a 2022 IEEE 19th Annual Consumer Communications & Networking Conference (CCNC), celebrat del 8 a l'11 de gener de 2022 de manera virtual., IEEE 802.11be networks (aka Wi-Fi 7) will have to cope with new bandwidth-hungry and low-latency services such as eXtended Reality and multi-party cloud gaming. With this goal in mind, transmit opportunity (TXOP) sharing between coordinated access points (APs) may contribute to alleviating inter-AP contention, hence increasing the overall network throughput. This paper evaluates two coordinated TXOP sharing strategies: coordinated time division multiple access (c-TDMA) and coordinatedTDMA with spatial reuse (c-TDMA/SR). We show that, while c-TDMA alone does not result in any significant improvement in terms of the WLAN throughput, it lays the groundwork to implement coordinated SR (c-SR) techniques. To evaluate the performance of c-TDMA/SR, we propose a fair scheduler able to select the best subset of parallel transmissions in WLAN deployments, as well as the appropriate power levels to be used by APs and stations (STAs), leading to maximum performance. The results obtained for c-TDMA/SR show significant throughput gains compared with c-TDMA, with values higher than 140% in 90% of the considered scenarios., The work of D. Nuñez and B. Bellalta was supported by WINDMAL PGC2018-099959-B-I00 (MCIU/AEI/FEDER,UE) and Cisco.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

ECTS: enhanced centralized TSCH scheduling with packet aggregation for industrial IoT

  • Michaelides, Michael
  • Adame, Toni
  • Bellalta, Boris
Comunicació presentada a 2021 IEEE Conference on Standards for Communications and Networking (CSCN), celebrat del 15 al 17 de desembre de 2021 de manera virtual., The Industrial Internet of Things (IoT) has gained a lot of momentum thanks to the introduction of Time Slotted Channel Hopping (TSCH) in IEEE 802.15.4. At last, we can enjoy collision-free, low-latency wireless communication in challenging environments. Nevertheless, the fixed size of time slots in TSCH provides an opportunity for further enhancements. In this paper, we propose an enhanced centralized TSCH scheduling (ECTS) algorithm with simple packet aggregation while collecting data over a tree topology. Having in mind that the payload of a sensor node is rather short, we attempt to put more than one payload in one packet. Thus, we occupy just one cell to forward them. We investigated the schedule compactness of ECTS in Matlab, and we evaluated its operation, after implementing it in Contiki-NG, using Cooja. Our results show that ECTS with packet aggregation outperforms TASA in terms of slotframe duration and imposes fairness among the nodes in terms of latency. A validation exercise using real motes confirms its successful operation in real deployments., This work was supported in part by grant WINDMAL PGC2018-099959-B-I00 (MCIU/AEI/FEDER,UE), and co-financed by the European Union Regional Development Fund within the framework of the ERDF Operational Program of Catalonia 2014–2020 (Looming Factory).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

An experimental study of latency for IEEE 802.11be multi-link operation

  • Carrascosa Zamacois, Marc
  • Geraci, Giovanni
  • Knightly, Edward
  • Bellalta, Boris
Comunicació presentada a IEEE International Conference on Communications (ICC 2022), celebrat del 16 al 20 de maig de 2022 a Seül, Corea del Sud., Will Multi-Link Operation (MLO) be able to improve the latency of Wi-Fi networks? MLO is one of the most disruptive MAC-layer techniques included in the IEEE 802.11be amendment. It allows a device to use multiple radios simultaneously and in a coordinated way, providing a new framework to improve the WLAN throughput and latency. In this paper, we investigate the potential latency benefits of MLO by using a large dataset containing 5 GHz spectrum occupancy measurements. Experimental results show that when the channels are symmetrically occupied, MLO can improve latency by one order of magnitude. In contrast, in asymmetrically occupied channels, MLO can sometimes be detrimental and increase latency. This is a result of packets being assigned to an interface before carrying out the backoff, which is more likely to be interrupted on the busier link. We overcome this issue by allowing multiple backoffs to run in parallel, assigning the packet to the particular interface where the backoff expires first, which also achieves lower latency overall., M. Carrascosa and B. Bellalta were supported by WINDMAL PGC2018- 099959-B-I00 (MCIU/AEI/FEDER,UE) and Cisco. G. Geraci was supported by MINECO’s Project RTI2018-101040 and by a “Ramón y Cajal” Fellowship from the Spanish State Research Agency. E. Knightly’s research was supported by Cisco, Intel, and NSF Grants CNS1955075 and CNS-1923782.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Wi-fi all-channel analyzer

  • Barrachina Muñoz, Sergio
  • Bellalta, Boris
  • Knightly, Edward
Comunicació presentada al MobiCom '20: The 26th Annual International Conference on Mobile Computing and Networking, celebrat el 15 de setembre de 2020 a Londres, Regne Unit., In this paper, we present WACA, the first system to simultaneously measure the energy in all 24 Wi-Fi channels that allow channel bonding at 5 GHz with microsecond scale granularity. With WACA, we perform a first-of-its-kind measurement campaign in areas including urban hotspots, residential neighborhoods, universities, and a sold-out stadium with 98,000 fans and 12,000 simultaneous Wi-Fi connections. The gathered dataset is a unique asset to find insights otherwise not possible in the context of multi-channel technologies like Wi-Fi. To show its potential, we compare the performance of contiguous and non-contiguous channel bonding using a trace-driven framework. We show that while non-contiguous outperforms contiguous channel bonding's throughput, occasionally bigger by a factor of 5, their average throughputs are similar., The work from S. BarrachinaMuoz and B. Bellalta has been partially supported by CISCO (CG#890107-SVCF), Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), WINDMAL PGC2018- 099959-B-I00 (MCIU/AEI/FEDER,UE), and SGR-2017- 1188. The research of E. Knightly was supported by Cisco, Intel, and by NSF grants CNS-1955075 and CNS-1824529.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Stateless reinforcement learning for multi-agent systems: the case of spectrum allocation in dynamic channel bonding WLANs

  • Barrachina Muñoz, Sergio
  • Chiumento, Alessandro
  • Bellalta, Boris
Comunicació presentada a 2021 Wireless Days (WD), celebrat del 30 de juny al 2 de juliol de 2021 a París, França., Spectrum allocation in the form of primary channel and bandwidth selection is a key factor for dynamic channel bonding (DCB) wireless local area networks (WLANs). To cope with varying environments, where networks change their configurations on their own, the wireless community is looking towards solutions aided by machine learning (ML), and especially reinforcement learning (RL) given its trial-and-error approach. However, strong assumptions are normally made to let complex RL models converge to near-optimal solutions. Our goal with this paper is two-fold: justify in a comprehensible way why RL should be the approach for wireless networks problems like decentralized spectrum allocation, and call into question whether the use of complex RL algorithms helps the quest of rapid learning in realistic scenarios. We derive that stateless RL in the form of lightweight multi-armed-bandits (MABs) is an efficient solution for rapid adaptation avoiding the definition of extensive or meaningless RL states., The work of Sergio Barrachina-Munoz and Boris-Bellalta was ˜ supported in part by Cisco, WINDMAL under Grant PGC2018- 099959-B-I00 (MCIU/AEI/FEDER,UE) and Grant SGR-2017-1188. Alessandro Chiumento is partially funded by the InSecTT project (https://www.insectt.eu/) which has received funding from the ECSEL Joint Undertaking (JU) under grant agreement No 876038.
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

An experimental study of latency for IEEE 802.11be multi-link operation

  • Carrascosa Zamacois, Marc
  • Geraci, Giovanni
  • Knightly, Edward
  • Bellalta, Boris
Comunicació presentada a IEEE International Conference on Communications (ICC 2022), celebrat del 16 al 20 de maig de 2022 a Seül, Corea del Sud., Will Multi-Link Operation (MLO) be able to improve the latency of Wi-Fi networks? MLO is one of the most disruptive MAC-layer techniques included in the IEEE 802.11be amendment. It allows a device to use multiple radios simultaneously and in a coordinated way, providing a new framework to improve the WLAN throughput and latency. In this paper, we investigate the potential latency benefits of MLO by using a large dataset containing 5 GHz spectrum occupancy measurements. Experimental results show that when the channels are symmetrically occupied, MLO can improve latency by one order of magnitude. In contrast, in asymmetrically occupied channels, MLO can sometimes be detrimental and increase latency. This is a result of packets being assigned to an interface before carrying out the backoff, which is more likely to be interrupted on the busier link. We overcome this issue by allowing multiple backoffs to run in parallel, assigning the packet to the particular interface where the backoff expires first, which also achieves lower latency overall., M. Carrascosa and B. Bellalta were supported by WINDMAL PGC2018- 099959-B-I00 (MCIU/AEI/FEDER,UE) and Cisco. G. Geraci was supported by MINECO’s Project RTI2018-101040 and by a “Ramón y Cajal” Fellowship from the Spanish State Research Agency. E. Knightly’s research was supported by Cisco, Intel, and NSF Grants CNS1955075 and CNS-1923782.
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Stateless reinforcement learning for multi-agent systems: the case of spectrum allocation in dynamic channel bonding WLANs

  • Barrachina Muñoz, Sergio
  • Chiumento, Alessandro
  • Bellalta, Boris
Comunicació presentada a 2021 Wireless Days (WD), celebrat del 30 de juny al 2 de juliol de 2021 a París, França., Spectrum allocation in the form of primary channel and bandwidth selection is a key factor for dynamic channel bonding (DCB) wireless local area networks (WLANs). To cope with varying environments, where networks change their configurations on their own, the wireless community is looking towards solutions aided by machine learning (ML), and especially reinforcement learning (RL) given its trial-and-error approach. However, strong assumptions are normally made to let complex RL models converge to near-optimal solutions. Our goal with this paper is two-fold: justify in a comprehensible way why RL should be the approach for wireless networks problems like decentralized spectrum allocation, and call into question whether the use of complex RL algorithms helps the quest of rapid learning in realistic scenarios. We derive that stateless RL in the form of lightweight multi-armed-bandits (MABs) is an efficient solution for rapid adaptation avoiding the definition of extensive or meaningless RL states., The work of Sergio Barrachina-Munoz and Boris-Bellalta was ˜ supported in part by Cisco, WINDMAL under Grant PGC2018- 099959-B-I00 (MCIU/AEI/FEDER,UE) and Grant SGR-2017-1188. Alessandro Chiumento is partially funded by the InSecTT project (https://www.insectt.eu/) which has received funding from the ECSEL Joint Undertaking (JU) under grant agreement No 876038.
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Wi-fi all-channel analyzer

  • Barrachina Muñoz, Sergio
  • Bellalta, Boris
  • Knightly, Edward
Comunicació presentada al MobiCom '20: The 26th Annual International Conference on Mobile Computing and Networking, celebrat el 15 de setembre de 2020 a Londres, Regne Unit., In this paper, we present WACA, the first system to simultaneously measure the energy in all 24 Wi-Fi channels that allow channel bonding at 5 GHz with microsecond scale granularity. With WACA, we perform a first-of-its-kind measurement campaign in areas including urban hotspots, residential neighborhoods, universities, and a sold-out stadium with 98,000 fans and 12,000 simultaneous Wi-Fi connections. The gathered dataset is a unique asset to find insights otherwise not possible in the context of multi-channel technologies like Wi-Fi. To show its potential, we compare the performance of contiguous and non-contiguous channel bonding using a trace-driven framework. We show that while non-contiguous outperforms contiguous channel bonding's throughput, occasionally bigger by a factor of 5, their average throughputs are similar., The work from S. BarrachinaMuoz and B. Bellalta has been partially supported by CISCO (CG#890107-SVCF), Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), WINDMAL PGC2018- 099959-B-I00 (MCIU/AEI/FEDER,UE), and SGR-2017- 1188. The research of E. Knightly was supported by Cisco, Intel, and by NSF grants CNS-1955075 and CNS-1824529.
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

TXOP sharing with coordinated spatial reuse in multi-AP cooperative IEEE 802.11be WLANs

  • Núñez Cuadrado, David
  • Wilhelmi Roca, Francesc
  • Avallone, Stefano
  • Smith, M.
  • Bellalta, Boris
Comunicació presentada a 2022 IEEE 19th Annual Consumer Communications & Networking Conference (CCNC), celebrat del 8 a l'11 de gener de 2022 de manera virtual., IEEE 802.11be networks (aka Wi-Fi 7) will have to cope with new bandwidth-hungry and low-latency services such as eXtended Reality and multi-party cloud gaming. With this goal in mind, transmit opportunity (TXOP) sharing between coordinated access points (APs) may contribute to alleviating inter-AP contention, hence increasing the overall network throughput. This paper evaluates two coordinated TXOP sharing strategies: coordinated time division multiple access (c-TDMA) and coordinatedTDMA with spatial reuse (c-TDMA/SR). We show that, while c-TDMA alone does not result in any significant improvement in terms of the WLAN throughput, it lays the groundwork to implement coordinated SR (c-SR) techniques. To evaluate the performance of c-TDMA/SR, we propose a fair scheduler able to select the best subset of parallel transmissions in WLAN deployments, as well as the appropriate power levels to be used by APs and stations (STAs), leading to maximum performance. The results obtained for c-TDMA/SR show significant throughput gains compared with c-TDMA, with values higher than 140% in 90% of the considered scenarios., The work of D. Nuñez and B. Bellalta was supported by WINDMAL PGC2018-099959-B-I00 (MCIU/AEI/FEDER,UE) and Cisco.
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

ECTS: enhanced centralized TSCH scheduling with packet aggregation for industrial IoT

  • Michaelides, Michael
  • Adame, Toni
  • Bellalta, Boris
Comunicació presentada a 2021 IEEE Conference on Standards for Communications and Networking (CSCN), celebrat del 15 al 17 de desembre de 2021 de manera virtual., The Industrial Internet of Things (IoT) has gained a lot of momentum thanks to the introduction of Time Slotted Channel Hopping (TSCH) in IEEE 802.15.4. At last, we can enjoy collision-free, low-latency wireless communication in challenging environments. Nevertheless, the fixed size of time slots in TSCH provides an opportunity for further enhancements. In this paper, we propose an enhanced centralized TSCH scheduling (ECTS) algorithm with simple packet aggregation while collecting data over a tree topology. Having in mind that the payload of a sensor node is rather short, we attempt to put more than one payload in one packet. Thus, we occupy just one cell to forward them. We investigated the schedule compactness of ECTS in Matlab, and we evaluated its operation, after implementing it in Contiki-NG, using Cooja. Our results show that ECTS with packet aggregation outperforms TASA in terms of slotframe duration and imposes fairness among the nodes in terms of latency. A validation exercise using real motes confirms its successful operation in real deployments., This work was supported in part by grant WINDMAL PGC2018-099959-B-I00 (MCIU/AEI/FEDER,UE), and co-financed by the European Union Regional Development Fund within the framework of the ERDF Operational Program of Catalonia 2014–2020 (Looming Factory).