Learning congestion state for mmWave channels

Talal Ahmad, Shiva R. Iyer, Luis Diez, Yasir Zaki, Ramón Agüero, Lakshminarayanan Subramanian

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Millimeter wave (commonly known as mmWave) is enabling the next generation of last-hop communications for mobile devices. But these technologies cannot reach their full potential because existing congestion control schemes at the transport layer perform sub-optimally over mmWave links. In this paper, we show how existing congestion control schemes perform sub-optimally in such channels. Then, we propose that we can learn early congestion signals by using end-to-end measurements at the sender and receiver. We believe that these learned measurements can help build a better congestion control scheme. We show that we can learn Explicit Congestion Notification (ECN) per packet with an F1-score as high as 97%. We achieve this by leveraging unsupervised learning on data obtained from sending periodic bursts of probe packets over emulated 60 GHz links (based on real-world WiGig measurements), with random background traffic.

Original languageEnglish (US)
Title of host publicationmmNets 2019 - Proceedings of the 3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, co-located with MobiCom 2019
PublisherAssociation for Computing Machinery
Pages19-25
Number of pages7
ISBN (Electronic)9781450369329
DOIs
StatePublished - Oct 7 2019
Event3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, mmNets 2019, co-located with MobiCom 2019 - Los Cabos, Mexico
Duration: Oct 25 2019 → …

Publication series

NameProceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM

Conference

Conference3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, mmNets 2019, co-located with MobiCom 2019
CountryMexico
CityLos Cabos
Period10/25/19 → …

Fingerprint

Unsupervised learning
Millimeter waves
Mobile devices
Communication

Keywords

  • Congestion control
  • ECN
  • Machine learning
  • MmWave

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Hardware and Architecture
  • Software

Cite this

Ahmad, T., Iyer, S. R., Diez, L., Zaki, Y., Agüero, R., & Subramanian, L. (2019). Learning congestion state for mmWave channels. In mmNets 2019 - Proceedings of the 3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, co-located with MobiCom 2019 (pp. 19-25). (Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM). Association for Computing Machinery. https://doi.org/10.1145/3349624.3356769

Learning congestion state for mmWave channels. / Ahmad, Talal; Iyer, Shiva R.; Diez, Luis; Zaki, Yasir; Agüero, Ramón; Subramanian, Lakshminarayanan.

mmNets 2019 - Proceedings of the 3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, co-located with MobiCom 2019. Association for Computing Machinery, 2019. p. 19-25 (Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ahmad, T, Iyer, SR, Diez, L, Zaki, Y, Agüero, R & Subramanian, L 2019, Learning congestion state for mmWave channels. in mmNets 2019 - Proceedings of the 3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, co-located with MobiCom 2019. Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM, Association for Computing Machinery, pp. 19-25, 3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, mmNets 2019, co-located with MobiCom 2019, Los Cabos, Mexico, 10/25/19. https://doi.org/10.1145/3349624.3356769
Ahmad T, Iyer SR, Diez L, Zaki Y, Agüero R, Subramanian L. Learning congestion state for mmWave channels. In mmNets 2019 - Proceedings of the 3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, co-located with MobiCom 2019. Association for Computing Machinery. 2019. p. 19-25. (Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM). https://doi.org/10.1145/3349624.3356769
Ahmad, Talal ; Iyer, Shiva R. ; Diez, Luis ; Zaki, Yasir ; Agüero, Ramón ; Subramanian, Lakshminarayanan. / Learning congestion state for mmWave channels. mmNets 2019 - Proceedings of the 3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, co-located with MobiCom 2019. Association for Computing Machinery, 2019. pp. 19-25 (Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM).
@inproceedings{8f03b8f787a447119fc7becce77029b5,
title = "Learning congestion state for mmWave channels",
abstract = "Millimeter wave (commonly known as mmWave) is enabling the next generation of last-hop communications for mobile devices. But these technologies cannot reach their full potential because existing congestion control schemes at the transport layer perform sub-optimally over mmWave links. In this paper, we show how existing congestion control schemes perform sub-optimally in such channels. Then, we propose that we can learn early congestion signals by using end-to-end measurements at the sender and receiver. We believe that these learned measurements can help build a better congestion control scheme. We show that we can learn Explicit Congestion Notification (ECN) per packet with an F1-score as high as 97{\%}. We achieve this by leveraging unsupervised learning on data obtained from sending periodic bursts of probe packets over emulated 60 GHz links (based on real-world WiGig measurements), with random background traffic.",
keywords = "Congestion control, ECN, Machine learning, MmWave",
author = "Talal Ahmad and Iyer, {Shiva R.} and Luis Diez and Yasir Zaki and Ram{\'o}n Ag{\"u}ero and Lakshminarayanan Subramanian",
year = "2019",
month = "10",
day = "7",
doi = "10.1145/3349624.3356769",
language = "English (US)",
series = "Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM",
publisher = "Association for Computing Machinery",
pages = "19--25",
booktitle = "mmNets 2019 - Proceedings of the 3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, co-located with MobiCom 2019",

}

TY - GEN

T1 - Learning congestion state for mmWave channels

AU - Ahmad, Talal

AU - Iyer, Shiva R.

AU - Diez, Luis

AU - Zaki, Yasir

AU - Agüero, Ramón

AU - Subramanian, Lakshminarayanan

PY - 2019/10/7

Y1 - 2019/10/7

N2 - Millimeter wave (commonly known as mmWave) is enabling the next generation of last-hop communications for mobile devices. But these technologies cannot reach their full potential because existing congestion control schemes at the transport layer perform sub-optimally over mmWave links. In this paper, we show how existing congestion control schemes perform sub-optimally in such channels. Then, we propose that we can learn early congestion signals by using end-to-end measurements at the sender and receiver. We believe that these learned measurements can help build a better congestion control scheme. We show that we can learn Explicit Congestion Notification (ECN) per packet with an F1-score as high as 97%. We achieve this by leveraging unsupervised learning on data obtained from sending periodic bursts of probe packets over emulated 60 GHz links (based on real-world WiGig measurements), with random background traffic.

AB - Millimeter wave (commonly known as mmWave) is enabling the next generation of last-hop communications for mobile devices. But these technologies cannot reach their full potential because existing congestion control schemes at the transport layer perform sub-optimally over mmWave links. In this paper, we show how existing congestion control schemes perform sub-optimally in such channels. Then, we propose that we can learn early congestion signals by using end-to-end measurements at the sender and receiver. We believe that these learned measurements can help build a better congestion control scheme. We show that we can learn Explicit Congestion Notification (ECN) per packet with an F1-score as high as 97%. We achieve this by leveraging unsupervised learning on data obtained from sending periodic bursts of probe packets over emulated 60 GHz links (based on real-world WiGig measurements), with random background traffic.

KW - Congestion control

KW - ECN

KW - Machine learning

KW - MmWave

UR - http://www.scopus.com/inward/record.url?scp=85076433859&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85076433859&partnerID=8YFLogxK

U2 - 10.1145/3349624.3356769

DO - 10.1145/3349624.3356769

M3 - Conference contribution

AN - SCOPUS:85076433859

T3 - Proceedings of the Annual International Conference on Mobile Computing and Networking, MOBICOM

SP - 19

EP - 25

BT - mmNets 2019 - Proceedings of the 3rd ACM Workshop on Millimeter-Wave Networks and Sensing Systems, co-located with MobiCom 2019

PB - Association for Computing Machinery

ER -