Publication:
vrAIn: a deep learning approach tailoring computing and radio resources in virtualized RANs

Simple item page
dc.affiliation.dptoUC3M. Departamento de Ingeniería Telemáticaes
dc.affiliation.grupoinvUC3M. Grupo de Investigación: Network Technologieses
dc.contributor.authorAyala Romero, José A.
dc.contributor.authorGarcia Saavedra, Andres
dc.contributor.authorGramaglia, Marco
dc.contributor.authorCosta-Pérez, Xavier
dc.contributor.authorBanchs Roca, Albert
dc.contributor.authorAlcaraz, Juan J.
dc.contributor.funderEuropean Commissionen
dc.date.accessioned2019-10-30T11:11:00Z
dc.date.available2019-10-30T11:11:00Z
dc.date.issued2019-09-05
dc.descriptionProceeding of: 25th Annual International Conference on Mobile Computing and Networking (MobiCom'19), October 21-25, 2019, Los Cabos, Mexico.en
dc.description.abstractThe virtualization of radio access networks (vRAN) is the last milestone in the NFV revolution. However, the complex dependencies between computing and radio resources make vRAN resource control particularly daunting. We present vrAIn, a dynamic resource controller for vRANs based on deep reinforcement learning. First, we use an autoencoder to project high-dimensional context data (traffic and signal quality patterns) into a latent representation. Then, we use a deep deterministic policy gradient (DDPG) algorithm based on an actor-critic neural network structure and a classifier to map (encoded) contexts into resource control decisions. We have implemented vrAIn using an open-source LTE stack over different platforms. Our results show that vrAIn successfully derives appropriate compute and radio control actions irrespective of the platform and context: (i) it provides savings in computational capacity of up to 30% over CPU-unaware methods; (ii) it improves the probability of meeting QoS targets by 25% over static allocation policies using similar CPU resources in average; (iii) upon CPU capacity shortage, it improves throughput performance by 25% over state-of-the-art schemes; and (iv) it performs close to optimal policies resulting from an offline oracle. To the best of our knowledge, this is the first work that thoroughly studies the computational behavior of vRANs, and the first approach to a model-free solution that does not need to assume any particular vRAN platform or system conditions.en
dc.description.sponsorshipThe work of University Carlos III of Madrid was supported by H2020 5GMoNArch project (grant agreement no. 761445) and H2020 5G-TOURS project (grant agreement no. 856950). The work of NEC Laboratories Europe was supported by H2020 5GTRANSFORMER project (grant agreement no. 761536) and 5GROWTH project (grant agreement no. 856709). The work of University of Cartagena was supported by Grant AEI/FEDER TEC2016-76465-C2-1-R (AIM) and Grant FPU14/03701.en
dc.description.statusPublicadoes
dc.format.extent16
dc.identifier.bibliographicCitationProceedings of the 25th Annual International Conference on Mobile Computing and Networking (MobiCom'19). New York: ACM, cop. 2019. Article nº 30, [16] pp.en
dc.identifier.doihttps://doi.org/10.1145/3300061.3345431
dc.identifier.isbn978-1-4503-6169-9
dc.identifier.publicationtitleProceedings of the 25th Annual International Conference on Mobile Computing and Networking (MobiCom'19)en
dc.identifier.urihttps://hdl.handle.net/10016/29093
dc.identifier.uxxiCC/0000029938
dc.language.isoengen
dc.publisherAssociation for Computing Machineryen
dc.relation.eventdateOctober 21-25, 2019en
dc.relation.eventnumber25
dc.relation.eventplaceLos Cabos, México.es
dc.relation.eventtitleAnnual International Conference on Mobile Computing and Networking (MobiCom'19)en
dc.relation.projectIDinfo:eu-repo/grantAgreement/EC/H2020/761445/5G-MoNArch
dc.relation.projectIDinfo:eu-repo/grantAgreement/EC/H2020/856950/5G-TOURS
dc.relation.projectIDinfo:eu-repo/grantAgreement/EC/H2020/761536/5G Transformer
dc.relation.projectIDinfo:eu-repo/grant/Agreeement/EC/H2020/856709/5GROWTH
dc.rights© 2019 Association for Computing Machinery.en
dc.rights.accessRightsopen accessen
dc.subject.ecienciaTelecomunicacioneses
dc.subject.otherRAN virtualizationen
dc.subject.otherResource managementen
dc.subject.otherMachine learningen
dc.subject.otherNetwork algorithmsen
dc.subject.otherMobile networksen
dc.titlevrAIn: a deep learning approach tailoring computing and radio resources in virtualized RANsen
dc.typeconference proceedings*
dc.type.hasVersionAM*
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
vrAln_MobiCom_2019_ps.pdf
Size:
2.48 MB
Format:
Adobe Portable Document Format
Download
Collections
DIT - RYSC - Comunicaciones en Conferencias Internacionales / International Conferences
OpenAIRE: Open Access Infrastructure for Research in Europe