English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

In spite of the new opportunities brought about by the Open RAN, advances inML-based network automation have been slow, mainly because of theunavailability of large-scale datasets and experimental testing infrastructure.This slows down the development and widespread adoption of Deep ReinforcementLearning (DRL) agents on real networks, delaying progress in intelligent andautonomous RAN control. In this paper, we address these challenges by proposingpractical solutions and software pipelines for the design, training, testing,and experimental evaluation of DRL-based closed-loop control in the Open RAN.We introduce ColO-RAN, the first publicly-available large-scale O-RAN testingframework with software-defined radios-in-the-loop. Building on the scale andcomputational capabilities of the Colosseum wireless network emulator, ColO-RANenables ML research at scale using O-RAN components, programmable basestations, and a "wireless data factory". Specifically, we design and developthree exemplary xApps for DRL-based control of RAN slicing, scheduling andonline model training, and evaluate their performance on a cellular networkwith 7 softwarized base stations and 42 users. Finally, we showcase theportability of ColO-RAN to different platforms by deploying it on Arena, anindoor programmable testbed. Extensive results from our first-of-its-kindlarge-scale evaluation highlight the benefits and challenges of DRL-basedadaptive control. They also provide insights on the development of wireless DRLpipelines, from data analysis to the design of DRL agents, and on the tradeoffsassociated to training on a live RAN. ColO-RAN and the collected large-scaledataset will be made publicly available to the research community.

ColO-RAN: Developing Machine Learning-based xApps for Open RAN Closed-loop Control on Programmable Experimental Platforms