Spider: Deep Learning-driven Sparse Mobile Traffic Measurement Collection and Reconstruction

Yini Fang, Alec Diallo, Chaoyun Zhang, Paul Patras

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

Abstract

Data-driven mobile network management hinges on accurate traffic measurements, which routinely require expensive specialized equipment and substantial local storage capabilities, and bear high data transfer overheads. To overcome these challenges, in this paper we propose Spider, a deep-learningdriven mobile traffic measurement collection and reconstruction framework, which reduces the cost of data collection while retaining state-of-the-art accuracy in inferring mobile traffic consumption with fine geographic granularity. Spider harnesses Reinforcement Learning and tackles large action spaces to train a policy network that selectively samples a minimal number of cells where data should be collected. We further introduce a fast and accurate neural model that extracts spatiotemporal correlations from historical data to reconstruct network-wide traffic consumption based on sparse measurements. Experiments we conduct with a real-world mobile traffic dataset demonstrate that Spider samples 48% fewer cells as compared to several benchmarks considered, and yields up to 67% lower reconstruction errors than state-of-the-art interpolation methods. Moreover, our framework can adapt to previously unseen traffic patterns.
Original languageEnglish
Title of host publicationIEEE Global Communications Conference (GLOBECOM)
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Publication statusAccepted/In press - 15 Aug 2021
Event2021 IEEE Global Communications Conference - Madrid, Spain
Duration: 7 Dec 202111 Dec 2021
https://globecom2021.ieee-globecom.org/

Conference

Conference2021 IEEE Global Communications Conference
Abbreviated titleGLOBECOM 2021
Country/TerritorySpain
CityMadrid
Period7/12/2111/12/21
Internet address

Fingerprint

Dive into the research topics of 'Spider: Deep Learning-driven Sparse Mobile Traffic Measurement Collection and Reconstruction'. Together they form a unique fingerprint.

Cite this