Weplananalytics

WEPLANANALYTICSManage supplier listing

Crowdsourced network-quality measurement corpus built specifically for regulatory use, and unusual in this pool because it is geostatistically interpolated rather than merely sampled. Daily crowdsourced performance measurements across all mobile operators in a market, with the stated regulatory applications: radio spectrum management (monitoring actual frequency usage and detecting under-utilisation), competition and transparency (comparable published service-quality information), verification of real coverage against licence and auction coverage commitments, and aligning network performance with citizen and sector connectivity needs. Two assets are genuinely scarce. First, the CDTI and ERDF-funded research programme (project IDI-20230364, prediction system of the coverage area of a mobile communications network based on accumulated volumes of massive data generated by smartphones, EUR 229,521, September 2022 to September 2023) produced an advanced GEOSTATISTICAL ALGORITHM that infers network quality where no measurement exists and delivered a scalable system for predicting total network coverage from crowdsourced data - meaning the licensable output is a gap-filled coverage surface, not a scatter of sparse sample points, which is the single biggest weakness of ordinary crowdsourced data. Second, that project's published results include an in-depth analysis of MOBILE OPERATOR NETWORK PARAMETERIZATION - a reconstruction from measurement of how each operator has configured its network, intelligence no operator publishes and no regulator holds. The separate UNICO 6G R&D Programme 2023 project (file TSI-064200-2023-5, eligible EUR 779,634, grant EUR 467,780) built a performance-evaluation and ANOMALY-DETECTION system for 5G using machine learning on crowdsourced data, so the corpus carries ML-derived anomaly and degradation labels alongside the raw measurements. A published methodology section for mobile quality measurement exists, which matters commercially: an authority-grade dataset is only licensable if the method is documented and defensible, and here it is.