towards a reference methodological framework for cdr c rs

Towards a Reference Methodological Framework for processing of Mobile Network Operator data for Official Statistics Fabio Ricciato, Albrecht Wirthmann (Eurostat), Martijn Tennekes (CBS - Statistics Netherlands), Benjamin Sakarovitch (INSEE - National Institute of Statistics and Economic Studies), Roberta Radini (ISTAT - Italian National Institute of Statistics) and David Salgado (INE - Spanish Statistical Office, Complutense University of Madrid) [email protected] Modular structure of the density estimation procedure We propose a general Reference Methodological Framework (RMF) intended to facilitate the use of Mobile Network Operator (MNO) in general, including signalling data, for Official Statistics. Why? Signalling data are more costly to extract and more complex to interpret than CDR. Furthermore, they yield a higher degree of heterogeneity across different MNOs. It is difficult for statisticians and researchers without a solid telco engineering background to interpret and manipulate such data directly. Furthermore, their format and semantic changes, following the evolution of MNO infrastructure. Development Production Algorithms designed by statisticians Algorithms designed by statisticians and telecom engineers Knowledge interface Algorithms executed at the National Statistical Institute (NSI) Algorithms executed at the Mobile Network Operator (MNO) Data export interface More informative higher spatial resolution, higher temporal frequency, better coverage More costly to extract for MNO to interpret for statisticans CDR CN signalling RAN signalling tower locations detailed radio configuration parameters BSA maps cell type & basic parameters Transactional data Auxiliary data CN: Core Network RAN: Radio Access Network BSA: Best Service Area The RMF is inspired by the principles of modularity, functional layering and the “hourglass model” that lie at the foundation of modern computer network architectures. The functional modularity of the RMF agrees with the current international statistical production standards such as the GSBPM. Thus, it paves the way for a natural combination of diverse data sources in an integrated production framework. F. Ricciato, Towards a Reference Methodological Framework for processing MNO data for Official Statistics., 2018 https ://tinyurl.com/ycgvx4m6 F. Ricciato, G. Lanzieri and A. Wirthmann Towards a methodological framework for estimating present population density from mobile network operator data, 2019 https ://tinyurl.com/yyo8c8pb References

Upload: others

Post on 03-Dec-2021

6 views

Category:

Documents

0 download

Report

Download

Embed Size (px):

TRANSCRIPT

Page 1: Towards a Reference Methodological Framework for CDR c rs

Towards a Reference Methodological Framework for processing of Mobile Network Operator data for Official Statistics

Fabio Ricciato, Albrecht Wirthmann (Eurostat), Martijn Tennekes (CBS - Statistics Netherlands), Benjamin Sakarovitch (INSEE - National Institute of Statistics and Economic Studies), Roberta Radini (ISTAT - Italian National Institute of Statistics) and David Salgado (INE - Spanish Statistical Office, Complutense University of Madrid)

[email protected]

Modular structure of the density estimation procedure

We propose a general Reference Methodological Framework (RMF) intended to facilitatethe use of Mobile Network Operator (MNO) in general, including signalling data, forOfficial Statistics.

Why? Signalling data are more costly to extract and more complex to interpret than CDR.Furthermore, they yield a higher degree of heterogeneity across different MNOs. It isdifficult for statisticians and researchers without a solid telco engineering background tointerpret and manipulate such data directly. Furthermore, their format and semanticchanges, following the evolution of MNO infrastructure.

Development Production

Algorithms designed bystatisticians

Algorithms designed bystatisticians andtelecom engineers

Knowledge interface

Algorithms executed at the National Statistical Institute (NSI)

Algorithms executed at the Mobile Network Operator (MNO)

Data export interface

More informativehigher spatial resolution, higher temporal frequency, better coverage

More costlyto extract for MNOto interpret for statisticans

CDR

CN signalling

RAN signalling

towerlocations

detailed radioconfigurationparameters

BSA maps

cell type &basic parameters

Transactional data

Auxiliary data

CN: Core NetworkRAN: Radio Access NetworkBSA: Best Service Area

The RMF is inspired by the principles of modularity, functional layering and the “hourglassmodel” that lie at the foundation of modern computer network architectures.

The functional modularity of the RMF agrees with the current international statisticalproduction standards such as the GSBPM. Thus, it paves the way for a naturalcombination of diverse data sources in an integrated production framework.

F. Ricciato, Towards a Reference Methodological Framework for processing MNO data for Official Statistics., 2018 https://tinyurl.com/ycgvx4m6F. Ricciato, G. Lanzieri and A. Wirthmann Towards a methodological framework for estimating present population density from mobile network operator data, 2019 https://tinyurl.com/yyo8c8pb

References

mailto:[email protected]

https://tinyurl.com/ycgvx4m6

https://tinyurl.com/yyo8c8pb

Comune di Roveredo in Piano - gea-pn.it · batteria di automobili batteria per cellulare batteria stilo cc se cdr cdr cdr se se cdr cdr pm or pm pm cdr cdr cdr cdr cdr pm cdr cdr

Methodological Appendix

methodological guide

Methodological issues

RS 33 - RS 37 - Sol Paragliders · RS 33 - RS 37 . Title: reserva2.cdr Author: Juliana Created Date: 7/2/2009 7:34:54 PM

Methodological Proposal

Optoelectronics & Photonics Electrical DEMUX TIA TIA TIA TIA MUX Optical Electrical CDR CDR CDR CDR CDR CDR CDR Quad Channel Drivers DAC DAC DAC DAC ADC ADC ADC ADC DSP Engine Coherent

files.vector-images.comfiles.vector-images.com/pdf/egypt.pdf · .cdr .cdr myth1C.adr egOû52.ùdr egvpt Cdr eg0030.cdr Cdr .cdr eg005E.cdr 390007 .cdr ay0032.cdr 390057 .cdr Cdr 990060