Correlation-Based Feature Mapping of Crowdsourced LTE Data

Kasper Apajalahti, Eremias Waleigne, Jukka Manner, Eero Hyvönen

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Abstract

There have been efforts taken by different research projects to understand the complexity and the performance of a mobile broadband network. Various mobile network measurement platforms are proposed to collect performance metrics for analysis. Data integration would provide more thorough data analyses as well as prediction and decision models from one dataset to another. The crucial part of the data integration is to find out, whether two datasets have corresponding features (performance metrics). However, finding common features across datasets is a challenging task. For example, features might: 1) have similar names but be different metrics, 2) have different names but be similar metrics, or 3) be same metrics but have differences in the underlying methodology. We designed a feature mapping methodology between two crowdsourced LTE measurement-based datasets. Our method is based on correlations between the features and the mapping algorithm is solving a maximum constraint satisfaction problem (CSP). We define our constraints as inequality patterns between the correlation coefficients of the measured features. Our results show that the method maps measurement features based on their correlation coefficients with high confidence scores (between 0.78 to 1.0 depending on the amount of features). We observe that mapping score increases as a function of the amount of features. Altogether, our results show that this methodology can be used as an automated tool in the measurement data integration.
Original languageEnglish
Title of host publication2018 IEEE 29th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Bologna, Italy
Number of pages7
PublisherIEEE
Publication date20 Dec 2018
Pages1-7
Article number8580999
ISBN (Print)978-1-5386-6009-6
DOIs
Publication statusPublished - 20 Dec 2018
Externally publishedYes
MoE publication typeA4 Article in conference proceedings
EventIEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC - Bologna, Italy
Duration: 9 Sep 201812 Sep 2018

Fields of Science

  • 213 Electronic, automation and communications engineering, electronics

Cite this

Apajalahti, K., Waleigne, E., Manner, J., & Hyvönen, E. (2018). Correlation-Based Feature Mapping of Crowdsourced LTE Data. In 2018 IEEE 29th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Bologna, Italy (pp. 1-7). [8580999] IEEE. https://doi.org/10.1109/PIMRC.2018.8580999
Apajalahti, Kasper ; Waleigne, Eremias ; Manner, Jukka ; Hyvönen, Eero. / Correlation-Based Feature Mapping of Crowdsourced LTE Data. 2018 IEEE 29th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Bologna, Italy. IEEE, 2018. pp. 1-7
@inproceedings{934bed9a88294eb297d71baef6aa7262,
title = "Correlation-Based Feature Mapping of Crowdsourced LTE Data",
abstract = "There have been efforts taken by different research projects to understand the complexity and the performance of a mobile broadband network. Various mobile network measurement platforms are proposed to collect performance metrics for analysis. Data integration would provide more thorough data analyses as well as prediction and decision models from one dataset to another. The crucial part of the data integration is to find out, whether two datasets have corresponding features (performance metrics). However, finding common features across datasets is a challenging task. For example, features might: 1) have similar names but be different metrics, 2) have different names but be similar metrics, or 3) be same metrics but have differences in the underlying methodology. We designed a feature mapping methodology between two crowdsourced LTE measurement-based datasets. Our method is based on correlations between the features and the mapping algorithm is solving a maximum constraint satisfaction problem (CSP). We define our constraints as inequality patterns between the correlation coefficients of the measured features. Our results show that the method maps measurement features based on their correlation coefficients with high confidence scores (between 0.78 to 1.0 depending on the amount of features). We observe that mapping score increases as a function of the amount of features. Altogether, our results show that this methodology can be used as an automated tool in the measurement data integration.",
keywords = "213 Electronic, automation and communications engineering, electronics",
author = "Kasper Apajalahti and Eremias Waleigne and Jukka Manner and Eero Hyv{\"o}nen",
year = "2018",
month = "12",
day = "20",
doi = "10.1109/PIMRC.2018.8580999",
language = "English",
isbn = "978-1-5386-6009-6",
pages = "1--7",
booktitle = "2018 IEEE 29th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Bologna, Italy",
publisher = "IEEE",
address = "International",

}

Apajalahti, K, Waleigne, E, Manner, J & Hyvönen, E 2018, Correlation-Based Feature Mapping of Crowdsourced LTE Data. in 2018 IEEE 29th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Bologna, Italy., 8580999, IEEE, pp. 1-7, IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC, Bologna, Italy, 09/09/2018. https://doi.org/10.1109/PIMRC.2018.8580999

Correlation-Based Feature Mapping of Crowdsourced LTE Data. / Apajalahti, Kasper; Waleigne, Eremias; Manner, Jukka; Hyvönen, Eero.

2018 IEEE 29th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Bologna, Italy. IEEE, 2018. p. 1-7 8580999.

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

TY - GEN

T1 - Correlation-Based Feature Mapping of Crowdsourced LTE Data

AU - Apajalahti, Kasper

AU - Waleigne, Eremias

AU - Manner, Jukka

AU - Hyvönen, Eero

PY - 2018/12/20

Y1 - 2018/12/20

N2 - There have been efforts taken by different research projects to understand the complexity and the performance of a mobile broadband network. Various mobile network measurement platforms are proposed to collect performance metrics for analysis. Data integration would provide more thorough data analyses as well as prediction and decision models from one dataset to another. The crucial part of the data integration is to find out, whether two datasets have corresponding features (performance metrics). However, finding common features across datasets is a challenging task. For example, features might: 1) have similar names but be different metrics, 2) have different names but be similar metrics, or 3) be same metrics but have differences in the underlying methodology. We designed a feature mapping methodology between two crowdsourced LTE measurement-based datasets. Our method is based on correlations between the features and the mapping algorithm is solving a maximum constraint satisfaction problem (CSP). We define our constraints as inequality patterns between the correlation coefficients of the measured features. Our results show that the method maps measurement features based on their correlation coefficients with high confidence scores (between 0.78 to 1.0 depending on the amount of features). We observe that mapping score increases as a function of the amount of features. Altogether, our results show that this methodology can be used as an automated tool in the measurement data integration.

AB - There have been efforts taken by different research projects to understand the complexity and the performance of a mobile broadband network. Various mobile network measurement platforms are proposed to collect performance metrics for analysis. Data integration would provide more thorough data analyses as well as prediction and decision models from one dataset to another. The crucial part of the data integration is to find out, whether two datasets have corresponding features (performance metrics). However, finding common features across datasets is a challenging task. For example, features might: 1) have similar names but be different metrics, 2) have different names but be similar metrics, or 3) be same metrics but have differences in the underlying methodology. We designed a feature mapping methodology between two crowdsourced LTE measurement-based datasets. Our method is based on correlations between the features and the mapping algorithm is solving a maximum constraint satisfaction problem (CSP). We define our constraints as inequality patterns between the correlation coefficients of the measured features. Our results show that the method maps measurement features based on their correlation coefficients with high confidence scores (between 0.78 to 1.0 depending on the amount of features). We observe that mapping score increases as a function of the amount of features. Altogether, our results show that this methodology can be used as an automated tool in the measurement data integration.

KW - 213 Electronic, automation and communications engineering, electronics

U2 - 10.1109/PIMRC.2018.8580999

DO - 10.1109/PIMRC.2018.8580999

M3 - Conference contribution

SN - 978-1-5386-6009-6

SP - 1

EP - 7

BT - 2018 IEEE 29th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Bologna, Italy

PB - IEEE

ER -

Apajalahti K, Waleigne E, Manner J, Hyvönen E. Correlation-Based Feature Mapping of Crowdsourced LTE Data. In 2018 IEEE 29th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Bologna, Italy. IEEE. 2018. p. 1-7. 8580999 https://doi.org/10.1109/PIMRC.2018.8580999