Over the past years, one of the main challenges for leading mobile operators was to keep high QoS and performance of both Radio Access Network (RAN) and backbone in order to satisfy customers demand and prevent customer churn problem. Nowadays, we live in a world full of data. According to the research performed by leading TELCOS, the amount of data used by people was 7 Exabytes per month in 2016 and it is forecasted to reach 49 Exabytes per month by 2021 [1]. Possession the ability to extract useful information from the raw data and benefit from it for the future purposes became crucial aspect. Since data analytics becomes more available, using these techniques to understand the huge number of LTE network measurements and diagnosis data, evaluation and prediction of LTE network capacity and performance becomes a very promising approach. The main objective of this work was evaluation of inter-cell interference and channel condition prediction via application of Data Analysis techniques. Data provided by international mobile operator, has been studied, evaluated and processed in order to analyze the existence of information about inter-cell interference for the 800MHz, 1800MHz, 2600MHz frequency bands and predict the channel condition in urban (Milan) environments. Two separate studies were performed: for inter-cell interference the Pearson correlation coefficients among two independent (CQI & Downlink Traffic) features selected from the data were calculated, while for channel condition forecasting two linear models were applied and the best one was chosen.
Negli anni precedenti, una delle sfide principali per i operatori di telefonia mobile primari era quello di tenere alta QoS e performance sia di Radio Access Network che di backbone per soddisfare la domanda dei client e prevenire fidelizzazione dei clienti. Attualmente viviamo in un mondo pieno di dati. Secondo le richerche eseguita da TELCOS principali, la quantità di dati utilizzati dalle persone era 7 Exabyte nel 2016. Questo numero si prevede raggiunga 49 Exabyte al mese entro il 2021 [1]. Possesso abilità estrarre informazioni utili dai dati grezzi e beneficiarne per scopi futuri ha diventato compito cruciale. Poiché l'analisi dei dati diventa più disponibile, utilizzando queste tecniche per comprendere l'enorme numero di misurazioni della rete LTE e dati di diagnosi, valutazione e predizione della capacità e performance della rete LTE diventa un approccio molto promettente. L'obiettivo principale di questo lavoro era utilizando tecniche di analisi dei dati per la valutazione dell'interferenza delle cellule e la previsione delle condizioni del canale. I dati forniti dall'operatore mobile internazionale sono stati studiati, valutati ed elaborati al fine di analizzare l'esistenza di informazioni sull'interferenza cellular per le bande di frequenza 800 MHz, 1800 MHz, 2600 MHz e per prevedere la condizione del canale negli ambienti urbani (Milano). Sono stati effettuati due studi separati: per interferenza tra le cellule sono stati calcolati i coefficienti di correlazione di Pearson tra due caratteristiche indipendenti (CQI e Downlink Traffic) selezionate dai dati, mentre per la previsione delle condizioni del canale sono stati applicati due modelli lineari e il migliore è stato scelto.
Channel quality analysis and prediction in LTE cellular networks
MUSTAFAYEV, MADAT
2017/2018
Abstract
Over the past years, one of the main challenges for leading mobile operators was to keep high QoS and performance of both Radio Access Network (RAN) and backbone in order to satisfy customers demand and prevent customer churn problem. Nowadays, we live in a world full of data. According to the research performed by leading TELCOS, the amount of data used by people was 7 Exabytes per month in 2016 and it is forecasted to reach 49 Exabytes per month by 2021 [1]. Possession the ability to extract useful information from the raw data and benefit from it for the future purposes became crucial aspect. Since data analytics becomes more available, using these techniques to understand the huge number of LTE network measurements and diagnosis data, evaluation and prediction of LTE network capacity and performance becomes a very promising approach. The main objective of this work was evaluation of inter-cell interference and channel condition prediction via application of Data Analysis techniques. Data provided by international mobile operator, has been studied, evaluated and processed in order to analyze the existence of information about inter-cell interference for the 800MHz, 1800MHz, 2600MHz frequency bands and predict the channel condition in urban (Milan) environments. Two separate studies were performed: for inter-cell interference the Pearson correlation coefficients among two independent (CQI & Downlink Traffic) features selected from the data were calculated, while for channel condition forecasting two linear models were applied and the best one was chosen.File | Dimensione | Formato | |
---|---|---|---|
Channel Quality Analysis and Prediction in LTE Cellular Networks.pdf
accessibile in internet per tutti
Descrizione: Thesis text
Dimensione
1.33 MB
Formato
Adobe PDF
|
1.33 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/140022