Klasifikacija pokretnih korisničkih profila telekomunikacijske mreže metodologijom strojnog učenja

master's thesis

• Title (english): Mobile user profile classification of a telecommunication network based on the machine learning methodology
• Author: Kristina Kušić
• Mentor: Tomislav Stipančić (mentor)
• Committee member: Biserka Runje (predsjednik povjerenstva)
• Committee member: Irena Žmak (član povjerenstva)
• Committee member: Tomislav Stipančić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2020-01-23, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering; Production Mechanical Engineering

Abstract

Tema ovog diplomskog rada je „Klasifikacija pokretnih korisničkih profila telekomunikacijske mreže metodologijom strojnog učenja“. U ovom radu opisani su koncepti rada mobilne telekomunikacijske mreže. Dvjema metodama strojnog učenja analizirani su određeni parametri pokretnih korisničkih profila između dva europska grada. Korištena je dostupna baza podataka s ciljem identifikacije i grupacije pojedinih ćelija unutar zadanog područja. Cilj je pronaći ćelije visokog stupnja mobilnosti, odnosno ćelije čije je područje pokrivanja usmjereno na autoput i prugu. Korištene metode su K-means i Gaussian Mixture metoda. Provedena je usporedba rezultata dobivenih objema metodama. Također, dobiveni rezultati uspoređeni su s podacima o radu mreže iz drugog vremenskog razdoblja kako bi se dokazala ponovljivost i sezonalnost rezultata.

Abstract (english)

The topic of this master thesis is “Mobile user profiles classification of a telecommunication network based on the machine learning”. This project describes the concept of how mobile network works. This research uses two unsupervised machine learning methods in order to analyze cell mobility patterns. For this purpose, cell trace data from two European cities are used. The goal is to find “high-mobility” cells, or cells whose area of coverage is focused on the highway and track. The methods used were K-means and Gaussian Mixture method. A comparison of the results obtained with both methods is performed. Furthermore, the results were compared with the network performance data from other time interval to prove the repeatability and seasonality of the results.

• Keywords: ćelija; bazna stanica; korisnički profil; grupiranje; klaster.
• Keywords (english): cell; base station; user profile; clustering; cluster.
• Language: croatian
• URN:NBN: urn:nbn:hr:235:381193
• Study programme: Title: Mechanical Engineering; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka strojarstva (magistar/magistra inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access; Embargo expiration date: 2020-01-23
• Created on: 2020-01-21 11:24:47