Analysis of the expansion of a GSM network using Gaussian Processes

the expansion plan of a Global System Mobile (GSM) network requires the analysis of some important variables known as key performance indicators (KPI) on the network. Network operators have tools for analyzing a KPI behavior on a particular network cell. This paper proposes a tool that illustrates graphically the behavior-in-time of a KPI in a whole geographical zone (including cell positions). A Gaussian process repressor is used over a real data set and time-space inference is performed. Finally we observe how a particular region presents high-KPI values most of the time. This alerts the network operator for including a solution in the formulation phase of the network expansion plan. Keywords— Gaussian Processes,GSM networks, KPIs, machine learning, regression.

[1]  Dieter Fox,et al.  Gaussian Processes for Signal Strength-Based Location Estimation , 2006, Robotics: Science and Systems.

[2]  Roberto Cusani,et al.  Network monitoring and performance evaluation in a 3.5G network , 2007, Comput. Networks.

[3]  Antoine Liutkus,et al.  Multi-dimensional signal separation with Gaussian processes , 2011, 2011 IEEE Statistical Signal Processing Workshop (SSP).

[4]  Felix Duvallet,et al.  WiFi position estimation in industrial environments using Gaussian processes , 2008, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[5]  Ajay R. Mishra,et al.  Advanced Cellular Network Planning and Optimisation: 2G/2.5G/3G...Evolution to 4G , 2006 .

[6]  Anton Schwaighofer,et al.  GPPS: A Gaussian Process Positioning System for Cellular Networks , 2003, NIPS.

[7]  Roderick Murray-Smith,et al.  Hierarchical Gaussian process mixtures for regression , 2005, Stat. Comput..

[8]  Kimmo Hätönen,et al.  Analysing 3G radio network performance with fuzzy methods , 2013, Neurocomputing.

[9]  Krzysztof Chalupka,et al.  Empirical evaluation of Gaussian Process approximation algorithms , 2011 .

[10]  Md. Baitul Al Sadi,et al.  Performance Analysis of UMTS Cellular Network using Sectorization Based on Capacity and Coverage , 2011 .

[11]  Kevin P. Murphy,et al.  Machine learning - a probabilistic perspective , 2012, Adaptive computation and machine learning series.

[12]  W. Usaha,et al.  Performance evaluation of anomaly detection in cellular core networks using self-organizing map , 2008, 2008 5th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology.

[13]  Ye Ouyang,et al.  A performance analysis for UMTS packet switched network based on multivariate KPIs , 2010, 2010 Wireless Telecommunications Symposium (WTS).