Analysis of mobile radio access network using the self-organizing map

Abstract

Mobile networks produce a huge amount of spatiMemporal data. The data consists of parameters of base stations and quality information of calls. The Self-organizing Map (SOM) is an efficient mol for visualization and clustering of multidimensional data. It transforms the input vectors on two-dimensional grid of prototype veeors and orden them. The ordered prototype yectDrs are easier to visualize and explore than the original data. Then are two possible ways to start the analysis. We can build either a model Of the network using state vectors with parameters from all mobile cells or a general one cell model trained using one cell state vectors from all cells. In both methods huther analysis is needed. In the Bnt method the distributions of parameters of one cell can be compared with the others and in the second it can be compared how well the general model represents each cell.

Sequence-based detection of sleeping cell failures in mobile networks

“…In [19,20] application of classification and clustering methods for detection and diagnosis of strangely behaving network regions is presented. Some studies also consider neural network algorithms for detection of malfunctions [57,48].…”

Cellular network coverage optimization through the application of self-organizing neural networks

“…One of the constraints of the hexagonal geometry for cellular radio applications is that each cluster is surrounded by six similar clusters with the same orientation. It is argued in [3][4] that, under this constraint, each cluster has a total area equivalent to what can be called a super-hexagon. Therefore, the number of cells in a cluster that satisfies the constraint is the ratio of the area in the super-hexagon to the area of one hexagon.…”

“…The performance ratio when determining the optimal setup of a wireless site cluster has been the subject of recent discussion, particularly in the emergence of third-generation wireless technology [1][2][3]. Current network planning and optimization tools employ software platforms, whose concept is illustarted in Fig.…”