Kimmo Raivio

Advanced analysis methods for 3G cellular networks

The operation and maintenance of the third generation (3G) mobile networks will be challenging. These networks will be strongly service driven, and this approach differs significantly from the traditional speech dominated in the second generation (2G) approach. Compared to 2G, in 3G, the mobile cells interact and interfere with each other more, they have hundreds of adjustable parameters, and they monitor and record data related to several hundreds of different variables in each cell. This paper shows that a neural network algorithm called the self-organizing map, together with a conventional clustering method like the k-means, can effectively be used to simplify and focus network analysis. It is shown that these algorithms help in visualizing and grouping similarly behaving cells. Thus, it is easier for a human expert to discern different states of the network. This makes it possible to perform faster and more efficient troubleshooting and optimization of the parameters of the cells. The presented methods are applicable for different radio access network technologies.

Combining linear equalization and self-organizing adaptation in dynamic discrete-signal detection

An adaptive algorithm combining traditional linear equalization techniques and a self-organizing neural learning algorithm is presented. The results show that the performance of the neural equalizer is insensitive to nonlinear learning distortions in dynamic discrete-signal detection. Stabilization of the self-organizing map during undistorted transmission has to be further considered to decrease the absolute mean-square error (MSE) rate of the neural equalizer. The error is due to oscillations in the self-organizing map, mainly caused by the neighborhood learning. The oscillations can be decreased by taking more samples to the map before adapting the mi values and by decreasing the neighborhood learning parameter β

Neural analysis of mobile radio access network

The self-organizing map (SOM) is an efficient tool for visualization and clustering of multidimensional data. It transforms the input vectors on two-dimensional grid of prototype vectors and orders them. The ordered prototype vectors are easier to visualize and explore than the original data. Mobile networks produce a huge amount of spatiotemporal data. The data consists of parameters of base stations (BS) and quality information of calls. There are two alternatives in starting the data analysis. We can build either a general one-cell-model trained using state vectors from all cells, or a model of the network using state vectors with parameters from all mobile cells. In both methods, further analysis is needed to understand the reasons for various operational states of the entire network.

A SOM based approach for visualization of GSM network performance data

In this paper, a neural network based approach to visualize performance data of a GSM network is presented. The proposed approach consists of several steps. First, a suitable proportion of measurement data is selected. Then, the selected set of multi-dimensional data is projected into two-dimensional space for visualization purposes with a neural network algorithm called Self-Organizing Map (SOM). Then, the data is clustered and additional visualizations for each data cluster are provided in order to infer the presence of various failure types, their sources and times of occurrence. We apply the proposed approach in the analysis of degradations in signaling and traffic channel capacity of a GSM network.

A knowledge-based model for analyzing GSM network performance

In this paper, a method to analyze GSM network performance on the basis of massive data records and application domain knowledge is presented. The available measurements are divided into variable sets describing the performance of the different subsystems of the GSM network. Simple mathematical models for the subsystems are proposed. The model parameters are estimated from the available data record using quadratic programming. The parameter estimates are used to find the input-output variable pairs involved in the most severe performance degradations. Finally, the resulting variable pairs are visualized as a tree-shaped cause-effect chain in order to allow user friendly analysis of the network performance.

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

Mobile networks produce a huge amount of spatio-temporal data. The data consists of parameters of base stations and quality information of calls. The self-organizing map (SOM) is an efficient tool for visualization and clustering of multidimensional data. It transforms the input vectors on a two-dimensional grid of prototype vectors and orders them. The ordered prototype vectors are easier to visualize and explore than the original data. There 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 vector from all cells. In both methods, further analysis is needed. In the first 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.

Neural detection of QAM signal with strongly nonlinear receiver

Abstract Neural receiver structures have been developed for adaptive discrete-signal detection in telecommunication applications. Neural networks combined with conventional equalizers improve the performance especially in compensating for nonlinear distortions. These distortions may result, for instance, from nonlinear amplification implemented for reducing the power consumption. In this paper, the behavior of the neural receiver in multipath channel with additive white Gaussian noise has been investigated. The transmitted signal is quadrature amplitude modulated (QAM). A receiver structure based on self-organizing map (SOM) is compared with a conventional decision feedback equalizer (DFE).

Self-organizing operator maps in complex system analysis

The growth in amount of data available today has encouraged the development of effective data analysis methods to support human decision-making. Neuro-fuzzy computation is a soft computing hybridisation combining the learning capabilities of the neural networks with the linguistic representation of data provided by the fuzzy models. In this paper, a framework to build temporally local neuro-fuzzy systems for the analysis of nonstationary process data using self-organizing operator maps is described.

Analysis of soft handover measurements in 3G network

A neural network based clustering method for the analysis of soft handovers in 3G network is introduced. The method is highly visual and it could be utilized in explorative analysis of mobile networks. In this paper, the method is used to find groups of similar mobile cell pairs in the sense of handover measurements. The groups or clusters found by the method are characterized by the rate of successful handovers as well as the causes of failing handover attempts. The most interesting clusters are those which represent certain type of problems in handover attempts. By comparing variable histograms of a selected cluster to histograms of the whole data set an application domain expert may find some explanations on problems. Two clusters are investigated further and causes of failing handover attempts are discussed.

PERFORMANCE EVALUATION OF SELF-ORGANIZING MAP BASED NEURAL EQUALIZERS IN DYNAMIC DISCRETE-SIGNAL DETECTION

Novel equalizer structures utilizing neural computation have recently been developed for adaptive discrete-signal detection. The equalizer structures combine the traditional transversal equalizer and the Self-Organizing Map algorithm in parallel or cascade. Extensive simulations have been run to investigate different parameter effects using a two-path channel model and 16-QAM modulation. The results have shown that the neural equalizer adapts very well to changing channel conditions, including both linear multipath and nonlinear distortions. Especially in difficult channels, the new structures are superior when compared with the traditional equalizers. The computational complexities of the combined structures are not significantly higher when compared to the practical linear equalizers.