Fedor Chernogorov

Detection of Sleeping Cells in LTE Networks Using Diffusion Maps

In mobile networks emergence of failures is caused by various breakdowns of hardware and software elements. One of the serious failures in radio networks is a Sleeping Cell. In our work one of the possible root causes for appearance of this network failure is simulated in a dynamic network simulator. The main aim of the research is to detect the presence of a Sleeping Cell in the network and to define its location. For this purpose Diffusion Maps data mining technique is employed. The developed fault identification framework is using the performance characteristics of the network, collected during its regular operation, and for that reason it can be implemented in real Long Term Evolution (LTE) networks within the Self-Organizing Networks (SON) concept.

An Approach for Network Outage Detection from Drive-Testing Databases

A data-mining framework for analyzing a cellular network drive testing database is described in this paper. The presented method is designed to detect sleeping base stations, network outage, and change of the dominance areas in a cognitive and self-organizing manner. The essence of the method is to find similarities between periodical network measurements and previously known outage data. For this purpose, diffusion maps dimensionality reduction and nearest neighbor data classification methods are utilized. The method is cognitive because it requires training data for the outage detection. In addition, the method is autonomous because it uses minimization of drive testing (MDT) functionality to gather the training and testing data. Motivation of classifying MDT measurement reports to periodical, handover, and outage categories is to detect areas where periodical reports start to become similar to the outage samples. Moreover, these areas are associated with estimated dominance areas to detected sleeping base stations. In the studied verification case, measurement classification results in an increase of the amount of samples which can be used for detection of performance degradations, and consequently, makes the outage detection faster and more reliable.

N-gram analysis for sleeping cell detection in LTE networks

Sleeping cell detection in a wireless network means to find the cells which are not working properly due to various reasons. The research in the area has mostly focused on cell outage detection, e.g. due to hardware failures at the base station antennas or non-optimal network planning. In this paper we extend the research into a more challenging setting which is overlooked in the literature: the case where no outages occur in the network. The essence of the proposed method for detection of problematic cells is to analyze the sequences of the events reported by the mobile terminals to the serving base stations. The suggested n-gram analysis includes dimensionality reduction and classification of the data and ends up with providing a set of abnormal users, which at the end reveal the location of the problematic cell. We verify the proposed framework with simulated LTE network data and using the minimization of drive testing (MDT) functionality to gather the training and testing data sets.

User satisfaction classification for Minimization of Drive Tests QoS verification

In wireless mobile networks quality of user experience changes dynamically and depends on large variety of factors. Because of that mobile operators are willing to timely and effectively evaluate provided Quality of Service (QoS) in their networks. Nowadays the main tool for monitoring network state and performance is drive testing. To replace this expensive and mostly manual procedure, concept of Minimization of Drive Tests (MDT) is being developed in 3GPP LTE standardization. One area where MDT is applied is QoS verification, where user satisfaction is evaluated on the basis of periodic reports which contain values of different Key Performance Indicators (KPIs). By analyzing these KPIs one could find users which are unsatisfied with the experienced QoS and optimize configuration parameters in the areas of the network where concentration of unsatisfied users is high. In this paper we introduce a data mining framework which allows to distinguish between satisfied and unsatisfied users in LTE mobile network on the basis of limited number of KPIs. In addition, we take into use a KPI ranking system which gives an ability to significantly reduce the number of analyzed variables without compromising the resulting accuracy.

QoS Verification for Minimization of Drive Tests in LTE Networks

Nowadays, operational quality and robustness of cellular networks are among the hottest topics wireless communications research. As a response to a growing need in reduction of expenses for mobile operators, 3rd Generation Partnership Project (3GPP) initiated work on Minimization of Drive Tests (MDT). There are several major areas of standardization related to MDT, such as coverage, capacity, mobility optimization and verification of end user quality [1]. This paper presents results of the research devoted to Quality of Service (QoS) verification for MDT. The main idea is to jointly observe the user experienced QoS in terms of throughput, and corresponding radio conditions. Also the necessity to supplement the existing MDT metrics with the new reporting types is elaborated.

Data mining framework for random access failure detection in LTE networks

Sleeping cell problem is a particular type of cell degradation. There are various software and hardware reasons that might cause such kind of cell outage. In this study a cell becomes sleeping because of Random Access Channel (RACH) failure. This kind of network problem can appear due to misconfiguration, excessive load or software/firmware problem at the Base Station (BS). In practice such failure might cause network performance degradation, which is hardly traceable by an operator. In this paper we present a data mining based framework for the detection of problematic cells. In its core is the analysis of event sequences reported by a User Equipment (UE) to a serving BS. The choice of N in N-gram feature selection algorithm is considered, because of its significant impact on computational efficiency. Moreover, qualitative and heuristic performance metrics have been developed to assess the performance of the proposed detection algorithm. Sleeping cell detection framework is verified by means of dynamic LTE (Long-Term Evolution) system simulator, using Minimization of Drive Testing (MDT) functionality. It is shown that sleeping cell can be determined with very high reliability even using 1-gram algorithm.

Sequence-based detection of sleeping cell failures in mobile networks

This article presents an automatic malfunction detection framework based on data mining approach to analysis of network event sequences. The considered environment is long term evolution (LTE) of Universal Mobile Telecommunications System with sleeping cell caused by random access channel failure. Sleeping cell problem means unavailability of network service without triggered alarm. The proposed detection framework uses N-gram analysis for identification of abnormal behavior in sequences of network events. These events are collected with minimization of drive tests functionality standardized in LTE. Further processing applies dimensionality reduction, anomaly detection with K-Nearest Neighbors, cross-validation, postprocessing techniques and efficiency evaluation. Different anomaly detection approaches proposed in this paper are compared against each other with both classic data mining metrics, such as F-score and receiver operating characteristic curves, and a newly proposed heuristic approach. Achieved results demonstrate that the suggested method can be used in modern performance monitoring systems for reliable, timely and automatic detection of random access channel sleeping cells.

Fairness vs. Performance in Rel-13 LTE Licensed Assisted Access.pdf

Wireless spectrum being a valuable resource, a large number of both non-profit and for-profit users rely on its unlicensed parts, usually to provide some sort of local area wireless coverage. When the cellular vendors and service providers expressed interest in the unlicensed spectrum, the incumbent community started to fear that the new technology would not respect the existing users. And although members of the incumbent community strongly influenced the design of licensed assisted access (LAA) in 3GPP, the fear has not dispersed yet. Representing a vendor with stakes in both communities, we attempt to disperse the fear by providing insight on the question of fairness and performance of LAA, openly discussing why it should generate less interference than WiFi and why it could be otherwise. We augment the discussion with insightful simulations, looking how detailed parameters of the channel access mechanism influence LAAs fairness and performance. Our investigations show that in common scenarios LAA is a fair neighbor to WiFi as is, and even in an extreme scenario only small adjustments to the channel access mechanism are needed to achieve the same outcome.

The Effect of Discontinuous Reception and RRC Release Timer Parameterization on Mobility

In 3rd Generation Partnership Project (3GPP), enhancements of Diverse Data Applications (eDDA) work item has targeted at improving always-on connectivity for smartphones, identifying and specifying mechanisms at the Radio Access Network (RAN) level to enhance the ability of 3GPP Long Term Evolution (LTE) to handle diverse traffic profiles. The key aspects are the Radio Resource Control (RRC) release timer for User Equipment (UE) state control between connected and idle modes, and Discontinuous Reception (DRX) targeted for minimizing the UE power consumption in connected state. There exist important tradeoffs between UE power consumption, Quality-of-Service and RAN/core network signalling load. Whats more, these tradeoffs may be dependent on the UE mobility/velocity. In this article we study the effect of DRX and RRC release timer parameterization on UE mobility performance in heterogeneous networks. These results can be applied to parameter optimization of DRX and RRC release.