Learning of model parameters for fault diagnosis in wireless networks

Abstract

Self-management is essential for Beyond 3G (B3G) systems, where the existence of multiple access technologies (GSM, GPRS, UMTS, WLAN, etc.) will complicate network operation. Diagnosis, that is, fault identification, is the most difficult task in automatic fault management. This paper presents a probabilistic system for auto-diagnosis in the radio access part of wireless networks, which comprises a model and a method. The parameters of the model are thresholds for the discretization of Key Performance Indicators (KPIs) and probabilities. In this paper, some techniques are proposed for the automatic learning of those model parameters. In order to support the theoretical concepts, experimental results are examined, based on data from a live network. It has been proven that calculating parameters from network statistics, instead of being defined by diagnosis experts, highly increases the performance of the diagnosis system. In addition, the proposed techniques enhance the results obtained with continuous diagnosis models previously exposed in the literature.

Automatic root cause analysis based on traces for LTE self-organizing networks

“…This system can easily be designed from the relation between the radio causes and indicators (e.g., Table 1), and it does not require big computational capacity. A more advanced automatic diagnosis system based on a probabilistic method such as Bayesian networks is proposed in [13] for SH in mobile networks. Both systems require the design of thresholds to analyze the input data.…”

“…Note that the chosen automatic classification system is the simplest one and uses crisp thresholds, which makes the proper identification more difficult. More sophisticated diagnosis systems, such as those in [4,13,14], should improve those results. In spite of this, the obtained error rate is comparable to that obtained by a human expert.…”

A method for identifying faulty cells using a classification tree-based UE diagnosis in LTE

“…In the field of self-healing, several research efforts have been devoted to the development of usable automatic detection and diagnosis systems [29]. On the one hand, various mathematical approaches have been applied to analyze network measurements, such as Bayesian networks [30,31], Neural Networks [5,8], Fuzzy Logic combined with Genetic Algorithms [32], linear prediction [33], correlation [34], and statistical analysis [35,36]. However, these data-driven algorithms have been exclusively evaluated with per-cell level measurements, which may not be sufficient to manage the new data services.…”

Using big data for more dependability

“…However, much of the work focuses on detection of failures [3,6,8,9], and identification of root cause of the failure [1,10]. Their goal is different from ours, which is to predict failures and proactively try to prevent them, or lessen the effects of failures on user experience in the short term.…”