Continuous versus Discrete Model in Autodiagnosis Systems for Wireless Networks

In the near future, several radio access technologies will coexist in Beyond 3G mobile networks (B3G), and they will be eventually transformed into one seamless global communication infrastructure. Self-managing systems (i.e., those that self-configure, self-protect, self-heal, and self-optimize) are the solution to tackle the high complexity inherent to these networks. In this context, this paper proposes a system for autodiagnosis in the Radio Access Network (RAN) of wireless systems. The malfunction of the RAN may be due not only to a hardware fault but also (and more difficult to identify) to a bad configuration. The proposed system is based on the analysis of Key Performance Indicators (KPIs) in order to isolate the cause of the network malfunction. In this paper, two alternative probabilistic systems are compared, which differ on how KPIs are modeled (continuous or discrete variables). Experimental results are examined in order to support the theoretical concepts, based on data from a live network. The drawbacks and benefits of both systems are studied, and some conclusions on the scenarios under which each model should be used are presented.

[1]  Gregory M. Provan,et al.  Why is diagnosis using belief networks insensitive to imprecision in probabilities? , 1996, UAI.

[2]  Malgorzata Steinder,et al.  Probabilistic fault localization in communication systems using belief networks , 2004, IEEE/ACM Transactions on Networking.

[3]  Kimmo Raivio,et al.  A Knowledge-Based Model for Analyzing GSM Network Performance , 2005, IDA.

[4]  R. Brennan,et al.  The Celtic Gandalf framework , 2006, MELECON 2006 - 2006 IEEE Mediterranean Electrotechnical Conference.

[5]  Hermann Wietgrefe.  Investigation and practical assessment of alarm correlation methods for the use in GSM access networks , 2002, NOMS 2002. IEEE/IFIP Network Operations and Management Symposium. ' Management Solutions for the New Communications World'(Cat. No.02CH37327).

[6]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[7]  Nir Friedman,et al.  Bayesian Network Classifiers , 1997, Machine Learning.

[8]  Volker Wille,et al.  System for automated diagnosis in cellular networks based on performance indicators , 2005, Eur. Trans. Telecommun..

[9]  Thomas J. Watson,et al.  An empirical study of the naive Bayes classifier , 2001 .

[10]  Bojan Cestnik,et al.  Estimating Probabilities: A Crucial Task in Machine Learning , 1990, ECAI.

[11]  K. Hatonen,et al.  Advanced analysis methods for 3G cellular networks , 2005, IEEE Transactions on Wireless Communications.

[12]  Zwi Altman,et al.  A Bayesian Approach for Automated Troubleshooting for UMTS Networks , 2006, 2006 IEEE 17th International Symposium on Personal, Indoor and Mobile Radio Communications.

[13]  Kimmo Raivio,et al.  A SOM Based Approach for Visualization of GSM Network Performance Data , 2005, IEA/AIE.

[14]  Z. Altman,et al.  Automated Troubleshooting in WLAN Networks , 2007, 2007 16th IST Mobile and Wireless Communications Summit.

[15]  Abbas Jamalipour,et al.  A tutorial on multiple access technologies for beyond 3G mobile networks , 2005, IEEE Communications Magazine.

[16]  Oscar Kipersztok,et al.  Another look at sensitivity of Bayesian networks to imprecise probabilities , 2001, AISTATS.

[17]  Usama M. Fayyad,et al.  Multi-Interval Discretization of Continuous-Valued Attributes for Classification Learning , 1993, IJCAI.

[18]  Catherine Blake,et al.  UCI Repository of machine learning databases , 1998 .

[19]  Geok See Ng,et al.  Using a qualitative probabilistic network to explain diagnostic reasoning in an expert system for chest pain diagnosis , 2000, Computers in Cardiology 2000. Vol.27 (Cat. 00CH37163).

[20]  Pedro M. Domingos,et al.  On the Optimality of the Simple Bayesian Classifier under Zero-One Loss , 1997, Machine Learning.

[21]  David Heckerman,et al.  Decision-theoretic troubleshooting , 1995, CACM.

[22]  Gregory M. Provan,et al.  The Sensitivity of Belief Networks to Imprecise Probabilities: An Experimental Investigation , 1996, Artif. Intell..

[23]  João Cesar M. Mota,et al.  Competitive Neural Networks for Fault Detection and Diagnosis in 3G Cellular Systems , 2004, ICT.