Hakki C. Cankaya

Reliability and availability evaluation of self-healing SONET mesh networks

For self-healing SONET mesh networks, the restoration process plays an important role in determining reliability and availability. Because of this, reliability and availability evaluation of the network, with the ability to quantify the differences between restoration techniques, have become essential. We first introduce a Markov state reward model to evaluate the reliability and availability of a network. We then propose a third metric, called restorability which represents the self restoration capability of the network. A numerical solution to the model is discussed along with an illustrative example.

Accelerated reliability analysis for self-healing SONET networks

Recently, a parametric State Reward Markov Model (SRMM/p) has been developed for the reliability and availability analysis of self-healing SONET mesh networks [2]. In this paper, we investigate the factors that affect the run-time complexity of the model presented in [2]. In order to accelerate the reliability and availability analysis, we present an approach that aggregates a set of states in the model based on 2-phase hypoexponential distribution. A comparison of the original and the reduced model, with respect to runtime complexity and accuracy, is carried out by applying the models for the analysis of few complex networks.

Customer Churn Prediction for Telecom Services

Customer churn is a big concern for telecom service providers due to its associated costs. This short paper briefly explains our ongoing work on customer churn prediction for telecom services. We are working on data mining methods to accurately predict customers who will change and turn to another provider for the same or similar service. Sample dataset we use for our experiments has been compiled by Orange Telecom from real data. They posted the sample dataset for 2009 Knowledge Discovery and Data Mining Competition. IBM has scored the highest on this dataset requiring significant amount of computational resources. We are aiming to find alternative methods that can match or improve the recorded highest score with more efficient use of resources. Dataset has very large number of features, examples and incomplete values. As the first step, we employ some methods to preprocess the dataset for its imperfections. Then, we compare and contrast various ensemble and single classifiers. We conclude the paper with future directions for the study.

Survivability analysis of self-healing SONET rings

Previously, a parametric state reward Markov model (SRMM/p), has been developed for the survivability analysis of self-healing SONET mesh networks. In order to handle the inherently high complexity of mesh networks, the model uses some complexity-reduction approximations. Since a ring is a simplified version of a mesh from the topological point of view, the SRMM/p is equally applicable to self-healing SONET rings (SONET SHRs). The uniformity and the simplicity of the rings allows one to include more details of the failures in the model so that the analysis is more accurate. To that end, we propose a probability-tree based evaluation of the probability of various system states resulting from all possible combinations of node and link failures. The resulting improvements in the reliability analysis are illustrated using experiments.

Hierarchical approach to performance analysis of self-healing SONET networks

The performance evaluation of large self-healing SONET mesh networks is of great interest to the telecom industry. In this paper, we first modify a model that has been proposed originally for the reliability analysis of SONET networks, for the performance evaluation of such networks. Then, we propose a hierarchical approach that will not only reduce the complexity of the analysis, but also improve the accuracy. The sensitivity of the performance metrics to the traffic distribution and the performance of individual subnetworks are studied experimentally.