Digital Twin Use Case in Telecommunications Transport Network

Digital Twin (DT) is a high-fidelity simulation technique that converts a real-world workplace into a virtual environment. Information technology and the industrial industry are now deeply integrated and developing as a result of the advent of DT technology. The application of this technology to telecommunications has come to the forefront. In this study, artificial intelligence methods were applied for failure root cause analysis in transport systems adapted to digital twin technology used in telecommunication systems. Three different classification methods were used to analyze the obtained data. The Synthetic Minority Oversampling Technique (SMOTE) model was used to stabilize the data. The study includes the analysis of a data pool created by correlating the data received from the devices belonging to the fiber optic cable (Fiber), Dense Wavelength Division Multiplexing (DWDM) systems, and Internet Protocol Multi-Protocol Label Switching (IP-MPLS) layer, which constitute the transport layer of telecommunication systems. The use of DT terminology in telecommunication technologies is directly related to determining which use case to address, which domains, and which parameters to take in the existing network. For this purpose, the DT structure created by topologically associating DWDM technology and IP-MPLS technology, and then matching them at fault and alarm level, is used in root cause analysis prediction using artificial intelligence algorithms. As the output of the data pool; 3 basic error conditions “MPLS Flap”, “Short Term Interruption”, and “Long Term Interruption”, which cause interruptions on the user side, are discussed. Random Forest (RF), Naive Bayes (NB), and K-Nearest Neighbors (KNN) methods were used as artificial intelligence classification methods. When the results are analyzed, it is seen that the results of the Random Forest (RF) method are the most accurate prediction.