Identification of Structural Reliability Parameters in Distributed Telecommunication System

Abstract

The article considers estimation of structural parameters for communication network. The article aims to calculate the key indicators of structural reliability. Proposed method consists of mathematical expectation estimation for number of connections and connectivity probability in different structured networks. The methodology is based on solution to the problem of network optimization structure for reliability view. Classification of equipment is proposed in terms of failure impact on quality management of network traffic. According to the accepted classification, reliability parameters are listed characterizing each of the classified equipment groups. Reliability parameters are divided into single and complex ones. The reliability and quality of the communication network functioning is assessed by the estimated values of these parameters. A probabilistic graph of network model and state diagram of communication lines for one hierarchical level were plotted. When plotting, it was assumed that switching nodes are absolutely reliable. The reliability of communication lines is expressed in terms of availability. The calculation of parameters was carried out on the assumption that communication line failures are statistically independent events. Reliability indicators are estimated, such as the probability of connectivity between each pair of telecommunication nodes and the absolute value of mathematical expectation for number of connections. For convenience, practical application of the proposed method, calculation results of the mathematical expectation of number of connections is expressed as a percentage. The article considers a mathematical model of the functioning of a separate communication line with the absolute reliability of the system for monitoring its technical condition. The mathematical model is presented in the form of a state diagram. It is assumed that the network monitoring system performs the functions of periodic and continuous monitoring. A method for calculating complex indicators of reliability, such as the availability factor and the downtime factor, is proposed. The calculation method is based on the matrix method for analyzing probabilistic systems. The calculation formulas include statistical data from the system for monitoring the technical condition of communication lines. Modelling is used to simulate the process of communication network functioning in experimental research. The process of communication network functioning is simulated. Depending on the incoming load, a search for free algorithmic resources and communication lines is performed. The calculation of the reliability indicators of the selected way of servicing calls is performed, taking into account the selection criterion set by the program. The monitor displays a dependency graph of the reliability indicators on the intensity of the incoming load. The calculation of the relative statistical frequency of the success of servicing requests is carried out, and the frequency of communication lines failure is also calculated. The program output is proposals for communication network reconstruction due to network resources redistribution.