Method of identification of data routes in wireless self-organized networks
Andrii Divitskyi, Serhii Salnyk, Vladyslav Hol, Антон Сторчак · Collection Information technology and security · 2021
Proposes a method for identifying data routes in wireless self-organized networks on the basis of genetic algorithms. The features of building networks of this class are described. The main tasks of the functioning of control systems of wireless self-organized networks were defined. It was emphasized that for complete functioning of wireless self-organized networks control systems was maintaining of adequate quality of their service, which included the process of changing data transmission routes and predicting the time of changes in routes. It was justified that forecasting allowed you to set up the network in time to prevent overloads, errors, failure,to predict changes in data transmission routes in different situations. The forecasting process was described. It was found out that to solve the forecasting tasks, it is advisable to use a genetic algorithm, in particular, the problems of multicritical optimization. This is due to the principle of multicritial optimization, which consists in searching for the optimal solution that simultaneously satisfies more than one target function. The routing system, its tasks and features of construction are described. The model of the forecasting subsystem is described, its importance is emphasized. The concept of identification and its methods (active, passive) are defined. It was considered the work of the rapid genetic algorithm in which due to the presence of a special elite population we can significantly reduce the time of searching for acceptable solutions on separate steps of measurements, compared to the classic genetic algorithm. The stages of the work of the rapid genetic algorithm are described and the corresponding calculations with graphical display are carried out. The essence of the proposed method is in using of a rapid genetic algorithm, which provides an acceptable quality of identification of unknown parameters of the wireless self-organized networks forecasting subsystem. On the other hand, due to the presence of a special elite population, it is possible to significantly reduce the time of searching for an acceptable solution during the processing of each measurement, turning on the classic genetic procedure of the loss optimization function only when its value exceeds some permissible threshold level. This algorithm will remain in operation even in case of non-functioning of the wireless self-organized networks forecasting subsystem.