МЕТОД ПІДВИЩЕННЯ ЕФЕКТИВНОСТІ КЕРУВАННЯ ЕНЕРГЕТИЧНИМ ПОТЕНЦІАЛОМ ЗАХИЩЕНИХ РАДІОЛІНІЙ ТЕРАГЕРЦОВОГО ДІАПАЗОНУ З ВИКОРИСТАННЯМ ШТУЧНОГО ІНТЕЛЕКТУ

Abstract

The growth of large volumes of information flows encourages the development of transmission and reception systems in the very high frequency range to ensure effective control of IR-UWB radio links of terahertz signals based on machine learning algorithms and neural networks, taking into account energy saving. For this purpose, the article proposes an algorithm for tracking a multipath signal of a system for receiving signals from spatially separated low-power transmitters, a feature of which is the refinement in the process of tracking the time positions of the components and their number. A feature of the developed algorithm is the use of the wavelet transform to obtain the input image of the neural network. A structural and functional model for constructing a receiving system for IR-UWB signals in the very high frequency range with intelligent control elements is proposed, which is based on separate control planes and physical infrastructure for automatic and operational control of the process of sharing physical infrastructure resources and artificial intelligence methods. Unlike existing models of IR-UWB receiving systems for terahertz signals, it provides protocol and infrastructure data collection for intelligent algorithms. The presented physical infrastructure has a training and optimization module that involves the use of an existing simulation model of a radio link in the terahertz range from 0.11 to 0.17 terahertz to test intelligent algorithms for controlling the energy potential of IR-UWB radio links. The developed data collection algorithm involves monitoring the state of the blocks of the receiving complex for rational data collection using the change in the values of both the Euclidean distance metrics and the metrics of functional technical parameters in relation to the number of clusters.