INFORMATION TECHNOLOGY OF DYNAMIC FREQUENCY RESOURCE MANAGEMENT IN COMPLEX RADIO-ELECTRONIC ENVIRONMENT

Abstract

The article proposes an advanced information technology for dynamic frequency resource management under complex radio-electronic conditions, where spectrum scarcity, high user density, and intentional electronic countermeasures create critical challenges. The developed model integrates the flexibility of software-defined radio (SDR) with the adaptive intelligence of cognitive radio, enabling autonomous detection of free frequency bands, resistance to interference, and rapid reconfiguration of transmission parameters. The functional basis of the approach is the “observation–analysis–decision–adaptation” cycle, which has been implemented in the GNU Radio environment with the use of SDR platforms such as USRP. Within this framework, the system continuously monitors the radio spectrum, applies spectral analysis methods, and adapts transmission according to real-time assessments of interference and noise levels. Special emphasis is given to the application of machine learning algorithms, including Q-learning, which provide the system with the capability to accumulate operational experience and improve decision-making efficiency in highly dynamic environments. Such mechanisms ensure a reduction of reaction time to less than 250 ms, allowing the system to instantly switch to optimal frequencies. Experimental validation confirmed that the proposed methodology maintains the bit error rate (BER) at levels below 10⁻⁴ and keeps average packet losses within 5–7%, even under strong jamming and hostile spectrum congestion. Additionally, the system achieves uniform spectrum utilization, which prevents the overload of specific frequency bands and reduces the risk of conflicts with other users. The research builds upon the pioneering works of J. Mitola and S. Haykin, who defined the theoretical foundations of cognitive radio and its learning-based adaptation cycle. It also incorporates findings by Zhou, Wang, Steyn, Pratt, as well as Ukrainian researchers Popov, Zaitsev, and Trysnyuk, who contributed to the study of SDR-based adaptive systems in the context of security and electronic warfare. The obtained results demonstrate that the proposed methodology not only increases resilience and efficiency of communications but also enables cost-effective virtual testing of electronic warfare scenarios, reducing the need for large-scale field experiments. The approach is particularly relevant for military applications, where reliable communications under hostile radio-electronic conditions are critical. At the same time, it provides perspectives for use in civilian domains with congested spectrum environments.