Information technology for passive location of dynamic events in the border zone

• Datum: 2025
• Hauptverfasser: Vasylenko, Vladyslav, Trysnyuk, Taras, Berchun, Yaroslav
• Format: Artikel
• Sprache: Ukrainian
• Veröffentlicht: Kyiv National University of Construction and Architecture 2025
• Schlagworte: passive localization; geographic information systems; Chernihiv region; time series analysis; forecasting; monitoring
• Online Zugang: https://es-journal.in.ua/article/view/343593
• Назва журналу: Environmental safety and natural resources

This article presents the results of a research project focused on developing an information technology for passive localization of dynamic events in the border areas of Chernihiv region, Ukraine – a territory continuously affected by artillery shelling and military activity due to ongoing armed aggression. The proposed system integrates multiple components: sensor-based measurement modules (seismic and acoustic detectors), digital catalogs of amplitude-frequency spectral patterns, geoinformation services, signal processing techniques, and predictive analytics. The research utilizes data from various sources – sensor networks, high-resolution satellite imagery, and open-source web platforms – all integrated into a multilayered system architecture powered by modern data processing tools.During the scientific research, the following results were obtained: the hardware and software complex of passive location was improved; digital catalogs of spectra of civil and military equipment were formed; GIS risk models were created for the border strip of Chernihiv region; modern forecasting methods were applied; a decision support system was developed; the possibilities of integrating the technology into the national early warning system were substantiated.The outcome includes dynamic risk maps, thermal maps of forest burnouts and agricultural degradation, polygonal models of affected territories, and spatio-temporal diagrams of threat activity. Forecasting modules were implemented using statistical models such as ARIMA and Prophet, as well as deep learning models based on LSTM networks, allowing accurate prediction of repeated shelling or vehicle movement patterns. The practical relevance of this work lies in the possibility of integrating the developed technology into Ukraine’s national early warning system. The system can be effectively used by local authorities, the State Emergency Service, and military units to monitor the security environment, ensure timely response, and minimize casualties and infrastructure damage in vulnerable border communities.