R.N. Kosyakov1, M.A. Isaev2, E.M. Celishcheva3, D.E. Losev4, A.P. Kovalev5

1-5 JSC “CNIRTI named after academician A.I. Berg” (Moscow, Russia)

1-5 post@cnirti.ru

Problem statement. The current stage in the development of electronic reconnaissance (ER) and passive direction-finding systems is characterized by the increasing complexity of the electronic environment due to the widespread use of digital signals with a low probability of detection, frequency and time diversity, as well as the widespread use of broadband and multichannel communication systems. Under these conditions, analytical signal processing methods based on deterministic algorithms and a priori models face fundamental performance limitations in conditions of uncertainty, noise, and intentional interference. In this context, the use of artificial intelligence (AI) technologies, including machine learning and deep learning methods, as tools for solving detection, classification, parametric estimation, and direction-finding tasks is attracting increasing attention. The ability to learn from large amounts of data, adapt to changing observation conditions, and the ability to identify hidden patterns in multidimensional signal spaces make AI approaches promising for improving the accuracy, noise immunity, and sensitivity of ER systems.

Goal. To develop a model of an amplitude-phase direction finder using AI technologies to improve the efficiency of ER systems in detecting and locating complex radio signals in conditions of a priori uncertainty, high levels of noise and interference.

Results. A neural network model of an amplitude-phase direction finder is presented. Graphs of comparison of estimates of the effectiveness of direction-finding methods using the analytical method and using an artificial neural network (NN) are shown.

Practical significance. The use of the development results will allow the modernization of existing and the creation of promising electronic control systems used in the interests of electronic warfare (EW). For example, in radio frequency spectrum monitoring systems, as well as part of electronic suppression and counteraction systems for unmanned aerial vehicles.

Kosyakov R.N., Isaev M.A., Celishcheva E.M., Losev D.E., Kovalev A.P. The use of artificial intelligence in electronic intelligence and passive direction-finding systems. Radiotekhnika. 2025. V. 89. № 10. P. 140−145. DOI: https://doi.org/10.18127/j00338486-202510-17
(In Russian)

1. Golikov V.N., Busygin I.N., Kostin G.A., Manzhos V.N., Minervin N.N., Najdenov B.V., Poljakov V.I., Chelpanov A.S. Teoreticheskie osnovy radiolokacii; Ucheb. posobie dlja vuzov. Pod red. Ja.D. Shirmana. M.: Sovetskoe radio. 1970. 560 s. (in Russian).
2. Denisov V.P. Antennye sistemy fazovyh radiopelengatorov: Metodicheskie rekomendacii po vypolneniju raschetnogo zadanija po kursu «Osnovy teorii sistem i kompleksov radiojelektronnoj bor'by». Tomsk: Tomskij gos. un-t sistem upravlenija i radiojelektroniki. 2019. https://studfile.net/preview/16873708/ (in Russian).
3. Metod Ujelforda i mnogomernaja linejnaja regressija [Jelektronnyj resurs] // habr.com. URL: https://habr.com/ru/artic-les/343752/.
4. Shapalin V.G., Nikolaenko D.V. Rassmotrenie sushhestvujushhih sposobov sozdanija jelementov sistemy raspoznavanija lic // Sb. nauch. trudov IV Mezhdunar. nauch.-praktich. konf. Doneck. 2022. T. 1. S. 236-240 (in Russian).
5. Realizuem i sravnivaem optimizatory modelej v glubokom obuchenii [Jelektronnyj resurs] // habr.com. URL: https://habr.com/ru/companies/skillfactory/articles/525214/ (in Russian).