Abstract
Formulation of the problem. The use of artificial neural networks to solve the problem of detecting and identifying signal structures as objects in the image is not oriented to work in a real, highly dynamic and saturated radio-electronic environment, characterized by intense fluctuations in radio signal parameters and changes in radio link parameters. The fundamental difficulty in creating an up-to-date, complete and reliable reference description of the radio technical parameters of detected radio links determines the need to create highly universal and adaptive algorithms for identifying radio links, which significantly limits the range of applications of the neural network approach to solving these problems. Target. Develop and analyze a method for detecting the boundaries of a radio signal in the time-frequency domain. Results. The paper proposes a new multidimensional matrix window operator and a corresponding statistical rule for adaptive detection of the corner of a rectangular signal region in a time-frequency panorama. An exact analytical expression is obtained for the distribution of the decisive detection statistics, the probabilities of false alarms and missed signals. Practical significance. The implementation of the proposed approach makes it possible to detect and distinguish radio signals from modern radio communication and data transmission lines in a real complex electromagnetic environment, characterized by mutual overlap of frequency-time domains of signals from different radio lines with a priori uncertainty regarding the noise level.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call