Answers are grounded in this paper and cite their sources.
The paper presents algorithms for identifying signals and determining the threshold of false identification based on the integral bispectrum conversion and the Euclidean distance calculation. The analytical calculation of statistical characteristics in the form of the average probability of identification error, identification error of a known signal and a new signal is carried out. The advantages of bispectral signal conversion over spectral power density in identifying signals with their strong mutual correlation (from 0.5 to 0.9) are shown. Mathematical and computer modeling of the signal identification procedure and the optimal threshold for determining a new signal is performed. The simulation results confirmed the coincidence with the theoretical values of the probability of signal identification error.
No takes yet. Share an insight, caveat, or question.
А. Е. Манохин (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: