Method of Automatic Parameters Estimation of Radio Signals with Frequency-Hopping Spread Spectrum

Abstract

Method of automatically estimation of frequency elements duration, first hop beginning time, last hop ending time, the number and nominals of target group frequencies and the bandwidth of radio signal with frequency-hopping spread spectrum in the presence of durable extraneous emissions in the frequency range of receiver is proposed in the article. The method is based on the analysis of the time-frequency distribution obtained using the Welch window transformation and consists of three main stages: extraneous emissions detection, time parameters estimation and frequency parameters estimation of frequency-hopping spread spectrum. The detection of extraneous emissions is carried out according to the time criterion using an iterative approach. On the basis of estimated emissions frequency parameters band-rejection filters are formed, which central frequencies are equal to the central interferences frequencies and stopband are equal to emissions spectrum width. The estimation of time and frequency parameters of the radio signal is carried out by analyzing the signal time-frequency distribution, taking into account the characteristics of the band-rejection filters.To exclude from the calculation of the parts of the signal containing harmonics associated with the change in the values of frequency elements, the time windows of the time-frequency distribution are limited and aligned with the centers of hops. An approach to the formation of operating frequencies grid using the statistical criterion and base frequencies, calculated by minimizing the sum of the differences between the current nominal frequency elements and the frequency grid step is proposed. An algorithm for calculation the last hop ending time based on a joint analysis of the current values of the central frequencies and the energies of the frequency elements is developed. The results of testing the operability and efficiency of the developed method by modeling in the MATLAB software environment with a signal-to-noise ratio of -25 dB are given.