Stochastic signal processing in adaptive measurement systems with rough space-time statistics: invertible spectral analysis method

A modified method for efficient spectral and correlation-based spatio-temporal signal processing is presented, designed for operation under conditions of high dynamic variability in input signal intensity and varying frequency-time resources of digital processing. The method is based on adapting algorithms and refining the goals and tasks of signal processing. A methodology has been developed to achieve high instrumental resolution of signals based on spectral and/or correlation features. The study addresses aspects of improving the efficiency of spatio-temporal signal processing, which are relevant for radio engineering measurement systems with digital phased antenna arrays and moving target selection systems. The previously proposed adaptive method of reversible spectral analysis, developed by the author, has been modernized for use in spatio-temporal signal processing. The updated method takes into account the variability (type changes) of processed signals and introduces measures to reduce the dynamic range requirements of the input signal stream by utilizing coarse (low-bit and binary) statistics. The frequency nomenclature has been expanded to include temporal and spatial spectra of distribution laws (characteristic functions). Technical (hardware and software) constraints, such as the use of coarse signal quantization, are also considered. Processing efficiency is achieved through the application of a whitening operation (rejection of dominant components) for passive interference prior to the main stage – the reversible spectral analysis method — and through the use of traditional stochastic algorithms. This includes increasing sample sizes (apertures, windows) and improving the convergence rate of measurements in the basic Monte Carlo method. The results obtained can be applied in radio engineering measurement complexes, including radar systems, for tasks such as radio and radio technical monitoring, as well as for measuring range and bearing coordinates.

stochastic linearization, randomization, stochastic measurement system, Monte Carlo method, rough statistics, averaging