D.A. Fofanov

One of the major radiolocation problems is determination of number of targets and estimation of their parameters. When targets are close to each other a problem of resolution-detection and resolution-estimation is originating. Signals are known to be resolved by parameter if the parameter of these signals differs as large as the width of ambiguity function. This threshold of resolution can be overcomed by detailed analysis of the response waveform. For example, two signals can be detected even when their responses are greatly overlapping. Response waveform depends on amplitudes, delays, Doppler frequencies and phases of received signals. These dependencies can be exploited to determine parameters of the signals. Correlator response can be described by a model representing an effect of receiving fully determined signals with known parameters. These parameters are being varied and a coincidence factor such as maximum correlation or minimum mean squared error is searched for. Optimal parameters are considered as the estimate of signals parameters. A simulation of such search was conducted and the coincidence factor was evaluated for every combination of model parameters. An approximation of correlator output principal maximum can be applied to reduce computational complexity. Parabolic and cosinusoidal approximation is shown to be suitable. Approximation functions parameters can be found with by least squares algorithm on basis of the response samples. High quality of approximation can be achieved by increasing of sample rate and threshold raising that leads to approximation area reduction. As a result of the research an interdependence of correlator output waveform and parameters of received signals is shown. This waveform can be recorded with high sample rate with further detailed analysis to provide resolution and parameters estimation. The results are supposed to be exploited to process signals in presence of noise and interference