Wideband spectrum estimation in frequency dense environments

The majority of new radar applications requires to analyse a large frequency spectrum in real time, in order to perform effective detection and recognition in frequency dense environments. When the spectrum is very large, the samples representing a wideband signal must be collected at a much lower rate than the Nyquist rate corresponding to the highest frequency of the spectrum. We focus our attention on the two most promising methodologies proposed to perform this task. The first one is based on the correlation of different sub-Nyquist asynchronous channels. The second method is based on a preliminary analog processing of the wideband signal, followed by digital sub-Nyquist sampling. Both methods reach fair performances when the spectrum density is not too high and/or the input signal to noise ratio is sufficiently high. This paper combines the two methods, in order to work efficiently also when the previous conditions are not satisfied.