Time-frequency approach to radar detection, imaging, and classification

One of the central problems in exploiting the radar data is the analysis of a time series. The problem at hand is how to extract the information present in the data and use it to its full potential. Traditionally, radar signals have been analysed in either the time or the frequency domain. The Fourier transform is at the heart of a wide range of techniques that are generally used in radar data analysis and processing. However, the change of frequency content with time is one of the main features we generally observe in radar data. As a result of this change of frequency content with time, radar signals belong to the class of non-stationary signals. The analysis of nonstationary signals requires a technique that extends the notion of a global frequency spectrum to a local frequency description. Joint time-frequency analysis using time-frequency transforms has improved the analysis of non-stationary signals by revealing time-varying information embedded in signals.