Based On Sparse Representation Of The Radar Signal

If a traditional radar system is required to enhance its resolution and work in real time, one faces a challenge of high sampling rate, fast processing speed. In order to solve these problems, a new radar sampling method is proposed. The theory of sparse representation and compressive sensing can be used for the reconstruct of sparse or compressible signals from a small set of linear measurements. With properly chosen sets, the sampling rate could be lower than Nyquist sampling rate.This paper consists of two parts, which mainly study the theory of sparse representation and compressive sensing. The first part mainly investigates the sparse representation of the signals. Firstly, we study constructing the dictionaries, including Gabor dictionary, matched dictionary and so on. Then, two different arithmetics about sparse decomposing are introduced, and their capabilities are validated by decomposing signals. By the end of this part, we construct a simple radar echo model and use the MP arithmetic to decompose the echo signals.The second part of this paper mainly discusses the theory and applications of compressive sensing. The theory of compressive sensing and arithmetic of construction are studied in the first place. Then we introduce three different applications about sparse representation and compressive sensing, including AIC, random filtering and CS radar imaging, all of which are used to lower the sampling rate essentially. The simulation results have verified the feasibility of the theory.