Research On Signal Processing Of Non-Linear Stepped-Frequency Chirp Pulse Train

High range resolution radar is an important aspect in the development of modern radar. It is possible to achieve the one-dimensional range profile of the targets in practical radar system by applying the synthetic wideband technology, which can decrease the difficulty of realization of the practical radar system and the system cost comparing to a traditional way. It is also one of the hot topics in recent years. This dissertation focuses on the theory analysis and signal processing technology of a kind of non-linear stepped-frequency chirp pulse train modified from linear stepped-frequency chirp pulse train.The research on the non-linear stepped-frequency technology is still a curtail step in the world. The non-linear stepped-frequency technology not only inherits the merits of linear stepped-frequency and chirp technology, but also provides a better suppression level of range side lobes and grating lobes of the one-dimensional range profile and a better Doppler resolution, which has great value in practical engineering. The dissertation presents a novel and systematic investigation on its characteristics and signal processing technology and proposes several methods to resolve the questions occurring in signal processing.The main research work is as follows:1. The characteristics of the non-linear stepped-frequency chirp pulse train were analyzed intensively and its ambiguity function and spectrum structure were derived. The process of achieving high range resolution by non-linear stepped- frequency chirp pulse train is also studied.2. The parameter selection rule of the approximate weighting by sub-pulse coherent integration and its compare to traditional weighting were analyzed. The ratio of main lobe to sidelobes, coefficient of main lobe wideness and signal-to-noise ratio improvement caused by approximate weighting were derived theoretically. The characteristics of grating lobes of non-linear stepped-frequency chirp pulse train were researched and two methods, systematic parameter selection and sub-pulse extended correlation, were provided to suppress and nullify them. 3. The Doppler Effect of the non-linear stepped-frequency chirp pulse train introduced by target motion and its effect on synthesizing the one-dimensional range profile were analyzed in detail. The theoretical Cramer-Rao bound of motion parameter estimation was also derived. Two novel and applicable methods, Doppler matched filtering algorithm and least step error algorithm, were proposed to estimate the motion parameter accurately and compensate the Doppler Effect based on the traditional methods applied to the linear stepped-frequency chirp pulse train. Their performance and effect on the estimation accuracy of different parameters and their effectiveness were analyzed and demonstrated by comparing to the theoretical Cramer-Rao bound.4. The reason of alias, redundancy and replication occurring in the one-dimensional range profile synthesized by the non-linear stepped-frequency chirp pulse train were analyzed. The principle of the parameter selection such as frequency step size, sub-pulse duration, pulse repetition time, sub-pulse bandwidth, sampling rate and etc. was summarized to avoid the above problems and a combination of parameter selection and target pick-up method was proposed to obtain a real and entire one-dimensional range profile.