Research On High-precision Range And Velocity Estimation For Narrowband Radar Systems

Narrowband radar can work all time,all weather and provide long-range detection;besides,it can finish the tasks that are difficult for some other systems(such as optics system)in a complex environment,which makes radar more and more popular in military and civil fields.As a frontier research topic in radar signal processing,high-precision range and velocity estimation plays a significant role in target recognition,radar imaging and other fields and has important military value and scientific research significance.The main research in this paper is focus on the high-precision range and velocity estimation in narrowband radar and the main work is outlined as follow:(1)Linear frequency modulated signal that is common in radar system is taken as an example.We describe the signal models of radar system in detail from the transmission to the subsequent signal processing,which paves the way to the motion parameter estimation discussed in the latter part of this dissertation.(2)Maximum likelihood estimation has the optimal estimation accuracy in theory,but there is serious velocity ambiguity when this method is applied in narrowband radar systems under low pulse repetition frequency.In order to solve this problem and set a velocity search scope for maximum likelihood estimation,a novel velocity estimation algorithm based on in-band cross-correlation is presented.Then the signal-to-noise ratio before and after in-band cross-correlation is analyzed,and the minimum mean square error of the algorithm is derived.Meantime,the optimal interval number of cross-correlation operation is obtained based on the minimum mean square error of the algorithm.Finally,the properties of the proposed algorithm is analyzed and verified through Monte Carlo simulations.(3)In order to reduce the computation load of the maximum likelihood estimation method in the range dimension,a coast range estimation method is introduced to set a range search scope for maximum likelihood estimation.Maximum likelihood estimation belongs to the classical point estimation methods,which has the contradiction between computation complexity and estimation accuracy.Then,inverse fast Fourier transform(IFFT)and Chirp-Z transform(CZT)is applied to simplify the calculation process of likelihood function,which makes the maximum likelihood estimation run faster.Finally,a single-target joint estimation algorithm for range,velocity and acceleration is proposed based on the above algorithms.Experimental results show that the performance of the joint estimation method is stable and can achieve Cramer-Rao lower bound.In order to adapt to the multi-target application scenarios,the multi-target joint estimation algorithm combined the novel single-objective joint estimation algorithm and the CLEAN algorithm is presented,which can achieve the joint estimation of range,velocity and acceleration for all motion targets by eliminating the strong target signal one by one.The numeric experiments show that this method can effectually estimate the range and motion parameters of multiple targets.