Research On Adaptive Processing Technology For Frequency Diverse Array Radar

As an emerging array radar,frequency diverse array(FDA)radar is formed by imposing an additional frequency increment across the elements of conventional phased array(PA)radar.Different from conventional PA radar transmitting an angle-dependent-only beampattern,FDA radar is capable of producing a jointly range-angle-dependent beampattern due to the use of frequency increment.Hence,besides the typical functionality of conventional PA,FDA finds applicability in range-dependent beampattern synthesis,target detection,interference suppression,electronic countermeasure and secure communications.Departing from the fundamental signal characteristics of FDA radar,this thesis has systematically investigated the critical techniques associated with FDA radar adaptive processing,including the coherent receiver design for FDA radar signal,adaptive moving target detection and tracking approaches using FDA radar.The main contributions of this thesis are summarized as follows.(1)Theoretical analysis for the impact of frequency increment on the frequency decorrelation and resolution capability of FDA radar signal.This thesis proposes an empirical criterion for frequency increment selection by considering the frequency de-correlation of echo signal amplitude,in the application scenarios of high-resolution ranging and clutter suppression,respectively.Besides,this thesis analyzes the impact of frequency increment on the resolution capability in range,angle and Doppler shift through the generalized ambiguity function of FDA radar.The associated issues,such as inner-bin range ambiguity,transmit beam broadening and Doppler ambiguity,are discussed,which provides insights for subsequent selection of frequency increment.(2)General coherent receiver design approach for FDA radar.For general FDA radar echo signals,this thesis designs a coherent FDA receiver structure based on multi-channel matched filtering(MCMF).This MCMF receiver is generic,since it applies to both spectrally overlapping(correlated waveforms)and non-overlapping(orthogonal waveforms)cases.Based on the receiver data model,this thesis analyzes the performance of FDA radar in terms of virtual transmit beampattern,output signal-to-interference-plus-noise ratio(SINR)and Cramér-Rao bound(CRB)for parameter estimation,and then compares it with those of conventional PA and multi-input and multi-output(MIMO)radars,to show the application superiority of FDA radar in high-precision range estimation and rangedependent interference suppression.(3)Adaptive moving target detection using FDA radar.For adaptive detection of targets with conventional and high speeds,this thesis proposes a low-complexity detection algorithm without extra training data required,and an adaptive detection approach of blind-speed target based on mainlobe clutter suppression,respectively.The former is proposed to overcome the difficulty that the training data from adjacent range bins cannot directly be used to estimate the interference covariance matrix(ICM)due to the range dependency of FDA data model.It is derived from an unstructured generalized likelihood ratio test,which estimates the ICM using the target-present data.Moreover,the algorithm is computationally efficient since it requires a one-dimensional Doppler-only search,instead of a three-dimensional search in the joint range-angle-Doppler domain.As the target speed increases,the Doppler spreading due to the use of frequency increment becomes increasingly significant.The latter approach exactly utilizes those additional Doppler spreading phases to differentiate the blind-speed target and mainlobe clutters,thereby improving the detection performance of blind-speed target.Compared to traditional approaches,this FDA-based one does not need to change the pulse repetition frequency.(4)Adaptive moving target tracking using FDA radar.In order to achieve the automatic avoidance of spectral interferences and real-time focus of transmit beam,this thesis,inspired by the concept of cognitive radar,proposes two moving target tracking approaches for FDA radar,using adaptive transmit power allocation and adaptive subaperture,respectively.The former can adaptively allocate the transmit power of each element according to a real-time spectrum sensing result,so that the FDA radar can automatically avoid narrowband spectral interferences.As compared to the one with fixed power allocation,the proposed approach improves the output SINR and thus decreases the tracking errors.The latter divides the transmit array into multiple phased subarrays and employs a frequency increment between adjacent subarrays,in order to overcome the drawback that the FDA radar with orthogonal frequency increments cannot enjoy the directional gain.It enables the FDA radar to automatically focus the transmit beams of subarrays toward the target direction,thereby achieving matched illumination.The resulting tracking errors are lower than those of orthogonal and sum-beam transmissions.