Detection,estimation,and Discrimination Of True And False Targets For Frequency Diverse Array Radar

Different from conventional phased array(PA)radar system,frequency diverse ar-ray(FDA)systems provide a range-and-angle dependent transmit beampattern,caused by the used frequency increments.Hence,besides the typical functionality of a PA,FDA finds applicability in areas such as range-dependent clutter suppression,low-altitude tar-get detection,and target estimation.Considering the distinctive characteristic of the fre-quency increments,this dissertation primarily analyzes the impact of these frequency in-crements on target detection,estimation,and discrimination,including multi-path perfor-mance analysis,adaptive target detection,parameter estimation,frequency increment op-timization design,and discrimination of deceptive interference when using an FDA radar.The main contributions are summarized as follows:(1)Persymmetric adaptive detection for FDA radar.Based on the structural knowl-edge of the covariance matrix for the FDA radar,this dissertation proposes a novel per-symmetric adaptive detector for signals measured in Gaussian noise.In order to detail how the FDA radar cross section(RCS)may vary due to the frequency increments across space and time,this dissertation formulates a statistical FDA RCS model.Then,this disserta-tion derives the analytical detection probability for an evaluate the FDA radar measuring fluctuating targets.The detection performance is evaluated using extensive simulations,demonstrating that the used frequency increments imply a higher detection probability as compared to the FDA systems with non-frequency increment.This dissertation also shows that,with the knowledge of structure of the FDA,the detection performance is improved by using a persymmetric detector.(2)Moving target detection for FDA radar.The used frequency increments in an FDA will cause an increase of the Doppler spread for moving targets.This dissertation analyzes the performance of an FDA radar experiencing Doppler spread in terms of de-tection performance and the corresponding Cramér-Rao lower bound(CRLB),and then compare these with what may be expected when using an FDA radar model which ignores the Doppler spread,illustrating the resulting superiority of the FDA system in handling a Doppler spread.Moreover,this dissertation proposes a novel approach for low-altitude target adaptive detection and for multi-path clutter suppression using the FDA multi-path propagation model,taking into account the effects of the atmosphere.The resulting de-tection performance for multi-path propagation signals is better than that of traditional approaches,with a demonstrated smoother average power characteristics.Furthermore,the results show how the system may be potentially useful for anti-clutter and anti-multi-path fading signals.(3)Designing optimal frequency increments for FDA to optimize the location and detection performance.The frequency increments affect the transmit beampattern for the FDA system,suggesting the need to carefully design the used frequency increments.This dissertation proposes an optimal transmit frequency selection scheme for an FDA system,which selects the used carrier frequencies for each transmitter in order to optimize the location and detection performance.The proposed scheme forms an A-optimal minimiza-tion of the corresponding CRLB over a range of candidate frequency increments,thereby obtaining the set of carriers most suitable for locating a target given the prior information.When the prior information of the parameters are known to come from given prior distribu-tions,the optimal FDA frequency design scheme is instead based on the Bayesian CRLB(BCRLB).The used frequency increments are found as those minimizing the correspond-ing CRLBs.Comparative simulation results show that the proposed selection schemes significantly improve the estimation and detection performance as compared with exist-ing frequency selection approaches.The results also show that the Bayesian formulation offers better detection performance as compared to the non-Bayesian formulation.(4)The discrimination of deception jammers using FDA.This dissertation designs an FDA system that can discriminate between a target and a speed or range deception jamming signal given the unique time-frequency and deception jamming characteristics caused by the used frequency increments.For speed deceptive jammers,the time-frequency characteristics of the echo signal during the stopping periods are different from those of the true target echo signal for the different carrier frequencies,allowing a speed decep-tive jammer signal to be identified without estimating the Doppler frequency of the target.Furthermore,the location of a point target is shown to be obtainable by the multiple signal classification(MUSIC)algorithm,after which the identification of the speed deceptive jammer may be completed by comparing the resulting estimates for the point target.The difference between the carrier frequencies of the received signal and the carrier frequen-cies of the matched filter in the FDA causes the loss of the desired signal.Therefore,this dissertation proposes a discrimination method using a generalized likelihood estimation technique for identifying range deceptive jammers by the use of pulse-cooperation be-tween signal with different frequency increments.The experimental results show that the characteristics caused by the frequency deviation of the FDA with different increments can effectively identify the speed and range deceptive jammer signal.