A Research Of Interferences Suppression Based On Frequency Diverse Array

Due to the special far-field energy distribution caused by frequency increments between the transmit antennas,frequency diverse array(FDA)has been given much attention these years.Given the range-angle dependent transmit beampattern,FDA has a promising applications on interference suppression and target localization.The present FDA literature mainly focus on FDA transmit beampattern design and parameter estimation while little of them systematically discusses FDA's application on interference cancellation.In this work,transmit beampattern design and adaptive beamforming are mainly used for FDA interference suppression while parameter estimation in the interference-plus-noise environment is also considered.Thus,this work can be generally divided into three following parts:(1)The first part mainly focuses on FDA transmit beampattern design for interference suppression.In the scenario of a basic FDA system,conventional delay-and-sum beamformers are applied to generalize FDA static beampattern aiming to suppress some interferences.Furthermore,the new degree-of-freedom(DOF)can also be utilized to modified the beampattern such that mainlobe interference can be suppressed.Next,considering the range-angle coupling in FDA's beampattern,the condition for decoupling range and angle is derived with several specific decoupling methods proposed and compared.In order to achieve a low probability of intercept beampattern,the uniform elemental power constraint is added into the criterion of beampattern design,resulting in frequency diverse array multiple-input-multiple-output radar(FDA-MIMO)system.The derivation of FDA-MIMO itself proves that FDA-MIMO has the ability of low probability of intercept.(2)Interference suppression algorithms at the receiver are systematically discussed in the second part.Since the mathematical structure of FDA's signal observed by the array contains receive steering vector,conventional adaptive beamforming that performs on angle dimension can be utilized to suppress interferences which can be separated from the real target in angle.In addition,solutions to non-ideal scenarios like array calibration error,snapshot deficiency and mainlobe interference are given.On the other hand,in the scenario of a FDA-MIMO system,deceptive jamming that can be identified from range dimension are suppressed using range-angle adaptive beamforming.The per-formance of the range-angle beamformer is measured by the distance between the real target and false target.In the case where the false target is located in the mainlobe of FDA-MIMO's beampattern,the mainlobe interference suppression algorithm based on blocking matrix is proposed to cancel the mainlobe jamming successfully.For snapshot deficiency case where signal is present in the training data,the resulted adaptive beampattern can barely focus on the target area and suppress jamming.By reconstruction of interference-plus-noise covariance matrix,the output signal to interference-plus-noise ratio(SINR)of the beamformer can quickly converge to the optimal value.(3)Parameter estimation of FDA-MIMO in the interference-plus-noise environment is discussed.Conventional maximum likelihood(ML)estimator for joint range,angle and Doppler is of quite high complexity and sensitive to array errors.To solve this problem,the original data model is changed into the unstructured model then ML is performed resulting in the low-complexity unstructured maximum likelihood(UML)estimator.During UML estimation,the covariance of array error and interference is compensated into the covariance matrix to make UML estimator robust to array error and interference.Along with the computational efficiency brought by UML,performance loss is also introduced and thus UML range and angle estimators are no longer consistent.However,the numerical analysis verifies that UML estimation could greatly reduce the complexity brought by conventional ML method while suffers little performance loss.