Study On Joint Processing Of Transmitter And Receiver In Cognitive Radar System

Due to the growing complexity of electromagnetic environment,it becomes increasingly difficult to achieve the desired radar performance simply by using the adaptive processing algorithms in the receiver.To improve the flexibility and adaptability of radar system,the exploitation of the degrees-of-freedom(DOFs)in the transmitter has gradually received intensive attention from the radar scholars.With the concepts of multiple-input multiple-output(MIMO)radar and cognitive radar,the knowledge-based adaptive processing method has been regarded as an important approach to improve the radar performance in complicated environments.So far,numerous efforts have been devoted to verify the advantages of this technology.However,there are some problems to be further studied,such as the establishment of model,the acquirement of information,and the design of waveform,etc.Based on the requirements of radar in different applications,this dissertation aims at the joint design of the transmitter and receiver for cognitive radar systems.The main contributions are summarized as follows:1.The waveform design problem based on the prior knowledge of target and environment is studied.To satisfy the requirement of the multifunction radar systems,a design algorithm based on two criteria is proposed to synthesize the phase-modulated waveform.The energy spectral density(ESD)of ideal waveform is presented with the criteria of maximizing the output signal to clutter plus noise ratio(SCNR)and mutual information(MI)between the received signal and the target response,respectively.Then a multi-objective function is constructed to approximate the optimized ESD waveform to the ideal ESD waveform as much as possible.To solve it,a cyclic iterative algorithm is proposed to obtain the time-domain optimized waveform.Simulation results show that the synthesized waveform can increase the SCNR and MI simultaneously.2.To improve the amplitude estimation accuracy of fluctuant targets in unknown clutter,a new method with the joint transmit-receiver in cognitive radar is proposed based on the minimum mean square error(MMSE)criterion.First the received signal model is constructed in the multiple-pulse mode.Then by analyzing the problems of existing methods,the proposed method is presented in three stages.Firstly,radar transmits a group of probing waveform to estimate the amplitudes of the out-of-window scatterers,based on which the waveform for the next transmission is designed to suppress the sidelobes of large out-of-window scatterers.Finally,adaptive filtering is realized with the statistical amplitude estimation of scatterers.Simulation results show that the proposed method can obtain better estimation performance than the adaptive filtering methods processed only in the receiver.3.A new method with the joint transmit-receiver in cognitive radar is proposed for the detection of unknown fluctuant targets based on the high-resolution range profile(HRRP).It is well known that the radar detection performance can be improved by transmitting the waveform matched with the HRRP of target.However,the optimal waveform cannot be obtained directly with an unknown target.To provide the updated waveform and target HRRP,the detection mode of estimation-while-optimization is required.Furthermore,there exists serious degradation of the autocorrelation property in the matched waveform,which will decrease/damage the estimation accuracy of the HRRP.To overcome this,a generalized model of HRRP that consists of the deterministic term and random term is introduced under the consideration of variations in the target response.Based on the derivation of the generalized likelihood ratio test(GLRT)detector and the objective function for waveform design,a new algorithm is proposed with the criteria of maximizing the signal to noise ratio(SNR)and MMSE.Simulation results show that the proposed method can make a tradeoff between the estimation error and SNR,and steadily improve the estimation and detection performance for the unknown target.4.To suppress the range-ambiguous clutter in forward-looking array radar,a new method with the joint transmit-receiver is proposed by using the frequency diverse array(FDA)technology.Based on the analysis of the characteristics of FDA beampattern and clutter space-time spectrum,the design principle of the frequency increment in transmitter is established to distinguish the range-ambiguous clutter and avoid signal decorrelation.Then the subspace projection algorithm is used in receiver to suppress the range-ambiguous clutter.Due to the range dependent property in FDA,the proposed method can suppress the clutter effectively,independent of the DOFs in vertical dimension.Simulation results show that the proposed method performs better than the conventional elevation filtering method with smaller elevation DOFs.5.To improve the suppression performance of the deceptive jamming in forward-looking array radar,a new method with the joint transmit-receiver is proposed by using the vertical frequency diverse array(VFDA)technology.First,a beamforming method based on the array configuration is presented to solve the problems of the conventional FDA beampattern,i.e.the coupling between the azimuth angle and range,and the grating lobes.In the VFDA,the frequency increment is applied across the vertical array elements,which avoids the coupling between the azimuth angle and range.Then the grating lobes are suppressed with the parameter design of array element.Furthermore,due to the vertical velocity component of the diving movement,the phase information of azimuth angle and range is introduced into the vertical steering vector frequency increment.Using this property,an adaptive elevation filter is proposed to suppress the deceptive jamming.Simulation results show that the proposed method can provide better suppression performance and lower computational complexity compared with the method based on the conventional FDA technology in the dense deceptive jamming circumstance.