Research On Jamming Suppression And Waveform Design In Cognitive Radars

With the rapid development of digital radio frequency memory,the modern jammers always can precisely interrupt,copy and reproduce the radar signal.The generated deceptive false targets against radar systems are prone to obtain larger coherent processing power at the radar receive ending.Meanwhile,the characteristics of these false targets in the time domain,airspace and frequency domain are very similar to the ones of true targets.The false targets would severely affect the true target detection,tracking and parameter estimation performance.When the radar is jammed,the true targets are apt to lose,the radar resource is severely consumed and the unusual warning is prone to be triggered.Employing the cognitive radar technology,the radar system would have the ability to deeply cognize the environment,intelligently utilize the environment information and adaptively optimize the transmitting(including the waveforms,polarization,frequency and so on).Then,the anti-jamming performance of radar system can be further improved in the complicated electromagnetic environment.For the anti-jamming of cognitive radar,this dissertation focuses on the jamming suppression methods and waveform design.Some works have been carried out and the proposed algorithms and schemes are beneficial to enhance the information interaction between the radar system and the electromagnetic environment,and improve the target detection performance of cognitive radar system.In summary,the main contents and some novelties of this thesis are listed as follows(1)Employing the polarization charactristics,a spatial-polarizational oblique projection filter operator is proposed to suppress the deceptive jamming for Pulse-Doppler(PD)radars.It is noted that the spatial angles of true targets and deceptive false targets can be different or same,however the polarizational characteristics are different.Based on the polarization sensitive array,the trilinear alternate least squares algorithm is first used to estimate the 2-dimensional(2D)spatial angles and polarizational characteristic parameters.Then a spatial-polarizational oblique projection filter operator is constructed based on the joint estimation of the above parameters.By employing the filter operator,the mixed return of radar is projected onto the target echo sublinear space to further reserve the target echo and suppress the deceptive jamming.Compared to the conventional polarizational filter,our spatial-polarizational oblique projection filter operator not only can suppress the sidelobe jamming,but also can suppress the mainlobe jamming.The robustness of the filter operator is also better.Moreover the filter operator is effective for the different radar waveforms with different pulse width and duty ratio.(2)For the suppression of wideband coherent interrupted sampling repeater jamming(ISRJ),a pass-band filter is designed in the thesis.Based on the time-frequency(TF)characteristic differences between target echo and ISRJ returns,the echo during the period,when the jammer receives and samples the radar signal,is firstly extracted.Then an efficient band-pass filter and a feasible jamming suppression scheme are proposed,respectively.The high resolution rang profile of the received echo followed with pulse suppression would be similar to that of target,on basis of large correlation coefficient and signal-to jamming-plus-noise ratio(SJNR)improvement factor.Four simulation cases with different order,delayed time,signal-to-noise ratio(SNR)and jamming duty ratio demonstrate the effectiveness of our proposed pass-band filter against the wideband ISRJ.Simulation results show that when the output SNR>25d B in De-Chirp ISAR and the ISRJ order is different,the correlation coefficient of the filter output echo is all larger than 0.8,and SJNR improvement factor is all larger than 15 d B.(3)Three design schemes of spatial filter measurement matrix(SFMM)are discussed for the colocated multiple-input multiple-output(MIMO)radar to jointly suppress the jamming in the target sector-of-interest(SOI)and the sidelobe interferences/clutter,when the compressive sensing(CS)-based method is employed for the target direction-of-arrive(DOA)estimation.It is shown that these schemes not only can suppress the above jamming and interferences,but also can precisely estimate the target DOA,even the DOAs of distributed jammers are same to the ones of true targets.Besides,these schemes are more general and also effective to suppress the sidelobe interferences/clutter.Simulation results show that the third designed SFMM can obtain larger than85 d B null attenuation gains for jamming,and larger than 30 d B sidelobe attenuation gains for interferences.When SNR>5d B,the mean square root error of target DOA estimation is smaller than1°.(4)A new cognitive radar waveform design strategy is outlined based on the adaptive initial phases to counter the velocity deception jamming for PD radars.The design kernel of adaptive initial phases is to adapt the Doppler spectrum of radar echo and suppress the jamming in some stopband notches,where the true targets exist.Meanwhile,a multi-channel processing trick with different integral multiple pulse repetition interval(PRI)delay is employed to estimate the parameters of true targets and false targets,especially the lagged PRI number of repeat-back jammer.The algorithm is computationally efficient and would make the waveform real-time updating possible,which benefits from that the main steps are based on fast Fourier transformation(FFT).Simulation results show that our waveform design scheme can generate frequency notches at the target frequencies,and further improve the SJNR of constant false alarm rate detection(CFAR).Meanwhile,it is also beneficial for the weak target detection.(5)Focusing on the partially correlated waveform design of colocated MIMO radar,two waveform design schemes and a kind of transmit waveform covariance matrix(TWCM)with the form of symmetrical Toeplitz matrix are outlined,based on the design constraints of transmit beampattern synthesis and the SINR maximum of output signal at the receive ending,respectively.In this thesis,the TWCM optimization model is firstly given based on the ideal transmit beampattern,and then is relaxed to a convex optimization question to solve.To overcome the disadvantages of traditional TWCM iterative optimization algorithms,such as being subject to terminate on a local optimum and being sensitive to the initial value,we transform the TWCM to the square root matrix for the indirect construction,and transfer the TWCM non-convex optimization model to an objective optimization function of the square root matrix.Then the particle swarm optimization(PSO)algorithm is employed to code and solve the optimal square root matrix.Simulation results show that there is more than 3d B improvement in comparison to the TWCM iterative optimization algorithm.Besides,a kind of TWCM with the form of symmetrical Toeplitz matrix is proposed and the full rank characteristic is proved mathematically,which guarantees that the waveform diversity advantage of MIMO radar can be sufficiently exploited and the maximum number of interfering sources can be further suppressed.Moreover,we demonstrate that the proposed TWCM can be synthesized with binary phase shift keying(BPSK)waveforms in closed form.The relationship between the TWCM control parameter and the maximum SINR of the output signal is also given.The cognitive colocated MIMO radar can adopt a proper TWCM to design and update the transmit waveform online based on the environment cognition capability,such as the prior knowledge of DOA and echo power estimation of the targets,interfering sources and clutter.Therefore,it can be seen that the proposed waveform design scheme has more flexibility and applicability to counter the sidelobe interferences.