Partial Knowledge Aided Beamforming And Interference Suppression

Space-time interference suppresion is an important branch of the array signal processing.To deal with specific application scenarios,it requires targeted strategies and schemes to suppress the interference.In some scenarios,such as large-scale arrays and high resolution radar,huge computational complexity always exists.Meanwhile,the output Signal-to-Interference-plus-noise ratio(SINR)may suffer performance degradation due to many non-ideal factors.The system robustness can be effectively improved via introducing the prior knowledge.In this work,we deeply study the influence of the signal and the system's parameters on the performance of the Generalized sidelobe canceller(GSC),the robust algorithm for interference suppression under the Direction of Arrival(DOA)mismatch of the desired signal,wideband beampattern synthesis algorithm with low computational complexity,and the wideband multi-beamforming method.The contributions are mainly as follows:(1)Study of the influences of the signal and the system's parameters on the performance of the GSCThis work presents the performance analysis of the GSC in finite sample size.The auxiliary array perturbation weight model for the finite sample data is proposed.By researching on the performance of the GSC based on this model,closed-form expressions of the interference cancellation ratio and the null level in the direction of the interference are proposed.Meanwhile we analyse the anti-interference performance impacts of the signal and the system's parameters further,including the number of the snapshots,the input SNR,the correlation coefficient between the desired signal and the interference and the element number of the auxiliary array.(2)Robust GSC design against the DOA mismatch of the desired signalOne steering vector estimation model via the quadratic constraint is established.In this model,the steering vector is estimated by maximizing the residue output power of the GSC.Meanwhile,one iterative estimation scheme of the main array weights and the steering vector the auxiliary array is proposed in order to improve the output SINR further.On this basis,a novel subspace rotating algorithm which has lower computational complexity and closed-form solution is proposed with the knowledge of the subspace dimension.Additionally,we consider the online implementation of the subspace rotating method to meet the realization requirement.(3)Partial knowledge aided robust wideband beamformingOne novel robust wideband beamforming algorithm based on the steering vector estimation is proposed.Firstly,we improve the tradional Linearly Constrained Minimum Variance(LCMV)beamformer.By introducing the frequency invariant constraint.Consequently,the wideband beam steering constraint is cast as the constraint on the array response at the reference frequency.On this basis,the space-time steering vector is estimated utilizing the algorithms based on the quadratic constraint method and the subspace rotation method as the result of the output SINR improvement.Additionally,we propose the online implementation of the subspace rotating algorithm in order to meet the realization need.(4)Research on the wideband beampattern synthesisThis work presents the analysis of the wideband beampattern synthesis.We firstly compare the performance of several classical wideband beampattern synthesis algorithm in terms of computational complexity and beampattern performance.In this basis,we propose a wideband beampattern synthesis method for the centrosymmetric array.By introducing the conjugate symmetry constraint,the computational compexity of the weights optimization is reduced.Whereafter,the transformation relationship of the weights between the space frequency structure and the space time structure is analysed.On this basis,we propose one efficient wideband beampattern synthesis method.Firstly the subband weights are optimized.Then,The weights of the space-time structure are obtained utilizing this transformation relationship.(5)Research on the wideband simultaneous multi-beamforming based on the Farrow structureIn this work,one low sidelobe level wideband multi-beamfoming algorithm is proposed for the centrosymmetric array by introducing the conjugate symmetric constraint.Whereafter,we present sparse optimization scheme of the weights for the Farrow structure.Finally,one sparse optimization algorithm with priority of the time-domain weights is proposed in order to meet some realization needs.