Researches Of Anti-interference Algorithms Based On Adaptive Arrays

Over the past several decades,with the urgent demand of infrastructure construction,arrays based anti-interference techniques are being bettered.However,with the improvement of levels of modern science and technique,the increasing requirement of anti-interference performance is necessary and some new anti-interference scenarios emerge.For example,in the scenario of high-dynamic navigation interference suppression,conventional methods cannot perfectly suppress the interferences with direction perturbations.Moreover,conventional distributed beamforming methods present low real-time property and high costs.We conclude the conventional anti-interference algorithms based on the polarized space-time signal model.We give corresponding(un)supervised batch and adaptive algorithms,respectively.We compare these algorithms through several illustrative simulations based on the navigation signals.We propose a null broadening algorithm for the dual-polarized sensitive arrays in space-time domain.We first derive the covariance matrix taper based on the prior estimated perturbation parameters,then estimate the covariance matrix of the received signal in high-dynamic scenario by incorporating that in low-dynamic scenario with the derived covariance matrix taper,and ultimately solve the weight vector by applying the power inversion criterion.The proposed null broadening algorithm could improve the accuracy of positioning by mitigating the severe performance deterioration of conventional anti-jamming algorithms due to the null mismatch caused by the high relative velocity between the receiver and interferences.We propose a tensor based distributed collaborative beamforming algorithm.We extrapolate the tensor model of a stand-alone array to that of the array network with massive arrays,considering both scalar and vector-sensor arrays in a unified model.Then,incorporating conventional diffusion strategies,we propose a tensor based distributed collaborative beamforming algorithm,suitable for both scalar arrays and polarizationsensitive arrays.We also analyze its mean stability and give a sufficient condition for its convergence.The proposed distributed collaborative beamforming algorithm presents low computational complexity,accelerated convergence rate and low communication costs as compared to the conventional beam coordination algorithm.We propose a diffusion entropy least-mean-square(DE-LMS)algorithm based on Gaussion entropy minimization for the distributed interference suppression scenarios in terms of non-circular signals,with superior performance as compared to the conventional diffusion least-mean-square algorithm(D-LMS).We further analyze the mean and meansquare stability of the DE-LMS algorithm incorporated with an asymptotically unbiased circularity coefficient estimation method.By deriving the bounds of the performance of DE-LMS algorithm,we further analyze the relation between the performance and the related parameters.In this article,all the proposed algorithms have been validated by illustrative simulations.The results show that the proposed null broadening algorithm could efficiently suppress interferences with both direction perturbations and polarization information perturbations.Moreover,the proposed tensor based collaborative beamforming algorithm could be realized with notably reduced computational complexity,markedly accelerated convergence rate and enhanced communication efficiency.Finally,the proposed diffusion entropy LMS algorithm take into account the non-circular property of complex-valued measurement noises,and thus outperforms conventional distributed beamforming algorithms.