Research On The Robust Beamforming Methods Of Adaptive Antenna Array Under Non-ideal Conditions

Due to the rapid development of wireless communication technology,adaptive antenna array has been widely used in communication.As the core technology of adaptive antenna array,adaptive beamforming technology can achieve good spatial filtering performance under ideal conditions by aligning the mainlobe with the desired signal direction and forming a narrow null at the sidelobe interference position.In practical application,the environment of adaptive antenna array is complex,which will be affected by various non-ideal conditions,resulting in the sharp decline of its output performance.Therefore,in view of the above situation,considering all kinds of errors and interference faced by adaptive antenna array in practical application,this thesis studies the non-ideal conditions such as desired signal steering vector mismatch,mainlobe interference,sidelobe motion interference and array calibrated errors respectively,so as to improve the output performance of adaptive beamforming method in a single non-ideal condition;Then a robust adaptive beamforming(RAB)method for a variety of errors and special interference is further proposed to improve the comprehensive robustness of the adaptive antenna array.The main research contents of this thesis are summarized as follows:1.Aiming at the mismatch of desired signal steering vector,this thesis proposes a RAB based on automatic variable loading firstly.The core idea of the method is to construct an automatic variable loading matrix that can change with the input SNR,so as to improve the output performance in the whole SNR environment.Because the covariance matrix contains the desired signal,the above method is still sensitive to the desired signal steering vector.Although the output performance is significantly better than that of similar methods,there is still a gap between the theoretical optimal value of the output SINR distance in the high input SNR environment.For this problem,this thesis proposes a robust adaptive beamforming method of INCM based on interference power simple estimation.This method simply estimates the power of the interference source through the eigenvalue of the covariance matrix and matches the interference source from all angles with the power value,so as to construct the INCM.Compared with the existing related methods,it has excellent performance for the mismatch of desired signal steering vector caused by various non-ideal conditions,and the running time of the method is ideal.2.Aiming at the problem of special interference,this thesis mainly studies the suppression methods of mainlobe interference and sidelobe motion interference.For the former,a robust adaptive beamforming method against mainlobe interference based on eigenprojection and covariance matrix reconstruction is proposed.The method uses the mainlobe area except the desired signal to construct the mainlobe interference suppression matrix.The matrix is used to project the received signal and optimizes the sidelobe INCM,so that the adaptive antenna array can suppress the mainlobe interference without offset of the sidelobe null.Compared with the existing methods,because the conventional sampling received signal is used to solve the weighted vector,the method is more suitable for practical application scenarios and has good output performance.For the latter,this thesis proposes two kinds of null broadening methods.First,by adding virtual interference sources near the interference,the INCM containing virtual interference sources is constructed to achieve the purpose of null width broadening;Secondly,the correlation matrix of interference azimuth broadening information is constructed,and the INCM containing null broadening information is constructed by using projection and diagonal loading technology.Both methods can suppress the sidelobe motion interference.Compared with the existing methods,the proposed methods are relatively robust to the desired signal steering vector mismatch and have better output performance.3.Aiming at the problem of array calibrated error,this thesis studies the robustness improvement of adaptive antenna array beamforming method in the case of mutual coupling and amplitude and gain-phase errors respectively.Firstly,for the problem of mutual coupling,a robust adaptive beamforming method based on unknown mutual coupling suppression is proposed.This method uses the characteristics of mutual coupling between uniform linear array elements to construct the transformation matrix,and constructs INCM to suppress mutual coupling by optimizing the eigenvector of covariance matrix.Compared with other similar methods,it has good output SINR in the case of mutual coupling and desired signal steering vector mismatch.The computational complexity can be reduced by expanding the spatial sampling angle interval,which is suitable for the real-time processing of spatial signal sources by adaptive antenna array.Then,for the problem of gain-phase errors,a robust adaptive beamforming based on partially calibrated uniform linear array is proposed.Based on the array of partly calibrated elements,the method selects the signal source with high power value for gain-phase errors estimation,and uses the estimation results to optimize the INCM,so as to improve the robustness of the adaptive antenna array to gain-phase errors.Compared with other methods,this method estimates the gain-phase errors more accurately and has higher output SINR in the environment of gain-phase errors and the desired signal steering vector mismatch.4.For the comprehensive error in non-ideal conditions,the RAB against the comprehensive error based on partially corrected array is proposed in this thesis.Firstly,PCARAB method deduces the array manifold error model,and converts the steering vector affected by the mismatch into the form of the product of conversion matrix and error matrix.Then,the transformation matrix is constructed through the array of partly calibrated elements to optimize the eigenvalues and eigenvectors corresponding to the interference,so as to construct the INCM.The method inherits the advantages of the two methods proposed for the array calibrated errors.Although its output performance is slightly worse than the high-performance method suitable for this situation in the case of some individual errors,it has good robustness for all kinds of errors and the comprehensive performance is the best.In addition,the method can be combined with the above anti-special interference method to suppress the special interference under the comprehensive error.The simulation results show its effectiveness.