| Adaptive antenna arrays have been extensively researched and applied in a number of fields due to the rapid growth of wireless communication technology and antenna technology.Array antennas are dependent upon adaptive beamforming technology to ensure accurate signal reception and to resist interference in real-time.However,in practical applications,adaptive beamforming methods under the traditional array configuration are not able to meet some higher performance requirements like high-precision beam pointing and multi-interference suppression.Recent research has introduced a new way to configure sparse arrays called the coprime array.Its unique sparse arrangement of array elements significantly improves the array’s aperture dramatically,enabling narrower beams,increased directivity,higher resolution,and more concentrated gains.However,the traditional adaptive beamforming method can’t reflect its performance advantage directly on the coprime array.Further research is also needed on how to ensure the robustness of adaptive beamforming in non-ideal environments,as the application scenarios for array antennas are complex and changeable.This paper discusses and studies the adaptive beamforming method for coprime arrays.Utilizing the advantages and characteristics of coprime arrays,the ability of array antennas to withstand multiple interferences simultaneously as well as their robustness in a non-ideal environment will be further enhanced.This paper contains the following main research contents:1.In view of the problems that conventional beamforming methods can’t directly match dimensions with virtual arrays extended by coprime arrays and the computational complexity is high,firstly,an adaptive beamforming method of coprime arrays based on compressed sensing is proposed.By using the sparseness of signals in the spatial domain and the sparseness of the structure of coprime arrays,the optimization problem with the smallest fitting error is designed by sparse reconstruction.It is possible to obtain the exact incoming direction of the received signal.Then,a fast interference power estimation method is adopted to obtain the estimated interference power.Combined with the known physical structure of the coprime array,the interference plus noise covariance matrix whose dimensions match the number of physical elements of the coprime array is reconstructed.Then,the steering vector of the desired signal is estimated and corrected,and finally,the beamforming weight vector of the physical elements of the coprime array is obtained.Compared with other existing methods,the proposed method has better beamforming output performance and lower computational complexity.Especially,when there are multiple interference signals or the interference sources are close,the output performance advantage of the proposed method is obvious compared with other methods.2.As a means of improving the comprehensive robustness of beamformers against signaloriented vector mismatches and array-popular errors,a robust adaptive beamforming method of coprime array based on covariance matrix reconstruction is proposed.According to the structural characteristics of coprime array,the received data sampling covariance matrix of coprime array is transformed into a single fast-row received data vector by vectorization,and the vector is de-redundant according to the array element position of coprime array differential optimization array.Then,by means of topelitz matrix reconstruction,an adaptive beamforming method is constructed.A more accurate signal direction can be obtained by estimating the signal parameters through the covariance matrix with increased dimensions.Combined with the position information of the physical elements of the coprime matrix,the steering vector of the signal is constructed.Furthermore,considering some mismatches that may occur in practical engineering applications,the estimated signal steering vector is corrected by uncertainty set constraint,and the power value of the interference signal is estimated accurately according to the orthogonality of steering vectors of incident signals from different directions.To obtain the accurate desired signal steering vector,the desired signal correlation matrix must be constructed in the approximate angle range of the desired signal after residual noise is removed.Finally,the beamforming weight vector matched with the physical unit of the coprime matrix is calculated.By comparing the proposed method to the existing methods,it is demonstrated that it is able to produce high-quality beamforming results,and in a variety of mismatched environments,such as DOA estimation error,array position error,and array popular vector error.Especially,the proposed method has obvious advantages in output performance when multiple interference signals and array popular errors occur simultaneously.The proposed method is clearly superior in terms of output performance,as well as being robust. |
