应用半正定规划的目标方位超分辨方法

针对水下目标方位超分辨估计问题,提出了一种基于半正定规划(Sdp)的常规波束(CBF)方位超分辨算法(SdpCBF).Sdp-CBF算法基于常规波束形成获得多目标方位谱数据,利用阵列响应矩阵和半正定规划技术,精确估计目标数量和波达角方向.该算法的本质是利用阵列特性和信号能量信息获得超分辨方位估计,不用进行子空间分解,通过卷积反演的方式将阵列孔径的有限效应消除,在L₂范数约束条件下重构空间谱.仿真表明,Sdp-CBF算法具有较强的噪声抑制能力,对非相干和相干信号均具有目标方位超分辨能力,在低信噪比环境下的方位分辨性能超过多重信号分类(MUSIC)等经典高分辨算法。对消声水池以及湖上实验数据的处理结果显示,Sdp-CBF算法在复杂环境中对相干信号及微弱信号具有较强的分辨能力。 

Super-resolution azimuth estimation using conventional beamforming and semi-positive programming techniques

The direction-Of-Arrival(DOA) estimation algorithm named Sdp-CBF combining conventional beamforming and semidefinite programming is proposed in this paper,which can be used for the azimuth super-resolution estimation of underwater targets.The Sdp-CBF algorithm obtains multi-target azimuth spectrum data based on conventional beamforming,and uses the array response matrix and semi-definite programming techniques to accurately estimate the target number and direction of arrival.By calculating the conventional beampattern matrix and azimuth spectrum of a given array,Sdp-CBF azimuth spectrum is obtained by quadratic programming.The algorithm does not require subspace decomposition and can be applied to any noise signal model.Sdp-CBF algorithm is suitable for arbitrary noise model,and super-resolution bearing estimation is obtained based on array characteristics and signal energy information.The simulation results show that Sdp-CBF algorithm also has strong noise suppression ability and can effectively reduce the background level of azimuth spectrum,the azimuth resolution performance in low SNR environment is better than that of the classical high-resolution algorithms such as MUSIC.The results of the anechoic pool test data and lake test data show that the algorithm has a strong resolution for weak signals in complex environments.