# All matrix from the original matrix, a window function

All the available methods for DOA estimation, are based on the eigen decomposition of the sample covariance matrix, which is calculated over the entire sample data set. However, the received data can be used to provide more information.Researchers and Scholars have been using the sensor arrays for obtaining sample covariance matrix as a whole and then proceeding further based on this matrix. Here, we will make full use of the received data, achieving more accurate estimation. Jackknifing is an effective strategy used in the statistical area to estimate sample statistics 29. The approach is to use subsets of available data covariance matrix to improve the estimation performance. We propose array signal processing method based on the concept of jackknifing. Our primary assumption is that a large proportion of the available data contains roughly the same amount of information as the whole available data set does.The data covariance matrix is obtained /generated by implementing a large number of snapshots for increasing the resolution of the output. Thus, a  MxM sample covariance matrix is obtained. For obtaining resampled covariance matrix from the original matrix, a window function is devised. This resampled matrix should be a symmetric matrix to obtain real eigen values after performing eigen decomposition of the same. Thus, the window function has to be a square matrix and should vary along the diagonal of the original covariance matrix for resampling, to be symmetric in nature. It has been proposed by … that, for N number of sources, minimum N+1 sensor arrays are necessary for approximate estimation. Thus, the window at least should be of size (N+1)x(N+1). For improved efficiency, it is suggested that square matrix size be more than N+1.This is repeated until the entire original sample covariance matrix is utilized and accordingly, we will have the number of Jackknifing resamples.The newly obtained matrices are then decomposed for obtaining eigenvalues and eigenvectors. After decomposition, MUSIC algorithm is implemented as proposed by ….. The algorithm for above procedure is shown in fig.1.Then, obtain the estimated value of DOA by selecting the searched peaks which are repeated a maximum number of times.