The ground penetrating radar( GPR) detection data is a wide band signal,always disturbed by some noise,such as ambient random noise and multiple reflection waves. The noise affects the target identification of underground medium seriously. A method based on principal component analysis( PCA) was proposed to extract the target signal and remove the uncorrelated noise. According to the correlation of signal,the authors get the eigenvalues and corresponding eigenvectors by decomposing the covariance matrix of GPR data and make linear transformation for the GPR data to get the principal components( PCs). The lower-order PCs stand for the strong correlated target signals of the raw data,and the higher-order ones present the uncorrelated noise.Thus the authors can extract the target signal and filter uncorrelated noise effectively by the PCA. This method was demonstrated on real ultra-wideband through-wall radar data and simulated GPR data. Both of the results show that the PCA method can effectively extract the GPR target signal and remove the uncorrelated noise. The ground penetrating radar (GPR) detection data is a wide band signal, always disturbed by some noise, such as ambient random noise and multiple reflection waves. PCA) was proposed to extract the target signal and remove the uncorrelated noise. According to the correlation of signal, the authors get the eigenvalues ​​and corresponding eigenvectors by decomposing the covariance matrix of GPR data and make linear transformation for the GPR data to get the principal components (PCs). The lower-order PCs stand for the strong correlated target signals of the raw data, and the higher-order ones present the uncorrelated noise .hus the authors can extract the target signal and filter uncorrelated noise effectively by the PCA. This method was demonstrated on real ultra-wideband through-wall radar data and simulated GPR data. Both of the results show that the PCA method can ef fectively extract the GPR target signal and remove the uncorrelated noise.