Abstract: In shallow water environment, weak targets are always masked in the sidelobes of strong interference, thus the targets can not be detected. Studying how to suppress the strong interference and increase the output signal-to-noise ratio (SNR) is especially significant for improving the detection performance of sonar. In this paper, it is assumed that the interference energy is far greater than that of the target. Firstly, the eigen-decomposition of received data covariance matrix is made, and the eigenvector, which corresponds to the largest eigenvalue, is considered as the interference's ei-genvector. Then the array received data are projected to the orthogonal subspace of interference subspace by using the orthogonal projection method. The data of interference is removed so that the purpose of suppressing strong interference can be achieved. Numerical simulation and experimental results show that this method can effectively suppress the strong interference, increase the output SNR, and enhance the reliability of directions of arrival (DOA) estimates of the targets. The performance of passive ranging will be possibly improved in this condition.