Abstract: In this paper, a new method of DOA estimation and sensor array error calibration based on FastICA is proposed to address the issue of performance deterioration in traditional high-resolution DOA estimation algorithms caused by sensor array error. Our method utilizes the characteristics of the diagonal elements of the covariance matrix to estimate the gain error. The mixing matrix, composed of the array manifold matrix and array error matrix, is extracted from the snapshot samples using the FastICA algorithm. The mixing matrix is then projected onto the signal subspace to eliminate unwanted signals and obtain the true steering vector. The algorithm utilizes the true steering vector to construct an oblique projection operator to acquire the single-component received data. The DOA estimation is obtained through subspace orthogonality spectral search, and the phase error is estimated based on it. This method completes array error calibration without setting calibration sources, and the DOA estimation performance is independent of the array phase error. Moreover, the proposed method avoids the issue of suboptimal convergence that occurs in the WF method. Numerical simulations validate the effectiveness of the proposed method.