The direction-of-arrival (DOA) estimation method for noncircular sources with maximum noncircularity rate in multiple subarrays is proposed when there exists mutual coupling between sensors of each subarray.Based on the banded and symmetric Toeplitz character of the coupling matrix of uniform linear arrays (ULAs) and the block diagonal character of the coupling matrix of multiple subarrays

the proposed method can avoid multidimensional search and iterative computation like the conventional rank reduction estimator (RARE).We extend the data model by using the noncircular feature of the sources

and thus the proposed method outperforms the conventional RARE in terms of estimation accuracy and the number of sources that can be distinguished.The performance study provides a necessary condition for unique estimation

and the expression of noncircular Gaussian Cramr-Rao bound (CRB) matrix for DOAs is presented.The simulation results illustrate that the proposed algorithm is more robust than the conventional RARE with respect to lower signal-to-noise ratio and fewer number of samplings.