Abstract: Based on uniform circular arrays (UCA), a direction of arrival (DOA) estimation algorithm of coherent signals (DD-EVD) with single snapshot data was proposed. The algorithm used mode excitation to transform the manifold matrix of UCA into uniform linear array. Then, a Toeplitz matrix which was reconstructed with the data vectors of beam space was decomposed in accordance with array manifold. It was proved that the rank of this Toeplitz matrix was regained and determined merely by the number of incident sources. Accurate signal subspace and noise subspace can be acquired by performing eigenvalue decomposition of the matrix. Combined with the subspace kind algorithms, the DOA of coherent signals was estimated. Because of using single snapshot data to reconstruct matrix, the algorithm does not need correlation calculation and cumulation of snapshots. In view of this, the algorithm is intended to estimate the DOA of non-stationary signals and the complexity of the algorithm is reduced. Finally, the simulation verifies the effectiveness.