Abstract: This paper proposes a Modified Reduced-Dimensional MUSIC (MRD-MUSIC) Direction Of Arrival (DOA) estimation algorithm to solve the computation-intensive problem in the two-dimensional DOA estimations. The MRD-MUSIC algorithm can distinctly reduce the computational load, since the two-dimensional DOA estimation can be decomposed into two stages of one-dimensional DOA estimations by the quadratic optimization method. In this algorithm, a one-dimensional DOA is obtained by the direct derivation of the quadratic optimization function, and the steering vectors are strongly constrained in the solving process, as a result, the derivation is closer to the optimal solution than the others. The simulation results verify that the theoretical derivations of this algorithm is of correctness, it does not need pair matching processes, and it has higher successful rate and precision of the angle estimation.