The quaternion multiple signal classification(Q-MUSIC)algorithm generally requires four-dimensional spectral peak search to estimate the direction of arrival(DOA)and polarization parameters,which would result in the huge computation burden.A dimension reduction Q-MUSIC algorithm(DRQ-MUSIC)based on L-shaped array is presented to reduce the computational complexity in this paper.The proposed algorithm divides the steering vector into three parts,and estimates each part separately,thus DOA and polarization parameters can be estimated only by N times one-dimensional spectral peak search,where N denotes the sources number.Besides,pair match is not required.Finally,simulation examples demonstrate the effectiveness and feasibility of the proposed algorithm. The quaternion multiple signal classification (Q-MUSIC) algorithm requires 4-dimensional spectral peak search to estimate the direction of arrival (DOA) and polarization parameters, which would result in the huge computation burden. A dimension reduction Q-MUSIC algorithm -MUSIC) based on L-shaped array is presented to reduce the computational complexity in this paper.The proposed algorithm divides the steering vector into three parts, and estimates each part separately, thus DOA and polarization parameters can be estimated only by N times one -dimensional spectral peak search, where N denotes the sources number.Besides, pair match is not required. Finally, simulation examples demonstrate the effectiveness and feasibility of the proposed algorithm.