Estimation of mobile station location is made difficult by nonsymmetric contamination of measured data caused by NLOS propagation effects. In this paper, a novel algorithm based on kriging method is proposed to reduce the effects of NLOS propagation on the location error. Simulation results show that the proposed algorithm clearly outperforms two other algorithms (non-parametric kernel method and parametric least square method) and comes close to meeting Cramer Rao lower bound (CRLB).