Abstract: This paper first analyses the proper ties of impulsive noise and its affecting on the traditional subspace techniques employing second-or der or higher-or der statistics. Then we present a new subspace-based direction finding algor ithm in the presence of impulsive noise. According to the basic idea behind TLS-ESPRIT, the method estimatest he bear ing information of the incoming signals by performing singular value decomposition ( SVD) on the covariationcoefficient matrices of array outputs. Finally, the simulation results show the feasibility and effectiveness of the algorithm.

Key words: direction of arrival, subspace, lower order moment, TLS ESPRIT, impulsive n oise