Abstract: A novel interference suppression algorithm for Global Navigation Sattellite System(GNSS) signal is proposed in this paper. The spreading codes in GNSS signal exhibit strong selfcoherence and the correlation matrix of the GNSS signal can be estimated using several peaks of correlation values of received signal and reference signal. The optimum weight is obtained under the maximum Signal to interference plus noise ratio(SINR) criterion using the covariance matrix of precorrelation and postcorrelation signal. No priori knowledge of the directions of GNSS signals is needed for the algorithm. Simulations and experiments demonstrate that, compared with the power inversion method and spectral selfcoherent restoral SCORE algorithm, the proposed algorithm can provide a similar performance of inference suppression, but the output SINR is significantly increased.