In order to solve the problem that some satellites are not sensitive in the fault detection algorithm, a Slope-Weighted Least Square (SWLS) algorithm is proposed.This method theoretically derives the expression of the satellite feature slope, constructs the slope-weighted matrix according to the size of the feature slope, and gives greater weight to the satellite with smaller feature slope.The slope-weighted test statistics are constructed, and the satellite fault detection is realized by comparing the new test statistics with the detection threshold.This method is verified by using real data, and the results show that the fault detection rate of the slope-weighted least square RAIM algorithm is better than that of the least square RAIM algorithm.