In power system state estimation, the accuracy of different type measurements will excert a great impact on the state estimation results, and thus to determine the weight of the measurement is particularly important. Based on the weighted least squares method, this paper proposes a weighted least squares method that the weight of measurements is determined by the Analytic Hierarchy Process (AHP). The AHP method decomposes the power system state estimation into the minimum square of the difference between the measurements and estimated values(target), different types of measurement value (guideline) and relative accuracy of each type measurement (scheme), and based on which qualitative and quantitative analysis is conducted. Experimental results of IEEE 14-bus system show that using the proposed method the state estimation values of the power system can be estimated effectively. Through comparing the proposed method with basic weighted least squares method the effectiveness and applicability of the proposed algorithm are validated.