This paper describes a GPS (Global Positioning System) signal acquisition method using sample-based shifting and FFT (Fast Fourier Transform). Before the digital IF (Intermediate Frequency) signal is processed with the correlation algorithm, an averaging algorithm is applied to reduce the FFT operation. The principle of sample-based shifting is presented and its influence on the maximum correlation value is also simulated and analyzed. The result of simulation indicates that this method can increase the maximum correlation value and reduce the number of correlation operations.