Rolling bearings are key components in rotating machinery and its running status has an important effect on the performance of the equipment and even the entire system. Rolling bearing failure produces a high frequency stress wave signal, but Peakvue technique could detect effectively, so a fault diagnosis of rolling bearings using Peakvue technique can be used. Acceleration sensor is used to collect vibration signal of rolling bearings, and the high-pass filter is used to filter unwanted low frequency signal from the output signal of acceleration sensor, and peak extraction is used on high frequency signal and stress wave signal according to a certain time interval, and envelope detection is used on peak signal to analyze the fault type. Datasets from the Case Western Reserve University was used for validation, and the results indicate that this method can accurately detect the fault type of rolling bearings.