Abstract: A novel non-uniform sampling interval calibration method is proposed for the Domino Ring Sampler (DRS4). This method calculates the calibration value of the sampling interval by maintaining a fixed proportionality (equal slope) between the amplitude difference of the ramp pulse and the sampling interval of each two consecutive sampling cells, utilizing the ramp pulse as the calibration source. The signal generator produces two ramp pulses with a period of 200 ns and a slope of ±4.5 mV/ns. These ramp pulses are then sent to the waveform sampling digitizer to calculate the sampling intervals of 1024 sampling cells in the linear increase/decrease process. The average sampling interval of the sampling cells is found to be 200.288 ps, with an RMS of 15.603 ps. Test results comparing this method with the "zero-crossing of sine wave" and "sawtooth wave" methods show that the relative errors of the average sampling interval are 0.14%, 2.49%, and 2.48%, respectively. These results indicate that the calibration accuracy of the ramp method is the highest. Furthermore, the ramp method has no frequency range limitation for the calibration source signal, and conducting 3000 iterations is adequate to achieve the necessary measurement accuracy. This makes it more suitable for applications with real-time calibration requirements.