Dispersion measurement of surface-waves is critical for utilizing surface-waves. However, the traditional dispersion measurements of multi-channel seismic data are usually based on linear signal comparison and achieve a low resolution at low frequencies. To improve the accuracy of surface-wave dispersion measurement for multi-channel seismic data, we propose a novel measurement that can produce a high and adjustable resolution among a wide detected frequency range. This measurement is based on a nonlinear signal comparison using multi-channel seismic data. We evaluated the performance of this proposed multi-channel nonlinear signal comparison (MNLSC) on several datasets: one synthetic shotgather containing only surface waves, one synthetic shotgather with full elastic waves, and one field shotgather. These examples show that the MNLSC is promising to be a reliable and effective tool for measuring the surface dispersion from the multi-channel seismic data. (C) 2018 Elsevier B.V. All rights reserved.