The Terrestrial Ecosystem Carbon Inventory Satellite (TECIS) is the first Chinese satellite in orbit equipped with both waveform LiDAR and multi-angle imaging capabilities, dedicated to forest carbon monitoring. It has collected a large volume of waveform data. To accurately estimate canopy height using waveform data, two critical steps are identifying the effective signal range based on signal threshold settings and selecting the optimal waveform metrics processing method. The choice of waveform threshold settings and metrics calculation methods directly affects the accuracy of canopy height estimation.To optimize threshold settings and identify suitable waveform metrics calculation methods, this study focused on the northern forest region of China, using airborne laser scanning (ALS) data as a reference to evaluate TECIS canopy height estimation approaches. First, effective TECIS footprints were selected and spatially matched with ALS data. Then, 24 waveform threshold combinations and 3 metrics calculation methods were applied to derive waveform metrics, and the bias between the Relative Height (RH) metrics and ALS percentile heights was analysed. Subsequently, seven sets of waveform metrics were used to construct linear regression models for estimating the 99th percentile ALS height (P99), representing the observed maximum canopy height. Finally, the bias between RH100 and ALS P99 under different conditions was evaluated.The results showed that using a six-times start signal threshold and a twelve-times end signal threshold, combined with RH metrics calculated based on waveform decomposition and signal cut-off points, yielded the highest consistency with ALS height data. For RH metrics calculated based solely on signal cut-off points, stricter threshold settings—especially at the signal end—resulted in better agreement. In ALS P99 regression modelling, the optimal waveform metrics model achieved a Pearson correlation coefficient (r) of 0.86 when slope was not considered. Under low-slope, medium canopy cover, and night-time data acquisition conditions, the estimation bias was further reduced.This study provides a new perspective from a signal processing standpoint for improving TECIS waveform-based canopy height estimation. The findings are expected to support the development of TECIS canopy height products and promote large-scale application of wide-footprint waveform LiDAR data.