提出一种基于Hilbert-Huang变换的THz时域光谱分析方法，将THz时域脉冲信号分解成有限数目的单分量信号之和，利用Hilbert变换求得瞬时频率来获得幅值的时频分布——Hilbert-Huang变换谱，实现了通过水蒸气的THz脉冲信号的时频分析，揭示了THz波与水蒸气相互作用的频谱时域分布特性，并与基于小波变换的时频图进行了对比分析。结果表明，该方法可以同时提高THz脉冲时频分布的时间分辨率和频率分辨率， 具有局部化分析和自适应选择的特点，还能直观地表现出各频率成分之间的相对时间延迟。

A method for analyzing THz time-domain spectroscopy based on Hilbert-Huang transform(HHT) was presented. THz pulse of vapor can be decomposed into a number of signals with single frequency, and using Hilbert transform to those components can acquire the instantaneous frequency. The final presentation of those results is an energy-frequency-time distribution, designated as the HHT spectrum, which is compared with the time-frequency distribution based on wavelet transform. The results indicate that the method can improve the temporal resolution and frequency resolution, and it has the feature of localization analysis and adaptive selection. The HHT based THz-TDS is feasible in researching the mechanism on interaction of THz with materials by time-domain and frequency-domain.