根据瞬态通信信号和非高斯噪声的特点,建立了相应的信号模型,并利用希尔伯特-黄变换(HHT)处理非线性非平稳信号的优势,提出了基于改进HHT的非高斯噪声中瞬态通信信号的检测算法。该检测算法分为集合经验模式分解(EEMD)和固有模态函数(IMF)分量筛选两部分,首先经过加入随机白噪声多次试验取均值得到待检测信号的IMF分量,再结合各个分量与原信号的能量差异和相关性剔除虚假IMF分量,从而实现对混叠在非高斯噪声中的瞬态通信信号的有效检测。仿真在不同的条件下对比了本文算法与其他算法对信号的检测效果,结果证明本文算法能够有效克服HHT中存在的缺陷,实现对瞬态信号更为准确的分析和检测。 According to the characteristics of transient communication signal and non-Gaussian noise, the corresponding signal model is established and the advantage of using Hilbert-Huang Transform (HHT) to process non-linear non-stationary signals is proposed. The non-Gaussian noise based on improved HHT Transient communication signal detection algorithm. The detection algorithm is divided into EEMD (Set Empirical Mode Decomposition) and IMF (Intrinsic Mode Function) components screening. Firstly, the IMFs of signals to be detected are obtained by adding mean of random white noise. Then, The energy differences and correlations of the signals reject false IMF components, thereby enabling efficient detection of transient communication signals that are aliased in non-Gaussian noise. Simulation results show that the proposed algorithm can effectively overcome the shortcomings in HHT and achieve more accurate analysis and detection of transient signals under different conditions.