基于形态学预处理的短波猝发信号检测算法

摘要

There are problems of great fluctuation of channel noise floor,plenty of multi⁃peak signals and so on in burst sig⁃nal detection under the conditions of short wave wideband receiving. To solve the two problems,a short wave detection method based on spectrum gray⁃scale morphology pre⁃processing theory is proposed. The noise floor of wideband spectrum is well estimated by taking the wideband spectrum chart as the one⁃dimensional gray image to perform morphology pre⁃processing. The problem of multi⁃peak signal identification is solved by the combined identification method of multi⁃threshold,such as adaptive threshold, amplitude threshold and bandwidth threshold. The experimental results indicate this detecting method can realize high probability captur of the shortwave burst signal under wideband condition,and has the characters of strong robustness and practicability.%在短波宽带接收条件下的猝发信号检测中，存在信道噪底起伏较大、多峰信号较多等问题。针对这两个问题，提出一种基于频谱灰度形态学预处理原理的短波检测方法，通过将宽带频谱图作为一维灰度图像进行形态学预处理，较好地估计了宽带频谱的噪声基底；并采用自适应门限、幅度门限和带宽门限等多门限联合判证方法，有效解决了多谱峰信号的判证问题。大量实验表明，该检测方法能够在宽带条件下实现对短猝发信号的高概率捕获，具有很强的鲁棒性和实用性。