摘要： 在突发信号处理环境中，传统的检验方法检验统计量计算复杂，不利于实时检测。为此，提出一种基于拟合优度检验的突发信号检测算法。采用Kolmogorov检验代替传统的参数统计检验，通过计算接收到样本的经验分布函数与标准的噪声分布函数，并将其垂直方向上的最大距离(Kolmogorov统计量)与特定虚警率下的门限作对比，实现对突发信号的检测。仿真结果表明，该算法的检测性能较好。

关键词: 拟合优度检验, Kolmogorov检验, 能量检测, 虚警率, 突发信号, 检测概率

Abstract: In the burst signal processing environment, the traditional test statistic of the computational complexity is not conducive to real-time detection. So this paper gives a burst signal detection algorithm based on the goodness of fit test by applying Kolmogorov test instead of parametric test methods. By calculating Kolmogorov statistics variable which is the maximum distances between the experience function of distribution and the standard function of noise distribution, and contrasting with the threshold on the specific probability of false alarm, it can achieve detecting for burst signals. Simulation results show that the algorithm can obtain better detection performance.

Key words: goodness of fit test, Kolmogorov test, energy detection, false alarm rate, burst signal, detection probability

中图分类号: 

• TP396