关键词: 存在性检测, 突发信号, 谱方差, 自适应门限

Abstract: An adaptive blind presence detection algorithm for burst signals was proposed in this paper. A spectral variance function was chosen as the detection function, the change of background noise was tracked by establishing a noise function set with the threshold adjusted adaptively in real time. Then, the length of control and a stat-transition mechanism were used to mitigate the effect of dithering of detection function on the decision results. The simulation results show that the proposed method has a better robustness and a higher detection probability for burst signals compared with short-time energy method and spectral entropy method.

Key words: presence detection, burst signal, spectral variance, adaptive threshold