基于不变矩的微弱信号检测方法

由于混沌振子从混沌状态转变到大尺度周期状态时其相图的对称性将发生较大变化,提出应用Hu氏不变矩对混沌振子相图的状态进行描述,通过不变矩-策动力关系曲线确定混沌振子相图处于大尺度周期状态的左临界阈值,根据混沌振子混沌状态和大尺度周期状态下不变矩的差异实现混沌振子相图的自动识别。对仿真和工程实测信号进行了分析,结果表明,不变矩值能够准确识别混沌振子相图的状态,并具有良好的抗噪能力。

New Method for Weak Signal Detection Based on Invariant Moment

When chaotic oscillator changing from chaotic state to large-scale periodic state, its phase diagram symmetry will obviously changed. Hu invariant moment is adopted to describe the different state of chaotic oscillator phase diagram. The threshold value of chaos oscillator under the large-scale periodic state is ascertained by the curve of invariant moment-periodic driving force. According to the difference of Hu invariant moment value for different chaotic oscillatr phase diagram,the state of chaos oscillator can be automatically identified. Finally, the simulated and engineering vibration signals are analyzed, and the results show that invariant moment can accurately identify the state of chaotic os cillator and have better noise-tolerance.