基于伪哈密顿量的变尺度Duffing振子弱信号检测  

Weak signal detection based on Pseudo-Hamiltonian using the scale transformation duffing oscillator

作　　者：张刚[1] 王颖[1] 王源[1] 

机构地区：[1]重庆邮电大学信号与信息处理重庆市重点实验室,重庆400065

出　　处：《电子技术应用》2014年第8期101-104,108,共5页Application of Electronic Technique

基　　金：国家自然科学基金项目(61371164;61071196;61102131);教育部新世纪优秀人才支持计划项目(NCET-10-0927);信号与信息处理重庆市市级重点实验室建设项目(CSTC2009CA2003);重庆市杰出青年基金项目(CSTC2011jjjq40002);重庆市自然科学基金项目(CSTC2010BB2398;CSTC2010BB2409;CSTC2010BB2411;CSTC2012JJA40008);重庆市教育委员会科研项目(KJ120525;KJ130524)

摘　　要：针对强噪声背景中微弱工程信号检测问题,在传统检测方法基础上,提出了基于伪哈密顿量的变尺度Duffing振子弱信号检测方法。待测高频工程信号经尺度变换为固定低频信号,从而唯一确定了相变阈值,并克服了传统方法中低频参数信号的限制。搭建自相关和小波阈值变换联合去噪系统,避免了噪声对检测结果的不利影响。构造Duffing系统伪哈密顿量实时地表征系统动力学行为,解决了定量判断系统状态时计算量大、效率低的难题。仿真测试表明,该检测方法可以快速检测任意频率、任意相位的低信噪比周期信号,改善了湮没在强噪声下的微弱信号检测技术。For the detection of weak engineering signal under strong noise background, a novel method was proposed in terms of conventional detection. This method referred to the scale transformation Duffing oscillator based on Pseudo-Hamihonian. The high-frequency engineering signal was changed into a fixed low-frequency signal by scale transformation. Thus the unique state transition threshold was determined, and the problem of low-frequency parameter ]imitation was overcome for the signal. The joint denoising system of autocorrelation and wavelet threshold transformation was composed, so the negative impact from noise was avoid- ed for the detection result. Pseudo-Hamiltonian was constructed to depic, t the dynamic behavior of Dulling system in real time, therefore reducing the calculation complexity and raising the efficiency when judging the system state quantitatively. Simulation results indicate that this method can be used to detect low SNR period signal with any frequency and phase quickly. So the weak signal detection method is improved under strong noise background.

关 键 词：微弱信号 DUFFING振子 尺度变换 伪哈密顿量 

分 类 号：TN911.7[电子电信—通信与信息系统]