论文部分内容阅读
针对微弱周期信号提出小波阈值去噪和混沌系统相结合的微弱周期信号检测新方法,该方法利用小波变换的平滑作用对包含噪声的信号进行有限离散处理,并根据小波自适应分解尺度确定阈值去噪深度,然后根据混沌系统对噪声的免疫性和对周期信号的敏感性,把重构的信号作为周期策动力的摄动并入混沌系统,由混沌系统完成微弱信号检测;并改进了Duffing方程,可用于不同频率信号的检测,同时使混沌系统的相轨迹由临界状态变为大尺度周期运动更灵敏。测试结果表明,本方法克服了以往小波分解对尺度确定的盲目性和阈值选择的不合理性;同时通过调节混沌系统频率对周期信号的敏感性提高了信号的检测精确度,其检测的信噪比下限达到-64.7 dB;完全适用于毫伏级下信号的检测应用,证明了提出的方法是合理和有效的。
Aiming at the weak periodic signal, a new method of weak periodic signal detection based on the combination of wavelet threshold denoising and chaotic system is proposed. This method uses the smoothing effect of wavelet transform to process the signal containing the noise with finite discretization and determines the threshold according to the wavelet adaptive decomposition scale According to the immunity of the chaotic system to the noise and the sensitivity to the periodic signal, the reconstructed signal is incorporated into the chaos system as the perturbation of the periodic power and the weak signal is detected by the chaotic system. The Duffing equation , Which can be used for the detection of different frequency signals and make the phase trajectory of the chaotic system more sensitive from the critical state to the large-scale periodic motion. The test results show that this method overcomes the irrationality of blindness and threshold selection of the scale determination by the wavelet decomposition in the past and improves the signal detection accuracy by adjusting the sensitivity of the chaotic system frequency to the periodic signal. The detected signal-to-noise The lower limit is -64.7 dB. It is fully suitable for the detection of signals at millivolts level, which proves that the proposed method is reasonable and effective.