中文摘要：

常规Duffing振子检测方法只能确定待测弱信号的有无及幅值大小却无法给出其在时域上的分布信息,且待测信号包络有起伏时会引起Duffing振子系统的误判,针对上述问题提出可变幅值系数的短时Duffing振子阵列检测方法。应用互模糊函数推导出随机振幅包络起伏信号通过短时处理可以降低包络起伏因素对Duffing振子检测性能的影响,对信号加时间窗的短时处理可以保证窗内信号近似平稳,并可同时获得待测信号在时域上的分布信息。为克服相图判别法定性分析的不足,降低待测信号与内置策动力合成的总策动力幅值小于跃变阈值时出现的漏报误判概率,提出可变幅值系数的Poincare映射集判别方法,给出了量化统计判别结果。仿真及实测水声信号检测结果表明,在-33d B的低信噪比背景下仍可实现信号的检测判别。

英文摘要：

The conventional detection method for a Duffing oscillator can determine if a weak signal exists and can determine the amplitude value of that signal but gives no distribution information in the time domain. wrongful judg-ments of the system are caused by the envelope fluctuation of this signal. A detection method of a short-time Duffing oscillator array detection method of a with variable amplitude coefficients is proposed for the above problems. By u-sing a cross-ambiguity function, we observed that the influence of the envelope fluctuation factors on the detection performance of the Duffing oscillator can be reduced by short-time processing. Adding a short-time window ensures that the signal in the window is approximately stable. Simultaneously, distribution information on the measured sig-nal in the time-domain is obtained by window analysis. To overcome the shortcomings of the phase diagram discrimi-nation method in quantitative analysis and to reduce the error probability caused by the amplitude of the total driv-ing force being less than the jump threshold, a Poincare mapping set identification method with variable amplitu decoefficients is proposed and quantitative statistical results are given. Simulation and experimental results show that the signal can be detected even when the signal to noise ratio is -33 dB.