To eliminate the aliasing that appeared during the measurement of multi-components nonstationary signals, a novel kind of anti-aliasing algorithm based on the short time Fourier transform （STFT） is brought forward. ...To eliminate the aliasing that appeared during the measurement of multi-components nonstationary signals, a novel kind of anti-aliasing algorithm based on the short time Fourier transform （STFT） is brought forward. First the physical essence of aliasing that occurs is analyzed; second the interpolation algorithm model is setup based on the Hamming window; then the fast implementation of the algorithm using the Newton iteration method is given. Using the numerical simulation the feasibility of algorithm is validated. Finally, the electrical circuit experiment shows the practicality of the algorithm in the electrical engineering.展开更多
在无线通信环境中压制式复合干扰信号对通信系统的正常工作有着严重的影响,针对其特征提取和识别较为困难的问题,提出一种基于短时傅里叶变换(short time Fourier transform,STFT)和残差卷积网络的复合干扰识别算法。该算法将STFT得到...在无线通信环境中压制式复合干扰信号对通信系统的正常工作有着严重的影响,针对其特征提取和识别较为困难的问题,提出一种基于短时傅里叶变换(short time Fourier transform,STFT)和残差卷积网络的复合干扰识别算法。该算法将STFT得到的时频域信息作为输入,同时对复合干扰信号的种类和干噪比进行识别,为了使模型更加适合部署在移动端上,采用幻影卷积代替普通卷积。仿真结果表明,在干噪比为-15~10 dB的范围内,该算法在5种单一干扰及其复合而成的10种复合干扰信号种类识别任务上准确率可以达到99.97%,在干噪比识别任务上准确率可以达到99.04%。相比于残差卷积网络,该算法在几乎不降低准确率的前提下可以使模型参数量减小38.4%,计算复杂度降低46.6%,更加符合移动端的要求。展开更多
针对旋转机械故障振动信号的降噪问题,提出一种基于短时Fourier变换(short time Fourier transform,简称STFT)时频谱系数收缩的信号降噪方法。先将信号进行STFT,得到其时频谱。由于谱系数为复数,故根据模值大小进行谱系数收缩,并利用步...针对旋转机械故障振动信号的降噪问题,提出一种基于短时Fourier变换(short time Fourier transform,简称STFT)时频谱系数收缩的信号降噪方法。先将信号进行STFT,得到其时频谱。由于谱系数为复数,故根据模值大小进行谱系数收缩,并利用步长迭代算法在0到谱系数最大模值的区间内估计最优阈值。迭代运算过程中,首先,分别采用基本的硬阈值函数和软阈值函数进行系数收缩;然后,以改进风险函数为阈值评价标准,估计最优阈值;最后,利用最优阈值重新进行谱系数收缩,对得到的新谱进行STFT逆变换,重构降噪后的时域信号。仿真信号与试验数据的处理结果表明,利用所估计的最优阈值,STFT时频谱系数硬、软阈值函数收缩方法均能够实现噪声混合信号的降噪。展开更多
基金the National Natural Science Foundation of China (90407007 60372001).
文摘To eliminate the aliasing that appeared during the measurement of multi-components nonstationary signals, a novel kind of anti-aliasing algorithm based on the short time Fourier transform （STFT） is brought forward. First the physical essence of aliasing that occurs is analyzed; second the interpolation algorithm model is setup based on the Hamming window; then the fast implementation of the algorithm using the Newton iteration method is given. Using the numerical simulation the feasibility of algorithm is validated. Finally, the electrical circuit experiment shows the practicality of the algorithm in the electrical engineering.
文摘在无线通信环境中压制式复合干扰信号对通信系统的正常工作有着严重的影响,针对其特征提取和识别较为困难的问题,提出一种基于短时傅里叶变换(short time Fourier transform,STFT)和残差卷积网络的复合干扰识别算法。该算法将STFT得到的时频域信息作为输入,同时对复合干扰信号的种类和干噪比进行识别,为了使模型更加适合部署在移动端上,采用幻影卷积代替普通卷积。仿真结果表明,在干噪比为-15~10 dB的范围内,该算法在5种单一干扰及其复合而成的10种复合干扰信号种类识别任务上准确率可以达到99.97%,在干噪比识别任务上准确率可以达到99.04%。相比于残差卷积网络,该算法在几乎不降低准确率的前提下可以使模型参数量减小38.4%,计算复杂度降低46.6%,更加符合移动端的要求。
文摘针对旋转机械故障振动信号的降噪问题,提出一种基于短时Fourier变换(short time Fourier transform,简称STFT)时频谱系数收缩的信号降噪方法。先将信号进行STFT,得到其时频谱。由于谱系数为复数,故根据模值大小进行谱系数收缩,并利用步长迭代算法在0到谱系数最大模值的区间内估计最优阈值。迭代运算过程中,首先,分别采用基本的硬阈值函数和软阈值函数进行系数收缩;然后,以改进风险函数为阈值评价标准,估计最优阈值;最后,利用最优阈值重新进行谱系数收缩,对得到的新谱进行STFT逆变换,重构降噪后的时域信号。仿真信号与试验数据的处理结果表明,利用所估计的最优阈值,STFT时频谱系数硬、软阈值函数收缩方法均能够实现噪声混合信号的降噪。
