四阶累积量谱线增强方法的改进仿真研究

摘要

Based on the aim of improving the tracing performance of adaptive line enhancer and reducing the computational complexity of cumulant estimate a new method of higher-order cumulant-based variable step size ISA algorithm iteration adaptive dynamic spectrum enhancement was proposed method taking the advantages of variable step size algorithm and ISA algorithm and basing on higher-order cumulant-based energy power function type variable step size sign iteration adaptive dynamic spectrum enhancement algorithm.The proposed method utilizes the energy power function as variable step size to amend the tracing performance and it uses piecewise amplitude limiting function of the input signal to reduce the computation complexity of higher-order cumulant estimate.The simulation shows that the proposed method has outstanding performance in suppressing Gauss noise and Laplace noise that is better than energy power function type variable step size iteration adaptive dynamic spectrum enhancement algorithm.And it can maintain improving signal-to-noise ratio 3 dB steadily.%为提高自适应谱线增强器跟踪性能，减小高阶累计量的计算量，综合了变步长算法与变符号函数法（ISA）算法的优点，在高阶累积量的能量幂函数变步长极性迭代自适应谱线增强算法的基础上，首次提出了一种计算正弦信号基于高阶累积量变步长ISA算法的迭代自适应谱线增强方法。该方法用输入信号的能量幂函数作变步长来改善跟踪性能，用非输入信号的分段性限幅函数来减小高阶累计量的计算量。仿真结果表明该算法在抑制高斯噪声和非高斯对称分布（拉普拉斯）噪声上优越于基于能量幂函数变步长极性迭代自适应谱线增强算法，能稳定保持信噪比提高3 dB。