一种空间相关高斯噪声背景下的时变时延估计算法

在空间相关高斯噪声的背景下,基于二阶统计量的时延估计方法会失效,该文提出了一种基于三阶统计量的自适应时变时延估计算法,并分析了算法的收敛性,最后的仿真结果表明该算法可以有效地抑制相关高斯噪声。     

非高斯噪声下基于Wilcoxon范数的变步长符号扩散式仿射投影算法

扩散式仿射投影算法(DAPA)是实现分布式网络参数自适应估计的一种重要方法,该算法在输入信号存在相关性时仍快速收敛,但抑制具有脉冲特性的非高斯噪声能力弱,且固定步长对收敛性有所限制.为此,该文提出了基于Wilcoxon范数的变步长符号扩散式仿射投影算法(VSS-DWAPA).首先,引入稳健估计理论中抗异常值能力强的Wi...     

自适应序贯M估计算法及其性能分析

针对复杂噪声环境下的参数估计问题,提出了一种稳健的自适应序贯M估计算法(Adaptive Recursive M-Estimation,ARME),并从理论分析和Monte Carlo实验仿真两方面分析了该算法的收敛性、渐进无偏特性和稳健性.理论分析和仿真试验表明:在高斯白噪声背景下,ARME具有与序贯最小二乘算法(Recursive Least Square,RLS)相近的性能;在有突出干扰等非高斯噪声背景下,与RLS相比,ARME的参数估计收敛速度更快,估计误差更小,而且在稳健性上大大优于RLS.     

强干扰环境下水声时延估计技术研究

水声定位系统是现代深海作业必备的高精度水下定位装备,精确的时延估计是实现高精度水声定位的基础,但由于信号远距离传输以及强干扰的影响,水声定位系统时延估计精度较低。针对此问题该文提出一种基于子空间理论的宽带强干扰抑制方法,首先通过贝叶斯信息量准则估计子空间维度,然后推导了不同信号假设下的概率密度函数,求解未知参数的最大似然估计,构造广义似然比并通过最优匹配广义似然比检测法估计与期望信号最匹配的子空间,然后以此构造空间投影算子对接收数据进行线性投影,最终抑制干扰和噪声,提高时延估计精度。仿真结果表明该方法能够有效抑制干扰和噪声的影响,提高定位系统时延估计精度。     

强干扰环境下水声时延估计技术研究

水声定位系统是现代深海作业必备的高精度水下定位装备,精确的时延估计是实现高精度水声定位的基础,但由于信号远距离传输以及强干扰的影响,水声定位系统时延估计精度较低.针对此问题该文提出一种基于子空间理论的宽带强干扰抑制方法,首先通过贝叶斯信息量准则估计子空间维度,然后推导了不同信号假设下的概率密度函数,求解未知参数的最大似然估计,构造广义似然比并通过最优匹配广义似然比检测法估计与期望信号最匹配的子空间,然后以此构造空间投影算子对接收数据进行线性投影,最终抑制干扰和噪声,提高时延估计精度.仿真结果表明该方法能够有效抑制干扰和噪声的影响,提高定位系统时延估计精度.     

非高斯噪声下基于Wilcoxon范数的变步长符号扩散式仿射投影算法

扩散式仿射投影算法(DAPA)是实现分布式网络参数自适应估计的一种重要方法,该算法在输入信号存在相关性时仍快速收敛,但抑制具有脉冲特性的非高斯噪声能力弱,且固定步长对收敛性有所限制.为此,该文提出了基于Wilcoxon范数的变步长符号扩散式仿射投影算法(VSS-DWAPA).首先,引入稳健估计理论中抗异常值能力强的Wilcoxon范数作为代价函数并根据其取值特点进行了符号量化,推导出了新的迭代方程;其次,针对固定步长的局限性,采用迭代方式实现了误差信号对步长的控制,在初始阶段和接近收敛阶段选择不同的步长,使算法具有更好的适应性.仿真结果表明,在非高斯噪声下本文的VSS-DWAPA算法在收敛性、跟踪性等方面均优于现有一些扩散式自适应滤波算法,同时在高斯噪声环境下也具有较好的性能.     

一种基于四阶累积量的约束自适应多径时延估计

在存在多径信号和空间相关性未知的背景高斯噪声情况下,不考虑多径信号传输的传统时延估计方法的性能会受到影响,甚至恶化。针对此问题,提出了一种基于四阶累积量的约束自适应多径时延估计算法,并对该算法的多径时延估计性能进行了收敛性能分析。该算法能够有效抑制空间相关性未知噪声的影响,在低信噪比的情况下能够直接、准确地进行自适应多径时延估计,克服了传统算法不能直接估计非整数倍采样间隔时延的缺点。计算机仿真试验验证了新算法的有效性。     

基于循环平稳性的约束自适应多径时延估计

在有多径信号、多通道非平稳干扰信号以及平稳背景噪声的情况下,不考虑多径信号传输的时延估计方法不能准确地估计时延,甚至估计性能会恶化.为此,本文提出了基于循环平稳性的约束自适应多径时延估计算法,并对算法的收敛性能进行了分析.该算法可以有效地抑制干扰和噪声的影响,在低信噪比的情况下直接地、准确地进行自适应多径时延估计,特别对噪声是空间相关的情形也适用,克服了传统算法不能直接估计非整数倍采样间隔的时延和多径时延的缺点.计算机仿真试验验证了新方法的有效性.     

脉冲噪声环境下基于分数低阶循环相关的自适应时延估计方法

以α稳定分布作为噪声模型,研究了非高斯噪声对传统的二阶循环统计量的影响,提出了分数低阶循环相关的概念,研究并证明了其性质,对传统意义上的二阶循环统计量进行了广义化,并在此基础上结合自适应技术提出了一种基于分数低阶循环相关的自适应时延估计方法。计算机模拟表明,该方法可有效估计高斯噪声和脉冲噪声条件下的时变和非时变时延值,其性能不仅优于基于二阶循环相关的自适应时延估计算法,而且优于最小平均p范数(LMP)自适应时延估计方法。     

基于三阶累积量的多径时延估计

研究多径传输条件下的时延估计问题。利用三阶累积量的一维切片作为高阶统计量,结合相关算法原理,提出一种新的时延估计算法。为提高时延估计精度,对相关数据进行了加权处理。该算法可有效抑制空间相关高斯噪声或对称分布噪声,得到非高斯信号准确的时延估计。算法具有计算量小,易于实现的优点。仿真结果表明了该算法的有效性。     

复杂环境下基于时延估计的声源定位技术研究简

为了改善在复杂环境下声源定位算法的性能,提出了一种新的时延估计（TDE）方法,即基于传递函数比的统计模型方法（ATFR-SM）。该方法采用统计模型去除噪声对传递函数（ATF）的影响,在计算传递函数时对功率谱密度（PSD）进行平滑和“白化”,以去除混响对传递函数的影响。同时,算法中引入话音激活检测（VAD）去除对求取传递函数无用的噪声段,以提高时延估计的准确性。此外,将所提时延估计方法与线性定位法相结合,构成一套完整的声源定位方法。实验结果表明,在复杂环境下,时延估计方法具有更低的异常点百分比（PAP）和均方根误差（RMSE）,且明显优于传统的参考算法,同时声源定位方法具有更高的定位精度。     

Robust time delay estimation of bioelectric signals using least absolute deviation neural network

The time delay estimation (TDE) is an important issue in modern signal processing and it has found extensive applications in the spatial propagation feature extraction of biomedical signals as well. Due to the extreme complexity and variability of the underlying systems, biomedical signals are usually nonstationary, unstable and even chaotic. Furthermore, due to the limitations of the measurement environments, biomedical signals are often noise-contaminated. Therefore, the TDE of biomedical signals is a challenging issue. A new TDE algorithm based on the least absolute deviation neural network (LADNN) and its application experiments are presented in this paper. The LADNN is the neural implementation of the least absolute deviation (LAD) optimization model, also called unconstrained minimum L1-norm model, with a theoretically proven global convergence. In the proposed LADNN-based TDE algorithm, a given signal is modeled using the moving average (MA) model. The MA parameters are estimated by using the LADNN and the time delay corresponds to the time index at which the MA coefficients have a peak. Due to the excellent features of L1-norm model superior to Lp-norm (p > 1) models in non-Gaussian noise environments or even in chaos, especially for signals that contain sharp transitions (such as biomedical signals with spiky series or motion artifacts) or chaotic dynamic processes, the LADNN-based TDE is more robust than the existing TDE algorithms based on wavelet-domain correlation and those based on higher-order spectra (HOS). Unlike these conventional methods, especially the current state-of-the-art HOS-based TDE, the LADNN-based method is free of the assumption that the signal is non-Gaussian and the noises are Gaussian and, thus, it is more applicable in real situations. Simulation experiments under three different noise environments, Gaussian, non-Gaussian and chaotic, are conducted to compare the proposed TDE method with the existing HOS-based method. Real application experiment is conducted to extract time delay information between every two adjacent channels of gastric myoelectrical activity (GMA) to assess the spatial propagation characteristics of GMA during different phases of the migrating myoelectrical complex (MMC).     

Real-time DSP-based time delay estimation using two sensors

There are few available approaches for time delay estimation (TDE) which offer with a potential to be implemented as real-time-embedded systems. Most of these lose accuracy in reverberant and highly noisy environments. In this paper, we initially proposed a 2-dimensional (2D) signal based on the cross-correlation. Then, we proposed two approaches for TDE, based on this 2D signal. There are two main advantages for using the proposed approaches. The first merit is that using a large amount of information about the correlation between the received signals allows a more robust TDE against noise and reverberation, when compared to the other possible techniques. The second one is that the number of samples to calculate the TDE algorithms is decreasing significantly, which is reducing the computational complexity. This paper also proposes a feasible hardware structure to implement the proposed approaches on digital signal processor embedded systems in real time.     

脉冲噪声环境下改进的顽健循环时延估计算法

研究了脉冲噪声环境下循环平稳信号的时延估计问题,针对脉冲噪声环境中基于传统二阶谱相关函数的时延估计方法性能退化问题,提出了基于分数低阶循环谱的改进顽健算法。相对于传统算法,新算法对脉冲噪声、高斯噪声、干扰信号都具有较好的抑制作用。仿真结果证明了算法的有效性和顽健性。     

Modified higher order spectral analysis based TDE algorithm for CFAR signal detection problem

With the conditions of small data size and low Signal-to-Noise Ratio (SNR), the application of Higher Order Statistics (HOS) is restrained not only by its high estimation variance, but also by its low estimation precision. Therefore, a modified HOS based Time Delay Estimation (TDE) algorithm is proposed to overcome these problems. Comparing with the conventional TDE algorithms, the estimation variance is improved greatly. A typical simulation example is completed in order to test the performance of the algorithm proposed, which shows that the proposed algorithm has advantages over the traditional ones in both detection performance and computation efficiency.     

Interference cancellation in respiratory sounds via amultiresolution joint time-delay and signal-estimation scheme

This paper is concerned with the problem of cancellation of heart sounds from the acquired respiratory sounds using a new joint time-delay and signal-estimation (JTDSE) procedure. Multiresolution discrete wavelet transform (DWT) is first applied to decompose the signals into several subbands. To accurately separate the heart sounds from the acquired respiratory sounds, time-delay estimation (TDE) is performed iteratively in each subband using two adaptation mechanisms that minimize the sum of squared errors between these signals. The time delay is updated using a nonlinear adaptation, namely the Levenberg-Marquardt (LM) algorithm, while the function of the other adaptive system-which uses the block fast transversal filter (BFTF)-is to minimize the mean squared error between the outputs of the delay estimator and the adaptive filter. The proposed methodology possesses a number of key benefits such as the incorporation of multiple complementary information at different subbands, robustness in presence of noise, and accuracy in TDE. The scheme is applied to several cases of simulated and actual respiratory sounds under different conditions and the results are compared with those of the standard adaptive filtering. The results showed the promise of the scheme for the TDE and subsequent interference cancellation     

AN IMPROVED OPTIMAL CHANNEL SHORTENING ALGORITHM FOR DMT MODULATION SYSTEMS

In Discrete Multi-Tone (DMT) modulation systems, the well-known technique to overcome the Inter-Carrier Interference (ICI)/Inter-Symbol Interference (ISI) caused by the inadequate Cyclic Prefix (CP) length is to use a Time-Domain Equalizer (TDE) at the receiver front-end. An algorithm used to calculate the coefficients of the optimal shortening Time Domain Equalizer (TDE) was given by Melsa. However, this algorithm requires that the length of the TDE must be smaller than or equal to the memory length of the target impulse response. This paper modifies this algorithm and makes it not only fit for calculating the coefficients of the TDE with arbitrary length, but also have a much less computational time.     

无先验参数条件下直扩信号时差估计研究

周期谱相关具有较强的抑制噪声和抗干扰的能力,适应于低信噪比时差估计的情况,但其抗噪声和干扰能力完全在于循环频率的选择。目前关于直扩信号时差估计的研究都是建立在循环频率已知的前提下进行的。运用直扩信号的周期谱特性,可在无先验参数条件下确定其循环频率,对时差进行估计,大大降低了谱相关时差估计算法的计算量。仿真结果表明,该方法是切实可行的。     

低信噪比下的LMS自适应无偏时延估计

 比较性研究了最小均方(LMS)时延估计器中有偏与无偏估计算法的时延估计性能,并基于Treichler的γ-LMS 算法提出了一种改进的无偏估计方法.利用自适应滤波器中最佳逼近原理的几何解释来估计输入噪声的功率,迭代过程中逐步去除输入噪声的影响,使得最优维纳解的真实峰值得到增强,在低信噪比或复杂噪声环境下显著改善了自适应时延估计性能.该方法无需假设输入与输出噪声功率相等或功率比已知、有用信号应为白过程等限制条件,因此具有广泛的应用价值.仿真与实际数据处理都验证了该方法的有效性.     

奇异值分解的HB加权广义互相关时延估计

传统广义互相关时延估计技术是直接基于测量数据,其精度受环境噪声及异常值波动影响显著下降。针对上述问题,提出了一种新的时延估计算法,即奇异值分解的HB(Hassab-Boucher)加权广义互相关法。首先,将接收到的信号进行奇异值分解处理,抑制环境噪声的影响并提高信号的信噪比;其次,采用降噪后的信号进行互功率谱计算时引入HB加权函数,达到锐化互相关函数峰值的目的;最后,在时延初值未知的情况下,提出了一种基于中位数与平均数结合的时延后处理方案,去除时延估计结果中的异常值波动,得到最优时延估计值。仿真实验结果表明,在低信噪比条件下,与传统的广义互相关和基于奇异值分解的广义互相关参考方法相比,本文提出方法的异常点百分比和均方根误差更低,时延估计正确率更高。