Second-order adaptive Volterra system identification based ondiscrete nonlinear Wiener model

The authors present the nonlinear LMS adaptive filtering algorithm based on the discrete nonlinear Wiener (1942) model for second-order Volterra system identification application. The main approach is to perform a complete orthogonalisation procedure on the truncated Volterra series. This allows the use of the LMS adaptive linear filtering algorithm for calculating all the coefficients with efficiency. This orthogonalisation method is based on the nonlinear discrete Wiener model. It contains three sections: a single-input multi-output linear with memory section, a multi-input, multi-output nonlinear no-memory section and a multi-input, single-output amplification and summary section. For a white Gaussian noise input signal, the autocorrelation matrix of the adaptive filter input vector can be diagonalised unlike when using the Volterra model. This dramatically reduces the eigenvalue spread and results in more rapid convergence. Also, the discrete nonlinear Wiener model adaptive system allows us to represent a complicated Volterra system with only few coefficient terms. In general, it can also identify the nonlinear system without over-parameterisation. A theoretical performance analysis of steady-state behaviour is presented. Computer simulations are also included to verify the theory     

Underdetermined-order recursive least-squares adaptive filtering:the concept and algorithms

Underdetermined recursive least-squares (URLS) adaptive filtering is introduced. In particular, the URLS algorithm is derived and shown to be a direct consequence of the principle of minimal disturbance. By exploiting the Hankel structure of the filter input matrix, the fast transversal filter (FTF) version of the URLS algorithm (URLS-FTF) is derived including sliding window and growing window types. The computational complexity is reduced to O(N)+O(m), where N is the adaptive filter length, and m is the order of the URLS algorithm. In addition, the efficient URLS (EURLS) algorithm, which does not compute the filter coefficients explicitly, thereby significantly reducing the computational load, is presented. Some earlier adaptive algorithms such as the averaged LMS, filtered-X LMS, and fast conjugate gradient are shown to be suboptimal approximations of the URLS algorithm. Instrumental variable approximations are also discussed. The URLS algorithm has a whitening effect on the input, signal, which provides immunity to the eigenvalue spread of the input signal correlation matrix. Although the algorithm is sensitive to observation noise, it has good tracking characteristics, and tradeoffs can be found by tuning the step size. The utility of the URLS algorithms, in its basic form and FTF realization, depends heavily on the practical applicability of the mth-order sliding window estimate of the covariance matrix and mth-order PTF relations. The feasibility of the URLS family in practical applications is demonstrated in channel equalization and acoustic echo cancellation     

Optimal variable step size for the LMS/Newton algorithm withapplication to subband adaptive filtering

An efficient approach for the computation of the optimum convergence factor for the LMS (least mean square)/Newton algorithm applied to a transversal FIR structure is proposed. The approach leads to a variable step size algorithm that results in a dramatic reduction in convergence time. The algorithm is evaluated in system identification applications where two alternative implementations of the adaptive filter are considered: the conventional transversal FIR realization and adaptive filtering in subbands     

Stochastic gradient based third-order Volterra system identification by using nonlinear Wiener adaptive algorithm

The nonlinear Wiener stochastic gradient adaptive algorithm for third-order Volterra system identification application with Gaussian input signals is presented. The complete self-orthogonalisation procedure is based on the delay-line structure of the nonlinear discrete Wiener model. The approach diagonalises the autocorrelation matrix of an adaptive filter input vector which dramatically reduces the eigenvalue spread and results in more rapid convergence speed. The relationship between the autocorrelation matrix and cross-correlation matrix of filter input vectors of both nonlinear Wiener and Volterra models is derived. The algorithm has a computational complexity of O(M/sup 3/) multiplications per sample input where M represents the length of memory for the system model, which is comparable to the existing algorithms. It is also worth noting that the proposed algorithm provides a general solution for the Volterra system identification application. Computer simulations are included to verify the theory.     

A class of order statistic LMS algorithms

Conventional gradient-based adaptive filters, as typified by the well-known LMS algorithm, use an instantaneous estimate of the error-surface gradient to update the filter coefficients. Such a strategy leaves the algorithm extremely vulnerable to impulsive interference. A class of adaptive algorithms employing order statistic filtering of the sampled gradient estimates is presented. These algorithms, dubbed order statistic least mean squares (OSLMS), are designed to facilitate adaptive filter performance close to the least squares optimum across a wide range of input environments from Gaussian to highly impulsive. Three specific OSLMS filters are defined: the median LMS, the average LMS, and the trimmed-mean LMS. The properties of these algorithms are investigated and the potential for improvement demonstrated. Finally, a general adaptive OSLMS scheme in which the nature of the order-statistic operator is also adapted in response to the statistics of the input signal is presented. It is shown that this can facilitate performance gains over a wide range of input data types     

DFT/LMS算法在DSSS中的应用及性能分析

本文分析了直接序列扩频(DSSS)系统中最小错误概率(MPE)意义下的最优滤波器,并依据矩阵求逆引理证明最小均方误差(MMSE)意义下的最优滤波——维纳滤波也是MPE意义下的最优滤波。在DSSS中应用自适应滤波,无须先验已知扩频码的码型和干扰的统计特性,就能一并完成解扩以及有效抑制干扰。离散傅立叶变换/最小均方(DFT/LMS)算法的收敛速度远快于LMS算法,而运算量、稳健性与LMS算法基本相同。基于DFT/LMS算法的自适应滤波大大简化DSSS系统接收机的设计,显著增强系统抗干扰能力,具有很强的实用性。     

基于LMS算法的自适应滤波器仿真实现

为了达到最佳的滤波效果,使自适应滤波器在工作环境变化时自动调节其单位脉冲响应特性,提出了一种自适应算法:最小均方算法(LMS算法)。这种算法实现简单且对信号统计特性变化具有稳健性,所以获得了极为广泛的应用。针对用硬件实现LMS算法的自适应滤波器存在的诸多缺点,采用Matlab工具对基于LMS算法的自适应滤波器进行了仿真试验。仿真结果表明,应用LMS算法的自适应滤波器不仅可以实现对信号噪声的自适应滤除,还能用于系统识别。     

Transform domain adaptive linear phase filter

Describes a new adaptive linear-phase filter whose weights are updated by the normalized least-mean-square (LMS) algorithm in the transform domain. This algorithm provides a faster convergence rate compared with the time domain linear phase LMS algorithm. Various real-valued orthogonal transforms are investigated such as the discrete cosine transform (DCT), discrete Hartley transform (DHT), and power of two (PO2) transform, etc. By using the symmetry property of the transform matrix, an efficient implementation structure is proposed. A system identification example is presented to demonstrate its performance     

Stationary and nonstationary learning characteristics of the LMS adaptive filter

This paper describes the performance characteristics of the LMS adaptive filter, a digital filter composed of a tapped delay line and adjustable weights, whose impulse response is controlled by an adaptive algorithm. For stationary stochastic inputs, the mean-square error, the difference between the filter output and an externally supplied input called the "desired response," is a quadratic function of the weights, a paraboloid with a single fixed minimum point that can be sought by gradient techniques. The gradient estimation process is shown to introduce noise into the weight vector that is proportional to the speed of adaptation and number of weights. The effect of this noise is expressed in terms of a dimensionless quantity "misadjustment" that is a measure of the deviation from optimal Wiener performance. Analysis of a simple nonstationary case, in which the minimum point of the error surface is moving according to an assumed first-order Markov process, shows that an additional contribution to misadjustment arises from "lag" of the adaptive process in tracking the moving minimum point. This contribution, which is additive, is proportional to the number of weights but inversely proportional to the speed of adaptation. The sum of the misadjustments can be minimized by choosing the speed of adaptation to make equal the two contributions. It is further shown, in Appendix A, that for stationary inputs the LMS adaptive algorithm, based on the method of steepest descent, approaches the theoretical limit of efficiency in terms of misadjustment and speed of adaptation when the eigenvalues of the input correlation matrix are equal or close in value. When the eigenvalues are highly disparate (λmax/λmin> 10), an algorithm similar to LMS but based on Newton's method would approach this theoretical limit very closely.     

自适应滤波器中的量化效应分析

自适应滤波算法是根据系统的输入信号和输出信号的数学统计特性,采用特定算法自动地调整滤波器某些参数,使其达到理想滤波特性的一种算法。而量化效应是由于软件或硬件的条件限制,使实际精度无法达到理想值而产生的误差对整个系统的影响。     

非线性Volterra系统的总体全解耦自适应滤波

研究输入、输出观测数据均受噪声干扰时的非线性Volterra系统的全解耦自适应滤波问题.基于总体最小二乘技术和Volterra滤波器的伪线性组合结构,运用约束优化问题的分析方法研究Volterra滤波过程,从而建立了一种总体全解耦自适应滤波算法.并建立了分析该算法收敛性能的参数反馈调整模型,分析表明,该算法可使各阶Volterra核稳定地收敛到真值.仿真实验的结果表明,当输入、输出观测数据均受噪声干扰时,总体全解耦自适应滤波算法的鲁棒抗噪性能和滤波精度均优于全解耦LMS自适应滤波算法.     

In the optimum design of the block adaptive FIR digital filter

A general optimum block adaptive (GOBA) algorithm for adaptive FIR (finite impulse response) filtering is presented. In this algorithm, the correction terms for the filter coefficients in each block, instead of the convergence factors, are optimized in a least squares sense. There are no constraints on the block length L and the filter tap number N. It is shown that the GOBA algorithm is reduced to the normalized LMS algorithm when L⩾N. The convergence of the GOBA algorithm can be assured if the correlation matrix of the input signal is positive definite. Computer simulations based on an efficient computing procedure confirm that the GOBA algorithm achieves faster convergence with slightly degraded convergence accuracy in stationary environments and better weight tracking capability in nonstationary environments as compared to existing block adaptive algorithms with no constraints on L and N     

Analysis of stochastic gradient tracking of time-varying polynomialWiener systems

This paper presents analytical and Monte Carlo results for a stochastic gradient adaptive scheme that tracks a time-varying polynomial Wiener (1958) system [i.e., a linear time-invariant (LTI) filter with memory followed by a time-varying memoryless polynomial nonlinearity]. The adaptive scheme consists of two phases: (1) estimation of the LTI memory using the LMS algorithm and (2) tracking the time-varying polynomial-type nonlinearity using a second coupled gradient search for the polynomial coefficients. The time-varying polynomial nonlinearity causes a time-varying scaling for the optimum Wiener filter for Phase 1. These time variations are removed for Phase 2 using a novel coupling scheme to Phase 1. The analysis for Gaussian data includes recursions for the mean behavior of the LMS algorithm for estimating and tracking the optimum Wiener filter for Phase 1 for several different time-varying polynomial nonlinearities and recursions for the mean behavior of the stochastic gradient algorithm for Phase 2. The polynomial coefficients are shown to be accurately tracked. Monte Carlo simulations confirm the theoretical predictions and support the underlying statistical assumptions     

Architectural Synthesis of Computational Engines for Subband Adaptive Filtering

Architectural synthesis of low-power computational engines (hardware accelerators) for a subband-based adaptive filtering algorithm is presented. The full-band least mean square (LMS) adaptive filtering algorithm, widely used in various applications, is confronted by two problems, viz., slow convergence when the input correlation matrix is ill-conditioned, and increased computational complexity for applications involving use of large adaptive filter orders. Both of these problems can be overcome by the use of a subband-based normalized LMS (NLMS) adaptive filtering algorithm. Since this algorithm is not amenable to pipelining, delayed coefficient adaptation in the NLMS updation is used, which provides the required delays for pipelining. However, the convergence speed of this subband-based delayed NLMS (DNLMS) algorithm degrades with increase in the adaptation delay. We first present a pipelined subband DNLMS adaptive filtering architecture with minimal adaptation delay for any given sampling rate. The architecture is synthesized by using a number of function preserving transformations on the signal flow graph (SFG) representation of the subband DNLMS algorithm. With the use of carry-save arithmetic, the pipelined architecture can support high sampling rates limited only by the delay of two full adders and a 2-to-1 multiplexer. We then extend this synthesis methodology to synthesize a pipelined subband DNLMS architecture whose power dissipation meets a specified budget. This low-power architecture exploits the parallelism in the subband DNLMS algorithm to meet the required computational throughput. The architecture exhibits a novel tradeoff between algorithmic performance (convergence speed) and power dissipation. Finally, we incorporate configurability for filter order, sample period, power reduction factor, number of subbands and decimation/interpolation factor in the low-power architecture, thus resulting in a low-power subband computational engine for adaptive filtering.     

Nonlinear effects in LMS adaptive equalizers

An adaptive transversal equalizer based on the least-mean-square (LMS) algorithm, operating in an environment with a temporally correlated interference, can exhibit better steady-state mean-square-error (MSE) performance than the corresponding Wiener filter. This phenomenon is a result of the nonlinear nature of the LMS algorithm and is obscured by traditional analysis approaches that utilize the independence assumption (current filter weight vector assumed to be statistically independent of the current data vector). To analyze this equalizer problem, we use a transfer function approach to develop approximate analytical expressions of the LMS MSE for sinusoidal and autoregressive interference processes. We demonstrate that the degree to which LMS may outperform the corresponding Wiener filter is dependent on system parameters such as signal-to-noise ratio (SNR), signal-to-interference ratio (SIR), equalizer length, and the step-size parameter     

Oversampled sigma-delta LMS adaptive FIR filters

Two filter designs for adaptive least mean squares (LMS) filtering with sigma-delta modulated input signals are described. One implementation is multibit multiplier-free and operates entirely at the oversampling frequency of the sigma-delta signals, in the other design only the FIR filter operates at the oversampled frequency while the adaptive filtering algorithm is performed at the Nyquist rate. To circumvent any aliasing problems that may be caused by the downsampling process in the architecture and ensure convergence of the adaptive FIR filter. It is necessary to attenuate the high-frequency sigma-delta quantisation noise that is present. To perform this task a multiplier-free, multistage IIR filter structure is used that requires considerably fewer computations than an equivalent FIR filter. The two adaptive LMS filter designs are analysed and their performance is compared with a conventional PCM system in terms of achievable minimum MSE and adaptation speed     

Convergence analysis of adaptive filtering algorithms with singulardata covariance matrix

The paper provides a rigorous analysis of the behavior of adaptive filtering algorithms when the covariance matrix of the filter input is singular. The analysis is done in the context of adaptive plant identification. The considered algorithms are LMS, RLS, sign (SA), and signed regressor (SRA) algorithms. Both the signal and weight behavior of the algorithms are considered. The signal behavior is evaluated in terms of the moments of the excess output error of the filter. The weight behavior is evaluated in terms of the moments of the filter weight misalignment vector. It is found that the RLS and SRA diverge when the input covariance matrix is singular. The steady-state signal behavior of the LMS and SA can be made arbitrarily fine by using sufficiently small step sizes of the algorithms. Indeed, the long-term average of the mean square excess error of the LMS is proportional to the algorithm step size. The long-term average of the mean absolute excess error of the SA is proportional to the square root of the algorithm step size. On the other hand, the steady-state weight behavior of both the LMS and SA have biases that depend on the weight initialization. The analytical results of the paper are supported by simulations     

A new approach to subband adaptive filtering

Subband adaptive filtering has attracted much attention lately. In this paper, we propose a new structure and a new formulation for adapting the filter coefficients. This structure is based on polyphase decomposition of the filter to be adapted and is independent of the type of filter banks used in the subband decomposition. The new formulation yields improved convergence rate when the LMS algorithm is used for coefficient adaptation. As we increase the number of bands in the filter, the convergence rate increases and approaches the rate that can be obtained with a flat input spectrum. The computational complexity of the proposed scheme is nearly the same as that of the fullband approach. Simulation results are included to demonstrate the efficacy of the new approach     

Analytical model for the first and second moments of an adaptiveinterpolated FIR filter using the constrained filtered-X LMS algorithm

The authors present an analytical model for the mean weight behaviour and weight covariance matrix of an adaptive interpolated FIR filter using the LMS algorithm to adapt the filter weights. The particular structure of this adaptive filter determines that special analytical considerations must be used. First, the introduction of an interpolating block cascaded with the adaptive sparse filter requires that the input signal correlations must be considered. It is well known that such correlations are disregarded by the independence theory, which is the basis for the analysis of the LMS algorithm adapting FIR structures. Secondly a constrained analysis is used to deal mathematically with the sparse nature of the adaptive section. Experimental results demonstrate the effectiveness of the proposed analytical models as compared with the results obtained by classical analysis     

A Single-Channel Acoustic Echo Cancellation Scheme Using Gradient-Based Adaptive Filtering

In this paper, a single-channel acoustic echo cancellation (AEC) scheme is proposed using a gradient-based adaptive least mean squares (LMS) algorithm. Unlike the conventional dual-channel problem, by considering a delayed version of the echo-suppressed signal as a reference, a modified objective function is formulated and thereby an LMS update equation is derived. It is shown that the resulting update equation converges to the optimum Wiener–Hopf solution. Based on the commonly used assumption of negligibility of cross correlation between the reference and the current speech signals, a multi-step stopping criterion is introduced, which not only provides efficient control of the LMS update sequence but also ensures a faster convergence. The proposed control criterion is validated by considering all possible scenarios which arise due to the variation of speech properties at the reference and current samples of the adaptive filter. From extensive experimentation on several real-life echo-corrupted speech signals in different acoustic environments, it is found that the proposed algorithm can efficiently handle the problem of single-channel AEC and provide satisfactory performance in terms of both subjective and objective measures.