In this brief, two robust constrained affine-projection-like M-estimate (CAPLM) adaptive filtering algorithms are proposed, which solve the problem that the traditional constrained affine projection (CAP) algorithm is...
详细信息
In this brief, two robust constrained affine-projection-like M-estimate (CAPLM) adaptive filtering algorithms are proposed, which solve the problem that the traditional constrained affine projection (CAP) algorithm is not robust to impulsive interference, and realize unbiased output in some applications where the desired signal is unavailable or unnecessary. Specifically, a modified Huber function (MHF) based robust AP-like (APL) minimization problem with constraints is defined and can be transformed into two different unconstrained optimization problems by using Lipschitz continuity and two different methods, and then CAPLM-I and CAPLM-II algorithms are obtained respectively. Both CAPLM algorithms can avoid a certain amount of computational complexity caused by the inversion of the input signal matrix in classical AP algorithm, and realize the robustness to impulsive interference. In addition, the mean square stabilities of them are analyzed, and the corresponding stable step size ranges are also given. Simulation results show that the proposed CAPLM-I and CAPLM-II algorithms perform well in system identification and beamforming applications in impulsive noise environment, and provide lower steady-state error and faster convergence speed than other compared algorithms.
Surface electromyography (sEMG) is crucial in sports science, offering insights into muscle activation patterns. However, standard devices like the Delsys Trigno Wireless System are relatively costly. This study prese...
详细信息
ISBN:
(数字)9798350350821
ISBN:
(纸本)9798350350838
Surface electromyography (sEMG) is crucial in sports science, offering insights into muscle activation patterns. However, standard devices like the Delsys Trigno Wireless System are relatively costly. This study presents a prototype that simultaneously obtains sEMG and IMU signals to measure muscle activity. The presence of noise necessitates signal processing techniques to be applied to the raw signal, including feed-forward comb, adaptive, and wavelet filters. Evaluation against a reference signal revealed that the wavelet filter performed best, exhibiting the lowest MSE and highest SNR scores. Specifically, it achieved MSE scores of 2.62E-10 in the time domain, 6.25E-22 in PSD, and SNR scores 13.96 for squats. These findings underscore the effectiveness of the wavelet filter in reducing sEMG signal noise.
In nonstationary environments, existing frequency-domain adaptive filtering algorithms would exhibit poor tracking performance. To solve this issue, this paper focuses on developing new frequency-domain adaptive filte...
详细信息
ISBN:
(数字)9798350350920
ISBN:
(纸本)9798350350937
In nonstationary environments, existing frequency-domain adaptive filtering algorithms would exhibit poor tracking performance. To solve this issue, this paper focuses on developing new frequency-domain adaptive filtering algorithms based on single data. Using the circular matrix of the regression vector, we first establish a model and cost function suitable for a nonstationary system. Next, with resort to the stochastic gradient descent and power normalized methods, the frequency-domain least mean-square algorithm based on single data (SFDLMS) and its normalized version (named SFDNLMS) are derived. Even in the presence of correlated input signals, the proposed SFDNLMS algorithm can provide fast tracking/convergence performance. The transient and steady-state behavior is also studied. Finally, experiment results illustrate the advantages of the proposed algorithms and the reliability of the theoretical analysis.
The aim of this work evaluates the performance of three speech improvement algorithms in hearing aids, Noise Reduction via Spectral Subtraction Enhancement (NRSE), Adaptive Frequency-Domain Range Compression (AFDRC), ...
详细信息
ISBN:
(数字)9798331505134
ISBN:
(纸本)9798331505141
The aim of this work evaluates the performance of three speech improvement algorithms in hearing aids, Noise Reduction via Spectral Subtraction Enhancement (NRSE), Adaptive Frequency-Domain Range Compression (AFDRC), and Beamforming Noise Suppression (BNS). The action focuses on key objective metrics such as Speech Intelligibility Index (SII), Signal-to-Noise Ratio (SNR) Improvement, and Total Harmonic Distortion (THD). Outcomes indicate that all three algorithms perform similarly in no-noise environments, maintaining speech clarity and low distortion. However, in high-noise conditions, BNS is consistently profitable for both NRSE and AFDRC. BNS shows greater noise reduction capabilities, improving SNR by 10% compared to NRSE and 8% over AFDRC. Additionally, BNS reduces distortion more effectively, reducing THD by 9% and 7% compared to the other two methods. These findings help to highlight BNS's superior ability to improve speech clarity and reduce background noise, particularly in challenging environments. The work highlights the importance of using objective metrics like SII, SNR Improvement, and THD to evaluate and effective speech filtering algorithms in hearing aids. By improving performance in noisy conditions, BNS offers significant benefits for individuals with hearing impairments, enhancing overall hearing aid effectiveness in real-world scenarios.
This paper proposes wavelet domain adaptive filtering algorithms for sparse system identification under impulsive noise and correlated input conditions. Different algorithms are obtained by first transforming the inpu...
详细信息
ISBN:
(数字)9798350377200
ISBN:
(纸本)9798350377217
This paper proposes wavelet domain adaptive filtering algorithms for sparse system identification under impulsive noise and correlated input conditions. Different algorithms are obtained by first transforming the input to wavelet domain and then combining proportionate adaptation with error nonlinear adaptive filtering techniques. A comprehensive comparative study of the derived algorithms is carried out considering two important parameters which are steady state mean square deviation (MSD) and rate of convergence to show that the proposed algorithms significantly outperform their time domain counterparts.
In this paper, two new correntropy-based data-selective algorithms using the prediction error method and proportionate principle are proposed for acoustic feedback cancellation (AFC). The proposed data selective appro...
详细信息
ISBN:
(数字)9798350362510
ISBN:
(纸本)9798350362527
In this paper, two new correntropy-based data-selective algorithms using the prediction error method and proportionate principle are proposed for acoustic feedback cancellation (AFC). The proposed data selective approach is applied for the improved practical variable step size proportionate normalized least mean square algorithm (IPNLMS-IPVSS) and its novel tanh-based version. It is shown that the proposed data-selective algorithms can achieve close performance to the original algorithms at a reduced average numerical complexity.
We focus on the estimation error of a type of filtering algorithm in the scalar case, which is applicable to both linear and nonlinear systems. Under some regularity conditions, we construct a surrogate process that h...
详细信息
ISBN:
(纸本)9781665441971
We focus on the estimation error of a type of filtering algorithm in the scalar case, which is applicable to both linear and nonlinear systems. Under some regularity conditions, we construct a surrogate process that has a moment dominance property with respect to the true filtering error process. Then, moment-based probability inequalities can be used to compute probabilistic bounds for the filtering error. The sharpness of the bounds is tested on a simulated epidemic model with both Gaussian and non-Gaussian noise.
Suppose L simultaneous independent stochastic systems generate observations, where the observations from each system depend on the underlying parameter of that system. The observations are unlabeled (anonymized), in t...
详细信息
Suppose L simultaneous independent stochastic systems generate observations, where the observations from each system depend on the underlying parameter of that system. The observations are unlabeled (anonymized), in the sense that an analyst does not know which observation came from which stochastic system. How can the analyst estimate the underlying parameters of the L systems? Since the anonymized observations at each time are an unordered set of L measurements (rather than a vector), classical stochastic gradient algorithms cannot be directly used. By using symmetric polynomials, we formulate a symmetric measurement equation that maps the observation set to a unique vector. We then construct an adaptive filtering algorithm that yields a statistically consistent estimate of the underlying parameters.
Typically, speech signals interfere with other types of signals. In cases where noise performance is enhanced, its continuation can be modified, analyzed, or the results of speech evaluation changed. In other cases, s...
详细信息
ISBN:
(纸本)9781665432597
Typically, speech signals interfere with other types of signals. In cases where noise performance is enhanced, its continuation can be modified, analyzed, or the results of speech evaluation changed. In other cases, such as the analysis of noisy recordings for forensic purposes or the restoration of audio recordings in archives, the task of filtering the signal from noise are the main purpose of the research work. Therefore, the development of methods to filter the signal from noise is a very topical area of research.
This paper presents a variational representation of the Bayes’ law using optimal transportation theory. The variational representation is in terms of the optimal transportation between the joint distribution of the (...
详细信息
ISBN:
(数字)9781665467612
ISBN:
(纸本)9781665467629
This paper presents a variational representation of the Bayes’ law using optimal transportation theory. The variational representation is in terms of the optimal transportation between the joint distribution of the (state, observation) and their independent coupling. By imposing certain structure on the transport map, the solution to the variational problem is used to construct a Brenier-type map that transports the prior distribution to the posterior distribution for any value of the observation signal. The new formulation is used to derive the optimal transport form of the Ensemble Kalman filter (EnKF) for the discrete-time filtering problem and propose a novel extension of EnKF to the non-Gaussian setting utilizing input convex neural networks. Finally, the proposed methodology is used to derive the optimal transport form of the feedback particle filler (FPF) in the continuous-time limit, which constitutes its first variational construction without explicitly using the nonlinear filtering equation or Bayes’ law.
暂无评论