Robust distributed Lorentzian adaptive filter with diffusion strategy in impulsive noise environment

Wilson, Annet Mary; Panigrahi, Trilochan; Dubey, Ankit

doi:10.1016/j.dsp.2019.102589

Cited by 30 publications

(8 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the failure of least mean square-based diffusion algorithms to cope with impulse noise, researchers have been working on robust cost functions [38]. The resemblance of correntropy-based adaptive algorithms to robust M-estimation has been the motivation to adapt the Geman-McClure objective function to a distributed network and compare its efficiency with correntropy-based algorithms.…”

Section: Proposed Diffusion Geman-mcclure Algorithm and Computational Complexitymentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Geman-McClure Estimator for Robust Distributed Channel Estimation

et al. 2021

Self Cite

View full text Add to dashboard Cite

Communication systems are affected by channel distortions. Impulsive noise is one of the significant factors for channel impairments. The standard additive white Gaussian noise (AWGN) channel model and conventional estimation algorithms like least mean square (LMS) and its variants tend to be ineffective under such conditions. This paper presents a robust adaptive channel estimation algorithm using the Geman-McClure estimator in a diffusion-based distributed network. The analytical study on mean stability and mean square analysis is carried out under two separate noise statistics: Symmetric αstable (SαS) and Bernoulli-Gaussian (BG) distribution. The computer simulations confirm the proposed algorithm's competitive robustness compared to the Maximum Correntropy Criterion and Minimum Kernel Risk Sensitive Loss algorithms at a high impulsive noise environment without exponential cost function. Further, the efficiency is also verified by simulating the bit error rate by designing a minimum mean square error (MMSE) equalizer with the estimated coefficients.

show abstract

Section: Proposed Diffusion Geman-mcclure Algorithm and Computational Complexitymentioning

confidence: 99%

“…A windowbased Lorentzian adaptive algorithm that does not have an exponential factor is proposed, where the window length parameter regulates the convergence rate [37]. It has been extended to distributed channel estimation [38]. Despite their limitations, these algorithms accomplish their intended aim in robust estimation.…”

Section: Introductionmentioning

confidence: 99%

Adaptive Geman-McClure Estimator for Robust Distributed Channel Estimation

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…To handle outliers in the desired data, many robust techniques based on robust statistics have been reported in the literature. The techniques based on Wilcoxon norm [25–29 ], Huber loss [30–32 ], error non‐linearity [33, 34 ], Lorentzian norm [35 ], maximum correntropy criterion [36 ], the least logarithmic absolute difference [37 ], least mean p‐power [38, 39 ], minimum disturbance [40 ] and mixed p ‐norm [41, 42 ] are found to be robust against outliers in the desired data. However, all these methods assume that the input data is uncontaminated, which may not be true in the practical scenarios.…”

Section: Introductionmentioning

confidence: 99%

Sparse distributed learning based on diffusion minimum generalised rank norm

Modalavalasa

Sahoo

2020

IET signal process.

View full text Add to dashboard Cite

“…We are interested in developing signal detection methods for a large of class of random processes which include the Gaussian process as a special element. Studies and experiments results show that the class of alpha-stable distributions is better for modeling impulsive noise than Gaussian distribution in real life applications [12][13][14][15][16]. By the generalized central limit theorem, they are the only class of distributions that can be the limit distributions for sums of i.i.d.…”

Section: Introductionmentioning

confidence: 99%

“…By the generalized central limit theorem, they are the only class of distributions that can be the limit distributions for sums of i.i.d. random variables [16]. Another important property of the stable laws is the stability property.…”

Section: Introductionmentioning

confidence: 99%

Robust Signals Detection Algorithm Based on Cyclostationarity in Impulsive Noise

2020

International Journal of Circuits, Systems and Signal Processing

View full text Add to dashboard Cite

A robust method for detecting the communication signals impinging on an antenna with interference and non-Gaussian impulsive noise is introduced in this paper. Degradation of the conventional cyclic detector which based on max-output-SNR criterion in impulsive noise is shown both theoretically and experimentally. By fusing second-order cyclostationarity and fractional lower-order statistics, a type of cyclic fractional lower-order statistics is developed which is defined for exploiting cyclostationarity property. Then, a new robust type of detection algorithm is developed using the theory of optimal filtering based on max-output-SNR criterion and alpha-stable distribution, including the fractional lower-order cyclic matched filter, which is formulated for detecting the communication signals in the presence of interference and non-Gaussian alpha-stable distribution impulsive noise. It is shown that the new method is robust to Gaussian and non-Gaussian impulsive noises, and is immune to the interfering signals which occupy the same spectral band as that of the received signal. Simulation results show the robustness and effectiveness of the proposed algorithm.

show abstract

Robust distributed Lorentzian adaptive filter with diffusion strategy in impulsive noise environment

Cited by 30 publications

References 36 publications

Adaptive Geman-McClure Estimator for Robust Distributed Channel Estimation

Adaptive Geman-McClure Estimator for Robust Distributed Channel Estimation

Sparse distributed learning based on diffusion minimum generalised rank norm

Robust Signals Detection Algorithm Based on Cyclostationarity in Impulsive Noise

Contact Info

Product

Resources

About