A Generalized Normalized Gradient Descent Algorithm

Mandic, Danilo P.

doi:10.1109/lsp.2003.821649

Cited by 226 publications

(129 citation statements)

References 9 publications

Supporting

Mentioning

128

Contrasting

Unclassified

Order By: Relevance

“…Figure 1 shows the MSD curves of APA, VSS-APA in [17], proposed VSS-APA, ES [19] and GNGD algorithm [16] for colored Gaussian input and M = 20. The parameter K was set to 4 and different values for μ were used in APA.…”

Section: Resultsmentioning

confidence: 99%

“…As we can see the performance of proposed VSS-APA has both fast convergence speed and low steady-state MSD features. [16] (f) ES [19] (a) (b) (c) Input : Colored Gaussian signal M=20, K=4 (e) (f) (d) Figure 1 Comparing the MSD curves of APA with high and low step-sizes, variable step-size proposed in [17], proposed VSS-APA, GNGD [16]and ES [19]algorithms with M = 20, K = 4 and colored Gaussian signal as input.…”

Section: Resultsmentioning

confidence: 99%

“…The parameter K and the number of blocks were set to 10, and 5 respectively. In this figure the parameter S was set to 3, and different values [17] (d) Proposed-VSS-APA (e) GNGD, 0 =0.5, =0.15 [16] (f) ES [19] (a) (b) (c) (e) (f) (…”

Section: Resultsmentioning

confidence: 99%

“…In [16], a generalized normalized gradient descent (GNGD) algorithm for linear finite-impulse response (FIR) adaptive filters was introduced that represents an extension of the NLMS algorithm by means of an additional gradient adaptive term in the denominator of the learning rate of NLMS. The simulation results show that the GNGD algorithm is robust to significant variations of initial values of its parameters.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A family of variable step-size affine projection adaptive filter algorithms using statistics of channel impulse response

Abadi

Arani

2011

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

This paper extends the recently introduced variable step-size (VSS) approach to the family of adaptive filter algorithms. This method uses prior knowledge of the channel impulse response statistic. Accordingly, optimal stepsize vector is obtained by minimizing the mean-square deviation (MSD). The presented algorithms are the VSS affine projection algorithm (VSS-APA), the VSS selective partial update NLMS (VSS-SPU-NLMS), the VSS-SPU-APA, and the VSS selective regressor APA (VSS-SR-APA). In VSS-SPU adaptive algorithms the filter coefficients are partially updated which reduce the computational complexity. In VSS-SR-APA, the optimal selection of input regressors is performed during the adaptation. The presented algorithms have good convergence speed, low steady state mean square error (MSE), and low computational complexity features. We demonstrate the good performance of the proposed algorithms through several simulations in system identification scenario.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A family of variable step-size affine projection adaptive filter algorithms using statistics of channel impulse response

Abadi

Arani

2011

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

show abstract

“…In Section 3, we first apply a NLMS algorithm, similar to those developed by Rahman [7] Mandic [8] and Valin [9], to a sine wave embedded with random noise as a testing benchmark. We then filter out electromyography (EMG) noise from a CPS recorded ECG signal.…”

Section: Introductionmentioning

confidence: 99%

Adaptive Filtering for Detecting Myocardial Infarction using Noninvasive Conducting Polymer Composite Sensors

Montavon

Ergezer²,

Lozovyy³

et al. 2012

Imaging and Signal Processing in Health Care and Technology / 772: Human-Computer Interaction / 773: Communication, Internet An

View full text Add to dashboard Cite

Continuous electrocardiographic (ECG) monitoring using conducting polymer composite sensors (CPS) presents a non-invasive way to detect cardiac irregularities such as myocardial infarction (MI). Electromyography (EMG), which measures muscle activity in the human body, has a frequency range that overlaps that of the ECG wave. As a result, both EMG and ECG data are present when CPSs collect ECG signals. When measuring ECG waves of an individual during motion, we account for EMG by removing the motion artifact from the ECG signal. With the use of a normalized least mean square (NLMS) algorithm and known signal characteristics, we show that EMG noise can be successfully filtered from an ECG signal that is collected using our CPSs in the standard 12 lead ECG placement.Our software produces a diagnostic-friendly ECG signal and then determines the patient's heart rate.When applied to the arrhythmia database from the Massachusetts Institute of Technology and Beth Israel Hospital (MIT-BIH), our heartbeat detection logic has an accuracy of 99.6% with only 199 false beats and 240 missed beats out of 109,494 total heartbeats taken from 48 individual recordings. KEY WORDSSignal processing of physiological signals, wearable devices, biomedical signal processing, medical signal processing.

show abstract

A unifying framework for the analysis of proportionate NLMS algorithms

Jelfs

Mandic

2014

Adaptive Control & Signal

View full text Add to dashboard Cite

Despite being a de facto standard in sparse adaptive filtering, the two most important members of the class of proportionate normalised least mean square (PNLMS) algorithms are introduced empirically. Our aim is to provide a unifying framework for the derivation of PNLMS algorithms and their variants with an adaptive step-size. These include algorithms with gradient adaptive learning rates and algorithms with adaptive regularisation parameters. Convergence analysis is provided for the proportionate least mean square (PLMS) algorithm in both the mean and mean square sense and bounds on its parameters are derived. An alternative, more insightful approach to the convergence analysis is also presented and is shown to provide an estimate of the optimal step-size of the PLMS. Incorporating the so obtained step-size into the PLMS gives the standard PNLMS together with a unified framework for introducing other adaptive learning rates. Simulations on benchmark sparse impulse responses support the approach. ‡ Namely that the input signal and filter coefficient vectors are zero mean, stationary, jointly normal and with finite moments; the successive increments of tap weights are independent of one another, and the error and input vector sequences are statistically independent of one another [1]. ¶ Notice from (40)-(41) that as x.k/G.k/x.k/ ! 0, then effectively .k/ ! 1. By making and Á.0/ large, we achieve the same effect, allowing a convenient comparison of the fixed, linear and variable-" methods.

show abstract

A Generalized Normalized Gradient Descent Algorithm

Cited by 226 publications

References 9 publications

A family of variable step-size affine projection adaptive filter algorithms using statistics of channel impulse response

A family of variable step-size affine projection adaptive filter algorithms using statistics of channel impulse response

Adaptive Filtering for Detecting Myocardial Infarction using Noninvasive Conducting Polymer Composite Sensors

A unifying framework for the analysis of proportionate NLMS algorithms

Contact Info

Product

Resources

About