Moving Average Filter Research Articles

An Evaluation of a Novel Software Tool for Detecting Changes in Physiological Monitoring

We have developed a software tool (iAssist) to assist clinicians as they monitor the physiological data that guide their actions during anesthesia. The system tracks the statistical properties of multiple dynamic physiological processes and identifies new trend patterns. We report our initial evaluation of this tool (in pseudo real-time) and compare the detection of trend changes to a post hoc visual review of the full trend. We suggest a combination of criteria by which to evaluate the performance of monitoring devices that aim to enhance trend detection. Nineteen children and 28 adults consented to be included in the study, encompassing more than 68 h of anesthesia. In each surgical case, an anesthesiologist reported all perceived clinical changes in monitoring in real-time. A trained observer simultaneously documented the verbally reported changes and every anesthesiologist action. The same cases were subsequently evaluated offline (in pseudo real-time) by a novel software tool (iAssist). Heart rate, end-tidal carbon dioxide, exhaled minute ventilation, and respiratory rate were modeled using a dynamic linear growth model whose noise distribution was estimated by an adaptive Kalman filter based on a recursive expectation-maximization method. Changes were detected by adaptive local Cumulative Sum testing. Changes in the mean arterial noninvasive blood pressures and oxygen saturation were detected using adaptive Cumulative Sum testing on a filtered residual from an exponentially weighted moving averaging filter. In post hoc analysis, each change detected by iAssist was graded independently by two clinicians using a graphical display of the whole case. Missed changes were recorded. The iAssist software tool detected 869 true positive changes (at an average of 12.76/h) with a sensitivity of 0.91 and positive predictive value of 0.87. The post hoc review identified 91 missed changes (at an average of 1.34/h), resulting in an overall ratio of true positive rates to false-negative rates of 9.55. The clinicians in real-time reported 209 changes in trend (at an average of 3.07/h). The algorithms perform favorably compared with a visual inspection of the complete trend. Further research is needed to identify when and how to draw the clinician's attention to these changes.

Filtering electrocardiographic signals using an unbiased and normalized adaptive noise reduction system

We present a novel unbiased and normalized adaptive noise reduction (UNANR) system to suppress random noise in electrocardiographic (ECG) signals. The system contains procedures for the removal of baseline wander with a two-stage moving-average filter, comb filtering of power-line interference with an infinite impulse response (IIR) comb filter, an additive white noise generator to test the system’s performance in terms of signal-to-noise ratio (SNR), and the UNANR model that is used to estimate the noise which is subtracted from the contaminated ECG signals. The UNANR model does not contain a bias unit, and the coefficients are adaptively updated by using the steepest-descent algorithm. The corresponding adaptation process is designed to minimize the instantaneous error between the estimated signal power and the desired noise-free signal power. The benchmark MIT-BIH arrhythmia database was used to evaluate the performance of the UNANR system with different levels of input noise. The results of adaptive filtering and a study on convergence of the UNANR learning rate demonstrate that the adaptive noise-reduction system that includes the UNANR model can effectively eliminate random noise in ambulatory ECG recordings, leading to a higher SNR improvement than that with the same system using the popular least-mean-square (LMS) filter. The SNR improvement provided by the proposed UNANR system was higher than that provided by the system with the LMS filter, with the input SNR in the range of 5–20 dB over the 48 ambulatory ECG recordings tested.

Noise estimation in voice signals using short-term cepstral analysis

Cepstral-based estimation is used to provide a baseline estimate of the noise level in the logarithmic spectrum for voiced speech. A theoretical description of cepstral processing of voiced speech containing aspiration noise, together with supporting empirical data, is provided in order to illustrate the nature of the noise baseline estimation process. Taking the Fourier transform of the liftered (filtered in the cepstral domain) cepstrum produces a noise baseline estimate. It is shown that Fourier transforming the low-pass liftered cepstrum is comparable to applying a moving average (MA) filter to the logarithmic spectrum and hence the baseline receives contributions from the glottal source excited vocal tract and the noise excited vocal tract. Because the estimation process resembles the action of a MA filter, the resulting noise baseline is determined by the harmonic resolution (as determined by the temporal analysis window length) and the glottal source spectral tilt. On selecting an appropriate temporal analysis window length the estimated baseline is shown to lie halfway between the glottal excited vocal tract and the noise excited vocal tract. This information is employed in a new harmonics-to-noise (HNR) estimation technique, which is shown to provide accurate HNR estimates when tested on synthetically generated voice signals.

Estimation of thermal deformation in machine tools using the hybrid autoregressive moving-average - neural network model

A hybrid model, which is composed of an autoregressive moving-average (ARMA) filter and a feedforward neural network (FNN), is proposed to increase prediction accuracy and to reduce learning time for the estimation of thermal deformations in a machine tool. The ARMA filter is used to yield state variables which establish the relationship between the present and past states of thermal deformations for the reservation of the influences of past temperatures and deformation. Otherwise, the quantity of FNN inputs is very vast because of the data needed for the non-linear system. These state variables, which are estimated by past measured temperatures and past estimated deformation, serve as inputs of the FNN. The algorithms of this hybrid model are presented and verified by the experimental results; also, the prediction accuracy is compared with the ARMA and FNN independently for the same learning iterations.

Archery performance level and repeatability of event-related EMG

The purpose of the current study was to compare the repeatability of electromyographic linear envelopes (LE) of archery groups. Surface electromyography (EMG) signals of musculus flexor digitorum superficialis (MFDS) and extensor digitorum (MED) of 23 participants (seven skilled, six beginner archers and ten non-archers) were recorded during archery shooting. Two-second periods (clicker falls at first second) of 12 shots’ EMG data were recorded, full-wave rectified and filtered (60 ms moving-average filter) for each participant’s drawing arm. Repeatability was investigated by using a statistical criterion, variance ratio (VR). Archers’ performances were evaluated in terms of FITA scores. The results showed that FITA scores were significantly correlated to the VRs of MFDS and MED. EMG LEs were more repeatable among archers than non-archers. Therefore, we inferred that VRs of MFDS and MED might be important variables for (a) assessing shooting techniques, (b) evaluation of archers’ progress, and (c) selection of talented archers.

MFL signals and artificial neural networks applied to detection and classification of pipe weld defects

This work evaluates the use of artificial neural networks (ANNs) for pattern recognition of magnetic flux leakage (MFL) signals in weld joints of pipelines obtained by intelligent pig. Initially the ANNs were used to distinguish the pattern signals with non-defect (ND) and signals with defects (D) along of the weld bead. In the next step the ANNs were applied to classify signal patterns with three types of defects in the weld joint: external corrosion (EC), internal corrosion (IC) and lack of penetration (LP). The defects were intentionally inserted in the weld bead of a pipeline of API 5L-X65 steel with an outer diameter of 304.8 mm. In this way, the MFL signal itself, digitized with 1025 points, was used as the ANN input. Initially the signals were used as inputs for the neural network without any type of pre-processing, later four types of pre-processing were applied to the signals: Fourier analysis, Moving-average filter, Wavelet analysis and Savitzky–Golay filter. Signal processing techniques were employed to improve the performance of the neural networks in distinguishing between the defect classes. The results showed that it is possible to classify signals of classes D and ND using ANN with very efficient results (94.2%), as well as for corrosion (CO) and LP signals (92.5%). Also it is possible to classify the defect pattern signals: EC, IC and LP using neural networks with an average rate of success of 71.7% for the validation set.

Filtering device and method for reducing noise in electrical signals, in particular acoustic signals and images

A neuro-fuzzy filter device that implements a moving-average filtering technique in which the weights for final reconstruction of the signal are calculated in a neuro-fuzzy network according to specific fuzzy rules. The fuzzy rules operate on three signal features for each input sample. The signal features are correlated to the position of the sample in the considered sample window, to the difference between a sample and the sample at the center of the window, and to the difference between a sample and the average of the samples in the window. The filter device for the analysis of a voice signal includes a bank of neuro-fuzzy filters. The signal is split into a number of sub-bands, according to wavelet theory, using a bank of analysis filters including a pair of FIR QMFs and a pair of downsamplers; each sub-band signal is filtered by a neuro-fuzzy filter, and then the various sub-bands are reconstructed by a bank of synthesis filters including a pair of upsamplers, a pair of FIR QMFs, and an adder node.

Operation of Static Series Compensator Under Distorted Utility Conditions

The static series compensator (SSC) has shown a significant capability to mitigate voltage dips, which are the most severe problem to sensitive loads. Also, it has been declared that the function of the SSC can be extended to work as a series active filter. This work proposes a moving-average filter to detect the fundamental component of the measured voltages and currents (needed to control the SSC) while using a double vector control algorithm to improve the transient performance of the SSC. This is made in order to accurately control the fundamental voltage component at the load terminals in the case of distorted grid voltage. Furthermore, a selective harmonic compensation strategy is applied to filter out the grid harmonics. The operation of the SSC under distorted utility conditions and voltage dips is discussed. The validity of the proposed controller is verified by experiments, carried out on a 10-kV SSC laboratory setup. Experimental results have shown the ability of the SSC to mitigate voltage dips and harmonics. It is also shown that the proposed controller has improved the transient performance of the SSC even under distorted utility conditions.

GPS fault detection and exclusion using moving average filters

A new approach based on the moving average (MA) is proposed to perform satellite failure detection and exclusion (FDE). By taking the average of the last few sums of the squares of the GPS range residual errors, the MA filter is applied to speed up failure detection. The detection threshold cannot be obtained directly because the cumulative distribution of a random process with an MA filter is unknown. Therefore, the Markov chain approach is applied to resolve the threshold. In addition, variations in the number of visible satellites may cause problems in data fusion. The probability integral transformation (PIT) method is adopted to overcome this. After a satellite failure is detected, the multivariate MA filter is used to reduce the incorrect exclusion rate (IER) by taking the average of the last few parity vectors. Simulation results show that, in comparison with the conventional least-squares-residuals method, the MA filter demonstrates higher performance in detecting small failures and a similar level of performance in detecting large failures. Moreover, simulation results also verify that the proposed method has lower IER than the conventional parity space method.

A unified method for optimizing linear image restoration filters

Image restoration from degraded images lies at the foundation of image processing, pattern recognition, and computer vision, so it has been extensively studied. A large number of image restoration filters have been devised so far. It is known that a certain filter works excellently for a certain type of original image or degradation. However, the same filter may not be suitable for other images, so the selection of filters is exceedingly important in practice. Moreover, if a filter includes adjustable parameters such as the regularization parameter or threshold, its restoration performance relies heavily on the choice of the parameter values. In this paper, we therefore discuss the problem of optimizing the filter type and parameter values. Our method is based on the subspace information criterion (SIC), which is an unbiased estimator of the expected squared error between the restored and original images. Since SIC is applicable to any linear filters, one can optimize the filter type and parameter values in a consistent fashion. Our emphasis in this article is laid on the practical concerns of SIC, such as the noise variance estimation, computational issues, and comparison with existing methods. Specifically, we derive an analytic form of the optimal parameter values for the moving-average filter, which will greatly reduce the computational cost. Experiments with the regularization filter show that SIC is comparable to existing methods in the small degradation case, and SIC tends to outperform existing methods in the severe degradation case.

A simple optimally unbiased MA filter for timekeeping.

In this paper, we address the design of an optimally unbiased moving average (MA) filter for time error online estimating and synchronization in Global Positioning System (GPS)-based timekeeping. We proceed from the slowly changing nature of a time error generated by a local clock, assuming its function to be linear for N points in the average. Employing the inherent property of a simple MA to produce the lowest noise for the given averaging interval and using linear regression as a proper stochastic approximation to evaluate the bias, we find the optimal weighting coefficients. The designed optimally unbiased MA filter and its improved modification produce two times more noise than a simple MA when N >> 1. Both the numerical simulation and the filtering of the real GPS-based time error processes are provided and discussed in detail.

Run by run advanced process control of metal sputter deposition

Metal sputter deposition processes for semiconductor manufacturing are characterized by a decrease in deposition rate from run to run as the sputter target degrades. The goal is to maintain a desired deposition thickness from wafer to wafer and lot to lot. Run by run (RbR) model-based process control (MBPC) has been applied to metal sputter deposition processes at Texas Instruments. RbR MBPC, based on the exponentially weighted moving-average filter, provides the ability to track and compensate for process drifts without a priori assumptions on their magnitude or consistency (from sputter target to sputter target or collimator to collimator). The application of RbR MBPC resulted in an improved C/sub pk/ of 44% for aluminum sputter deposition, while reducing the number of lot-based monitor wafers by a factor of three. The application of RbR MBPC to the titanium sputter deposition process eliminated look-ahead test runs and reduced the number of monitor wafers by a factor of three. At the same time, C/sub pk/ was improved by 10% with the application of RbR MBPC.

Application of non-linear filters based on the median filter to experimental and simulated multiunit neural recordings

Two non-linear, high-pass filters based on the median filter are proposed and tested as substitutes for linear filtering in applications involving multiunit neural recordings. The first, the median-based high-pass (MH) filter, operates by subtracting the output from the input of the median filter; it is aimed at preserving the shape of the impulses. The second, the negative median-based high-pass (NMH) filter, sets at zero the positive values in the output of the MH filter; it is aimed at transforming the impulses into monophasic waves placed on a flat baseline. When applied to experimental recordings and to a template action potential, the two median-based filters clearly outperformed two corresponding procedures based on a linear filter (moving-average filter). They did not produce appreciable distortions of the impulses, whereas their two counterparts induced or enlarged lateral lobes, as is the rule for linear high-pass filters. The recording display was much improved and impulse identification was made easier. When the two filters were applied to simulated recordings and the mean output was estimated by averaging and cross-correlation, a certain degree of performance deterioration was assessed in conditions of sustained activity and/or noise, with a resulting growing similarity to the mean output of the two corresponding, moving-average-based filters.

Measuring Eye Movements During Locomotion: Filtering Techniques for Obtaining Velocity Signals From a Video-based Eye Monitor

Video-based eye-tracking systems are especially suited to studying eye movements during naturally occurring activities such as locomotion, but eye velocity records suffer from broad band noise that is not amenable to conventional filtering methods. We evaluated the effectiveness of combined median and moving-average filters by comparing prefiltered and postfiltered records made synchronously with a video eye-tracker and the magnetic search coil technique, which is relatively noise free. Root-mean-square noise was reduced by half, without distorting the eye velocity signal. To illustrate the practical use of this technique, we studied normal subjects and patients with deficient labyrinthine function and compared their ability to hold gaze on a visual target that moved with their heads (cancellation of the vestibulo-ocular reflex). Patients and normal subjects performed similarly during active head rotation but, during locomotion, patients held their eyes more steadily on the visual target than did subjects.

Does seasonal adjustment distort tests of stationarity? Some small-sample evidence

Starting from the idea that seasonal adjustment via a moving-average filter distorts tests for a unit root, it is investigated whether seasonal adjustment distorts tests that have stationarity as the null. The test due to Kahn and Ogaki (1992) appears to be only marginally distorted by seasonal filtering, but its power is low. In contrast, the test developed by Kwiatowski et al. (1992) has severe size distortions, but its power is relatively high when the distortion is minor.

The relation between the northern polar cap and auroral electrojet geomagnetic indices in the wintertime

The polar cap (PC) index is a measure of the high‐latitude geomagnetic disturbances due to Hall and field‐aligned currents. The index is well correlated with the auroral electrojet AL and AUindices (correlation with the PC index is 76% and 66%, resp.). Here we obtain several types of data‐based models that relate the PC to the AL and AU indices in the wintertime, when the ionospheric conductivity is mostly due to the precipitating particles of the field‐aligned currents. The new models predict AL and AU from PC with correlations much higher than those found by earlier studies. Thus linear moving‐average filters reproduce the observed AL with a correlation of 88% (AU: 75%) while linear autoregressive moving‐average (ARMA) models based on the PC index produce in‐sample single‐step predictions with 98% and 97% correlations with AL and AU respectively. For long‐term, out‐of‐sample prediction, the linear ARMA prediction from the PC index has an asymptotic prediction error which is at least 25% more accurate than the prediction from solar wind input. Nonlinear models are slightly more accurate than their linear counterparts, indicating a weak nonlinearity in the relation between the polar cap and auroral zone indices. The prediction‐observation correlations are sufficiently high that models based on the PC index can be used for specification of the auroral geomagnetic activity.

SIMULATION AND ESTIMATION OF LONG MEMORY CONTINUOUS TIME MODELS

Abstract. Some general properties of long memory continuous time processes are recalled or proved. Methods of simulation are studied. A comparison with the usual discrete time autoregressive fractionally integrated moving‐average filter is made and illustrations are provided. Then, two methods of estimation of the parameters of such a model from a discrete sample are studied, both theoretically and empirically, with Monte Carlo experiments.

Noise, filters and detection limits

The “noise” of Chromatographic baselines has been investigated in regard to the detector, the nature and extent of filtering or smoothing, and the methodologies of qualitative and quantitative assessment: all in order to clarify the role such factors play in the determination and interconversion of some common types of detection limits. This study scrutinizes baselines from the flame photometric detector in single-channel continuous and ten-channel multiplexing versions; it also examines baselines from flame ionization and electron-capture detectors. It makes use of finite impulse response and non-weighted moving-average digital smoothing, as well as three-pole analog filtering. Baseline fluctuations are quantified by the standard deviation derived from the common root-mean-square (RMS) calculation, or from the less common least-squares Gaussian fit; peak-to-peak noise ( N p-p) is estimated by procedures including or excluding presumed outliers. Individual results are expressed as the ratio of N p-p measurement and RMS calculation performed on the same data set. A wide variety of such ratios are then assembled from different detectors, filters, and smoothing conditions. They prove conclusively that -contrary to common belief—the conversion factor between the two types of measurements does vary: usually between 4 and 10, but occasionally even farther. Consequently, the conversion factor between the corresponding two types of detection limits varies as well. The N p-p/RMS ratio depends largely on the detector-of-origin, its condition, and the extent to which noise has been filtered. In contrast, the nature and sophistication of the filter hardly matters: either for the N p-p/RMS ratio or for the practical detection limit. This is because the slow undulations characteristic of heavily filtered baselines represent —at least in the detectors we used-dampened fast noise rather than aboriginally slow noise. Corresponding computer simulations, based on amplitudinally random noise smoothed by stationary boxcar or non-weighted moving-average filters, produce results strikingly similar to actual baselines. Simulated fast RMS noise correlates, as expected, with the square root (log-log slope = 1 2 ) of the filter's time constant. The corresponding slopes for experimental noise are usually close to 1 2 as well. Most importantly, though, the simulated N p-p/RMS ratio varies strongly with the extent of smoothing -thus mimicking and thereby explaining the behavior of the experimental ratio.

Integration of absolute ventricular fibrillation voltage correlates with successful defibrillation.

Previous work has suggested that at higher absolute ventricular fibrillation voltages (AVFV), the heart is more amenable to defibrillation. This study investigated in a canine model whether voltage integration of the AVFV is associated with the defibrillation success rate. The moving-average filter was used to process the ventricular fibrillation (VF) waveform recorded from Lead II of the electrocardiogram (ECG). In seven animals, defibrillation trials were analyzed using a dc shock (DCS) successful approximately 50% of the time when delivered randomly. For each of a total of 84 DCS (40% successes, 60% failures), the fibrillation waveform just prior to DCS was analyzed. The integration of the AVFV waveform was performed over various sample sizes including 1, 4, 8, 16, 64, and 128 ms, as well as the time equal to the mean VF cycle length. The results suggest that dc shocks delivered at instants of higher values of integrated AVFV over the various window sizes are associated with successful defibrillation. Window sizes less than 16 ms appeared to offer the best discrimination. The integration of AVFV over the entire VF cycle length was significantly higher for successful rather than unsuccessful DCS. This interesting observation is consistent with the clinical observation that "coarse" VF (high AVFV) is easier to defibrillate than "fine" VF (low AVFV). The use of voltage integration of AVFV may have potential implications in the improvement of defibrillation success in implantable devices.

ESTIMATION AND BLIND DECONVOLUTION OF AUTOREGRESSIVE SYSTEMS WITH NONSTATIONARY BINARY INPUTS

Abstract. The problem of parameter estimation and blind deconvolution of auto‐regressive (AR) systems with independent nonstationary binary inputs is considered. The estimation procedure consists of applying a moving‐average filter (equalizer) to the observed data and adjusting the parameters of the filter so as to minimize a criterion that measures the binariness of its output. The output sequence itself serves as an estimate of the unobservable binary input of the AR system. Without assuming stationarity of the inputs, it is shown that the proposed method produces a consistent estimator of the AR system not only in the sense of converging to the true parameter as the sample size increases, but also in the sense of attaining the true parameter of the AR system for a sufficiently large sample size. For noisy data, the estimation criterion is modified on the basis of an asymptotic analysis of the effect of the noise. It is shown that the modified criterion is also consistent (in the usual sense) and its variability depends upon the filtered noise. Some simulation results are presented to demonstrate the performance of the proposed method for parameter estimation as well as for blind deconvolution.