Adaptive Whitening Research Articles

Adaptive Whitening and Feature Gradient Smoothing-Based Anti-Sample Attack Method for Modulated Signals in Frequency-Hopping Communication

Adaptive Whitening and Feature Gradient Smoothing-Based Anti-Sample Attack Method for Modulated Signals in Frequency-Hopping Communication

A Matched Filter for Spaceborne GNSS-R Based Sea-Target Detection

The motion of any potential sea-targets in a sequence of delay Doppler maps (DDMs) collected by a spaceborne global navigation satellite system-reflectometry (GNSS-R) receiver can be computed as a priori information due to the receiver velocity dominating that of any realistic sea-targets. This can be exploited to design a matched filter to detect sea-targets in a DDM sequence without the requirement for any (pre)detection and can be considered a track-before-detect technique. In this article, such a filter is developed and applied to DDMs collected by TechDemoSat-1, and the detection of sea-ice is demonstrated. A frequency analysis of simulated and actual DDM sequences containing a sea-target is carried out. We show that the stationary sea-clutter is bound to the zero time–frequency plane in the 3-D frequency response of the DDM time series, while only $\approx $ 13.8% of the target’s energy is located in this plane. This finding motivates the application of an adaptive whitening filter to blindly suppress the temporally correlated clutter in the DDM sequence before applying the matched filter to the prewhitened sequence containing a spatially fixed point-like response arising from the edge of an ice sheet, which is equivalent to a point-like sea-target such as a ship. This results in a signal-to-clutter ratio of $\approx $ 12 dB being achieved when processing only the weak target responses at nonspecular locations originating from the spatially fixed edge of the ice sheet. This is the first technique to exploit the delay and Doppler dynamics of a target to obtain a coherent integration gain.

Recursive Spatial Multiplexing with Adaptive Interference Whitening

This article describes how recursive spatial multiplexing (RSM) is a closed-loop multiple-input multiple-output (MIMO) structure for achieving the capacity offered by MIMO channels with a low-complexity detector. The authors investigate how to make RSM able to provide a bit-error rate performance, which is robust against different types and levels of interference. The interference arising from simultaneous transmission of information signals is taken into account in the RSM scheme at the receiver using a whitening approach. Here, the covariance matrix is estimated and used subsequently for defining the retransmission subspace identifier to be fed back to the transmitter. The performance of this adaptive RSM scheme is compared with standard linear detection schemes like zero-forcing and minimum mean-squared error receivers. It turns out that the adaptive interference whitening substantially improves the bit-error rate performance. Moreover, adaptive RSM leads to a performance being independent of the correlation coefficient of the interference signals.

Body posture differentially impacts on visual attention towards tool, graspable, and non-graspable objects.

Viewed objects have been shown to afford suitable actions, even in the absence of any intention to act. However, little is known as to whether gaze behavior (i.e., the way we simply look at objects) is sensitive to action afforded by the seen object and how our actual motor possibilities affect this behavior. We recorded participants' eye movements during the observation of tools, graspable and ungraspable objects, while their hands were either freely resting on the table or tied behind their back. The effects of the observed object and hand posture on gaze behavior were measured by comparing the actual fixation distribution with that predicted by 2 widely supported models of visual attention, namely the Graph-Based Visual Saliency and the Adaptive Whitening Salience models. Results showed that saliency models did not accurately predict participants' fixation distributions for tools. Indeed, participants mostly fixated the action-related, functional part of the tools, regardless of its visual saliency. Critically, the restriction of the participants' action possibility led to a significant reduction of this effect and significantly improved the model prediction of the participants' gaze behavior. We suggest, first, that action-relevant object information at least in part guides gaze behavior. Second, postural information interacts with visual information to the generation of priority maps of fixation behavior. We support the view that the kind of information we access from the environment is constrained by our readiness to act. (PsycINFO Database Record

Adaptive whitening of ambient ocean noise with narrowband signal preservation.

Passive underwater listening devices are often deployed to listen for narrowband signals of interest in time-varying background ocean noise. Such tonals are generated mechanically by ships, submarines, and machines, or acoustically by aquatic wildlife. Quantization of the sensor data for storage or low bit-rate transmission adds white noise which can overwhelm weak narrowband signals if the background noise is sufficiently colored. Whitening the background noise prior to quantization can reduce the detrimental effects, but the whitening process must preserve any tonals in the signal for maximum effectiveness. Existing adaptive whitening techniques make no effort to avoid suppressing tonals in the whitening process, while existing spectral separation methods fail to whiten background noise. The proposed methods perform adaptive whitening of background ambient noise while preserving narrowband tones at their original signal-to-noise ratios. The proposed methods are shown to outperform combinations of existing partial solutions both subjectively and by evaluating the objective criteria introduced. The stability and convergence properties of the proposed algorithms match or surpass those of existing well-known adaptive algorithms.

Dynamic Whitening Saliency.

General dynamic scenes involve multiple rigid and flexible objects, with relative and common motion, camera induced or not. The complexity of the motion events together with their strong spatio-temporal correlations make the estimation of dynamic visual saliency a big computational challenge. In this work, we propose a computational model of saliency based on the assumption that perceptual relevant information is carried by high-order statistical structures. Through whitening, we completely remove the second-order information (correlations and variances) of the data, gaining access to the relevant information. The proposed approach is an analytically tractable and computationally simple framework which we call Dynamic Adaptive Whitening Saliency (AWS-D). For model assessment, the provided saliency maps were used to predict the fixations of human observers over six public video datasets, and also to reproduce the human behavior under certain psychophysical experiments (dynamic pop-out). The results demonstrate that AWS-D beats state-of-the-art dynamic saliency models, and suggest that the model might contain the basis to understand the key mechanisms of visual saliency. Experimental evaluation was performed using an extension to video of the well-known methodology for static images, together with a bootstrap permutation test (random label hypothesis) which yields additional information about temporal evolution of the metrics statistical significance.

A high-resolution processing technique for improving the energy of weak signal based on matching pursuit

<p>This paper proposes a new method to improve the resolution of the seismic signal and to compensate the energy of weak seismic signal based on matching pursuit. With a dictionary of Morlet wavelets, matching pursuit algorithm can decompose a seismic trace into a series of wavelets. We abstract complex-trace attributes from analytical expressions to shrink the search range of amplitude, frequency and phase. In addition, considering the level of correlation between constituent wavelets and average wavelet abstracted from well-seismic calibration, we can obtain the search range of scale which is an important adaptive parameter to control the width of wavelet in time and the bandwidth of frequency. Hence, the efficiency of selection of proper wavelets is improved by making first a preliminary estimate and refining a local selecting range. After removal of noise wavelets, we integrate useful wavelets which should be firstly executed by adaptive spectral whitening technique. This approach can improve the resolutions of seismic signal and enhance the energy of weak wavelets simultaneously. The application results of real seismic data show this method has a good perspective of application.</p>

Horizontal tuning for faces originates in high-level Fusiform Face Area

Recent work indicates that the specialization of face visual perception relies on the privileged processing of horizontal angles of facial information. This suggests that stimulus properties assumed to be fully resolved in primary visual cortex (V1; e.g., orientation) in fact determine human vision until high-level stages of processing. To address this hypothesis, the present fMRI study explored the orientation sensitivity of V1 and high-level face-specialized ventral regions such as the Occipital Face Area (OFA) and Fusiform Face Area (FFA) to different angles of face information.Participants viewed face images filtered to retain information at horizontal, vertical or oblique angles. Filtered images were viewed upright, inverted and (phase-)scrambled. FFA responded most strongly to the horizontal range of upright face information; its activation pattern reliably separated horizontal from oblique ranges, but only when faces were upright. Moreover, activation patterns induced in the right FFA and the OFA by upright and inverted faces could only be separated based on horizontal information. This indicates that the specialized processing of upright face information in the OFA and FFA essentially relies on the encoding of horizontal facial cues.This pattern was not passively inherited from V1, which was found to respond less strongly to horizontal than other orientations likely due to adaptive whitening. Moreover, we found that orientation decoding accuracy in V1 was impaired for stimuli containing no meaningful shape.By showing that primary coding in V1 is influenced by high-order stimulus structure and that high-level processing is tuned to selective ranges of primary information, the present work suggests that primary and high-level levels of the visual system interact in order to modulate the processing of certain ranges of primary information depending on their relevance with respect to the stimulus and task at hand.

Discrimination between the move and stop of a lift based on accelerometer signals

A solution to the problem of reliable discrimination between the move and stop states of a mine lift is proposed based on processing signals of an accelerometer mounted in the lift cabin. The accelerometer signal is processed using a combination of adaptive whitening algorithms and a nonparametric Neyman–Pearson test that provides an effective solution of the problem under a priori nonparametric uncertainty conditions.

Wavelet-Based Adaptive Detection of Magnetic Anomaly Signal Contaminated by 1/f Noise

Magnetic anomaly detection is a passive method for detection of a ferromagnetic target, and its performance is often limited by external noise with a power spectral density of 1/fa, (0<a<2). In consideration of this kind of noise is non-stationary, self-similarity and long-range correlation, an effective adaptive detection based on the wavelet transform is proposed in this paper. The discrete wavelet transform of the noisy signal is chosen as the inputs of the adaptive whitening filter, and then decomposed by the orthonormal basis functions (OBFs) and the energy signal was taken out for threshold detection. The simulation results show that the algorithm improves the effect for detecting weak magnetic anomaly signal contaminated by 1/fanoise.

Sigmoidal radial basis function ANN for QRS complex detection

Abstract This work proposes a QRS complex detector which works on high amplitude. An optimized nonlinear adaptive whitening filter based on sigmoidal radial basis function (SRBF) is used to cleverly accentuate the QRS complexes and suppress the noises present in the ECG signal. An M-estimator based objective function and a new contextually based centering technique are used to control the behavior of the whitening filter so that it totally suppresses low frequencies segments and preserves almost unchanged the high frequencies ones. The residual issued from the whitening filter is then fed to a matched filter to enhance the signal-to-noise ratio (SNR) which is yet an important step that allows achieving good detection performance. Additional nonlinear transformations are performed before the decision stage namely squaring and moving average filtering. The validation of the detector is performed over the whole MIT/BIH database files. The algorithm has produced an average detection error rate of 0.28%, sensitivity of 99.82% and specificity of 99.91%.

Stochastic Gradient Based Variable Momentum Factor Algorithm for Adaptive Whitening

In view of the convergence speed in adaptive whitening technique, by adding a momentum term with a variable momentum factor, a fast adaptive whitening algorithm is proposed. First, the presented algorithm incorporates a momentum term into the adaptive whitening algorithm to accelerate the convergence speed. Then, a variable momentum factor to improve the allover performance of the whitening system is obtained based on the stochastic gradient of the cost function. Experimental results demonstrate the good performance of the proposed fast algorithm both in stationary and non-stationary conditions.

Saliency from hierarchical adaptation through decorrelation and variance normalization

This paper presents a novel approach to visual saliency that relies on a contextually adapted representation produced through adaptive whitening of color and scale features. Unlike previous models, the proposal is grounded on the specific adaptation of the basis of low level features to the statistical structure of the image. Adaptation is achieved through decorrelation and contrast normalization in several steps in a hierarchical approach, in compliance with coarse features described in biological visual systems. Saliency is simply computed as the square of the vector norm in the resulting representation. The performance of the model is compared with several state-of-the-art approaches, in predicting human fixations using three different eye-tracking datasets. Referring this measure to the performance of human priority maps, the model proves to be the only one able to keep the same behavior through different datasets, showing free of biases. Moreover, it is able to predict a wide set of relevant psychophysical observations, to our knowledge, not reproduced together by any other model before.

Physically feasible adaptive whitening filter

Feasibilities of constructing a whitening filter for additive interference with an arbitrary energy spectrum have been considered under conditions of eliminating the spectrum shape distortions of useful signal in the proposed filter.

Source separation and tracking for time-varying systems

Many situations arise when it is desired to recover from a noisy mixture of data both an estimate of a propagating source signal as well as its angular position. Accordingly, the problem of source separation and tracking for time-varying systems of moving sources is studied. Using techniques borrowed from independent component analysis (ICA) and numerical linear algebra, a processing solution is formulated. The ULV decomposition (ULVD), a member of the complete orthogonal decompositions (CODs) family is used to normalize and decorrelate the data. Since the ULVD is known to stably update and downdate with time-varying data, this step is named ULVD adaptive whitening. The ICA-based EASI (equivariant adaptive separation based on independence) algorithm is adapted to provide not only separated sources but direction-of-arrival (DOA) estimates. Developed improvements to this technique are an adaptive filter method for estimating the mixing matrix from the newly estimated signals as well as a simple cross-over detection scheme. Finally, the developed solutions are tested in simulation examples using meaningful performance measures.

Adaptive MLSD receiver employing noise correlation

A per-survivor processing (PSP) maximum likelihood sequence detection (MLSD) receiver is developed for a fast time-varying frequency-selective Rayleigh fading channel with coloured additive noise, which follows an autoregressive (AR) model with unknown parameters. The correlation between noise samples is exploited to considerably enhance the performance of the communications. The maximum likelihood criterion is employed based on unknown noise parameters. This criterion has some desired properties, e.g. it has a unique joint minimum at the true values of the channel and the noise parameters. The new PSP–MLSD algorithm detects the input data and jointly estimates the noise and the channel parameters all together. The proposed structure can be viewed as a traditional PSP–MLSD receiver combined with an adaptive whitening filter. In a coloured noise environment, this scheme offers a faster tracking property, more accurate estimation of the channel and a substantially lower error probability compared with the traditional PSP–MLSD structure. The signal-to-noise ratio (SNR) improvement achieved by the proposed receiver, which can be called the noise whitening gain (NWG), is almost equal to the ratio of the energy of the additive noise to the energy of the unpredictable noise component. The square of the NWG gives also an accurate approximation for the bit error rate (BER) improvement ratio obtained by using the proposed algorithm compared with the traditional one.

Abrupt change detection in power system fault analysis using adaptive whitening filter and wavelet transform

This paper describes the application of the adaptive whitening filter and the wavelet transform used to detect the abrupt changes in the signals recorded during disturbances in the electrical power network in South Africa. Main focus has been to estimate exactly the time-instants of the changes in the signal model parameters during the pre-fault condition and following events like initiation of fault, circuit-breaker opening, auto-reclosure of the circuit-breakers. The key idea is to decompose the fault signals, de-noised using the adaptive whitening filter, into effective detailed and smoothed version using the multiresolution signal decomposition technique based on discrete wavelet transform. Then we apply the threshold method on the decomposed signals to estimate the change time-instants, segmenting the fault signals into the event-specific sections for further signal processing and analysis. This paper presents application on the recorded signals in the power transmission network of South Africa.

Nonlinear adaptive blind whitening for MIMO channels

A nonlinear adaptive whitening method is proposed for blind deconvolution of MIMO systems by whitening the received signals in both time and space, with a highly nonlinear function of the past output data. The whitened signals are ISI-free and can be viewed as outputs of a memoryless paraunitary mixing system. The convergence of the proposed recursive algorithm is proved. Numerical simulation shows that the whitening method proposed in the paper works well, even if the output signal is corrupted by additive noise.

Detecting landmines with ground-penetrating radar using feature-based rules, order statistics, and adaptive whitening

An approach to detecting landmines using ground-penetrating radar (GPR) based on feature-based rules, order statistics, and adaptive whitening (FROSAW) is described. FROSAW relies on independent adaptation of whitening statistics in different depths and combining feature-based methods with anomaly detection using rules. Constant false alarm rate (CFAR) detectors are used for anomaly detection on the depth-dependent adaptively whitened data. A single CFAR confidence measure is computed via a function of order statistics. Anomalies are detected at locations with high CFAR confidence. Depth-dependent features are computed on regions containing anomalies. Rules based on the features are used to reject alarms that do not exhibit mine-like properties. The utility of combining the CFAR and feature-based methods is evaluated. The algorithms and analysis are applied to data acquired from tens of thousands of square meters from several outdoor test sites with a state-of-the-art array of GPR sensors.

Multiresolution techniques for the detection of gravitational-wave bursts

We present two search algorithms that implement logarithmic tiling of the time–frequency plane in order to efficiently detect astrophysically unmodelled bursts of gravitational radiation. The first is a straightforward application of the dyadic wavelet transform. The second is a modification of the windowed Fourier transform which tiles the time–frequency plane for a specific Q. In addition, we also demonstrate adaptive whitening by linear prediction, which greatly simplifies our statistical analysis. This is a methodology paper that aims to describe the techniques for identifying significant events as well as the necessary pre-processing that is required in order to improve their performance. For this reason we use simulated LIGO noise in order to illustrate the methods and to present their preliminary performance.