Showing papers on "Adaptive filter published in 2004"

Visual tracking and recognition using appearance-adaptive models in particle filters

Abstract: We present an approach that incorporates appearance-adaptive models in a particle filter to realize robust visual tracking and recognition algorithms. Tracking needs modeling interframe motion and appearance changes, whereas recognition needs modeling appearance changes between frames and gallery images. In conventional tracking algorithms, the appearance model is either fixed or rapidly changing, and the motion model is simply a random walk with fixed noise variance. Also, the number of particles is typically fixed. All these factors make the visual tracker unstable. To stabilize the tracker, we propose the following modifications: an observation model arising from an adaptive appearance model, an adaptive velocity motion model with adaptive noise variance, and an adaptive number of particles. The adaptive-velocity model is derived using a first-order linear predictor based on the appearance difference between the incoming observation and the previous particle configuration. Occlusion analysis is implemented using robust statistics. Experimental results on tracking visual objects in long outdoor and indoor video sequences demonstrate the effectiveness and robustness of our tracking algorithm. We then perform simultaneous tracking and recognition by embedding them in a particle filter. For recognition purposes, we model the appearance changes between frames and gallery images by constructing the intra- and extrapersonal spaces. Accurate recognition is achieved when confronted by pose and view variations.

Real time self-adjusting calibration algorithm

Abstract: A method and a system for verifying the integrity of sensor data. Values are received (2100) from a sensor. A parameter related to the values is compared to a threshold (2110). For example the second-order derivative of the values may be compared to a threshold. Additional threshold violations are monitored for further values received from the sensor (2130). The receipt of data is terminated when two parameters exceed the threshold (2170). A method and a system for filtering data based on a quantifier of a variance derived from sensor data. The quantifier being used as input to an adaptive filter for filtering the sensor data. A method and a system for calibrating a sensor by determining the reliability of the sensor data, discarding data values that are unreliable, filtering the data values that have not been discarded and adjusting the output of the sensor using the filtered data values.

Variable step-size NLMS and affine projection algorithms

Abstract: This letter proposes two new variable step-size algorithms for normalized least mean square and affine projection. The proposed schemes lead to faster convergence rate and lower misadjustment error.

Dynamic analysis of neural encoding by point process adaptive filtering

Abstract: Neural receptive fields are dynamic in that with experience, neurons change their spiking responses to relevant stimuli. To understand how neural systems adapt their representations of biological information, analyses of receptive field plasticity from experimental measurements are crucial. Adaptive signal processing, the well-established engineering discipline for characterizing the temporal evolution of system parameters, suggests a framework for studying the plasticity of receptive fields. We use the Bayes' rule Chapman-Kolmogorov paradigm with a linear state equation and point process observation models to derive adaptive filters appropriate for estimation from neural spike trains. We derive point process filter analogues of the Kalman filter, recursive least squares, and steepest-descent algorithms and describe the properties of these new filters. We illustrate our algorithms in two simulated data examples. The first is a study of slow and rapid evolution of spatial receptive fields in hippocampal neurons. The second is an adaptive decoding study in which a signal is decoded from ensemble neural spiking activity as the receptive fields of the neurons in the ensemble evolve. Our results provide a paradigm for adaptive estimation for point process observations and suggest a practical approach for constructing filtering algorithms to track neural receptive field dynamics on a millisecond timescale.

Adaptive stream resource management using Kalman Filters

Abstract: To answer user queries efficiently, a stream management system must handle continuous, high-volume, possibly noisy, and time-varying data streams. One major research area in stream management seeks to allocate resources (such as network bandwidth and memory) to query plans, either to minimize resource usage under a precision requirement, or to maximize precision of results under resource constraints. To date, many solutions have been proposed; however, most solutions are ad hoc with hard-coded heuristics to generate query plans. In contrast, we perceive stream resource management as fundamentally a filtering problem, in which the objective is to filter out as much data as possible to conserve resources, provided that the precision standards can be met. We select the Kalman Filter as a general and adaptive filtering solution for conserving resources. The Kalman Filter has the ability to adapt to various stream characteristics, sensor noise, and time variance. Furthermore, we realize a significant performance boost by switching from traditional methods of caching static data (which can soon become stale) to our method of caching dynamic procedures that can predict data reliably at the server without the clients' involvement. In this work we focus on minimization of communication overhead for both synthetic and real-world streams. Through examples and empirical studies, we demonstrate the flexibility and effectiveness of using the Kalman Filter as a solution for managing trade-offs between precision of results and resources in satisfying stream queries.

Removal of ocular artifacts from electro-encephalogram by adaptive filtering

Abstract: The electro-encephalogram (EEG) is useful for clinical diagnosts and in biomedical research. EEG signals, however, especially those recorded from frontal channels, often contain strong electro-oculogram (EOG) artifacts produced by eye movements. Existing regression-based methods for removing EOG artifacts require various procedures for preprocessing and calibration that are inconvenient and timeconsuming. The paper describes a method for removing ocular artifacts based on adaptive filtering. The method uses separately recorded vertical EOG and horizontal EOG signals as two reference inputs. Each reference input is first processed by a finite impulse response filter of length M (M=3 in this application) and then subtracted from the original EEG. The method is implemented by a recursive leastsquares algorithm that includes a forgetting factor (λ=0.9999 in this application) to track the non-stationary portion of the EOG signals. Results from experimental data demonstrate that the method is easy to implement and stable, converges fast and is suitable for on-line removal of EOG artifacts. The first three coefficients (up to M=3) were significantly larger than any remaining coefficients.

Mean-square performance of a family of affine projection algorithms

Abstract: Affine projection algorithms are useful adaptive filters whose main purpose is to speed the convergence of LMS-type filters. Most analytical results on affine projection algorithms assume special regression models or Gaussian regression data. The available analysis also treat different affine projection filters separately. This paper provides a unified treatment of the mean-square error, tracking, and transient performances of a family of affine projection algorithms. The treatment relies on energy conservation arguments and does not restrict the regressors to specific models or to a Gaussian distribution. Simulation results illustrate the analysis and the derived performance expressions.

Improving convergence of the NLMS algorithm using constrained subband updates

Abstract: We propose a new design criterion for subband adaptive filters (SAFs). The proposed multiple-constraint optimization criterion is based on the principle of minimal disturbance, where the multiple constraints are imposed on the updated subband filter outputs. Compared to the classical fullband least-mean-square (LMS) algorithm, the subband adaptive filtering algorithm derived from the proposed criterion exhibits faster convergence under colored excitation. Furthermore, the recursive tap-weight adaptation can be expressed in a simple form comparable to that of the normalized LMS (NLMS) algorithm. We also show that the proposed multiple-constraint optimization criterion is related to another known weighted criterion. The efficacy of the proposed criterion and algorithm are examined and validated via mathematical analysis and simulation.

Speckle reduction in optical coherence tomography images by use of a spatially adaptive wavelet filter

Abstract: A spatially adaptive two-dimensional wavelet filter is used to reduce speckle noise in time-domain and Fourier-domain optical coherence tomography (OCT) images. Edges can be separated from discontinuities that are due to noise, and noise power can be attenuated in the wavelet domain without significantly compromising image sharpness. A single parameter controls the degree of noise reduction. When this filter is applied to ophthalmic OCT images, signal-to-noise ratio improvements of >7 dB are attained, with a sharpness reduction of <3%.

An adaptive notch filter for frequency estimation of a periodic signal

Abstract: Online frequency estimation of a pure sinusoidal signal is a classical problem that has many practical applications. Recently an ANF with global convergence property has been developed for this purpose. There exist some practical applications in which signals are not pure sinusoidal and contain harmonics. Therefore, online frequency estimation of periodic but not necessarily sinusoidal signals espoused by such applications becomes quite important. This note presents an alternative stability analysis for a modified ANF that permits the presence of harmonics in the incoming signal. Also, this stability analysis is simpler and alleviates the problem complexity even in the case of pure sinusoidal signal. Simulation results confirm theoretical issues.

Real-time particle filters

Abstract: Particle filters estimate the state of dynamic systems from sensor information. In many real-time applications of particle filters, however, sensor information arrives at a significantly higher rate than the update rate of the filter. The prevalent approach to dealing with such situations is to update the particle filter as often as possible and to discard sensor information that cannot be processed in time. In this paper we present real-time particle filters, which make use of all sensor information even when the filter update rate is below the update rate of the sensors. This is achieved by representing posteriors as mixtures of sample sets, where each mixture component integrates one observation arriving during a filter update. The weights of the mixture components are set so as to minimize the approximation error introduced by the mixture representation. Thereby, our approach focuses computational resources on valuable sensor information. Experiments using data collected with a mobile robot show that our approach yields strong improvements over other approaches.

A generalized normalized gradient descent algorithm

Abstract: A generalized normalized gradient descent (GNGD) algorithm for linear finite-impulse response (FIR) adaptive filters is introduced. The GNGD represents an extension of the normalized least mean square (NLMS) algorithm by means of an additional gradient adaptive term in the denominator of the learning rate of NLMS. This way, GNGD adapts its learning rate according to the dynamics of the input signal, with the additional adaptive term compensating for the simplifications in the derivation of NLMS. The performance of GNGD is bounded from below by the performance of the NLMS, whereas it converges in environments where NLMS diverges. The GNGD is shown to be robust to significant variations of initial values of its parameters. Simulations in the prediction setting support the analysis.

Least mean square adaptive digital background calibration of pipelined analog-to-digital converters

Abstract: We present an adaptive digital technique to calibrate pipelined analog-to-digital converters (ADCs). Rather than achieving linearity by adjustment of analog component values, the new approach infers component errors from conversion results and applies digital postprocessing to correct those results. The scheme proposed here draws close analogy to the channel equalization problem commonly encountered in digital communications. We show that, with the help of a slow but accurate ADC, the proposed code-domain adaptive finite-impulse-response filter is sufficient to remove the effect of component errors including capacitor mismatch, finite op-amp gain, op-amp offset, and sampling-switch-induced offset, provided they are not signal-dependent. The algorithm is all digital, fully adaptive, data-driven, and operates in the background. Strong tradeoffs between accuracy and speed of pipelined ADCs are greatly relaxed in this approach with the aid of digital correction techniques. Analog precision problems are translated into the complexity of digital signal-processing circuits, allowing this approach to benefit from CMOS device scaling in contrast to most conventional correction techniques.

Predicting respiratory motion for four-dimensional radiotherapy.

Abstract: Adapting radiation delivery to respiratory motion is made possible through corrective action based on real-time feedback of target position during respiration. The advantage of this approach lies with its ability to allow tighter margins around the target while simultaneously following its motion. A significant hurdle to the successful implementation of real-time target-tracking-based radiation delivery is the existence of a finite time delay between the acquisition of target position and the mechanical response of the system to the change in position. Target motion during the time delay leads to a resultant lag in the system's response to a change in tumor position. Predicting target position in advance is one approach to ensure accurate delivery. The aim of this manuscript is to estimate the predictive ability of sinusoidal and adaptive filter-based prediction algorithms on multiple sessions of patient respiratory patterns. Respiratory motion information was obtained from recordings of diaphragm motion for five patients over 60 sessions. A prediction algorithm that employed both prediction models-the sinusoidal model and the adaptive filter model-was developed to estimate prediction accuracy over all the sessions. For each session, prediction error was computed for several time instants (response time) in the future (0-1.8 seconds at 0.2-second intervals), based on position data collected over several signal-history lengths (1-7 seconds at 1-second intervals). Based on patient data included in this study, the following observations are made. Qualitative comparison of predicted and actual position indicated a progressive increase in prediction error with an increase in response time. A signal-history length of 5 seconds was found to be the optimal signal history length for prediction using the sinusoidal model for all breathing training modalities. In terms of overall error in predicting respiratory motion, the adaptive filter model performed better than the sinusoidal model. With the adaptive filter, average prediction errors of less than 0.2 cm (1sigma) are possible for response times less than 0.4 seconds. In comparing prediction error with system latency error (no prediction), the adaptive filter model exhibited lesser prediction errors as compared to the sinusoidal model, especially for longer response time values (>0.4 seconds). At smaller response time values (<0.4 seconds), improvements in prediction error reduction are required for both predictive models in order to maximize gains in position accuracy due to prediction. Respiratory motion patterns are inherently complex in nature. While linear prediction-based prediction models perform satisfactorily for shorter response times, their prediction accuracy significantly deteriorates for longer response times. Successful implementation of real-time target-tracking-based radiotherapy requires response times less than 0.4 seconds or improved prediction algorithms.

Adaptive detection for infrared small target under sea-sky complex background

Abstract: An adaptive Butterworth highpass filter is presented to detect a small target under a sea-sky complex background. By calculating the weighted information entropy of different infrared images, the cutoff frequency of the filter can be changed adaptively. Experimental results show that it is a robust small target detection method.

A recursive least M-estimate algorithm for robust adaptive filtering in impulsive noise: fast algorithm and convergence performance analysis

Abstract: This paper studies the problem of robust adaptive filtering in impulsive noise environment using a recursive least M-estimate algorithm (RLM). The RLM algorithm minimizes a robust M-estimator-based cost function instead of the conventional mean square error function (MSE). Previous work has showed that the RLM algorithm offers improved robustness to impulses over conventional recursive least squares (RLS) algorithm. In this paper, the mean and mean square convergence behaviors of the RLM algorithm under the contaminated Gaussian impulsive noise model is analyzed. A lattice structure-based fast RLM algorithm, called the Huber Prior Error Feedback-Least Squares Lattice (H-PEF-LSL) algorithm is derived. Part of the H-PEF-LSL algorithm was presented in ICASSP 2001. It has an order O(N) arithmetic complexity, where N is the length of the adaptive filter, and can be viewed as a fast implementation of the RLM algorithm based on the modified Huber M-estimate function and the conventional PEF-LSL adaptive filtering algorithm. Simulation results show that the transversal RLM and the H-PEF-LSL algorithms have better performance than the conventional RLS and other RLS-like robust adaptive algorithms tested when the desired and input signals are corrupted by impulsive noise. Furthermore, the theoretical and simulation results on the convergence behaviors agree very well with each other.

Static Analysis of Digital Filters

Abstract: We present an Abstract Interpretation-based framework for automatically analyzing programs containing digital filters. Our framework allows refining existing analyses so that they can handle given classes of digital filters. We only have to design a class of symbolic properties that describe the invariants throughout filter iterations, and to describe how these properties are transformed by filter iterations. Then, the analysis allows both inference and proofs of the properties about the program variables that are tied to any such filter.

Relative transfer function identification using speech signals

Abstract: An important component of a multichannel hands-free communication system is the identification of the relative transfer function between sensors in response to a desired source signal. In this paper, a robust system identification approach adapted to speech signals is proposed. A weighted least-squares optimization criterion is introduced, which considers the uncertainty of the desired signal presence in the observed signals. An asymptotically unbiased estimate for the system's transfer function is derived, and a corresponding recursive online implementation is presented. We show that compared to a competing nonstationarity-based method, a smaller error variance is achieved and generally shorter observation intervals are required. Furthermore, in the case of a time-varying system, faster convergence and higher reliability of the system identification are obtained by using the proposed method than by using the nonstationarity-based method. Evaluation of the proposed system identification approach is performed under various noise conditions, including simulated stationary and nonstationary white Gaussian noise, and car interior noise in real pseudo-stationary and nonstationary environments. The experimental results confirm the advantages of proposed approach.

A 0.18-/spl mu/m CMOS 3.5-gb/s continuous-time adaptive cable equalizer using enhanced low-frequency gain control method

Abstract: This paper describes a high-speed CMOS adaptive cable equalizer using an enhanced low-frequency gain control method. The additional low-frequency gain control loop enables the use of an open-loop equalizing filter, which alleviates the speed bottleneck of the conventional adaptation method. In addition, combined adaptation of low-frequency gain and high-frequency boosting improves the adaptation accuracy while supporting high-frequency operation. The open-loop equalizing filter incorporates a merged-path topology and offers infinite input impedance, which are suitable for higher frequency operation and cascaded design. This equalizing filter controls its common-mode output voltage level in a feedforward manner, thereby improving bandwidth. A prototype chip was fabricated in 0.18-/spl mu/m four-metal mixed-mode CMOS technology. The realized active area is 0.48/spl times/0.73 mm/sup 2/. The prototype adaptive equalizer operates up to 3.5 Gb/s over a 15-m RG-58 coaxial cable with 1.8-V supply and dissipates 80 mW. Moreover, the equalizing filter in manual adjustment mode operates up to 5 Gb/s over a 15-m RG-58 coaxial cable.

Estimation of nonstationary EEG with Kalman smoother approach: an application to event-related synchronization (ERS)

Abstract: An adaptive spectrum estimation method for nonstationary electroencephalogram by means of time-varying autoregressive moving average modeling is presented The time-varying parameter estimation problem is solved by Kalman filtering along with a fixed-interval smoothing procedure Kalman filter is an optimal filter in the mean square sense and it is a generalization of other adaptive filters such as recursive least squares or least mean square Furthermore, by using the smoother the unavoidable tracking lag of adaptive filters can be avoided Due to the properties of Kalman filter and benefits of the smoothing the time-frequency resolution of the presented Kalman smoother spectra is extremely high The presented approach is applied to estimation of event-related synchronization/desynchronization (ERS/ERD) dynamics of occipital alpha rhythm measured from three healthy subjects With the Kalman smoother approach detailed spectral information can be extracted from single ERS/ERD samples

Recursive least squares constant modulus algorithm for blind adaptive array

Abstract: We consider the problem of blind adaptive signal separation with an antenna array, based on the constant modulus (CM) criterion. An approximation to the CM cost function is proposed, which allows the use of the recursive least squares (RLS) optimization technique. A novel RLS constant modulus algorithm (RLS-CMA) is derived, where the modulus power of the array output can take on arbitrary positive real values (i.e., fractional values allowed). Simulations are performed to compare the performance of the proposed RLS-CMA to other well-known algorithms for blind adaptive beamforming. Results indicate that the RLS-CMA has a significantly faster convergence rate and better tracking ability.

Practical analog and digital filter design

Abstract: Introduction to Filters and Filter Design Software. Analog Filter Approximation Functions. Analog Lowpass, Highpass, Bandpass, and Bandstop Filters. Analog Filter Implementation Using Active Filters. Introduction to Discrete-Time Systems. Infinite Impulse Response Digital Filter Design. Finite Impulse Response Digital Filter Design. Digital Filter Implementation Using C. Digital Filtering Using the FFT. Appendices.

Adaptive equalization and data recovery in a dual-mode (PAM2/4) serial link transceiver

Abstract: To achieve high bit rates link designers are using more sophisticated communication techniques, often turning to 4PAM transmission or decision-feedback equalization (DFE). Interestingly, with only minor modification the same hardware needed to implement a 4PAM system can be used to implement a loop-unrolled single-tap DFE receiver. To get the maximum performance from either technique, the link has to be tuned to match the specific channel it is driving. Adaptive equalization using data based update filtering allows continuous updates while minimizing the required sampler front-end hardware and significantly reduces the cost of implementation in multi-level signaling schemes. A transceiver chip was designed and fabricated in 0.13 /spl mu/m CMOS process to investigate dual-mode operation and the modifications of the standard adaptive algorithms necessary to operate in high-speed link environments.

Partial-update NLMS algorithms with data-selective updating

Abstract: In this paper, we present mean-squared convergence analysis for the partial-update normalized least-mean square (PU-NLMS) algorithm with closed-form expressions for the case of white input signals. The formulae presented here are more accurate than the ones found in the literature for the PU-NLMS algorithm. Thereafter, the ideas of the partial-update NLMS-type algorithms found in the literature are incorporated in the framework of set-membership filtering, from which data-selective NLMS-type algorithms with partial-update are derived. The new algorithms, referred to herein as the set-membership partial-update normalized least-mean square (SM-PU-NLMS) algorithms, combine the data-selective updating from set-membership filtering with the reduced computational complexity from partial updating. A thorough discussion of the SM-PU-NLMS algorithms follows, whereby we propose different update strategies and provide stability analysis and closed-form formulae for excess mean-squared error (MSE). Simulation results verify the analysis for the PU-NLMS algorithm and the good performance of the SM-PU-NLMS algorithms in terms of convergence speed, final misadjustment, and computational complexity.

Adaptive BCI based on variational Bayesian Kalman filtering: an empirical evaluation

Abstract: This paper proposes the use of variational Kalman filtering as an inference technique for adaptive classification in a brain computer interface (BCI). The proposed algorithm translates electroencephalogram segments adaptively into probabilities of cognitive states. It, thus, allows for nonstationarities in the joint process over cognitive state and generated EEG which may occur during a consecutive number of trials. Nonstationarities may have technical reasons (e.g., changes in impedance between scalp and electrodes) or be caused by learning effects in subjects. We compare the performance of the proposed method against an equivalent static classifier by estimating the generalization accuracy and the bit rate of the BCI. Using data from two studies with healthy subjects, we conclude that adaptive classification significantly improves BCI performance. Averaging over all subjects that participated in the respective study, we obtain, depending on the cognitive task pairing, an increase both in generalization accuracy and bit rate of up to 8%. We may, thus, conclude that adaptive inference can play a significant contribution in the quest of increasing bit rates and robustness of current BCI technology. This is especially true since the proposed algorithm can be applied in real time.

Mitigation techniques for non-Gaussian sea clutter

Abstract: This paper deals with the problem of coherent radar detection of targets embedded in clutter modeled as a compound-Gaussian process. We first provide a survey on clutter mitigation techniques with a particular emphasis on adaptive detection schemes ensuring the constant false-alarm rate (CFAR) property with respect to all of the clutter parameters. Thus, we propose a novel decision rule based on a recursive covariance estimator, which exploits the persymmetry property of the clutter covariance matrix. Remarkably, the devised receiver is fully CFAR in that its threshold can be set independently of the clutter distribution as well as of its covariance, even if the environment is highly heterogeneous; namely, the disturbance distributional parameters vary from cell to cell. At the analysis stage, we compare the performance of the novel detector with some classical radar receivers such as that of Kelly and the adaptive matched filter both in the presence of simulated as well as on real radar data, which statistical analysis has shown to be compatible with the compound-Gaussian model. The results show that the new receiving structure generally provides higher detection performance than the others and, for a fluctuating target, it uniformly outperforms the counterparts. We also provide a discussion on the CFAR behavior of the analyzed receivers as well as on their computational complexity.

Data-aided and blind stochastic gradient algorithms for widely linear MMSE MAI suppression for DS-CDMA

Abstract: In this paper, three novel stochastic gradient algorithms for adjustment of the widely linear (WL) minimum mean-squared error (MMSE) filter for multiple access interference (MAI) suppression for direct-sequence code-division multiple access (DS-CDMA) are introduced and analyzed. In particular, we derive a data-aided WL least-mean-square (LMS) algorithm, a blind WL minimum-output-energy (MOE) algorithm, and a WL blind LMS (BLMS) algorithm. We give analytical expressions for the steady-state signal-to-interference-plus-noise ratios (SINRs) of the proposed WL algorithms, and we also investigate their speed of convergence. Wherever possible, comparisons with the corresponding linear adaptive algorithms are made. Both analytical considerations and simulations show, in good agreement, the superiority of the novel WL adaptive algorithms. Nevertheless, all proposed WL algorithms require a slightly lower computational complexity than their linear counterparts.

Particle swarm optimization for adaptive IIR filter structures

Abstract: This paper introduces the application of particle swarm optimization techniques to infinite impulse response (IIR) adaptive filter structures. Particle swarm optimization (PSO) is similar to the genetic algorithm (GA) in that it performs a structured randomized search of an unknown parameter space by manipulating a population of parameter estimates to converge on a suitable solution. Unlike the genetic algorithm, particle swarm optimization has not emerged in adaptive filtering literature. Both techniques are independent of the adaptive filter structure and are capable of converging on the global solution for multimodal optimization problems, which makes them especially useful for optimizing IIR and nonlinear adaptive filters. This paper outlines PSO and provides a comparison to the GA for IIR filter structures.