Adaptive Filter Design Research Articles

Bird Species Recognizer using LMS Algorithm

Birds play a vital role in many ecosystems, acting as both predators and preys for other living organisms. Therefore its important to monitor the population of various bird species in the environment in order to maintain balance in the ecosystem. This process will become tedious if it is done manually as it involves handling large sets of data at the same instant. We can do this by developing an automatic bird species recognizer which identifies the bird species based on bird songs and voice signals. In this research, we have used a tenth-order LMS adaptive filter to remove noise from bird voice signals which are recorded in different environmental conditions where different noise frequencies are present. The design of a tenth-order LMS adaptive filter using MATLAB has been implemented. The performance and characteristics of the filter for five different methods of LMS has been shown. After removal of noise from the noisy bird voice signal using LMS algorithm, we have made use of cross correlation to identify the bird species that it corresponds to. Signal to Noise Ratio (SNR) and Mean Square Error (MSE) of the filtered bird signals obtained using the variants of LMS like Normalized LMS, Sign-Data LMS, Sign-Error LMS and Sign-Sign LMS have been estimated and compared. We have made use of signal processing tool kits and various noise parameter schemes have been computed to show the effectiveness of the designed filter in the field of bird recognition.

Empirical wavelet transform and dual feed-forward neural network for classification of power quality disturbances

This paper proposes a novel approach for classification of single and combined power quality (PQ) disturbances. The EWT-based adaptive filtering technique is employed first to decompose the signal into its individual frequency components by estimation of frequencies. The frequency estimation in this paper is done using a divide-to-conquer principle-based FFT technique and followed by an adaptive filter design. Then, some unique potential features reflecting the characteristics of disturbances are extracted from the mono-frequency components as well as the signal. A single classifier used for the classification of combined disturbances, whose characteristics are alike, gives less classification accuracy. Therefore, the use of a dual FFNN is proposed for the classification of single and combined PQ disturbances to effectively reduce the misclassification and improve the accuracy. The effectiveness of the proposed approach is evaluated on a broad range of timevarying power signals with varying degree of irregularities, noise, and fundamental frequency deviation. The results obtained for both the simulated as well as the real disturbance signals elucidate the efficiency and robustness of the proposed approach for classification of the most frequent disturbances.

Empirical wavelet transform and dual feed-forward neural network for classification of power quality disturbances

This paper proposes a novel approach for classification of single and combined power quality (PQ) disturbances. The EWT-based adaptive filtering technique is employed first to decompose the signal into its individual frequency components by estimation of frequencies. The frequency estimation in this paper is done using a divide-to-conquer principle-based FFT technique and followed by an adaptive filter design. Then, some unique potential features reflecting the characteristics of disturbances are extracted from the mono-frequency components as well as the signal. A single classifier used for the classification of combined disturbances, whose characteristics are alike, gives less classification accuracy. Therefore, the use of a dual FFNN is proposed for the classification of single and combined PQ disturbances to effectively reduce the misclassification and improve the accuracy. The effectiveness of the proposed approach is evaluated on a broad range of timevarying power signals with varying degree of irregularities, noise, and fundamental frequency deviation. The results obtained for both the simulated as well as the real disturbance signals elucidate the efficiency and robustness of the proposed approach for classification of the most frequent disturbances.

Performance Analysis of Karatsuba Vedic Multiplier and Computation Sharing Multiplier in the Adaptive Filter Design

As real time signals change continuously, adaptive filtering is required for noise cancellation. An adaptive filter is one whose characteristics can be modified by adjusting its parameters according to an optimization algorithm The adaptive filtering operations can be implemented as a sequence of logic operations on a Digital Signal Processing (DSP) chip,Gate Arrays such as FPGA or Application Specific Integrated Circuits. There is always a tradeoff in the parameters area, power and speed in VLSI. This paper provides the implementation of adaptive LMS Filter using different types of multiplier and its analysis for the various parameters. The LMS Filter is designed using conventional array multiplier, Computation Sharing Mutliplier(CSHM) and Karatsuba Vedic Multiplier. The results show 70% reduction in delay and 19% reduction in area on using Karatsuba Vedic Multiplier ofr adaptive filter design.

A Robust Tool for Monitoring and Synchronizing Smart Grid through Adaptive Comb Filter

The power system signals are often polluted with harmonics and noise as a result of nonlinear load. This non stationary signal has to be monitored carefuly before it propagates as a grid problem.This article describes the design of adaptive comb filter, extraction of amplitude, frequency and phase with respect to time for monitoring purpose and extraction of harmonic components for suppressing the contamination present in the signal for synchronization with smart grid. The adaptive comb filter algorithm is a synchronizing tool implemented in Matlab/Simulink environment.The response of the filter is compared with enhanced phased locked loop to describe the characteristics of adaptive comb filter.The algorithm tracks the transient (dynamic) and steady state behaviour of the signal effectively, efficiently and accurately.

Adaptive iir filter design using self-adaptive search equation based artificial beecolony algorithm

Infinite impulse response (IIR) system identification problem is defined as an IIR filter modeling to represent an unknown system. During a modeling task, unknown system parameters are estimated by metaheuristic algorithms through the IIR filter. This work deals with the self-adaptive search-equation-based artificial bee colony (SSEABC) algorithm that is adapted to optimal IIR filter design. SSEABC algorithm is a recent and improved variant of artificial bee colony (ABC) algorithm in which appropriate search equation is determined with a self-adaptive strategy. Moreover, the success of the SSEABC algorithm enhanced with a competitive local search selection strategy was proved on benchmark functions in our previous studies. The SSEABC algorithm is utilized in filter modelings which have different cases. In order to demonstrate the performance of the SSEABC algorithm on IIR filter design, we have also used canonical ABC, modified ABC (MABC), best neighbor-guided ABC, and an ABC with an adaptive population size (APABC) algorithms as well as other algorithms in the literature for comparison. The obtained results and the analysis on performance evolution of compared algorithms on several filter design cases indicate that SSEABC outperforms all considered ABC variants and other algorithms in the literature.

Design of Optimal Adaptive Filters for Two-Dimensional Filamentary Structures Segmentation

The problems of filamentary structures segmentation encompass retinal vessel detection in fundus images, reconstruction and tracing of neurons in microscopic images and segmentation of human vasculatures in two-dimensional digital angiography. In this letter, we focus on retinal vessel segmentation which is an important and challenging problem among others. Many retinal vessel segmentation algorithms have been developed, where most of them rely on the filtering technique. However, the available filters suffer from undesirable responses at some vessel and non-vessel structures. Thus, we propose a new framework based on optimal adaptive filters for retinal vessel segmentation. The filter coefficients are obtained by solving multiple inverse problems under a regularization framework. The performance of the proposed framework is evaluated on fundus images from the DRIVE and STARE datasets. In the experimental section, we show that the proposed framework outperforms all of the compared methods in four indicators namely sensitivity, F1-score, G-mean and Mathews correlation coefficient.

Design and implementation of Least Mean Square adaptive FIR filter using offset binary coding based Distributed Arithmetic

Distributed Arithmetic (DA) based architecture is an efficient technique to attain high throughput without hardware multiplier and also it is essential for bit serial operation. The DA based Finite Impulse Response (FIR) adaptive filter is well suited for hardware implementation in Field Programmable Gate Array (FPGA) device. In conventional DA the partial products of the filter coefficients have been pre-calculated and stored in Look up Table (LUT) which in turn will increase the logic elements and power. To overcome this problem DA based Least Mean Square (LMS) adaptive filter using offset binary coding (OBC) without LUT is proposed. The proposed method will reduce the logic elements by half when compared to the conventional DA based OBC filter. The Carry Save Accumulator (CSA) is used to carry out the operation of shift and accumulation. The proposed architecture is implemented in Quartus II 9.1 with the device as Stratix-EP2S15F484C3 which offers 13.72% high throughput, 56.92% reduction in logic elements, 42.84% reduction in power, 57.74% reduction in logical registers for N = 16 and for N = 32 the number of logical element is reduced to 80.87%, 66.66% reduction in power and 24.12% high throughput.

ASENKRON MOTORUN İNDİRGENMİŞ-DERECELİ UYARLAMALI GENİŞLETİLMİŞ KALMAN FİLTRESİ İLE DURUM VE PARAMETRE KESTİRİMİ

In this study, the design of reduced-order adaptive extended Kalman filter (EKF) for speed-sensorless control of induction motors (IMs) is performed, and its performance is tested using it in a speed-sensorless direct vector controlled drive system under simulations. The proposed observer estimates the stator stationary axis components of rotor fluxes and rotor mechanical speed required for vector control in addition to disturbance load torque. On the other hand, estimation performance of EKFs depends on the correct selection of system ( ) and measurement ( ) error covariance matrices. In the literature, these matrices are generally assumed as constant and determined by the trial-and-error method. However, those matrices are affected by operating conditions of IM and should be updated according to operating conditions in order to obtain higher performance estimations. Since the simultaneously update of both and may lead to divergence or tracking problems, only is updated considering operating conditions by adaptive EKF (AEKF) algorithm having a forgetting factor. In addition, AEKF has been designed as reduced-order with the aim of reduction of its computational burden for real-time applications.

Adaptive filter design for active noise cancellation using recurrent type-2 fuzzy brain emotional learning neural network

This article aims to develop a more efficient adaptive filter for the active noise cancellation (ANC). A novel recurrent interval type-2 fuzzy brain emotional learning filter (RT2BELF) is proposed for achieving favourable filtering performance. The ANC is a method to eliminate noise by creating an anti-noise signal which has the same magnitude but opposite phase with the unwanted noise. In order to adapt to the change of the noise, the parameters for the RIT2BELF are online updated based on the adaptive laws, which are derived by the steepest descent gradient approach. The performance of the proposed ANC design method is successfully demonstrated based on numerical simulation results in the real signals. Finally, the superiority of the proposed method is confirmed by the results comparison with some noise cancellation methods.

A Novel Approach of 2D Adaptive Filter Based on MPSO Technique for Biomedical Image

In this paper, extension of the 1-D adaptive filter schemes to 2D formation and the new 2D adaptive filters are designed. The results of proposed scheme are compared with 2D variable step-size normalized least mean squares, the 2D VSS affine projection algorithms, the 2D set-membership NLMS, and 2D SM APA. The performance of proposed scheme is compared with other reported methods for 2D adaptive filter design. Based on simulation results, it is demonstrated that the proposed method can achieve 85% and 90% reduction in normalized mean square error and normalized maximum error mean, respectively. Moreover, the proposed 2D-ANC filter applied for reconstruction of a biomedical image shows 6 dB signal-to-noise ratio improved as compared to recently reported algorithm.

Adaptive echo suppression based on LabVIEW and C# implementation

This paper deals with the using of adaptive filtering in the area of telecommunication technology. The study focuses on the possibilities of design and implementation of adaptive filters based on virtual instrumentation in the LabVIEW development environment from the American company National Instruments. The selected adaptive algorithms (LMS, NLMS, RLS, QR-RLS) are implemented both in LabVIEW and C# programming language in the form of dynamically linked DLL (Dynamic-Link Library) libraries. The experiments test the reliability and speed of the implemented algorithms using them for the suppression of the echo in sound signal. Based on the findings from the implementation, a comparison of both options is made by evaluation of the computation time and improvement of signal-to-noise ratio (SNR).

Adaptive fetal ECG signal extraction based on LabVIEW and C# implementation

This paper deals with the using of adaptive filters in the biological signal processing area. It focuses on the design and implementation of adaptive filters based on virtual instrumentation in the LabVIEW development environment from National Instruments. The experiments are devoted to certain adaptive algorithms (LMS, NLMS, RLS, QR-RLS), which are implemented in LabVIEW and C# programming language in the form of dynamically linked DLL libraries. The implemented algorithms are tested in extraction of fetal ECG signal from the abdominal ECG signal of mother. The algorithms are compared in their computation time and the improvement of signal-to-noise ratio (SNR) value.

A Novel Adaptive Non-Local Means-Based Nonlinear Fitting for Visibility Improving

The spatial-based method has become the most widely used method in improving the visibility of images. The visibility improving is mainly to remove the noise in the image, in order to trade off denoising and detail maintaining. A novel adaptive non-local means-based nonlinear fitting method is proposed in this paper. Firstly, according to the smoothness of the intensity around the central pixel, eight kinds of templates with different precision are exploited to approximate the central pixel through a novel adaptive non-local means filter design; the approximate weight coefficients of templates are derived from the approximation credibility. Subsequently, the fractal correction is used to smooth the denoising results. Eventually, the Rockafellar multiplier method is employed to generalize the smooth plane fitting to any geometric surface, thus yielding the optimal fitting of the center pixel approximation. Through a large number of experiments, it is clearly elucidated that compared with the classical spatial iteration-based methods and the recent denoising algorithms, the proposed algorithm is more robust and has better effect on denoising, while keeping more original details during denoising.

IIR filter optimization using improved chaotic harmony search algorithm

Due to the fact that the error surface of adaptive infinite impulse response (IIR) systems is generally nonlinear and multimodal, conventional derivative-based techniques fail when used in adaptive Filter design. In this sense, global optimization techniques are required in order to avoid local minima. Harmony search (HS), a musical inspired metaheuristic, is a recently introduced population-based algorithm that has been successfully applied to global optimization problems. In the present paper, adaptive IIR filtering is formulated as a nonlinear optimization problem and then an improved version of HS incorporating chaotic search (CIHS) is introduced to solve the identification problem of three benchmark IIR systems. Furthermore, the performance of the proposed methodology is compared with HS and two well-known metaheuristic algorithms, genetic algorithm (GA) and particle swarm optimization (PSO) and a modified version of PSO called PSOW (Particle Swarm Optimization with weight Factor). The results demonstrate that the proposed method has superior performance over the other above-mentioned algorithms in terms of convergence speed and accuracy.

An improved adaptive filtering approach for power quality analysis of time-varying waveforms

An improved adaptive technique is proposed to estimate the frequencies, thereby decomposing the nonstationary signal for the analysis of time-varying disturbances. The proposed method is based on the successive application of FFT and followed by the adaptive filter design. The FFT based frequency estimation has been performed based on divide to conquer principle for accurate estimation of harmonics and inter-harmonics. The estimated frequencies are processed and the empirical wavelet filters are designed accordingly for extracting the mono-frequency components. Thereafter, the instantaneous frequency and amplitude of each component have been estimated by applying Hilbert transform. The effectiveness of the proposed method has been verified by testing it on various simulated as well as measured power quality signals containing harmonics, inter-harmonics, fundamental frequency deviation, transients, sag and swell. The results confirm that the proposed method is able to analyse the time-varying power quality signals accurately and rapidly, making it suitable for real-time monitoring.

Algorithm and VLSI Architecture Design of Proportionate-Type LMS Adaptive Filters for Sparse System Identification

Proportionate-type normalized LMS (Pt-NLMS) family of adaptive filtering algorithms for sparse system identification pose significant implementation challenges due to their high computational complexity especially for real-time applications like network echo cancelation. In this paper, we make the first attempt to implement Pt-NLMS algorithms in hardware. Several reformulations are proposed to simplify the original Pt-NLMS algorithms, thereby making them amenable to real-time VLSI implementations and the reformulated algorithms referred as delayed $\mu $ -law proportionate LMS (DMPLMS) algorithm for white input and delayed wavelet MPLMS (DWMPLMS) for colored input are then implemented in hardware. Simulation studies demonstrate that the performance loss is very small for the proposed reformulations. We implemented the proposed designs considering 16-bit fixed point representation in hardware, and synthesis results show that the DMPLMS architecture with ${\approx } 30\%$ increase in hardware over the state-of-the-art conventional delayed LMS architecture achieves $3\times $ improvement in convergence rate for white input and the DWMPLMS architecture with ${\approx } 70\%$ increase in hardware achieves $10\times $ improvement in convergence rate for correlated input conditions.

Adaptive Event-Triggered Fault Detection for Fuzzy Stochastic Systems With Missing Measurements

This paper discusses adaptive event-triggered fault detection filter design for fuzzy stochastic models with missing measurements. First, a novel event-triggered strategy is introduced, while an adaptive law is provided to adjust communication threshold dynamically. Compared with traditional event-triggered methods with fixed threshold, the proposed strategy is more effective on saving network communication resources. Second, a Bernoulli stochastic process is proposed to describe the measurement missing phenomenon, which always appears in real network environment. Then, an integrated fault detection model for fuzzy stochastic systems is constructed by taking network-induced delays, adaptive event-triggered strategy and missing measurements into account. A new method is provided to achieve mean-square asymptotical stability of residual model with one desired fault detection objective. Finally, simulation cases are introduced to verify the validity of the designed strategy.

Adaptive filter design via a gradient thresholding algorithm for compressive spectral imaging.

Sensing a spectral image data cube has traditionally been a time-consuming task since it requires a scanning process. In contrast, compressive spectral imaging (CSI) has attracted widespread interest since it requires fewer samples than scanning systems to acquire the data cube, thus improving the sensing speed. CSI captures linear projections of the scene, and then a reconstruction algorithm estimates the underlying scene. One notable CSI architecture is the color coded aperture snapshot spectral imager (C-CASSI), which employs pixelated filter arrays as the coding patterns to spatially and spectrally encode the incoming light. Up to date works on C-CASSI have used non-adaptive color coded apertures. Non-adaptive sampling ignores prior information about the signal to design the coding patterns. Therefore, this work proposes a method to adaptively design the color coded aperture, such that the quality of image reconstruction is improved. In more detail, this work introduces a gradient thresholding algorithm, which computes the consecutive color coded aperture from a rapidly reconstructed low-resolution version of the data cube. The successive adaptive patterns enable recovering a data cube in the presence of Gaussian noise with higher image quality. Real reconstructions and simulations evidence an improvement of up to 3dB in the quality of image reconstruction of the proposed method in comparison with state-of-the-art non-adaptive techniques.

Design Methodology of N-Path Filters With Adjustable Frequency, Bandwidth, and Filter Shape

In this paper, an adaptive N-path filter design technique is presented. The filter designed by this method can be reconfigured to different shapes, e.g., Butterworth or Chebyshev with variable bandwidth (BW), based on the user requirements. In addition, an analytical synthesis procedure is introduced to realize high-order bandpass filters (BPFs) based on N-path architecture with a prescribed set of specifications. A proof-of-concept BPF is fabricated and measured in a 65-nm CMOS process. The filter can be reconfigured to realize different filter shapes by changing coupling capacitors. The BW of the proposed filter is also tunable. The passband of the filter is tunable from 0.2 to 1.2 GHz with a gain of 14.5–18 dB, a noise figure of 3.7–5.5 dB, and a total power consumption of 37.5–52.7 mW. The channel BW can be varied from 20 to 40 MHz and the filter out-of-band third-order intercept point ( $\Delta f=50$ MHz) is 25 dBm.