Journal of Professor & SPaC member

If you click the title of the journal, you can get more information about the journal.
Last Updated : 03/2016
all           2017           2016           2015           2014           2013           2012           ~2011          
  • An Efficient Line-Search Algorithm for Unbiased Recursive Least-Squares Filtering With Noisy Inputs

    BH Kang, PG Park

    Signal Processing Letters, IEEE 20 (7), 693-696

    Abstract This letter proposes a new algorithm for efficiently finding an unbiased RLS estimate of FIR models with noisy inputs. The unbiased estimate is obtained without knowing any a priori information via a new cost. Furthermore, to reduce computational complexity, the estimate is updated along the current input-vector direction and the corresponding gain is efficiently computed. In addition, to increase the convergence rate, the algorithm is extended to update the estimate along not only current but also past input-vector directions. Simulation results show that the proposed algorithm exhibits a fast convergence rate and an enhanced tracking performance with noisy correlated inputs.
  • A Normalized Least Mean Squares Algorithm With a Step-Size Scaler Against Impulsive Measurement Noise

    I Song, PG Park, RW Newcomb

    Circuits and Systems II: Express Briefs, IEEE Transactions on 60 (7), 442-445

    Abstract This brief introduces the concept of a step-size scaler by investigating and modifying the tanh cost function for adaptive filtering with impulsive measurement noise. The step-size scaler instantly scales down the step size of gradient-based adaptive algorithms whenever impulsive measurement noise appears, which eliminates a possibility of updating weight vector estimates based on wrong information due to impulsive noise. The most attractive feature of the step-size scaler is that this is easily applicable to various gradient-based adaptive algorithms. Several representative gradient-based adaptive algorithms are performed without or with the step-size scaler in impulsive-noise environments, which shows the improvement of robustness against impulsive noise.
  • A bias-compensated affine projection algorithm for noisy input data

    SM Jung, NK Kwon, P Park

    Control Conference (ASCC), 2013 9th Asian, 1-5

    Abstract This paper proposes a bias-compensated affine projection algorithm (BC-APA) to eliminate bias due to noisy input data and to reduce the performance degradation due to highly correlated input data. A new affine projection algorithm (new APA) using innovative input data is presented for highly correlated input data. We analyze the bias in this innovative new APA under noisy input data and remove it. To remove the bias, an estimation method for the input noise variance is presented and explained. In simulations, the BC-APA provided both fast convergence rate and small mean square deviation. Based on improved precision to estimate a finite impulse response of an unknown system, the BC-APA can be applied extensively in adaptive signal processing areas.
  • LPV controller design with multiple parameters for the nonlinear RTAC system

    NK Kwon, BY Park, SM Jung, P Park

    Control Conference (ASCC), 2013 9th Asian, 1-6

    Abstract This paper proposes linear parameter varying (LPV) model with multiple parameters (LPV-MP) and statefeedback controller for the nonlinear rotational and translational actuator (RTAC) benchmark problem. First, based on LPV-MP, the conditions used for designing the state-feedback controller are formulated in terms of parameterized linear matrix inequalities (PLMIs) and the state-feedback LPV controller using multiple parameters-dependent Lyapunov function (MPDLF) is designed. Then, PLMI conditions are converted into linear matrix inequalities (LMIs) by using a parameter relaxation technique. The proposed method results in the reduced decision variables and simulation results show good performance of the proposed method.
  • Bias-compensated normalised LMS algorithm with noisy input

    B Kang, J Yoo, P Park

    Electronics Letters 49 (8), 538-539

    Abstract A new bias-compensated normalised least mean square (NLMS) algorithm for parameter estimation with a noisy input is proposed. The algorithm is obtained from an approximated cost function based on the statistical properties of the input noise and involves a condition checking constraint to decide whether the weight coefficient vector must be updated. Simulation results show that the proposed algorithm is more robust and accurate than the conventional method.
  • Multicore and Mesh Network-based Parallel Performance Evaluation using Intra Prediction Algorithms

    Yungho Choi, Neungsoo Park

    International Journal of Control and Automation

    Abstract So far, many parallel algorithms have been developed under the assumption that a high performance multicore processor uses a bus for inter-core communications. However, this assumption begins to change as the number of processing cores is increased and thus, higher connectivity among cores is required. So, in this paper, three HEVC intra prediction algorithms are ported into a mesh network-based multicore system by using a wavefront-style parallelization. By analyzing parallel performance, this paper shows that UDIP best fits in the mesh network-based multicore system (almost 2 times faster than other algorithms).
  • A Consistent Binormalized Data-Reusing LMS Algorithm for Noisy FIR Models

    Byung Hoon Kang, Nam Kyu Kwon, Hyon-Taek Choi, Poo Gyeon Park

    International Journal of Computer and Electrical Engineering

    Abstract This paper proposes a consistent binormalized data-reusing least mean square (LMS) algorithm for identifying finite impulse response models whose input and output are corrupted by additive white noise. The proposed algorithm exploits the stochastic properties of the noisy input to compensate a bias of estimation which is occurred by input noise. Furthermore, by reusing the input signal, the algorithm overcomes a decline of convergence performance with highly correlated input signal. The experimental results show that the proposed algorithm achieves consistent estimation with noisy input signal. Furthermore, the proposed algorithm gets faster convergence rate and smaller steady-state estimation errors than the ordinary consistent LMS algorithms when the input signal is highly correlated.
  • A normalized least-mean-square algorithm based on variable-step-size recursion with innovative input data

    Insun Song, PooGyeon Park

    Signal Processing Letters, IEEE

    Abstract This letter presents a variable-step-size normalized least-mean-square algorithm, where the step size is updated only when the current input vector is innovative from the last updated input vector. The instant innovativeness of the two input vectors is investigated through the relation between the angle of the two input vectors and the condition number of the input covariance matrix. Once the condition number is obtained, the resulting algorithm performs an excellent transient and steady-state behavior with different correlations in inputs. To reduce the computational burden of obtaining the condition number, this letter also presents a simple method to determine the condition number based on the power method.
  • An Improved Least Mean Kurtosis (LMK) Algorithm for Sparse System Identification

    Jin Woo Yoo, PooGyeon Park

    International Journal of Information and Electronics Engineering

    Abstract This paper proposes an improved least mean kurtosis (LMK) algorithm based on l0-norm cost for enhancing the filter performance in a sparse system. The LMK adaptive filtering algorithm uses a kurtosis of an estimated error signal to improve the filter performance when the noise contamination is serious. Due to the influence of l0-norm cost, the proposed LMK algorithm ensures a fast convergence rate and a small steady-state error in sparse system environment. Simulation results verify that the proposed algorithm improves the filter performance for sparse system identification.
  • A Robust Variable Step-Size NLMS Algorithm Through A Combination of Robust Cost Functions

    Insun Song, Won Il Lee, Nam Kyu Kwon, PooGyeon Park

    International Journal of Information and Electronics Engineering

    Abstract This letter introduces a new gradient-based adaptive filtering algorithm based on a cost function that is constructed by combining two robust cost functions, which are a new tanh-type cost function and Vega’s cost function. Through the approach to combine robust cost functions, the robustness of the proposed algorithm outperforms that of other adaptive algorithms.Since the proposed algorithm is derived by combining two robust cost functions, it leads to an excellent transient and steady-state behavior in high probability of impulsive measurement noise. The proposed algorithm is tested in different probability of impulsive measurement noise.