Trimmed Diffusion Least Mean Squares for Distributed Estimation

Size: px
Start display at page:

Download "Trimmed Diffusion Least Mean Squares for Distributed Estimation"

Transcription

1 Trimmed Diffusion Least Mean Squares for Distributed Estimation Hong Ji, Xiaohan Yang, Badong Chen School of Electronic and Information Engineering Xi an Jiaotong University Xi an , P.R. China Abstract We consider the problem of distributed estimation, where a set of nodes is required to collectively estimate networ parameters from noisy measurements. The problem is important when modeling a wide class of real-time sensor networs, where efficiency, robustness, and low power consumption are desired features. In this wor, we focus on diffusion-based adaptive solutions that capable to avoid undue influence from outliers, especially in the presence of impulsive noise or dysfunction of certain nodes. We motivate and propose trimmed diffusion least mean square (TDLMS) algorithm that selects normal neighborhood to update the system estimation. We provide performance analysis together with simulation results comparing with existing methods. Index Terms Adaptive networs, diffusion adaptation, distributed estimation, diffusion least mean square. I. INTRODUCTION Distributed estimation deals with the in-networ information extraction where the data collected at nodes distributed over a geographic region. Such algorithms are useful in several contexts, including communications and sensor networs. In the centralized solution to the problem, every node transmits their data to a fusion center for processing, and then receives the resulting estimate from that center. This approach enables the calculation of the global optimization, however has the disadvantage to fully rely on the fusion center. Moreover, centralizing all measurements in a single node may require large amounts of energy and communication resources. The distributed solutions on the other hand, every node only communicates with its closest neighbors and the processing is done locally, which achieve higher robustness, consume fewer resources and handle lighter loads in computation resources. Owing to these merits, distributed estimation has received more and more attention recently and been widely used in cognitive radios [1], environmental monitoring [2], and industrial automation [3]. Diffusion is one of the fundamental governing phenomena in the physical world. In previous studies, many distributed estimation algorithms have been proposed based on diffusion strategies, such as diffusion RLS [4, 5], diffusion LMS [6 9], and diffusion sparse LMS [10]. Prominent solution among these methods is the diffusion least mean square (DLMS) strategy that offers a simple but effective way to implement This wor was supported by the 973 project (2015CB351703) and NNSFC ( , ). distributed adaptive filtering over networs. This wor goes after the problem formulation of the DLMS algorithm. Most of the existing algorithms assume that the errors are Gaussian, and they may perform poorly when the data are non-gaussian, particularly in the case of large outliers (observations that significantly deviates from the bul of data, heavy tails). There is clear interest to develop robust estimators to wor properly in inpulsive environments. For instance, different types of artificial noise in electronic devices, atmospheric noise and lighting spies in natural phenomena, which can be described more accurately using heavy tailed non-gaussian noise models [11, 12]. The least trimmed squares (LTS) estimator, which minimizes the sum of the M smallest squared residuals, is a breathrough technique for robust estimators [13, 14]. In this wor, we will apply the LTS criterion into the problem of distributed adaptive filtering and develop a new algorithm named the trimmed diffusion least mean square (TDLMS), which can avoid undue influence from outliers and achieves better performance when the system is disturbed by impulsive noises. The rest of the paper is organized as follows. In section 2, we briefly introduce the problem formulation and the DLMS algorithm. The trimmed diffusion LMS (TDLMS) are then developed in Section 3. In Section 4, some encouraging simulation results are presented and finally, the conclusion is given in section 5. II. DLMS ALGORITHMS Consider a set of N nodes distributed over some geographic region. At every time instant i, every node taes a scalar measurement d (i) of some random process d (i) and a 1 M regression vector, µ,i, corresponding to a realization of a random process µ,i, which is correlated with d (i) by linear model: d (i) = µ,i ω o + ν (i) (1) where ν (i) represents bacground noise and is independent of µ,i for all and i, and independent of ν l (j) for l or i j. The common goal of all nodes is to identify ω o from the observations within node s neighbor set N in an adaptive manner. The neighbor set N is defined as the set of

2 nodes that have a direct lin with node, including node itself. The cost function of DLMS for each node is defined as J DLMS l N c l, E e l,i 2 l N c l, E d l (i) µ l,i ω 2 (2) where E denotes the expectation operator and c l, are possibly combination weights which satisfy: c l, = 0 if l / N, C1 = 1, 1 T C = 1 T (3) where C with N N non-negative real entries denotes the combination weight matrix and 1 denotes N 1 vector with unit entries. The diffusion strategy is then performed in two stages [6, 9, 15]: combination and adaptation. According to the different order of these two stages, the DLMS is classified into two algorithms: Combine-then-Adapt (CTA) DLMS and Adapt-then-Combine (ATC) DLMS [6, 9]. Replacing the statistical moments by local instantaneous approximations, start with ω l, 1 = 0 for all l, for each time i 0 and for each node, the ATC diffusion DLMS algorithm can be described as follows [6]. ψ,i = ω,i 1 + µ c l, µ l,i (d l(i) µ l,i ω,i 1 ) l N ω,i a l, ψ l,i (diffusion step) l N (4) where µ is a positive step-size chosen by node, denotes complex conjugate-transposition, ψ,i is an intermediate estimate of ω o of node at time instant i, and {a l, } is the weighting coefficients satisfying the condition: a l, = 0 if l / N, 1 T A = 1 T (5) where A is also a real non-negative N N matrix with individual entries {a l, }. Similarly, the CTA diffusion DLMS is: ψ,i 1 = a l, ω l,i 1 (diffusion step) l N ω,i = ψ,i 1 + µ c l, µ l,i l N (d l(i) µ l,i ψ,i 1 ) (6) III. TRIMMED DLMS ALGORITHMS In previous study, most of the diffusion distributed estimation algorithms assume that the noise follows Gaussian distribution. However, the Gaussianity of measurement noise is not guaranteed in many applications. In this wor, we focus on avoiding undue influence from large outliers, such as in the presence of impulsive noise or dysfunction of neighbor nodes, where the DLMS will lose its optimality. As mentioned in section 2, the cost function of DLMS minimize the sum of squared residuals from random measurements of all the neighbors. If however, one or a part of the neighbors data are recorded erroneously, the evaluated coefficients will be biased towards these leverage points. We hereby motivate and Fig. 1. At time i, every node taes measurement {d (i), µ,i }, only the trimmed neighborhood N will be employed to update the system estimation. propose a trimmed DLMS algorithm (see Fig.1), that we see an estimate of ω o only by combining P neighbors of its Q neighbors for each node (Q > P ). By doing so, we trimmed Q P observations with large error residuals and achieves an estimator with higher robustness and accuracy. The cost function of TDLMS for each node is defined as J T DLMS c l, E e l,i 2 l N l N c l, E d l (i) µ l,i ω 2 (7) where N is the node s neighbor set and Ñ is the selected neighbors of node that satisfies the condition Ñ N and e 2 j(1),i e 2 j(2),i e 2 j(p ),i e2 j(q ),i are ordered square errors, P is the trimming constant of node, i.e. j(1),..., j(p ) Ñ and j(1),..., j(q ) N. Analogously, { c l, } are non-negative combination coefficients meeting the condition: c l, = 0 if l / Ñ, 1 T C = 1 T where C is a N N matrix with individual entries { c l, }. The LTS criteria was introduced in [13], it is very robust to outliers and simultaneously possesses desirable asymptotic properties [13, 14, 16]. Accordingly, we obtain following Adapt-then- Combine (ATC) TDLMS and Combine-then-Adapt (CTA) TDLMS algorithms. ψ,i = ω,i 1 + µ ω,i ã l, ψ l,i l N ψ,i 1 ã l, ω l,i 1 l N ω,i = ψ,i 1 + µ c l, µ l N l,i (d l(i) µ l,i ω,i 1 ) (diffusion step) (diffusion step) (8) (9) l N c l, µ l,i (d l(i) µ l,i ψ,i 1 ) (10) The derivations are based on [6] and the details are omitted here.

3 IV. EXPERIMENTAL RESULTS In this section, we present simulation results to show the effectiveness of the proposed TDLMS. We focus on CTA TDLMS algorithm for demonstration purpose (the performance with ATC TDLMS can be tested similarly). We use the measures of the transient networ (mean-square deviation) and steady-state networ for performance comparison under alpha-stable noise, whose characteristic function is [17]: ψ γ,α (ω) = exp( γ ω α ), γ > 0, 0 < α 2 (11) The is defined as 1 K E ω,i ω o 2 (12) K =1 In the simulation, we consider a networ composed of 20 nodes and the length of the unnown parameter vector is 9, which is initialized as ω o = [0.1, 0.2, 0.3, 0.4, 0.5, 0.4, 0.3, 0.2, 0.1] T (13) We let each node connect to Q = 5 neighbor nodes and each node selects P neighbors to update its coefficients. The weights {ã l,i } and { c l,i } are set following [18] { 1/N, if l Ñ {ã l,i } = { c l,i } = (14) 0, otherwise The data µ,i and the noise ν,i are chosen as white Gaussian random sequences and the power of µ,i is uniformly set as 1. The results are averaged from 500 independent experiments. In Fig.2, we show how the values of the trimming constant affect the performance of TDLMS with the averaged convergence curves. We choose four different values: 5, 4, 3 and 2. When P = 5, the TDLMS reduces to DLMS. The stepsize is set at µ T DLMS = 0.02 and the system is trained for 2000 epochs which is long enough to guarantee it reaches the steady-state. The noise parameters are α = 1.5, γ = 0.1. As expected, when P gets larger, the algorithm will converge faster but its steady-state will get worse. From the figure we could probably say that P = 4 is a good choice and in the rest of the simulations we let P = 4. Next, we compare the performance of different algorithms, including LMS without cooperation, DLMS, TDLMS as well as global solutions GLMS and TGLMS [15, 19]. The trimming constant in TGLMS is experimentally set to 16. The step-sizes of the five algorithms are chosen as: µ NoCooperation = 0.05, µ DLMS = 0.01, µ T DLMS = 0.023, µ GLMS = , µ T GLMS = such that they produce roughly the same convergence speed. The results are illustrated in Fig. 3, from which we observe that: 1) the trimmed algorithms (TDLMS, TGLMS) perform better than non-trimmed counterparts (DLMS, GLMS) due to their robustness to outliers; 2) the global adaptive algorithms (GLMS, TGLMS) perform better than diffusion adaptive algorithms (DLMS, TDLMS) due to that global optimization but cost more resources. The P=2 P=3 P=4 P=5(DLMS) iteration Fig. 2. Transient networ of TDLMS for different trimming constants No cooperation DLMS TDLMS GLMS TGLMS iteration Fig. 3. Transient networ for different algorithms testing MSEs at final iteration are summarized in Table 1. The TDLMS algorithm is a little worse than TGLMS however saves a lot computational and communications resources, we argue it be the best choice among all these algorithms. We then let the exponential parameter α in the noise model vary within the range 1 to 2 at a step change 0.1, in order to test the performances of algorithms under noise of different distributions. The parameter settings are the same as in the previous simulation. The s averaged over the last 100 Algorithm No cooperation ± DLMS ± GLMS ± TDLMS ± e-05 TGLMS ± e-05 TABLE I AT FINAL ITERATION

4 No cooperation DLMS TDLMS GLMS TGLMS alpha Fig. 4. Steady-state networ for different α. iterations for different values are plotted in Fig. 4. As we can see that when α < 1.8, the trimmed algorithms perform better than non-trimmed counterparts while little worse when α 1.8. It is mainly because that when α comes near to 2, the noise will become Gaussian whose distribution is not longtailed (hence there are few outliers). Anyhow, the results is just a little inferior when noise follows Gaussian distribution. V. CONCLUSION In most practical situations, the real-world data obtained from the environment are often contaminated by outliers. The DLMS strategy, which is based on MSE criteria, offers an effective way to implement distributed adaptive filtering under the Gaussian assumption while performs poorly when the data are heavy tailed non-gaussian. To address this problem, we apply the idea of LTS to develop a new distributed adaptive filtering algorithm, namely the trimmed diffusion least mean square (TDLMS). Compared with the DLMS algorithm, TDLMS algorithm performs better under impulsive noise with alpha-stable distribution in the context of diffusion distributed estimation. Meanwhile, the proposed algorithm brought no significant increase in computational complexity. REFERENCES [1] Paolo Di Lorenzo, Sergio Barbarossa, and Ali H Sayed, Bio-inspired swarming for dynamic radio access based on diffusion adaptation, in Proc. 20th European Signal Processing Conference, [2] Xianghui Cao, Jiming Chen, Yang Xiao, and Youxian Sun, Building-environment control with wireless sensor and actuator networs: Centralized versus distributed, Industrial Electronics, IEEE Transactions on, vol. 57, no. 11, pp , [3] Jiming Chen, Xianghui Cao, Peng Cheng, Yang Xiao, and Youxian Sun, Distributed collaborative control for industrial automation with wireless sensor and actuator networs, Industrial Electronics, IEEE Transactions on, vol. 57, no. 12, pp , [4] Federico S Cattivelli, Cassio G Lopes, and Ali H Sayed, A diffusion rls scheme for distributed estimation over adaptive networs, in Signal Processing Advances in Wireless Communications, SPAWC IEEE 8th Worshop on. IEEE, 2007, pp [5] Federico S Cattivelli, Cassio G Lopes, and Ali H Sayed, Diffusion recursive least-squares for distributed estimation over adaptive networs, Signal Processing, IEEE Transactions on, vol. 56, no. 5, pp , [6] Federico S Cattivelli and Ali H Sayed, Diffusion lms strategies for distributed estimation, Signal Processing, IEEE Transactions on, vol. 58, no. 3, pp , [7] CassioG Lopes and Ali H Sayed, Distributed processing over adaptive networs, in Proc. adaptive sensor array processing worshop, 2006, pp [8] Cassio G Lopes and Ali H Sayed, Diffusion least-mean squares over adaptive networs., in ICASSP (3), 2007, pp [9] Cassio G Lopes and Ali H Sayed, Diffusion least-mean squares over adaptive networs: Formulation and performance analysis, Signal Processing, IEEE Transactions on, vol. 56, no. 7, pp , [10] Ying Liu, Chunguang Li, and Zhaoyang Zhang, Diffusion sparse least-mean squares over networs, Signal Processing, IEEE Transactions on, vol. 60, no. 8, pp , [11] Konstantinos N Plataniotis, Dimitrios Androutsos, and Anastasios N Venetsanopoulos, Nonlinear filtering of non-gaussian noise, Journal of Intelligent and Robotic Systems, vol. 19, no. 2, pp , [12] Binwei Weng and Kenneth E Barner, Nonlinear system identification in impulsive environments, Signal Processing, IEEE Transactions on, vol. 53, no. 7, pp , [13] Peter J Rousseeuw and Annic M Leroy, Robust regression and outlier detection, vol. 589, John Wiley & Sons, [14] Er-Wei Bai, A random least-trimmed-squares identification algorithm, Automatica, vol. 39, no. 9, pp , [15] Federico S Cattivelli and Ali H Sayed, Diffusion lms algorithms with information exchange, in Signals, Systems and Computers, nd Asilomar Conference on. IEEE, 2008, pp [16] Badong Chen, Xiaohan Yang, Hong Ji, Hua Qu, Nanning Zheng, and Jose C Principe, Trimmed affine projection algorithms, in Neural Networs (IJCNN), 2014 International Joint Conference on. IEEE, 2014, pp [17] Chrysostomos L Niias and Min Shao, Signal processing with alpha-stable distributions and applications, Wiley- Interscience, [18] Vincent Blondel, Julien M Hendricx, Alex Olshevsy, J Tsitsilis, et al., Convergence in multiagent coordi-

5 nation, consensus, and flocing, in IEEE Conference on Decision and Control. IEEE; 1998, 2005, vol. 44, p [19] Federico S Cattivelli and Ali H Sayed, Distributed detection over adaptive networs using diffusion adaptation, Signal Processing, IEEE Transactions on, vol. 59, no. 5, pp , 2011.

Robust diffusion maximum correntropy criterion algorithm for distributed network estimation

Robust diffusion maximum correntropy criterion algorithm for distributed network estimation Robust diffusion maximum correntropy criterion algorithm for distributed networ estimation Wentao Ma, Hua Qu, Badong Chen *, Jihong Zhao,, Jiandong Duan 3 School of Electronic and Information Engineering,

More information

STEADY-STATE MEAN SQUARE PERFORMANCE OF A SPARSIFIED KERNEL LEAST MEAN SQUARE ALGORITHM.

STEADY-STATE MEAN SQUARE PERFORMANCE OF A SPARSIFIED KERNEL LEAST MEAN SQUARE ALGORITHM. STEADY-STATE MEAN SQUARE PERFORMANCE OF A SPARSIFIED KERNEL LEAST MEAN SQUARE ALGORITHM Badong Chen 1, Zhengda Qin 1, Lei Sun 2 1 Institute of Artificial Intelligence and Robotics, Xi an Jiaotong University,

More information

Diffusion Maximum Correntropy Criterion Algorithms for Robust Distributed Estimation

Diffusion Maximum Correntropy Criterion Algorithms for Robust Distributed Estimation Diffusion Maximum Correntropy Criterion Algorithms for Robust Distributed Estimation Wentao Ma a, Badong Chen b, *,Jiandong Duan a,haiquan Zhao c a Department of Electrical Engineering, Xi an University

More information

Analysis of incremental RLS adaptive networks with noisy links

Analysis of incremental RLS adaptive networks with noisy links Analysis of incremental RLS adaptive networs with noisy lins Azam Khalili, Mohammad Ali Tinati, and Amir Rastegarnia a) Faculty of Electrical and Computer Engineering, University of Tabriz Tabriz 51664,

More information

Diffusion LMS Algorithms for Sensor Networks over Non-ideal Inter-sensor Wireless Channels

Diffusion LMS Algorithms for Sensor Networks over Non-ideal Inter-sensor Wireless Channels Diffusion LMS Algorithms for Sensor Networs over Non-ideal Inter-sensor Wireless Channels Reza Abdolee and Benoit Champagne Electrical and Computer Engineering McGill University 3480 University Street

More information

DISTRIBUTED DIFFUSION-BASED LMS FOR NODE-SPECIFIC PARAMETER ESTIMATION OVER ADAPTIVE NETWORKS. & C.T.I RU-8, 26500, Rio - Patra, Greece

DISTRIBUTED DIFFUSION-BASED LMS FOR NODE-SPECIFIC PARAMETER ESTIMATION OVER ADAPTIVE NETWORKS. & C.T.I RU-8, 26500, Rio - Patra, Greece 2014 IEEE International Conference on Acoustic, Speech Signal Processing (ICASSP) DISTRIBUTED DIFFUSION-BASED LMS FOR NODE-SPECIFIC PARAMETER ESTIMATION OVER ADAPTIVE NETWORKS Niola Bogdanović 1, Jorge

More information

EUSIPCO

EUSIPCO EUSIPCO 3 569736677 FULLY ISTRIBUTE SIGNAL ETECTION: APPLICATION TO COGNITIVE RAIO Franc Iutzeler Philippe Ciblat Telecom ParisTech, 46 rue Barrault 753 Paris, France email: firstnamelastname@telecom-paristechfr

More information

STOCHASTIC INFORMATION GRADIENT ALGORITHM BASED ON MAXIMUM ENTROPY DENSITY ESTIMATION. Badong Chen, Yu Zhu, Jinchun Hu and Ming Zhang

STOCHASTIC INFORMATION GRADIENT ALGORITHM BASED ON MAXIMUM ENTROPY DENSITY ESTIMATION. Badong Chen, Yu Zhu, Jinchun Hu and Ming Zhang ICIC Express Letters ICIC International c 2009 ISSN 1881-803X Volume 3, Number 3, September 2009 pp. 1 6 STOCHASTIC INFORMATION GRADIENT ALGORITHM BASED ON MAXIMUM ENTROPY DENSITY ESTIMATION Badong Chen,

More information

Variable Learning Rate LMS Based Linear Adaptive Inverse Control *

Variable Learning Rate LMS Based Linear Adaptive Inverse Control * ISSN 746-7659, England, UK Journal of Information and Computing Science Vol., No. 3, 6, pp. 39-48 Variable Learning Rate LMS Based Linear Adaptive Inverse Control * Shuying ie, Chengjin Zhang School of

More information

A Quality-aware Incremental LMS Algorithm for Distributed Adaptive Estimation

A Quality-aware Incremental LMS Algorithm for Distributed Adaptive Estimation International Journal of Automation and Computing 11(6), December 2014, 676-682 DOI: 10.1007/s11633-014-0838-x A Quality-aware Incremental LMS Algorithm for Distributed Adaptive Estimation Wael M. Bazzi

More information

Maximum Likelihood Diffusive Source Localization Based on Binary Observations

Maximum Likelihood Diffusive Source Localization Based on Binary Observations Maximum Lielihood Diffusive Source Localization Based on Binary Observations Yoav Levinboo and an F. Wong Wireless Information Networing Group, University of Florida Gainesville, Florida 32611-6130, USA

More information

Growing Window Recursive Quadratic Optimization with Variable Regularization

Growing Window Recursive Quadratic Optimization with Variable Regularization 49th IEEE Conference on Decision and Control December 15-17, Hilton Atlanta Hotel, Atlanta, GA, USA Growing Window Recursive Quadratic Optimization with Variable Regularization Asad A. Ali 1, Jesse B.

More information

Sliding Window Recursive Quadratic Optimization with Variable Regularization

Sliding Window Recursive Quadratic Optimization with Variable Regularization 11 American Control Conference on O'Farrell Street, San Francisco, CA, USA June 29 - July 1, 11 Sliding Window Recursive Quadratic Optimization with Variable Regularization Jesse B. Hoagg, Asad A. Ali,

More information

Regression Analysis for Data Containing Outliers and High Leverage Points

Regression Analysis for Data Containing Outliers and High Leverage Points Alabama Journal of Mathematics 39 (2015) ISSN 2373-0404 Regression Analysis for Data Containing Outliers and High Leverage Points Asim Kumer Dey Department of Mathematics Lamar University Md. Amir Hossain

More information

Flocking of Discrete-time Multi-Agent Systems with Predictive Mechanisms

Flocking of Discrete-time Multi-Agent Systems with Predictive Mechanisms Preprints of the 8th IFAC World Congress Milano (Italy) August 28 - September 2, 2 Flocing of Discrete-time Multi-Agent Systems Predictive Mechanisms Jingyuan Zhan, Xiang Li Adaptive Networs and Control

More information

Distributed Signal Processing Algorithms for Wireless Networks

Distributed Signal Processing Algorithms for Wireless Networks Distributed Signal Processing Algorithms for Wireless Networks This thesis is submitted in partial fulfilment of the requirements for Doctor of Philosophy (Ph.D.) Songcen Xu Communications and Signal Processing

More information

CTA diffusion based recursive energy detection

CTA diffusion based recursive energy detection CTA diffusion based recursive energy detection AHTI AINOMÄE KTH Royal Institute of Technology Department of Signal Processing. Tallinn University of Technology Department of Radio and Telecommunication

More information

Multi-wind Field Output Power Prediction Method based on Energy Internet and DBPSO-LSSVM

Multi-wind Field Output Power Prediction Method based on Energy Internet and DBPSO-LSSVM , pp.128-133 http://dx.doi.org/1.14257/astl.16.138.27 Multi-wind Field Output Power Prediction Method based on Energy Internet and DBPSO-LSSVM *Jianlou Lou 1, Hui Cao 1, Bin Song 2, Jizhe Xiao 1 1 School

More information

Distributed Optimization over Networks Gossip-Based Algorithms

Distributed Optimization over Networks Gossip-Based Algorithms Distributed Optimization over Networks Gossip-Based Algorithms Angelia Nedić angelia@illinois.edu ISE Department and Coordinated Science Laboratory University of Illinois at Urbana-Champaign Outline Random

More information

2262 IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 47, NO. 8, AUGUST A General Class of Nonlinear Normalized Adaptive Filtering Algorithms

2262 IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 47, NO. 8, AUGUST A General Class of Nonlinear Normalized Adaptive Filtering Algorithms 2262 IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 47, NO. 8, AUGUST 1999 A General Class of Nonlinear Normalized Adaptive Filtering Algorithms Sudhakar Kalluri, Member, IEEE, and Gonzalo R. Arce, Senior

More information

BAYESIAN ESTIMATION OF UNKNOWN PARAMETERS OVER NETWORKS

BAYESIAN ESTIMATION OF UNKNOWN PARAMETERS OVER NETWORKS BAYESIAN ESTIMATION OF UNKNOWN PARAMETERS OVER NETWORKS Petar M. Djurić Dept. of Electrical & Computer Engineering Stony Brook University Stony Brook, NY 11794, USA e-mail: petar.djuric@stonybrook.edu

More information

Recursive Generalized Eigendecomposition for Independent Component Analysis

Recursive Generalized Eigendecomposition for Independent Component Analysis Recursive Generalized Eigendecomposition for Independent Component Analysis Umut Ozertem 1, Deniz Erdogmus 1,, ian Lan 1 CSEE Department, OGI, Oregon Health & Science University, Portland, OR, USA. {ozertemu,deniz}@csee.ogi.edu

More information

Research on Consistency Problem of Network Multi-agent Car System with State Predictor

Research on Consistency Problem of Network Multi-agent Car System with State Predictor International Core Journal of Engineering Vol. No.9 06 ISSN: 44-895 Research on Consistency Problem of Network Multi-agent Car System with State Predictor Yue Duan a, Linli Zhou b and Yue Wu c Institute

More information

On the Stability of the Least-Mean Fourth (LMF) Algorithm

On the Stability of the Least-Mean Fourth (LMF) Algorithm XXI SIMPÓSIO BRASILEIRO DE TELECOMUNICACÕES-SBT 4, 6-9 DE SETEMBRO DE 4, BELÉM, PA On the Stability of the Least-Mean Fourth (LMF) Algorithm Vítor H. Nascimento and José Carlos M. Bermudez + Abstract We

More information

A Hybrid Time-delay Prediction Method for Networked Control System

A Hybrid Time-delay Prediction Method for Networked Control System International Journal of Automation and Computing 11(1), February 2014, 19-24 DOI: 10.1007/s11633-014-0761-1 A Hybrid Time-delay Prediction Method for Networked Control System Zhong-Da Tian Xian-Wen Gao

More information

Distributed Randomized Algorithms for the PageRank Computation Hideaki Ishii, Member, IEEE, and Roberto Tempo, Fellow, IEEE

Distributed Randomized Algorithms for the PageRank Computation Hideaki Ishii, Member, IEEE, and Roberto Tempo, Fellow, IEEE IEEE TRANSACTIONS ON AUTOMATIC CONTROL, VOL. 55, NO. 9, SEPTEMBER 2010 1987 Distributed Randomized Algorithms for the PageRank Computation Hideaki Ishii, Member, IEEE, and Roberto Tempo, Fellow, IEEE Abstract

More information

Gas Detection System Based on Multi-Sensor Fusion with BP Neural Network

Gas Detection System Based on Multi-Sensor Fusion with BP Neural Network Sensors & Transducers 2013 by IFSA http://www.sensorsportal.com Gas Detection System Based on Multi-Sensor Fusion with BP Neural Network Qiu-Xia LIU Department of Physics, Heze University, Heze Shandong

More information

Performance Analysis of a Threshold-Based Relay Selection Algorithm in Wireless Networks

Performance Analysis of a Threshold-Based Relay Selection Algorithm in Wireless Networks Communications and Networ, 2010, 2, 87-92 doi:10.4236/cn.2010.22014 Published Online May 2010 (http://www.scirp.org/journal/cn Performance Analysis of a Threshold-Based Relay Selection Algorithm in Wireless

More information

Discrete-time Consensus Filters on Directed Switching Graphs

Discrete-time Consensus Filters on Directed Switching Graphs 214 11th IEEE International Conference on Control & Automation (ICCA) June 18-2, 214. Taichung, Taiwan Discrete-time Consensus Filters on Directed Switching Graphs Shuai Li and Yi Guo Abstract We consider

More information

An Improved Method of Power System Short Term Load Forecasting Based on Neural Network

An Improved Method of Power System Short Term Load Forecasting Based on Neural Network An Improved Method of Power System Short Term Load Forecasting Based on Neural Network Shunzhou Wang School of Electrical and Electronic Engineering Huailin Zhao School of Electrical and Electronic Engineering

More information

Sparse Least Mean Square Algorithm for Estimation of Truncated Volterra Kernels

Sparse Least Mean Square Algorithm for Estimation of Truncated Volterra Kernels Sparse Least Mean Square Algorithm for Estimation of Truncated Volterra Kernels Bijit Kumar Das 1, Mrityunjoy Chakraborty 2 Department of Electronics and Electrical Communication Engineering Indian Institute

More information

Local Strong Convexity of Maximum-Likelihood TDOA-Based Source Localization and Its Algorithmic Implications

Local Strong Convexity of Maximum-Likelihood TDOA-Based Source Localization and Its Algorithmic Implications Local Strong Convexity of Maximum-Likelihood TDOA-Based Source Localization and Its Algorithmic Implications Huikang Liu, Yuen-Man Pun, and Anthony Man-Cho So Dept of Syst Eng & Eng Manag, The Chinese

More information

Distributed Coordinated Tracking With Reduced Interaction via a Variable Structure Approach Yongcan Cao, Member, IEEE, and Wei Ren, Member, IEEE

Distributed Coordinated Tracking With Reduced Interaction via a Variable Structure Approach Yongcan Cao, Member, IEEE, and Wei Ren, Member, IEEE IEEE TRANSACTIONS ON AUTOMATIC CONTROL, VOL. 57, NO. 1, JANUARY 2012 33 Distributed Coordinated Tracking With Reduced Interaction via a Variable Structure Approach Yongcan Cao, Member, IEEE, and Wei Ren,

More information

Computing Maximum Entropy Densities: A Hybrid Approach

Computing Maximum Entropy Densities: A Hybrid Approach Computing Maximum Entropy Densities: A Hybrid Approach Badong Chen Department of Precision Instruments and Mechanology Tsinghua University Beijing, 84, P. R. China Jinchun Hu Department of Precision Instruments

More information

Distributed Estimation and Detection for Smart Grid

Distributed Estimation and Detection for Smart Grid Distributed Estimation and Detection for Smart Grid Texas A&M University Joint Wor with: S. Kar (CMU), R. Tandon (Princeton), H. V. Poor (Princeton), and J. M. F. Moura (CMU) 1 Distributed Estimation/Detection

More information

MANY real-word applications require complex nonlinear

MANY real-word applications require complex nonlinear Proceedings of International Joint Conference on Neural Networks, San Jose, California, USA, July 31 August 5, 2011 Kernel Adaptive Filtering with Maximum Correntropy Criterion Songlin Zhao, Badong Chen,

More information

KNOWN approaches for improving the performance of

KNOWN approaches for improving the performance of IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II: EXPRESS BRIEFS, VOL. 58, NO. 8, AUGUST 2011 537 Robust Quasi-Newton Adaptive Filtering Algorithms Md. Zulfiquar Ali Bhotto, Student Member, IEEE, and Andreas

More information

Dominant Pole Localization of FxLMS Adaptation Process in Active Noise Control

Dominant Pole Localization of FxLMS Adaptation Process in Active Noise Control APSIPA ASC 20 Xi an Dominant Pole Localization of FxLMS Adaptation Process in Active Noise Control Iman Tabatabaei Ardekani, Waleed H. Abdulla The University of Auckland, Private Bag 9209, Auckland, New

More information

A Robust Event-Triggered Consensus Strategy for Linear Multi-Agent Systems with Uncertain Network Topology

A Robust Event-Triggered Consensus Strategy for Linear Multi-Agent Systems with Uncertain Network Topology A Robust Event-Triggered Consensus Strategy for Linear Multi-Agent Systems with Uncertain Network Topology Amir Amini, Amir Asif, Arash Mohammadi Electrical and Computer Engineering,, Montreal, Canada.

More information

Recursive Least Squares for an Entropy Regularized MSE Cost Function

Recursive Least Squares for an Entropy Regularized MSE Cost Function Recursive Least Squares for an Entropy Regularized MSE Cost Function Deniz Erdogmus, Yadunandana N. Rao, Jose C. Principe Oscar Fontenla-Romero, Amparo Alonso-Betanzos Electrical Eng. Dept., University

More information

Zhang Neural Network without Using Time-Derivative Information for Constant and Time-Varying Matrix Inversion

Zhang Neural Network without Using Time-Derivative Information for Constant and Time-Varying Matrix Inversion Zhang Neural Network without Using Time-Derivative Information for Constant and Time-Varying Matrix Inversion Yunong Zhang, Member, IEEE, Zenghai Chen, Ke Chen, and Binghuang Cai Abstract To obtain the

More information

A Graph-Theoretic Characterization of Structural Controllability for Multi-Agent System with Switching Topology

A Graph-Theoretic Characterization of Structural Controllability for Multi-Agent System with Switching Topology Joint 48th IEEE Conference on Decision and Control and 28th Chinese Control Conference Shanghai, P.R. China, December 16-18, 29 FrAIn2.3 A Graph-Theoretic Characterization of Structural Controllability

More information

Model-based Correlation Measure for Gain and Offset Nonuniformity in Infrared Focal-Plane-Array Sensors

Model-based Correlation Measure for Gain and Offset Nonuniformity in Infrared Focal-Plane-Array Sensors Model-based Correlation Measure for Gain and Offset Nonuniformity in Infrared Focal-Plane-Array Sensors César San Martin Sergio Torres Abstract In this paper, a model-based correlation measure between

More information

RECENTLY, wireless sensor networks have been the object

RECENTLY, wireless sensor networks have been the object IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 55, NO. 4, APRIL 2007 1511 Distributed Sequential Bayesian Estimation of a Diffusive Source in Wireless Sensor Networks Tong Zhao, Student Member, IEEE, and

More information

Distributed Consensus Optimization

Distributed Consensus Optimization Distributed Consensus Optimization Ming Yan Michigan State University, CMSE/Mathematics September 14, 2018 Decentralized-1 Backgroundwhy andwe motivation need decentralized optimization? I Decentralized

More information

Method for Optimizing the Number and Precision of Interval-Valued Parameters in a Multi-Object System

Method for Optimizing the Number and Precision of Interval-Valued Parameters in a Multi-Object System Method for Optimizing the Number and Precision of Interval-Valued Parameters in a Multi-Object System AMAURY CABALLERO*, KANG YEN*, *, JOSE L. ABREU**, ALDO PARDO*** *Department of Electrical & Computer

More information

A STATE-SPACE APPROACH FOR THE ANALYSIS OF WAVE AND DIFFUSION FIELDS

A STATE-SPACE APPROACH FOR THE ANALYSIS OF WAVE AND DIFFUSION FIELDS ICASSP 2015 A STATE-SPACE APPROACH FOR THE ANALYSIS OF WAVE AND DIFFUSION FIELDS Stefano Maranò Donat Fäh Hans-Andrea Loeliger ETH Zurich, Swiss Seismological Service, 8092 Zürich ETH Zurich, Dept. Information

More information

Efficient Distributed State Estimation of Hidden Markov Models over Unreliable Networks

Efficient Distributed State Estimation of Hidden Markov Models over Unreliable Networks Efficient Distributed State Estimation of Hidden Marov Models over Unreliable Networs Amirhossein Tamjidi 1, Reza Oftadeh 2, Suman Charavorty 1, Dylan Shell 2 Abstract This paper presents a new recursive

More information

Multilayer Perceptron = FeedForward Neural Network

Multilayer Perceptron = FeedForward Neural Network Multilayer Perceptron = FeedForward Neural Networ History Definition Classification = feedforward operation Learning = bacpropagation = local optimization in the space of weights Pattern Classification

More information

Linear Prediction Theory

Linear Prediction Theory Linear Prediction Theory Joseph A. O Sullivan ESE 524 Spring 29 March 3, 29 Overview The problem of estimating a value of a random process given other values of the random process is pervasive. Many problems

More information

L-statistics based Modification of Reconstruction Algorithms for Compressive Sensing in the Presence of Impulse Noise

L-statistics based Modification of Reconstruction Algorithms for Compressive Sensing in the Presence of Impulse Noise L-statistics based Modification of Reconstruction Algorithms for Compressive Sensing in the Presence of Impulse Noise Srdjan Stanković, Irena Orović and Moeness Amin 1 Abstract- A modification of standard

More information

Time Synchronization in WSNs: A Maximum Value Based Consensus Approach

Time Synchronization in WSNs: A Maximum Value Based Consensus Approach 211 5th IEEE Conference on Decision and Control and European Control Conference (CDC-ECC) Orlando, FL, USA, December 12-15, 211 Time Synchronization in WSNs: A Maximum Value Based Consensus Approach Jianping

More information

A novel dual H infinity filters based battery parameter and state estimation approach for electric vehicles application

A novel dual H infinity filters based battery parameter and state estimation approach for electric vehicles application Available online at www.sciencedirect.com ScienceDirect Energy Procedia 3 (26 ) 375 38 Applied Energy Symposium and Forum, REM26: Renewable Energy Integration with Mini/Microgrid, 9-2 April 26, Maldives

More information

EM-algorithm for Training of State-space Models with Application to Time Series Prediction

EM-algorithm for Training of State-space Models with Application to Time Series Prediction EM-algorithm for Training of State-space Models with Application to Time Series Prediction Elia Liitiäinen, Nima Reyhani and Amaury Lendasse Helsinki University of Technology - Neural Networks Research

More information

Blind Source Separation Using Artificial immune system

Blind Source Separation Using Artificial immune system American Journal of Engineering Research (AJER) e-issn : 2320-0847 p-issn : 2320-0936 Volume-03, Issue-02, pp-240-247 www.ajer.org Research Paper Open Access Blind Source Separation Using Artificial immune

More information

Combination of M-Estimators and Neural Network Model to Analyze Inside/Outside Bark Tree Diameters

Combination of M-Estimators and Neural Network Model to Analyze Inside/Outside Bark Tree Diameters Combination of M-Estimators and Neural Network Model to Analyze Inside/Outside Bark Tree Diameters Kyriaki Kitikidou, Elias Milios, Lazaros Iliadis, and Minas Kaymakis Democritus University of Thrace,

More information

Transfer Function Identification from Phase Response Data

Transfer Function Identification from Phase Response Data ransfer Function Identification from Phase Response Data Luciano De ommasi, Dir Deschrijver and om Dhaene his paper introduces an improved procedure for the identification of a transfer function from phase

More information

Holdout and Cross-Validation Methods Overfitting Avoidance

Holdout and Cross-Validation Methods Overfitting Avoidance Holdout and Cross-Validation Methods Overfitting Avoidance Decision Trees Reduce error pruning Cost-complexity pruning Neural Networks Early stopping Adjusting Regularizers via Cross-Validation Nearest

More information

Fully-distributed spectrum sensing: application to cognitive radio

Fully-distributed spectrum sensing: application to cognitive radio Fully-distributed spectrum sensing: application to cognitive radio Philippe Ciblat Dpt Comelec, Télécom ParisTech Joint work with F. Iutzeler (PhD student funded by DGA grant) Cognitive radio principle

More information

A NOVEL OPTIMAL PROBABILITY DENSITY FUNCTION TRACKING FILTER DESIGN 1

A NOVEL OPTIMAL PROBABILITY DENSITY FUNCTION TRACKING FILTER DESIGN 1 A NOVEL OPTIMAL PROBABILITY DENSITY FUNCTION TRACKING FILTER DESIGN 1 Jinglin Zhou Hong Wang, Donghua Zhou Department of Automation, Tsinghua University, Beijing 100084, P. R. China Control Systems Centre,

More information

Sparse representation classification and positive L1 minimization

Sparse representation classification and positive L1 minimization Sparse representation classification and positive L1 minimization Cencheng Shen Joint Work with Li Chen, Carey E. Priebe Applied Mathematics and Statistics Johns Hopkins University, August 5, 2014 Cencheng

More information

EVALUATING SYMMETRIC INFORMATION GAP BETWEEN DYNAMICAL SYSTEMS USING PARTICLE FILTER

EVALUATING SYMMETRIC INFORMATION GAP BETWEEN DYNAMICAL SYSTEMS USING PARTICLE FILTER EVALUATING SYMMETRIC INFORMATION GAP BETWEEN DYNAMICAL SYSTEMS USING PARTICLE FILTER Zhen Zhen 1, Jun Young Lee 2, and Abdus Saboor 3 1 Mingde College, Guizhou University, China zhenz2000@21cn.com 2 Department

More information

Dynamic Data Modeling of SCR De-NOx System Based on NARX Neural Network Wen-jie ZHAO * and Kai ZHANG

Dynamic Data Modeling of SCR De-NOx System Based on NARX Neural Network Wen-jie ZHAO * and Kai ZHANG 2018 International Conference on Modeling, Simulation and Analysis (ICMSA 2018) ISBN: 978-1-60595-544-5 Dynamic Data Modeling of SCR De-NOx System Based on NARX Neural Network Wen-jie ZHAO * and Kai ZHANG

More information

REDUCING POWER CONSUMPTION IN A SENSOR NETWORK BY INFORMATION FEEDBACK

REDUCING POWER CONSUMPTION IN A SENSOR NETWORK BY INFORMATION FEEDBACK REDUCING POWER CONSUMPTION IN A SENSOR NETWOR BY INFORMATION FEEDBAC Mikalai isialiou and Zhi-Quan Luo Department of Electrical and Computer Engineering University of Minnesota Minneapolis, MN, 55455,

More information

EXTENDING PARTIAL LEAST SQUARES REGRESSION

EXTENDING PARTIAL LEAST SQUARES REGRESSION EXTENDING PARTIAL LEAST SQUARES REGRESSION ATHANASSIOS KONDYLIS UNIVERSITY OF NEUCHÂTEL 1 Outline Multivariate Calibration in Chemometrics PLS regression (PLSR) and the PLS1 algorithm PLS1 from a statistical

More information

Asymptotic Distribution of The Number of Isolated Nodes in Wireless Ad Hoc Networks with Unreliable Nodes and Links

Asymptotic Distribution of The Number of Isolated Nodes in Wireless Ad Hoc Networks with Unreliable Nodes and Links Asymptotic Distribution of The Number of Isolated Nodes in Wireless Ad Hoc Networs with Unreliable Nodes and Lins Chih-Wei Yi, Peng-Jun Wan, Kuo-Wei Lin and Chih-Hao Huang Department of Computer Science

More information

IN recent years, the problems of sparse signal recovery

IN recent years, the problems of sparse signal recovery IEEE/CAA JOURNAL OF AUTOMATICA SINICA, VOL. 1, NO. 2, APRIL 2014 149 Distributed Sparse Signal Estimation in Sensor Networs Using H -Consensus Filtering Haiyang Yu Yisha Liu Wei Wang Abstract This paper

More information

MODELING AND EXPERIMENTAL STUDY ON DRILLING RIG ANTI-JAMMING VALVE WITH BP NEURAL NETWORK

MODELING AND EXPERIMENTAL STUDY ON DRILLING RIG ANTI-JAMMING VALVE WITH BP NEURAL NETWORK Engineering Review, Vol. 3, Issue 2, 99-0, 20. 99 MODELING AND EXPERIMENTAL STUDY ON DRILLING RIG ANTI-JAMMING VALVE WITH BP NEURAL NETWORK Wei Ma * Fei Ma School of Mechanical Engineering, University

More information

Pattern Classification

Pattern Classification Pattern Classification All materials in these slides were taen from Pattern Classification (2nd ed) by R. O. Duda,, P. E. Hart and D. G. Stor, John Wiley & Sons, 2000 with the permission of the authors

More information

New Recursive-Least-Squares Algorithms for Nonlinear Active Control of Sound and Vibration Using Neural Networks

New Recursive-Least-Squares Algorithms for Nonlinear Active Control of Sound and Vibration Using Neural Networks IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 12, NO. 1, JANUARY 2001 135 New Recursive-Least-Squares Algorithms for Nonlinear Active Control of Sound and Vibration Using Neural Networks Martin Bouchard,

More information

Distributed Detection and Estimation in Wireless Sensor Networks: Resource Allocation, Fusion Rules, and Network Security

Distributed Detection and Estimation in Wireless Sensor Networks: Resource Allocation, Fusion Rules, and Network Security Distributed Detection and Estimation in Wireless Sensor Networks: Resource Allocation, Fusion Rules, and Network Security Edmond Nurellari The University of Leeds, UK School of Electronic and Electrical

More information

A Robust Strategy for Joint Data Reconciliation and Parameter Estimation

A Robust Strategy for Joint Data Reconciliation and Parameter Estimation A Robust Strategy for Joint Data Reconciliation and Parameter Estimation Yen Yen Joe 1) 3), David Wang ), Chi Bun Ching 3), Arthur Tay 1), Weng Khuen Ho 1) and Jose Romagnoli ) * 1) Dept. of Electrical

More information

Sparse Sensing for Statistical Inference

Sparse Sensing for Statistical Inference Sparse Sensing for Statistical Inference Model-driven and data-driven paradigms Geert Leus, Sundeep Chepuri, and Georg Kail ITA 2016, 04 Feb 2016 1/17 Power networks, grid analytics Health informatics

More information

ADMM and Fast Gradient Methods for Distributed Optimization

ADMM and Fast Gradient Methods for Distributed Optimization ADMM and Fast Gradient Methods for Distributed Optimization João Xavier Instituto Sistemas e Robótica (ISR), Instituto Superior Técnico (IST) European Control Conference, ECC 13 July 16, 013 Joint work

More information

Distributed and Recursive Parameter Estimation in Parametrized Linear State-Space Models

Distributed and Recursive Parameter Estimation in Parametrized Linear State-Space Models 1 Distributed and Recursive Parameter Estimation in Parametrized Linear State-Space Models S. Sundhar Ram, V. V. Veeravalli, and A. Nedić Abstract We consider a network of sensors deployed to sense a spatio-temporal

More information

ARTIFICIAL NEURAL NETWORK WITH HYBRID TAGUCHI-GENETIC ALGORITHM FOR NONLINEAR MIMO MODEL OF MACHINING PROCESSES

ARTIFICIAL NEURAL NETWORK WITH HYBRID TAGUCHI-GENETIC ALGORITHM FOR NONLINEAR MIMO MODEL OF MACHINING PROCESSES International Journal of Innovative Computing, Information and Control ICIC International c 2013 ISSN 1349-4198 Volume 9, Number 4, April 2013 pp. 1455 1475 ARTIFICIAL NEURAL NETWORK WITH HYBRID TAGUCHI-GENETIC

More information

Cooperative Spectrum Sensing for Cognitive Radios under Bandwidth Constraints

Cooperative Spectrum Sensing for Cognitive Radios under Bandwidth Constraints Cooperative Spectrum Sensing for Cognitive Radios under Bandwidth Constraints Chunhua Sun, Wei Zhang, and haled Ben Letaief, Fellow, IEEE Department of Electronic and Computer Engineering The Hong ong

More information

Learning Gaussian Process Models from Uncertain Data

Learning Gaussian Process Models from Uncertain Data Learning Gaussian Process Models from Uncertain Data Patrick Dallaire, Camille Besse, and Brahim Chaib-draa DAMAS Laboratory, Computer Science & Software Engineering Department, Laval University, Canada

More information

Diffusion based Projection Method for Distributed Source Localization in Wireless Sensor Networks

Diffusion based Projection Method for Distributed Source Localization in Wireless Sensor Networks The Third International Workshop on Wireless Sensor, Actuator and Robot Networks Diffusion based Projection Method for Distributed Source Localization in Wireless Sensor Networks Wei Meng 1, Wendong Xiao,

More information

LECTURE # - NEURAL COMPUTATION, Feb 04, Linear Regression. x 1 θ 1 output... θ M x M. Assumes a functional form

LECTURE # - NEURAL COMPUTATION, Feb 04, Linear Regression. x 1 θ 1 output... θ M x M. Assumes a functional form LECTURE # - EURAL COPUTATIO, Feb 4, 4 Linear Regression Assumes a functional form f (, θ) = θ θ θ K θ (Eq) where = (,, ) are the attributes and θ = (θ, θ, θ ) are the function parameters Eample: f (, θ)

More information

ARTIFICIAL NEURAL NETWORKS گروه مطالعاتي 17 بهار 92

ARTIFICIAL NEURAL NETWORKS گروه مطالعاتي 17 بهار 92 ARTIFICIAL NEURAL NETWORKS گروه مطالعاتي 17 بهار 92 BIOLOGICAL INSPIRATIONS Some numbers The human brain contains about 10 billion nerve cells (neurons) Each neuron is connected to the others through 10000

More information

Encoder Decoder Design for Event-Triggered Feedback Control over Bandlimited Channels

Encoder Decoder Design for Event-Triggered Feedback Control over Bandlimited Channels Encoder Decoder Design for Event-Triggered Feedback Control over Bandlimited Channels Lei Bao, Mikael Skoglund and Karl Henrik Johansson Department of Signals, Sensors and Systems, Royal Institute of Technology,

More information

Statistical Learning Theory and the C-Loss cost function

Statistical Learning Theory and the C-Loss cost function Statistical Learning Theory and the C-Loss cost function Jose Principe, Ph.D. Distinguished Professor ECE, BME Computational NeuroEngineering Laboratory and principe@cnel.ufl.edu Statistical Learning Theory

More information

Myoelectrical signal classification based on S transform and two-directional 2DPCA

Myoelectrical signal classification based on S transform and two-directional 2DPCA Myoelectrical signal classification based on S transform and two-directional 2DPCA Hong-Bo Xie1 * and Hui Liu2 1 ARC Centre of Excellence for Mathematical and Statistical Frontiers Queensland University

More information

An Adaptive Sensor Array Using an Affine Combination of Two Filters

An Adaptive Sensor Array Using an Affine Combination of Two Filters An Adaptive Sensor Array Using an Affine Combination of Two Filters Tõnu Trump Tallinn University of Technology Department of Radio and Telecommunication Engineering Ehitajate tee 5, 19086 Tallinn Estonia

More information

LIMBO Self-Test Method using binary input and dithering signals

LIMBO Self-Test Method using binary input and dithering signals LIMBO Self-Test Method using binary input and dithering signals Laurent Bourgois, Jérome Juillard To cite this version: Laurent Bourgois, Jérome Juillard. LIMBO Self-Test Method using binary input and

More information

DIFFUSION-BASED DISTRIBUTED MVDR BEAMFORMER

DIFFUSION-BASED DISTRIBUTED MVDR BEAMFORMER 14 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) DIFFUSION-BASED DISTRIBUTED MVDR BEAMFORMER Matt O Connor 1 and W. Bastiaan Kleijn 1,2 1 School of Engineering and Computer

More information

Alternative Biased Estimator Based on Least. Trimmed Squares for Handling Collinear. Leverage Data Points

Alternative Biased Estimator Based on Least. Trimmed Squares for Handling Collinear. Leverage Data Points International Journal of Contemporary Mathematical Sciences Vol. 13, 018, no. 4, 177-189 HIKARI Ltd, www.m-hikari.com https://doi.org/10.1988/ijcms.018.8616 Alternative Biased Estimator Based on Least

More information

On the Scalability in Cooperative Control. Zhongkui Li. Peking University

On the Scalability in Cooperative Control. Zhongkui Li. Peking University On the Scalability in Cooperative Control Zhongkui Li Email: zhongkli@pku.edu.cn Peking University June 25, 2016 Zhongkui Li (PKU) Scalability June 25, 2016 1 / 28 Background Cooperative control is to

More information

Multi-Robotic Systems

Multi-Robotic Systems CHAPTER 9 Multi-Robotic Systems The topic of multi-robotic systems is quite popular now. It is believed that such systems can have the following benefits: Improved performance ( winning by numbers ) Distributed

More information

ARock: an algorithmic framework for asynchronous parallel coordinate updates

ARock: an algorithmic framework for asynchronous parallel coordinate updates ARock: an algorithmic framework for asynchronous parallel coordinate updates Zhimin Peng, Yangyang Xu, Ming Yan, Wotao Yin ( UCLA Math, U.Waterloo DCO) UCLA CAM Report 15-37 ShanghaiTech SSDS 15 June 25,

More information

NON-LINEAR NOISE ADAPTIVE KALMAN FILTERING VIA VARIATIONAL BAYES

NON-LINEAR NOISE ADAPTIVE KALMAN FILTERING VIA VARIATIONAL BAYES 2013 IEEE INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING NON-LINEAR NOISE ADAPTIVE KALMAN FILTERING VIA VARIATIONAL BAYES Simo Särä Aalto University, 02150 Espoo, Finland Jouni Hartiainen

More information

Encoder Decoder Design for Event-Triggered Feedback Control over Bandlimited Channels

Encoder Decoder Design for Event-Triggered Feedback Control over Bandlimited Channels Encoder Decoder Design for Event-Triggered Feedback Control over Bandlimited Channels LEI BAO, MIKAEL SKOGLUND AND KARL HENRIK JOHANSSON IR-EE- 26: Stockholm 26 Signal Processing School of Electrical Engineering

More information

Application of Artificial Neural Networks in Evaluation and Identification of Electrical Loss in Transformers According to the Energy Consumption

Application of Artificial Neural Networks in Evaluation and Identification of Electrical Loss in Transformers According to the Energy Consumption Application of Artificial Neural Networks in Evaluation and Identification of Electrical Loss in Transformers According to the Energy Consumption ANDRÉ NUNES DE SOUZA, JOSÉ ALFREDO C. ULSON, IVAN NUNES

More information

A Discrete Robust Adaptive Iterative Learning Control for a Class of Nonlinear Systems with Unknown Control Direction

A Discrete Robust Adaptive Iterative Learning Control for a Class of Nonlinear Systems with Unknown Control Direction Proceedings of the International MultiConference of Engineers and Computer Scientists 16 Vol I, IMECS 16, March 16-18, 16, Hong Kong A Discrete Robust Adaptive Iterative Learning Control for a Class of

More information

An Optimal Tracking Approach to Formation Control of Nonlinear Multi-Agent Systems

An Optimal Tracking Approach to Formation Control of Nonlinear Multi-Agent Systems AIAA Guidance, Navigation, and Control Conference 13-16 August 212, Minneapolis, Minnesota AIAA 212-4694 An Optimal Tracking Approach to Formation Control of Nonlinear Multi-Agent Systems Ali Heydari 1

More information

Impulsive Noise Filtering In Biomedical Signals With Application of New Myriad Filter

Impulsive Noise Filtering In Biomedical Signals With Application of New Myriad Filter BIOSIGAL 21 Impulsive oise Filtering In Biomedical Signals With Application of ew Myriad Filter Tomasz Pander 1 1 Division of Biomedical Electronics, Institute of Electronics, Silesian University of Technology,

More information

SOS Boosting of Image Denoising Algorithms

SOS Boosting of Image Denoising Algorithms SOS Boosting of Image Denoising Algorithms Yaniv Romano and Michael Elad The Technion Israel Institute of technology Haifa 32000, Israel The research leading to these results has received funding from

More information

Learning features by contrasting natural images with noise

Learning features by contrasting natural images with noise Learning features by contrasting natural images with noise Michael Gutmann 1 and Aapo Hyvärinen 12 1 Dept. of Computer Science and HIIT, University of Helsinki, P.O. Box 68, FIN-00014 University of Helsinki,

More information

Comparative Performance Analysis of Three Algorithms for Principal Component Analysis

Comparative Performance Analysis of Three Algorithms for Principal Component Analysis 84 R. LANDQVIST, A. MOHAMMED, COMPARATIVE PERFORMANCE ANALYSIS OF THR ALGORITHMS Comparative Performance Analysis of Three Algorithms for Principal Component Analysis Ronnie LANDQVIST, Abbas MOHAMMED Dept.

More information