Hayes, M.H., 1996. Statistical digital signal processing and modeling. John Wiley & Sons, Inc. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
First 50 documents Next 50
Prediction of Mobile Radio Channels - Modeling and Design - Ekman (2002) (2 citations) (Correct)
....for prediction of the taps of a mobile radio channel. The optimal linear FIR predictor using past noisy observations to predict a signal is given by the Wiener Hopf equations. Derivations of the optimal linear predictor are found in most textbooks in statistical signal processing, i.e. 42] 58] [59]. We will in this section recapitulate these results and introduce the corresponding notation. 6.2.1 The FIR predictor The goal is to predict a signal L time instances ahead using a FIR predictor with M coe#cients. The complexity is thus limited by the choice of M.In a vector formulation of the ....
M.H. Hayes, Statistical Digital Signal Processing and Modeling, Wiley, New York, NY, 1996.
Hybrid Distributed Iterative Capacity Allocation Over.. - Son, Schiøler, .. (Correct)
....traffics and the permitted traffics to meet requirement on packet losses on routes, several methods can be used, which possibly base on mathematical modeling of the estimated traffics. However, we use simple LevinsonDurbin algorithm, which is rather efficient to solve the YuleWalker equation [14]: ## # # . ### # # . 16) is a Hermitian Toeplitz matrix, is prediction coefficient and # is error variance. To simplify, the prediction order was set to and the record history of previous measured data was applied. B. Numeric results Figure 3.b ....
M. H. Hayes, Statistical Digital Signal Processing and Modeling, John Wiley & Sons,1999.
The Role of Filter Banks in - Sinusoidal Frequency Estimation (Correct)
....subbands is superior to that in the fullband. Keywords: sinusoidal frequency estimation, pseudospectra, filter banks, subband estimation, spectral flatness measure Work supported in part by the ONR grant N00014 99 1 1002. I. Introduction A classical problem of statistical signal processing [16, 7, 2] is that of determining the frequencies of sinusoids buried in noise. Such a problem arises in array processing [6, 10, 3] for example, when we wish to estimate the direction of arrival (DOA) of a narrowband electromagnetic signal incident on a uniform linear array. In this application, the ....
....of the autocorrelation matrix R x for the purpose of frequency estimation. In his classic paper [11] he used an autocorrelation matrix of size N = 1 and estimated the frequencies as the peaks of the following frequency estimation function commonly referred to now as the pseudospectrum [16, 7, 2] corresponding to the Pisarenko harmonic decomposition. V P 1 , where VP 1 (z) v P 1 (n)z n Ideally, all of the zeros of VP 1 (z) lie on the unit circle at angles corresponding to the frequencies of the sinusoids sent. Thus, the frequencies are estimated as the locations of ....
[Article contains additional citation context not shown here]
M. H. Hayes. Statistical Digital Signal Processing and Modeling. John Wiley & Sons, Inc., New York, NY, 1996.
Efficient Modeling And Synthesis Of Bell-Like Sounds - Karjalainen, Välimäki, Esquef (2002) (Correct)
....but in most cases the LTI assumption works well. There are many methods available for estimating a parametric model based on the impulse response of a given LTI system: AR (autoregressive) modeling for all pole filter and ARMA (autoregressive moving average) modeling for pole zero filter design [5, 6, 7]. Each eigenmode should correspond to a complexconjugate pole pair in the resulting filter. Although powerful in many applications, direct application of these methods to bell sounds may turn out to be problematic as will be described below. Very closely located modal frequencies require ....
M. H. Hayes, Statistical Digital Signal Processing and Modeling. John Wiley & Sons, 1996.
A Linear System Approach to QoS Control - Paper Id Number (Correct)
....input power spectrum. In this section we present an estimator that we used in simulation studies for the estimation of these statis tics. 4.1 Input Mean For M samples, we estimate input mean by 1 a = a(n) 36) 4. 2 Input Power Spectrum For the power spectrum estimation, we use Welch s method [14] instead of well known periodogram method to acquire a more consistent estimate. For M samples, i i iD) e j 2 i o w(n)a(n (37) Pa( ICLU :0 v Iw(r01 and weuse where M = L D( 4 1) U = 0 Hanning window for w (n) 5. MEASUREMENT BASED QOS CONTROL There are several possible ways to ....
M. Hayes, Statistical Digital Signal Processing and Modeling, 1st Ed., John Wiley & Sons, Inc., 1996.
Speech-Gesture Driven Multimodal Interfaces for.. - Sharma, Yeasin.. (Correct)
....using dynamic programming using the Viterbi algorithm [82] Head and Hand Tracking: The algorithms for head and face tracking are based on similar but slightly different approaches. Both trackers are based on rectangular tracking windows whose location is continuously adapted using Kalman filters [124] to follow the users head and hand. While the head tracker relies solely on skin color image cues, the hand tracker is a continuous version of the palm detector and geared towards skin colored moving objects. Prior knowledge about the human body is utilized for avoiding and resolving conflicts and ....
M. H. Hayes, Statistical Digital Signal Processing and Modeling: John Wiley & Sons, Inc., 1996.
Power-Spectrum Condition for Energy-Efficient Watermarking - Su, Girod (2002) (9 citations) (Correct)
....the sixth column of the table shows the corresponding , which is always 0 dB. V. EXPERIMENTAL RESULTS A. Theoretical Performance for Signal Models For the purpose of analysis, correlated signals such as audio or images are often approximated by autoregressive (AR) random processes [23]. We denote the one dimensional (1 D) th order AR process by AR( which has the model , where is 1 D WSS white noise. Lowpass I and Lowpass II denote AR(1) models with and 0.90, respectively. Bandpass is an AR(2) model with , while Highpass refers to an AR(1) model with . For ....
....Fig. 3. Theoretical performance of watermarks for natural images. The WOR is 030 dB. Circles indicate the results of the removal attack. For Cameraman, PSNR = ONR 12:24 dB; for Lenna, PSNR = ONR 13:76 dB. Fig. 4. Original Cameraman image. original image was estimated using the periodogram [23], where is the 2 D FFT of . We remark that taking the full size transform of an image may not be the best implementation for actual watermarking schemes. Also, the periodogram produces an unbiased, but not a consistent, estimate of a signal s power spectrum [23] Nonetheless, these methods are ....
[Article contains additional citation context not shown here]
M. H. Hayes, Statistical Digital Signal Processing and Modeling.New York: Wiley, 1996.
Equalization of Time Errors in Time Interleaved ADC.. - Elbornsson.. (2003) (Correct)
.... r )d#A 1 (#)B l = MT s E 1 (#, # # The subsequences Z i (e ) can then be calculated as = Y )MT s E 1 (#, # j#l (5) Y Y 0 (e YM 1 (e # The TDFT of the time error compensated signal, can then be calculated from its subsequences [15] j(#MT s mod 2#) e jl#T (6) With the inverse Fourier transform we get the time error reconstructed signal [k] TDFT 1 (Z ) 7) In practice (5) 6) and (7) are calculated on finite sequences using the DFT instead of the Fourier transform. IV. TIME ERROR ....
M. Hayes, Statistical digital signal processing and modeling. Wiley, 1996.
Unknown - Vector Analysis Of (2002) (Correct)
....that minimizes may be found by setting the derivatives of with respect to w k equal to zero, another approach is to search for the solution using the method of steepest descent. The method of steepest descent is an iterative procedure that has been used to find extrema of nonlinear functions [2]. The applied LMS algorithm is developed by this steepest descent method. Let w be an estimate of the vector that minimizes the mean square error n at time n. At time n l, a new estimation is formed by adding a correction to w, which is designed to bring w closer to the desired solution. The ....
....n becomes unstable and unbounded. Considering the convergence of the LMS adaptive filter, choosing the step size should be very careful. A more conservative bound is given by (p I)E x(n) 2 (2 19) where p is the order of the filter. For more details of the step size , please consult reference [2]. The gradient corresponds to the derivative of rOe with respect tow k, given by = EIVen2 E = enVen (2 20) The gradient of the estimation error is given by en = Xn (2 21) thus, the gradient of the cost function becomes REPORT 22 (57) Uppgjord (ven faktaansvadg om annan) Prepared ....
Hayes, M., Statistical Digital Signal Processing and Modeling, Hohn Wiley &Sons, New York, 1996.
Recipes for the linear analysis of EEG - Lucas Parra Clay (Correct)
No context found.
Hayes, M.H., 1996. Statistical digital signal processing and modeling. John Wiley & Sons, Inc.
Convolutive Source Separation and Signal Modeling with ML - Lucas Parra Clay (Correct)
No context found.
H. Hayes, Monson. Statistical Digital Signal Processing and Modeling. Wiley, 1996.
Speech-Gesture Driven Multimodal Interfaces for.. - Sharma, Yeasin.. (2003) (Correct)
No context found.
M. H. Hayes, Statistical Digital Signal Processing and Modeling. New York: Wiley, 1996.
Identifiability Issues in Noisy ICA - Mike Davies Member (2004) (Correct)
No context found.
M. H. Hayes, Statistical Digital Signal Processing and Modeling.New York: Wiley, 1996.
An Automatic Beat Detection Algorithm - For Pressure Signals (2005) (Correct)
No context found.
M. H. Hayes, Statistical Digital Signal Processing and Modeling.New York: Wiley, 1996.
Characteristic Electrocardiograph Intervals as Indices of.. - Shouldice (2004) (Correct)
No context found.
M. Hayes, Statistical Digital Signal Processing and Modeling. New York: Wiley, 1996.
Identifiability, subspace selection and noisy ICA - Davies (2004) (Correct)
No context found.
M.H. Hayes, Statistical Digital Signal Processing and Modeling, John Wiley & Sons, 1996.
Identifiability Issues in Noisy ICA - Mike Davies Member (2004) (Correct)
No context found.
M. H. Hayes, Statistical Digital Signal Processing and Modeling.New York: Wiley, 1996.
Efficient Channel Shortening Equalizer Design - Martin, Ding, Evans, Johnson.. (2004) (Correct)
No context found.
M. H. Hayes, Statistical Digital Signal Processing and Modeling, John Wiley & Sons, Inc., New York, NY, 1996.
Improved Channel Estimation For Ofdm Based Wlan - Systems Rangaraj Raghavendra (Correct)
No context found.
Monson H. Hayes, Statistical Digital Signal Processing and Modeling, John Wiley Sons, Inc., 1996.
Performance of Adaptive Filtering Algorithms: A Comparative.. - Lau, Hussian, Harris (2003) (Correct)
No context found.
M. H. Hayes, Statistical Digital Signal Processing and Modeling, John Wiley & Sons, 1996.
On the Design of L_p IIR Filters with Arbitrary Frequency.. - Vargas, Burrus (Correct)
No context found.
Monson H. Hayes, Statistical Digital Signal Processing and Modeling, Wiley, New York, 1996.
Estimation of Oscillator Phase Noise for MPSK-based.. - Thesis For The (2004) (Correct)
No context found.
M. H. Hayes, Statistical Digital Signal Processing and Modeling, John Wiley and Sons, 1996.
A Time-Varying Convergence Parameter for the LMS Algorithm .. - Lau, Hussian, Harris (2003) (Correct)
No context found.
Monson H. Hayes, Statistical digital signal processing and modeling John Wiley & Sons, 1996.
Blind Equalization of Time Errors in a Time - Interleaved Adc System (Correct)
No context found.
M. Hayes, Statistical digital signal processing and modeling. Wiley, 1996.
An Efficient Approach to Enhance Capacity Allocation.. - Son, Schiøler, .. (Correct)
No context found.
M. H. Hayes, Statistical Digital Signal Processing and Modeling, John Wiley & Sons,1999.
First 50 documents Next 50
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC