Ruen MEYLAN Cenker ODEN Ayn ERTUZUN Aytul ERÇL
In this paper, a 2-D iteratively reweighted least squares lattice algorithm, which is robust to the outliers, is introduced and is applied to defect detection problem in textured images. First, the philosophy of using different optimization functions that results in weighted least squares solution in the theory of 1-D robust regression is extended to 2-D. Then a new algorithm is derived which combines 2-D robust regression concepts with the 2-D recursive least squares lattice algorithm. With this approach, whatever the probability distribution of the prediction error may be, small weights are assigned to the outliers so that the least squares algorithm will be less sensitive to the outliers. Implementation of the proposed iteratively reweighted least squares lattice algorithm to the problem of defect detection in textured images is then considered. The performance evaluation, in terms of defect detection rate, demonstrates the importance of the proposed algorithm in reducing the effect of the outliers that generally correspond to false alarms in classification of textures as defective or nondefective.
In this paper, we present a time-delay estimation algorithm in a closely spaced multipath environment for a direct-sequence code division multiple access (DS-CDMA) systems. The proposed scheme first converts the observed signal to the frequency domain by Fast Fourier Transform (FFT). We then formulate a nonlinear estimation problem, and convert it into a large scale linear least squares problem. We apply conjugate gradients iterations on the resulting normal equations to obtain the solution. Unlike the methods which invoke the criterion of nonlinear least squares, the proposed scheme can achieve higher resolution. The delay estimation is combined with a tracking algorithm based on a FIR prefilter. Simulation results confirm the effectiveness of the proposed algorithm in both AWGN and Rayleigh fading channels.
M. Shahidur RAHMAN Tetsuya SHIMAMURA
A new system identification based method has been proposed for accurate estimation of vocal tract parameters. An often encountered problem in using the conventional linear prediction analysis is due to the harmonic structure of the excitation source of voiced speech. This harmonic characteristic is coupled with the estimation of autoregressive (AR) coefficients that results in difficulties in estimating the vocal tract filter. This paper models the effective voice source from the residual obtained through the covariance analysis in the first-pass which is then used as input to the second-pass least-square analysis. A better source-filter separation is thus achieved. The formant frequencies and corresponding bandwidths obtained using the proposed method for synthetic vowels are found to be accurate up to a factor of more than three (in percent) compared to the conventional method. Since the source characteristic is taken into account, local variations due to the positioning of analysis window are reduced significantly. The validity of the proposed method is also examined by inspecting the spectra obtained from natural vowel sounds uttered by high-pitched female speaker.
The least squares (LS) and the weighted least squares (WLS) algorithms are well known procedures that are used in the design of quadrature mirror filters (QMFs). It is known that these design techniques suffer from pass-band anomaly under certain conditions. A recent method, that overcomes pass-band anomaly for LS QMFs using a frequency sampling design for the initial filter, is extended to WLS design in this letter. A comparison between the modified LS and WLS designs based on experimental results is presented. Although WLS designs have been reported to have superior near-equiripple stop-band performance as compared to LS designs, we find that this is not always true. Specifically, LS designs, with inherent computational and robustness advantages, also have better peak stop-band ripple and transition bandwidth at higher cut-off frequencies than WLS.
In this Letter, linear least squares (LLS) techniques for phase estimation of real sinusoidal signals with known or unknown amplitudes are studied. It is proved that the asymptotic performance of the LLS approach attains Cramér-Rao lower bound. For the case of a single tone, a novel LLS algorithm with unit-norm constraint is derived. Simulation results are also included for algorithm evaluation.
Hyeon-Ho KIM Sung-Hwan HAN Hyeon-Deok BAE
Recently, DOAS (differential optical absorption spectroscopy) has been used for nondestructive air monitoring, in which the LS (least squares) method is used to calculate trace gas concentrations due to its computational simplicity. This paper applies the ICA (independent component analysis) method to the DOAS system of air monitoring, since the LS method is insufficient to recover the desired spectra perfectly due to sparsity characteristic. If the sparsity of reference spectra in the DOAS system imposes the assumption of independence, the ICA algorithm can be used. The proposed method is used to regress the observed spectrum on the estimates of the reference spectra. The ICA algorithm can be seen as a preprocessing method where the ICs of the references are used as the input in the regression. The performance of the proposed method is evaluated in simulation studies using synthetic data.
Junji KAWATA Yuichi TANJI Yoshifumi NISHIO Akio USHIDA
In this paper, we propose a new algorithm for calculating the exact poles of the admittance matrix of RLCG interconnects. After choosing dominant poles and corresponding residues, each element of the exact admittance matrix is approximated by partial fraction. A procedure to obtain the residues that guarantee the passivity is also provided, based on experimental studies. In the procedure the residues are calculated by using the least squares method so that the partial fraction matches each element of the exact admittance matrix in the frequency-domain. From the partial fraction representation, the asymptotic equivalent circuit models which can be easily simulated with SPICE are synthesized. It is shown that an efficient model-order reduction is possible for short-length interconnects.
Masahiro OKUDA Masahiro YOSHIDA Masaaki IKEHARA Shin-ichi TAKAHASHI
In this paper, we present a new numerical method for the complex approximation of FIR digital filters. Our objective is to design FIR filters with equiripple magnitude and phase errors. The proposed method solves the least squares (LS) problem iteratively. At each iteration, the desired response is updated so as to have an equiripple error. The proposed methods do not require any time-consuming optimization procedure such as the quasi-Newton methods and converge to equiripple solutions quickly. We show some examples to illustrate the advantages of our proposed methods.
One conventional technique for source localization is to utilize the time-difference-of-arrival (TDOA) measurements of a signal received at spatially separated sensors. A simple TDOA-based location algorithm that combines the advantages of two efficient positioning methods is developed. It is demonstrated that the proposed approach can give optimum performance in geolocation via satellites at different noise conditions.
Homayoon ORAIZI Gholam-Reza GABARANZAD GHADIM
In this paper the method of least squares is employed to design an axially symmetric contradirectional multisection coupled - line coupler together with the impedance matching of real generator and load impedances. An error function is constructed for the required coupling (C) based on the squared magnitude of the ratio of the coupler voltage to that at the incident port. Another algorithm based on the reflected and transmitted wave amplitudes is developed by the method of least squares for the design of a coupled - line coupler with impedance matching of different input and output complex impedances and arbitrary coupling and length. The error functions are minimized to determine the coupler geometry, namely the normalized strip conductor widths (u=w/h) and separation (g=s/h) and the coupler length, where h is the substrate thickness. A procedure is presented to provide the initial values of u and g. The computer implementation of the proposed method shows that a proper coupler design is possible for any given coupler length. This is particularly interesting where space limitations impose contraints on the coupler length. The results are favorably compared with available computer simulation softwares.
Feng LIU Taiyi ZHANG Ruonan ZHANG
For suppressing inter symbol interference, the support vector machine mutliuser detector (SVM-MUD) was adopted as a nonlinear method in direct sequence code division multiple access (DS-CDMA) signals transmitted through multipath channels. To solve the problems of the complexity of SVM-MUD model and the number of support vectors, based on recursive least squares support vector machine (RLS-SVM) and Riemannian geometry, a new algorithm for nonlinear multiuser detector is proposed. The algorithm introduces the forgetting factor to get the support vectors at the first training samples, then, uses Riemannian geometry to train the support vectors again and gets less improved support vectors. Simulation results illustrated that the algorithm simplifies SVM-MUD model at the cost of only a little more bit error rate and decreases the computational complexity. At the same time, the algorithm has an excellent effect on suppressing multipath interference.
Huadong MENG Xiqin WANG Hao ZHANG Yingning PENG
The high-resolution frequency estimators most commonly used, such as Least Square (LS) method based on AR model, MVSE, MUSIC and ESPRIT, determine estimates of the sinusoidal frequencies from the sample noise-corrupted data. In this paper, a new frequency estimation method named Pole-Placement Least Square (PPLS) is presented, which is a modified LS method with a certain number of model poles restricted to the unit circle. The statistical performance of PPLS is studied numerically, and compared with the Cramer-Rao bound as well as the statistical performance corresponding to the LS methods. PPLS is shown to have higher resolution than the conventional LS method. The relationship between poles location and its resolution is also discussed in detail.
Hideyuki IMAI Yuying YUAN Yoshiharu SATO
It is widely known that the family of projection filters includes the generalized inverse filter, and that the family of parametric projection filters includes parametric generalized projection filters. However, relations between the family of parametric projection filters and constrained least squares filters are not sufficiently clarified. In this paper, we consider relations between parameter estimation and image restoration by these families. As a result, we show that the restored image by the family of parametric projection filters is a maximum penalized likelihood estimator, and that it agrees with the restored image by constrained least squares filter under some suitable conditions.
Montree BUDSABATHON Shinsuke HARA
In this paper, we discuss the applicability of H filtering algorithm for a decision feedback equalizer (DFE) in mobile communications environments. A comparison of bit error rate (BER) performance between a H2 filtering (recursive least squares) algorithm-based and an a priori H filtering algorithm-based fractionally-tap-spaced DFEs in various fading channels is presented. Against, our results are rather pessimistic of the H equalizer, namely, as compared with the H2 equalizer, at most the same or a little better BER performance of the H equalizer is obtained only when the ratio of the average received energy per bit to the white noise power spectral density and the normalized fading rate are large enough, especially in Rician fading channels. Moreover, the H equalizer has the problems of how to choose an appropriate prescribed positive value and computational complexity, therefore it may not be considered as an attractive alternative to the H2 equalizer for wireless communications systems.
A new class of least-squares algorithms is presented for adaptive filtering. The idea is to use a fixed set of directions and perform line search with one direction at a time in a cyclic fashion. These algorithms are called Euclidean Direction Search (EDS) algorithms. The fast version of this class is called the Fast-EDS or FEDS algorithm. It is shown to have O(N) computational complexity and a convergence rate comparable to that of the RLS algorithm. Computer simulations are presented to illustrate the performance of the new algorithm.
Space-variant approaches subject to local image characteristics are useful in practical image restoration because many natural images are nonstationary. Motivated by the success of denoising approaches in the wavelet domain, we propose a region-adaptive restoration approach which adopts a wavelet denoising technique in flat regions after an under-regularized constrained least squares restoration. Experimental results verify that this approach not only improves image quality in mean square error but also contributes to ringing reduction.
Hiroyuki UKIDA Katsunobu KONISHI
We suggest the method to recover the 3D shape of an object by using a color image scanner which has three light sources. The photometric stereo is traditional to recover the surface normals of objects using multiple light sources. In this method, it usually assumes distant light sources to make the optical models simple. But the light sources in the image scanner are so close to an object that the illuminant intensity varies with the distance from the light source, therefore these light sources should be modeled as the linear light sources. In this method, by using these models and two step algorithm; the initial estimation by the iterating computation and the optimization by the non-linear least square method, not only the surface normal but also the absolute distance from the light source to the surface can be estimated. By using this method, we can recover the 3D shape more precisely. In the experimental results, the 3D shape of real objects can be recovered and the effectiveness of the proposed method is shown.
Won Gi JEON Kyung Hyun PAIK Yong Soo CHO
In this paper, we propose an efficient channel estimation technique for orthogonal frequency-division multiplexing (OFDM) systems with transmitter diversity. The proposed technique estimates uniquely all channel frequency responses needed in space-time coded OFDM systems using "comb-type" training symbols. The computational complexity of the proposed technique is reduced dramatically, compared with the previous minimum mean-squared error (MMSE) technique, due to the processing made all in the frequency-domain. Also, several other techniques for mitigating random noise effect and tracking channel variation are discussed to further improve the performance of the proposed approach. The performances of the proposed approach are demonstrated by computer simulation for mobile wireless channels.
Hak-Young KIM Won-Sik YOON Dae Jin KIM Young Han KIM
In this paper we propose a mobile positioning method based on a recursive least squares (RLS) algorithm for suppressing the non-line of sight (NLOS) effects in cellular systems. The proposed method finds the position of a mobile station from TOAs measured by three BSs. Simulation results show that the proposed method has a fast convergence time and greatly reduces the positioning error especially in NLOS situations. Thus it is expected that the proposed method can be effectively used in a dense urban environment.
Takafumi FUJIMOTO Kazumasa TANAKA Mitsuo TAGUCHI
The wall admittance of an arbitrarily shaped microstrip antenna is generally formulated. As examples, elliptical microstrip antennas with and without a circular slot are calculated. The wall admittance is determined by the spectral domain analysis in order to consider the effect of the dielectric substrate. The electromagnetic fields within the cavity are expanded in terms of the eigenfunctions in the cylindrical coordinate system and their expansion coefficients are determined by applying the impedance boundary condition at the aperture in the sense of the least squares. The calculated input impedance and axial ratio agree fairly well with the experimental data. The proposed method is valid for the microstrip antennas with a patch whose geometry deviates from the particular coordinate system, such as single-feed circularly polarized microstrip antennas.