In this paper, we propose a spatially adaptive gradient-projection algorithm for the H.264 video coding standard to remove coding artifacts using local statistics. A hybrid method combining a new weighted constrained least squares (WCLS) approach and the projection onto convex sets (POCS) approach is introduced, where weighting components are determined on the basis of the human visual system (HVS) and projection set is defined by the difference between adjacent pixels and the quantization index (QI). A new visual function is defined to determine the weighting matrices controlling the degree of global smoothness, and a projection set is used to obtain a solution satisfying local smoothing constraints, so that the coding artifacts such as blocking and ringing artifacts can be simultaneously removed. The experimental results show the capability and efficiency of the proposed algorithm.

In order to reduce the convergence time in an iterative procedure, some gradient based preliminary processes are employed to eliminate outliers. The adaptive variable block size is also introduced to balance the accuracy and computational complexity. Moreover, the use of Canberra distance instead of Euclidean distance illustrates higher performance in measuring motion similarity.

This paper discusses an efficient discrete model for nonlinear RF power amplifier (PA) with long-term memory effects and analyzes its error. The procedure of converting RF signals and systems into a discrete domain is explained for a discrete baseband memory polynomial model. Unlike a previous simple memory polynomial model, the proposed discrete model has two different sampling frequencies: one for nonlinear system with long-term memory effects and one for input signal. A method to choose an optimal sampling frequency for the system and a discrete memory depth is proposed to minimize the sensitivity of the system for perturbation of the measured data. A two-dimensional sensitivity function which is a product of relative residual and matrix condition number is defined for least square problem of the proposed model. Examples with a wideband WiBro 3FA signal and a WCDMA 4FA signal for nonlinear transmitters are presented to describe the overall procedure and effectiveness of the proposed scheme.

This letter deals with mobile location estimation based on a minimax mean-squared error (MSE) algorithm using time-of-arrival (TOA) measurements for mitigating the nonline-of-sight (NLOS) effects in cellular systems. Simulation results are provided for illustrating the minimax MSE estimator yields good performance than the other least squares and weighted least squares estimators under relatively low signal-to-noise ratio and moderately NLOS conditions.

Based on the least square (LS) approximation of sinusoidal signal in frequency domain by sample data, a frequency estimator is derived. Since sinusoidal signals are narrow-banded whereas white noise spreads equally in the whole spectrum, only narrow-band approximation around the actual tone is needed, and thus the influence of noise can be decreased significantly with high computational efficiency. Experimental results show that, without any iterations, the performance of the proposed estimator is close to the Cramer-Rao Bound (CRB), and has a lower SNR threshold compared with other existing estimators.

This paper presents a pilot-aided channel estimation method which is particularly suitable for mobile WiMAX 802.16e Downlink Partial Usage of Subchannel mode. Based on this mode, several commonly used channel estimation methods are studied and the method of least squares line fitting is proposed. As data of users are distributed onto permuted clusters of subcarriers in the transmitted OFDMA symbol, the proposed channel estimation method utilizes these advantages to provide better performance than conventional approaches while offering remarkably low complexity in practical implementation. Simulation results with different ITU-channels for mobile environments show that depending on situations, enhancement of 5 dB or more in term of SNR can be achieved.

In this paper, estimation accuracy of channel frequency response (CFR) according to least squared (LS) criterion with two transmit antennas for the time domain synchronous-orthogonal frequency division multiplexing (TDS-OFDM) system is investigated. To minimize the estimation variance, the conditions to guide the pseudo-noise (PN) sequence design are discussed and three training sequence design schemes are proposed accordingly. Simulations show that the proposed PN sequence design scheme is effective, while the implementation complexity for the channel estimation is low.

In this paper, we present and analyze a predictive closed-loop power control (CLPC) scheme which employs a comb-type sample arrangement to effectively compensate multiple power control group (PCG) delays over mobile fading channels. We consider both least squares and recursive least squares filters in our CLPC scheme. The effects of channel estimation error, prediction filter error, and power control bit transmission error on the performance of the proposed CLPC method along with competing non-predictive and predictive CLPC schemes are thoroughly investigated. Our results clearly indicate the superiority of the proposed scheme with its improved robustness under non-ideal conditions. Furthermore, we carry out a Monte-Carlo simulation study of a 55 square grid cellular network and evaluate the user capacity. Capacity improvements up to 90% are observed for a typical cellular network scenario.

This letter presents an efficient adaptive beamformer to deal with the multipath environments created by signal source scatterings. To improve the performance possible with the fixed forgetting factor, the regular adaptive forgetting factor algorithm is derived and applied to the subarray recursive least squares (RLS) beamforming. Simulations confirm that the proposed scheme has better performance than not only the conventional RLS algorithm but also the subarray RLS and adaptive forgetting factor RLS algorithms.

As a technique for calibrating electric-field probes used in standardized SAR (Specific Absorption Rate) assessment, we have studied the technique using the Friis transmission formula in the tissue-equivalent liquid. It is difficult to measure power transmission between two reference antennas in the far-field region due to large attenuation in the liquid. This means that the conventional Friis transmission formula cannot be applied to our measurement so that we developed an extension of this formula that is valid in the near-field region. In this paper, the method of weighted least squares is introduced to reduce the effect of the noise in the measurement system when the gain of the antenna operated in the liquid is determined by the curve-fitting technique. And we examine how to choose the fitting range to reduce the uncertainty of the estimated gain.

MIMO (Multiple-Input Multiple-Output) technologies have attracted much interest for high-rate and high-capacity wireless communications. MIMO technologies under frequency-selective fading environments (wideband MIMO technologies) have also been studied. A wideband MIMO system is affected by ISI (Inter Symbol Interference) and CCI (Co-Channel Interference). Hence, we need a MIMO signal detection technique that simultaneously suppresses ISI and CCI. The OFDM system and SC-FDE (Single Carrier-Frequency Domain Equalization) techniques are often used for suppressing ISI. By employing these techniques with the ZF (Zero Forcing) or the MMSE (Minimum Mean Square Error) spatial filtering technique, we can cancel both ISI and CCI. To use ZF or MMSE, we need channel state information for calculating the receive weights. Although an LS (Least Square) channel estimation technique has been proposed for MIMO-OFDM systems, it needs a large estimation matrix at the receiver side to obtain sufficient estimation performance in heavy multipath environments. However, the use of a large matrix increases computational complexity and the circuit size. We use frequency domain channel estimation to solve these problems and propose an iterative method for achieving better estimation performance. In this paper, we assume the use of a MIMO-UWB system that employs a UWB-IR (Ultra-Wideband Impulse Radio) scheme with the FDE technique as the wideband wireless transmission scheme for heavy multipath environments, and we evaluate the iterative frequency domain channel estimation through computer simulations and computational complexity calculations.

In this letter, we propose a new type of recursive least squares (RLS) algorithms without using the initial information of a parameter or a state to be estimated. The proposed RLS algorithm is first obtained for a generic linear model and is then extended to a state estimator for a stochastic state-space model. Compared with the existing algorithms, the proposed RLS algorithms are simpler and more numerically stable. It is shown through simulation that the proposed RLS algorithms have better numerical stability for digital computations than existing algorithms.

Complex speech analysis for an analytic speech signal can accurately estimate the spectrum in low frequencies since the analytic signal provides spectrum only over positive frequencies. The remarkable feature makes it possible to realize more accurate F0 estimation using complex residual signal extracted by complex-valued speech analysis. We have already proposed F0 estimation using complex LPC residual, in which the autocorrelation function weighted by AMDF was adopted as the criterion. The method adopted MMSE-based complex LPC analysis and it has been reported that it can estimate more accurate F0 for IRS filtered speech corrupted by white Gauss noise although it can not work better for the IRS filtered speech corrupted by pink noise. In this paper, robust complex speech analysis based on ELS (Extended Least Square) method is introduced in order to overcome the drawback. The experimental results for additive white Gauss or pink noise demonstrate that the proposed algorithm based on robust ELS-based complex AR analysis can perform better than other methods.

This paper proposes several cepstral statistics compensation and normalization algorithms which alleviate the effect of additive noise on cepstral features for speech recognition. The algorithms are simple yet efficient noise reduction techniques that use online-constructed pseudo-stereo codebooks to evaluate the statistics in both clean and noisy environments. The process yields transformations for both clean speech cepstra and noise-corrupted speech cepstra, or for noise-corrupted speech cepstra only, so that the statistics of the transformed speech cepstra are similar for both environments. Experimental results show that these codebook-based algorithms can provide significant performance gains compared to results obtained by using conventional utterance-based normalization approaches. The proposed codebook-based cesptral mean and variance normalization (C-CMVN), linear least squares (LLS) and quadratic least squares (QLS) outperform utterance-based CMVN (U-CMVN) by 26.03%, 22.72% and 27.48%, respectively, in relative word error rate reduction for experiments conducted on Test Set A of the Aurora-2 digit database.

A method for extracting a sound signal from a particular area that is surrounded by multiple ambient noise sources is proposed. This method performs several fixed beamformings on a pair of small microphone arrays separated from each other to estimate the signal and noise power spectra. Noise suppression is achieved by applying spectrum emphasis to the output of fixed beamforming in the frequency domain, which is derived from the estimated power spectra. In experiments performed in a room with reverberation, this method succeeded in suppressing the ambient noise, giving an SNR improvement of more than 10 dB, which is better than the performance of the conventional fixed and adaptive beamforming methods using a large-aperture microphone array. We also confirmed that this method keeps its performance even if the noise source location changes continuously or abruptly.

In this paper, a sparse multipath channel estimation algorithm is proposed for multiple-input multiple-output (MIMO) single-carrier systems with frequency-domain decision feedback equalizer (FD-DFE). This algorithm exploits the orthogonality of an optimal MIMO preamble based on repeated, phase-rotated, Chu (RPC) sequences, leading to a dramatic reduction in computation. Furthermore, the proposed algorithm employs an improved non-iterative procedure utilizing the Generalized AIC criterion (GAIC), resulting in further computational saving and performance improvement. The proposed scheme is simulated for 802.16d SCa-PHY and SUI-5 sparse channel model under a 22 spatial multiplexing scenario, with the MIMO FD-DFE as the receiver. The result shows that the channel estimation accuracy is significantly improved, and the performance loss compared to the known channel case is only 0.7 dB at the BER of 10-3.

Using the recursive generalized eigendecomposition method, we develop a recursive form solution to the data least squares (DLS) problem in which the error is assumed to lie in the data matrix only. We apply it to a linear channel equalizer. Simulations shows that the DLS-based equalizer outperforms the ordinary least squares-based one in a channel equalization problem.

A two-stage least square identification method is proposed for estimating ARMA (autoregressive moving average) coefficients from speech signals. A pulse-train like input sequence is often employed to account for the source effects in estimating vocal tract parameters of voiced speech. Due to glottal and radiation effects, the pulse train, however, does not represent the effective voice source. The authors have already proposed a simple but effective model of voice source for estimating AR (autoregressive) coefficients. This letter extends our approach to ARMA analysis to wider varieties of speech sounds including nasal vowels and consonants. Analysis results on both synthetic and natural nasal speech are presented to demonstrate the analysis ability of the method.

In this paper a new method for the design and optimization of microstrip parallel coupled-line bandpass filters is presented which allows for the specification of frequency bandwidths and arbitrary source and load impedance transformation. The even- and odd-mode theory and the relationships between impedance, transmission and scattering matrices and their properties are used to construct a positive definite error function using the insertion losses at discrete frequencies in the pass, transition and stop bands. The dispersion relations for the coupled line are also taken into account. The minimization of the error function determines the widths, gap spacings and lengths of the coupled-line filter, for the optimum design and realization of filter specifications. The proposed filter design and optimization method is coded by computer programs and the results of simulation, fabrication and testing of sample filters together with comparisons with available full-wave analysis softwares, indicate the efficacy of the proposed method. Filter design with up to 50% bandwidth and the design of shorter lengths of coupled line sections are achievable by the proposed method in part due to the incorporation of impedance matching.

The multi-step prediction model based on partial least squares (PLS) is established to predict short-term load series with high embedding dimension in this paper, which refrains from cumulative error with local single-step linear model, and can cope with the multi-collinearity in the reconstructed phase space. In the model, PLS is used to model the dynamic evolution between the phase points and the corresponding future points. With research on the PLS theory, the model algorithm is put forward. Finally, the actual load series are used to test this model, and the results show that the model plays well in chaotic time series prediction, even if the embedding dimension is selected a big value.