Search

The wavelet transform, the S-transform, the Gabor transform, and the Wigner distribution function are popular techniques for power quality (PQ) analysis in electrical power systems. They are mainly used to identify power harmonics and power disturbances and to estimate power quantities in the presence of nonstationary power components such as root-mean-square values and total harmonic distortions. Recently, the Hilbert-Huang transform has been also used in PQ analysis. These techniques have proven to be useful in PQ analysis; however, their performances depend on the types of PQ events. In this paper, a novel frequency-shifting wavelet decomposition via the Hilbert transform is introduced for PQ analysis. The proposed algorithm overcomes the spectra leakage problem in the discrete wavelet packet transform and can be used for estimating power quantities accurately and for detecting flickers. The effectiveness of the proposed algorithm was verified by computer simulations and experimental tests.

Parameter estimation of noisy damped sinusoidal signals in the frequency domain is presented in this paper. The advantage of the frequency domain approach is having the spectral energy concentrated in frequency domain samples. However, the least squares criterion for frequency estimation using frequency domain samples is nonlinear. A low complexity three-sample estimation algorithm (TSEA) for solving the nonlinear problem is proposed. Using the TSEA for initialization, a frequency domain nonlinear least squares (FD-NLS) estimation algorithm is then proposed. In the case of white Gaussian noise, it yields maximum likelihood estimates, verified by simulation results. A time domain NLS (TD-NLS) estimation algorithm is also proposed for comparison. The Cramer-Rao lower bound (CRLB) of the frequency domain estimation algorithms is derived. The theoretical analysis shows that the FD-NLS can yield a near-optimal performance with few energy-concentrated samples. On the other hand, the TD-NLS does not have the energy concentration property and requires more time domain samples to perform satisfactory estimation. Simulation results verify that the frequency domain estimation algorithms provide better tradeoff between computational complexity and estimation accuracy than time domain algorithms.

A new structured total least squares (STLS) based frequency estimation algorithm for real sinusoids corrupted by white noise is devised. Numerical results are included to contrast the estimator performance with an existing STLS frequency estimation method as well as the Cramer-Rao lower bound in different signal-to-noise ratio conditions.

The problem of frequency estimation for noisy sinusoidal signals from multiple segments or channels, which are referred to as gapped data, is addressed. Based on linear prediction and weighted least squares techniques, an iterative relaxation-based frequency estimator is devised and analyzed. The proposed algorithm is also extended to harmonically related frequencies. Computer simulations are conducted to compare the estimation performance of the developed approach with an existing multichannel frequency estimator and Crame?r-Rao lower bound.

The problem of parameter estimation of a single sinusoid with unknown offset in additive Gaussian noise is addressed. After deriving the linear prediction property of the noise-free signal, the maximum likelihood estimator for the frequency parameter is developed. The optimum estimator is relaxed according to the iterative quadratic maximum likelihood technique. The remaining parameters are then solved in a linear least squares manner. Theoretical variance expression of the frequency estimate based on high signal-to-noise ratio assumption is also derived. Simulation results show that the proposed algorithm can give optimum estimation performance and is superior to the nonlinear least squares approach.

In this paper, frequency estimation of a two-dimensional (2D) cisoid in the presence of additive white Gaussian noise is addressed. By utilizing the rank-one property of the 2D noise-free data matrix, the frequencies are estimated in a separable manner from the principal left and right singular vectors according to an iterative weighted least squares procedure. We have also derived the mean and variance expressions for the frequency estimates, which show that they are approximately unbiased and their accuracy achieves Cramer-Rao lower bound (CRLB) at sufficiently high signal-to-noise ratio conditions. Computer simulation results are included to corroborate the theoretical development as well as to contrast the performance of the proposed algorithm with the weighted phase averager and iterative quadratic maximum likelihood method as well as CRLB.

The ideal theoretical spectral analysis for PWM has been available for decades. However, PWM actually suffers from non-ideal switching characteristics and their effects on the output spectrum have not been thoroughly studied theoretically due to the complexity involved. Transient ringing on both rising and falling edges is a common phenomenon during switching and it has been held responsible for harmonic distortion, judging from experimental results. In this paper, we present a theoretical technique to model the PWM with transient ringing using a second-order circuit. This theoretical solution is the first of its kind available in the literature and it helps us to understand the distortion effects on the spectrum due to transient ringing. Simulations have been carried out to verify the correctness of the theoretical solution.

The detailed derivation of a compact generalized solution to determine the spectral components for multilevel uniformly sampled PWM (MPWM) is presented in this paper. By decomposing the multilevel PWM into a series of sub-PWM and making use of double Fourier series with variable translation, a compact solution without involving complex integrals has been developed. Solutions for both single-sided and double-sided sampled MUPWM are given in this paper. The compact solution has been verified with simulations.

This paper presents the performance analysis of a recursive least square algorithm with error-saturation in mixture noise. The algorithm is referred to as nonlinear RLS (NRLS). A generalized clipping function is considered for the error-saturation nonlinearity. An improved mean square behavior of NRLS is carried out. It is shown that the theoretical analysis and the simulation results are close to each other. From the analysis, we can relate the convergence and the mean square error in terms of the slope and clipping level of the nonlinear function. Based on the normalized mse, an instrumental variable is derived for yielding a variable clipping function to provide fast convergence and small mean square error.

This paper presents a simple methodology to convert a hard-switched pulse-width-modulated (PWM) H-bridge converter used in classical digital audio amplifiers into a zero-voltage-switching (ZVS) converter. ZVS is simply achieved by connecting an L-C series branch across the converter output. ZVS occurs during the dead time interval of the PWM signals, giving an effective improvement in the conversion efficiency, output frequency spectrum, and total harmonic distortion (THD). Simplified design procedures are given for choosing the value of the components. A prototype having output power of 20 W and with switching frequency of 705.6 kHz has been built and the experimental results are presented.

A new control strategy for a five-level neutral-point-clamped (NPC) H-bridge PWM inverter is proposed. It is shown that a three-level class BD PWM output can be obtained from one leg of the inverter if the phases of both the modulating signal and carrier are appropriately chosen. This new control strategy is called three-level NPC PWM and is very simple. By combining the three-level outputs from both legs of the NPC H-bridge PWM inverter, a five-level PWM output is achieved. A theoretical harmonic analysis of the proposed inverter has been carried out. It is shown that the inverter exhibits superior harmonics suppression property and only multiples of the fourth-order families of the cross-modulated harmonics are present.

This paper presents a frequency weighted generalized total least squares linear prediction for estimating closely spaced sinusoids. In this method, the received data is first processed by a pole-zero prefilter and then a generalized total least squares linear prediction is applied to the prefiltered signal. A procedure of optimizing the generalized solution is introduced. By computer simulations, it is shown that the solution can outperform the existing well known total least squares solutions especially in low signal-to-noise ratio.

This paper presents a new searching method, the Subspace Reference Neighbourhood Search (SRNS), to reduce the computational requirement in encoding image so that only a small number of distance calculations is needed to locate the closest codeword in a codebook and, at the same time, to preserve a good image quality. One significant advantage of this method is that the reference codeword can be located directly from the algorithm. The performances of SRNS in simulated known data sources and practical images are discussed and analyzed.

This paper presents a general total least squares (GTLS) solution for linear prediction to estimate closely spaced sinusoids. It is found that the TLS prediction error is not a good criterion to provide a robust solution. In this paper, a frequency weighted prediction error approach is introduced. Experimental results show that the GTLS solution based on the frequency weighted prediction error can give a robust performance even in very low SNR.

This paper proposes a new method to compute the third-order cumulants for a sequential data stream of arbitrary length. The new method is based on both the overlapping and recursive techniques to accommodate the missing correlations in the indirect method. The method is referred to as the Recursive-Averaging Cumulant Estimation (RACE). The new method has been used in signal detection problems and satisfactory results are obtained.

A modified frequency estimation method for use in the presence of additive complex white Gaussian noise is proposed. This method is basically an extension of Kay's method, but includes the use of autocorrelation functions. It retrieves the frequency value from the optimal linear combination of the differenced phases of the autocorrelation functions. It yields a considerably lower variance threshold than Kay's method while remaining unbiased and retaining the same frequency range.

A modified time-domain least-squares (MTLS) finite impulse response filter design method for transforming musical notes is introduced. The method is based on a new criterion called total harmonic gain to noise ratio (THGNR) in addition to the least-squares criterion. The THGNR criterion aims to reduce the noise gain of the filter. Compared to the time-domain least-squares finite impulse response filter design method, the MTLS offers a better solution, which has low noise gain and small time-domain distortion.

A general total least squares (GTLS) solution for linear prediction to estimate closely spaced sinusoids is presented. A frequency weighted prediction error approach is introduced. Experimental results show that the GTLS solution based on the frequency weighted prediction error can give a robust performance even in very low SNR.

A new two-stage nonlinear detector for synchronous DS/SSMA communications in impulsive channels is presented and analysed. In the detector, an interference cancellation scheme is used such that the effect of the multiple-access on the system performance can be alleviated. For a multiuser system with long spreading sequences an approximate error probability is derived and shows that the new detector is insensitive to the 'near-far' effect and gives a near-single-user performance. For cases with short spreading sequences Monte-Carlo simulations are carried out to estimate the error probability and the results show that the detector performs robustly in various mixtures of noise.

Waves arise in many physical phenomena which have applications such as describing the voltage along a transmission line and medical imaging modality of elastography. In this paper, estimating the parameters for two forms of lossy wave equations, which correspond to multi-mode and multi-dimensional waves, are tackled. By exploiting the linear prediction property of the noise-free signals, an iterative quadratic maximum likelihood (IQML) approach is devised for accurate parameter estimation. Simulation results show that the estimation performance of the proposed IQML algorithms can attain the optimal benchmark, namely, Cramer-Rao lower bound, at su-ciently high signal-to-noise ratio and/or large data size conditions.

A transient dc-link boost technique for avoiding pulse dropping in digital amplifiers is presented. The concept is based on increasing the dc link voltage and reducing the modulation index momentarily, when the modulating signal is large. Apart from effectively reducing harmonic distortion at the output, the proposed method can also widen the dynamic modulation range and thus reduce the required supply voltage. A 10 W prototype with an H-bridge inverter has been built. Some pulses are artificially dropped in the gate signals to simulate the pulse-dropping effect. The inverter output harmonic spectra with and without the transient boost have been compared.

This paper presents the derivation of a generic analytical solution for calculating the harmonic characteristics of multilevel sinusoidal pulse width modulation (SPWM) inverter. This generic solution has advantages over previously published solution since the phase shifts of all the carriers can be made arbitrary. This feature gives a high degree of freedom for evaluating the harmonic structure of a multilevel SPWM inverter with different carrier phase combinations. A multilevel SPWM waveform is decomposed into a series of positive and negative DC biased PWM (DBPWM) waveforms. A 3-dimensional model for describing each individual DBPWM is first derived. After applying the double Fourier series to each waveform, the multilevel SPWM output is then obtained by summing all the DBPWM outputs. The validity of the solution is verified using Pspice simulations.

A nonlinear extended recursive least square algorithm for the joint estimation of the amplitude and delay of the user signals in CDMA system over impulsive channels is presented. Simulation results show that by using a nonlinear function to palliate the effect of the impulse noise, the new algorithm can provide a robust performance quite close to the Cramer-Rao lower bound.