M. Z. R. Khan

Iterative noise power subtraction technique for improved speech quality

An iterative spectral subtraction technique for speech enhancement that gives improved performance has been presented. In all the speech enhancement methods, a predetermined average power is subtracted in the spectral domain or a gain function less than unity is generated for the individual harmonics to reduce noise power. What ever the process is, it is a common observation that some of the power spectrums become negative after the enhancement process and is avoided by forcing the coefficients to zero. Under such a condition the actual amount of noise power subtracted is less than the amount intended. If the power subtraction is done in an iterative way such that the predetermined power is correctly subtracted from the noisy signal, the segmental SNR improves appreciably. The results show that an improvement in average segmental SNR by more than 0.5 dB can be achieved by adopting this technique.

A face recognition scheme based on spectral domain cross-correlation function

This paper presents a simple yet efficient face recognition technique, where a feature extraction algorithm is proposed based on the principle of spectral domain cross-correlation. Instead of considering the spatial variation of a face image as a whole, first we concentrate on spectral variation of each row of the image individually, which is obtained using discrete cosine transform (DCT). As each of these rows could carry distinct characteristic of the face image, considering all of them would ensure the extraction of the variation in face geometry without discarding any information even in a minute scale. It is shown that the cross-correlations in the DCT-domain considering pairs of consecutive rows provide a signature of the particular face image reflecting the variation in the face geometry along the vertical direction. In a similar fashion, a horizontal signature can be obtained considering DCT-domain cross-correlations along consecutive columns. However, it is observed that in comparison to horizontal features, vertical features offer better within-class compactness and between-class separation. In the proposed method, these two signatures are utilized combinedly in order to obtain a distinguishable feature space. From extensive simulations on standard databases, it is found that the proposed feature extraction algorithm offers advantages of simple practical implementation with a high degree of face recognition accuracy.

An efficient face recognition algorithm based on frequency domain cross-correlation function

This paper presents a simple yet efficient feature extraction algorithm for face recognition based on the principle of spectral domain cross-correlation. Instead of considering the spatial data of a face image as a whole, spectral feature is extracted from the each row of the spatial data individually. As each of these rows bears distinct characteristic of the face image, considering row-wise Fourier domain representation of all of them ensures the extraction of detail variation in face geometry. It is shown that the cross-correlations obtained considering pairs of spectral representations of consecutive rows provide a signature of the particular face image reflecting the variation in the face geometry along the vertical direction. In a similar fashion, a horizontal signature can be obtained considering spectral cross-correlations along consecutive columns. In the proposed method, both of these horizontal and vertical signatures are utilized in order to obtain a distinguishable feature space for a particular person. It is found that the proposed feature extraction algorithm offers advantages of simple practical implementation with a high degree of face recognition accuracy.

A face recognition scheme based on spatial correlation function

This paper presents a simple yet efficient face recognition technique, where a feature extraction algorithm is proposed based on the principle of spatial cross-correlation. Instead of considering the spatial data of a face image as a whole, we concentrate on each row and column of the data individually. As each of these rows and columns could carry distinct characteristic of the face image, considering all of them would ensure the extraction of the variation in face geometry without discarding any information even in a minute scale. It is shown that the cross-correlations obtained considering pairs of consecutive rows provide a signature of the particular face image reflecting the variation in the face geometry along the vertical direction. In a similar fashion, a horizontal signature can be obtained considering the change in cross-correlations along the horizontal directions. In the proposed method, both of these horizontal and vertical signatures are utilized in order to obtain a distinguishable feature space for the face image of a particular person. It is found that the proposed feature extraction algorithm can provide a satisfactory face recognition performance for different standard databases even utilizing a simple distance based classifier. Thus, the advantage of this proposed method is that it offers a very simple real life implementation without sacrificing the recognition accuracy.

Parameter estimation of AR systems at a very low SNR using prefiltering in the autocorrelation domain

A simple prefiltering technique in the autocorrelation of a noisy signal in the Fourier domain has been presented here to identify the system parameters using the high-order Yule-Walker (HOYW) equations. Presence of noise in the autocorrelation domain is the main reason for failure of the conventional system identification techniques at low signal to noise ratio (SNR). As system modes are observed in the Fourier spectrum of the autocorrelation sequence, proper threshold reduces the randomly distributed noise spectra leaving the significant portion of the peaks of the system spectra undistorted. Reconstruction of the autocorrelation sequence from the filtered spectrum reduces the unwanted fluctuations making the system identification more consistent even at low SNR. Results for estimated system parameters at a low SNR of -5 dB has been presented to show the efficacy of the proposed technique.

Threshold voltage sensitivity reduction of SOI four gate transistor

Threshold voltage of a SOI four gate transistor is studied to determine its dependency on different device parameters. A surface potential based analytical model is used for studying threshold voltage and an Atlas/Silvaco 3-D numerical model is also developed for the validation of the analytical model. The numerical model incorporates non-ideal effects like Shockley-Read-Hall recombination, concentration dependent mobility, Auger recombination and bandgap narrowing effect. Threshold voltage sensitivity on channel length variation is reduced by controlling device width (W) and silicon layer thickness (tsi). The idea is justified by both analytical model and numerical model.

Mathematical modelling of accumulation layer thickness of fully depleted G 4 -FETs

A physics based mathematical model is proposed here to determine the accumulation layer thickness in thin film fully depleted G4-FETs. The model was derived as a function of surface potential which was calculated by self-consistently solving two dimensional Poissons equation. The dependence of accumulation layer thickness on gate biases as well as the device structural parameters are also analysed here. The purpose of proposing this model is to determine the range of structural parameters for designing the devices, within which the accumulation layer thickness remains small enough to be approximated as a sheet of charge of zero thickness so that popular charge sheet model can be used to analyse transistor characteristics. Moreover, the biasing range can be determined also for which charge-sheet model can be used to design circuits.

Fast Maximum Power Point Tracker for photovoltaic arrays

A Maximum Power Point Tracking algorithm with only one current sensor is presented. The output current provided by the photovoltaic (PV) panel is used as the signal controlling the duty cycle of the PWM converter that regulates PV panel power output. Very rapid tracking, acute oscillations around MPP and low implementation cost makes it a very good candidate for PV system controller. The validity and performance is verified by simulation.

Analytical modeling of the SOI four-gate transistor using conformal mapping

Conformal mapping technique is used to model potential distribution of SOI four gate transistor in sub threshold region. The modeling is based on the solution of Laplaces equation. Potential variation between the lateral junction-gates is assumed to be parabolic. The potential variation between the MOS gates and junction gates are also studied for different structural parameters.

Analytical modelling of an asymmetric pocket implanted n-channel MOSFETs

Analytical model for surface potential of a pocket implanted asymmetric n-channel MOSFET device has been derived upon solving one dimensional Poissons equation. The continuity of potential and electric field in lateral direction is satisfied and appropriate boundary conditions are used while deriving the model. The dependence of the surface potential on different device parameters and biasing conditions has also been studied. The result of the proposed model is compared with the data available in the literature and good agreement was obtained.