Discrete Cosine Transform

The spatial resolution of numerical weather prediction and climate models is generally determined by their grid spacing (∆x) or spectral truncation and the numerical implementation of dynamical core and model parametrisations. For example features of the scale 2∆x and 3∆x are smoothed to avoid numerical instabilities (e.g., aliasing effects) and parameterisations in connection to advection, pressure gradient force, and subgrid-scale diffusion can only be well represented at dimensions of at least four times the grid spacing. Some parametrisations, however, generate energy at the grid-spacing scale. These multiple effects on the effective resolution of models are investigated in this study for three high resolution regional climate models (RCMs) in dependence of their grid spacing.

In this paper, we present the IRIS architectural synthesis system for high-performance digital signal processing. This tool allows non-specialists to automatically derive VLSI circuit architectures from high-level, algorithmic representations, and provides a quick route to silicon implementation. By incorporating a novel synthesis methodology, called the Modular Design Procedure, within the IRIS system, parameterised models of complex and innovative DSP hardware can be derived and automatically assembled to create new DSP systems. ...

This paper examines finite field trigonometry as a tool to construct digital trigonometric transforms. In particular, by using properties of k-cosine function over a Galois field, the finite field discrete cosine transform is introduced. The finite field DCT pair in GF(p) is defined, having blocklengths that are divisors of (p+1)/2. A special case is the Mersenne finite field DCT, defined when p is Mersenne prime. In this instance block lengths that are power of two are possible and radix-two fast algorithms can be used to compute the transform.

This paper introduces a real-time data compression and transmission algorithm between e-health terminals for a periodic ECGsignal. The proposed algorithm consists of five compression procedures and four reconstruction procedures. In order to evaluate the performance of the proposed algorithm, the algorithm was applied to all 48 recordings of MIT-BIH arrhythmia database, and the compress ratio (CR), percent root mean square difference (PRD), percent root mean square difference normalized (PRDN), rms, SNR, and quality score (QS) values were obtained. The result showed that the CR was 27.9:1 and the PRD was 2.93 on average for all 48 data instances with a 15% window size. In addition, the performance of the algorithm was compared to those of similar algorithms introduced recently by others. It was found that the proposed algorithm showed clearly superior performance in all 48 data instances at a compression ratio lower than 15:1, whereas it showed similar or slightly inferior PRD performance for a data compression ratio higher than 20:1. In light of the fact that the similarity with the original data becomes meaningless when the PRD is higher than 2, the proposed algorithm shows significantly better performance compared to the performance levels of other algorithms. Moreover, because the algorithm can compress and transmit data in real time, it can be served as an optimal biosignal data transmission method for limited bandwidth communication between e-health devices.

A Phonocardiogram or PCG is a plot of high fidelity recording of the sounds and murmurs made by the heart with the help of the machine called phonocardiograph. It has developed continuously to perform an important role in the proper and accurate diagnosis of the defects of the heart. As usually with the stethoscope, it requires highly and experienced physicians to read the phonocardiogram. A diagnostic system based on Artificial Neural Networks (ANN) is implemented as a detector and classifier of heart diseases. The output of the system is the classification of the sound as either normal or abnormal, if it is abnormal what type of abnormality is present. In this paper, Based on the extracted time domain and frequency domain features such as energy, mean, variance and Mel Frequency Cepstral Coefficients (MFCC) various heart sound samples are classified using Support Vector Machine (SVM), K Nearest Neighbour (KNN), Bayesian and Gaussian Mixture Model (GMM) Classifiers. The data used in this paper was obtained from Michigan university website.

This paper aims at development and performance analysis of a speaker dependent speech recognition system using MATLAB®. The issues that were considered are 1) Can Matlab, be effectively used to complete the aforementioned task, 2) Accuracy of the Gaussian Mixture Model used for parametric modelling, 3) Performance analysis of the system, 4) Performance of the Gaussian Mixture Model as a parametric modelling technique as compared to other modelling technique and 5) Can a Matlab® based Speech recognition system be ported to a real world environment for recording and performing complex voice commands. The aforementioned system is designed to recognize isolated utterances of digits 0-9. The system is developed such that it can easily be extended to multisyllabic words as well.

Abstract – The development of digital imaging especially digital imaging cause greater storage utilization information. That resulting digital image quality is better, but also create new problems which the amount of storage required. With greater storage, data exchange is also increasingly difficult. Compression is a way to overcome this problem. Compression is generally divided into two kinds, lossless compression and lossy compression. Discrete Cosine Transform is one of the Fourier Transform method that provides a discrete function by simply taking the cosine of complex exponentials. In image processing, Discrete Cosine Transform array data intensity is converted into an array of data to determine how fast the frequency distribution of intensity.

Keywords : Lossy Compression, Discrete Cosine Transform

In the source-filter model of speech production, physiologically, the source corresponds to the vocal fold vibrations and the filter corresponds to the spectrum-shaping vocal tract. Vocal tract-based features like the mel-frequency cepstral coefficients (MFCCs) have been shown to contain speaker information. However, voice source (VS)-based features have also been shown to perform well in speaker recognition tasks, thereby revealing that the VS does contain speaker information. Moreover, a combination of the vocal tract and VS-based features has been shown to give an improved performance, showing that the latter contains supplementary speaker information.
In this study, the existing techniques for extracting speaker information from the VS are reviewed, and it is observed that parametric features perform poorly than non-parametric features. Here, an attempt is made to propose an alternate way of characterizing the VS to extract speaker information, and to study the merits and shortcomings of the proposed speaker-specific features.
The integrated linear prediction residual (ILPR) is used as the VS estimate. It is hypothesized here that a speaker’s voice may be characterized by the relative proportions of the harmonics present in the VS. The pitch synchronous discrete cosine transform (DCT) is shown to capture these, and the gross shape of the ILPR in a few coefficients. The ILPR and hence its DCT coefficients are visually observed to have both inter and intra-speaker variability, and thus it is hypothesized that the distribution of the DCT coefficients may capture speaker information, and this distribution is modeled by a Gaussian mixture model (GMM).
The DCT coefficients of the ILPR (termed the DCTILPR) are directly used as a feature vector in speaker identification (SID) tasks. By conducting SID experiments on three standard databases, it is found that the proposed DCTILPR features fare comparably with the existing VS-based features. It is also found that the gross shape of the VS contains most of the speaker information, and the very fine structure of the VS does not help in distinguishing speakers, and instead leads to more confusion between speakers. The major drawbacks of the DCTILPR are the session and handset variability, but they are also present in the existing state-of-the-art speaker-specific VS-based features and the MFCCs, and hence seem to be common problems. There are techniques to compensate these variabilities, which need to be used when the systems using these features are deployed in an actual application.
The DCTILPR is found to improve the SID accuracy of a system trained with MFCC features by 12%, indicating that the DCTILPR features capture speaker information which is missed by the MFCCs. It is also found that a combination of MFCC and DCTILPR features on a speaker verification task gives significant performance improvement in the case of short test utterances. Thus, on the whole, this study proposes an alternate way of extracting speaker information from the VS, and adds to the evidence for speaker information present in the VS.

Document Image processing and Optical Character Recognition (OCR) have been a frontline research area in the field of human-machine interface for the last few decades. Recognition of Indian language characters has been a topic of interest for quite some time. The earlier contributions were reported in [1] and [2]. A more recent work is reported in [3] and [9]. The need for efficient and robust algorithms and systems for recognition is being felt in India, especially in the post and telegraph department where OCR can assist the staff in sorting mail. Character recognition can also form a part in applications like intelligent scanning machines, text to speech converters, and automatic language-to-language translators.

Abstract The extension of the Fourier transform operator to a fractional power has received much attention in signal theory and is finding attractive applications. The paper introduces and develops the fractional discrete cosine transform (DCT) on the same lines, discussing multiplicity and computational aspects. Similarities and differences with respect to the fractional Fourier transform are pointed out

Real-time implementation of an assistive human-machine interface system based around tongue-movement ear pressure (TMEP) signals is presented, alongside results from a series of simulated control tasks. The implementation of this system into an online setting involves short-term energy calculation, detection, segmentation and subsequent signal classification, all of which had to be reformulated based on previous off-line testing. This has included the formulation of a new classification and feature extraction method. This scheme utilises the discrete cosine transform to extract the frequency features from the time domain information, a univariate Gaussian maximum likelihood classifier and a two phase cross-validation procedure for feature selection and extraction. The performance of this classifier is presented alongside a real-time implementation of the decision fusion classification algorithm, with each achieving 96.28% and 93.12% respectively. The system testing takes into consideration potential segmentation of false positive signals. A simulation mapping commands to a planar wheelchair demonstrates the capacity of the system for assistive robotic control. These are the first real-time results published for a tongue-based human-machine interface that does not require a transducer to be placed within the vicinity of the oral cavity.