Acoustic Analysis and Classification of Infant Cry: A New Approach

Volume 5 Issue 2 April - June 2017

Research Paper

Acoustic Analysis and Classification of Infant Cry: A New Approach

T. Shreekanth*, S. Saraswathi**

* Assistant Professor, Department of Electronics and Communication Engineering, Sri Jayachamarajendra College of Engineering, Mysuru, India.

** Lecturer, JSS Institute of Speech and Hearing, Mysuru, India.

Shreekanth, T., and Saraswathi, S. (2017). Acoustic Analysis and Classification of Infant Cry: A New Approach. i-manager’s Journal on Digital Signal Processing, 5(2), 1-7. https://doi.org/10.26634/jdp.5.2.13734

Abstract

In this work, a computational platform has been provided for a specific case study of infant cry analysis for identifying the abnormal infant cry, which is based on the principle of supervised classification, requiring the design of a proper knowledge base with a prior known normal infant cry samples called control sample. This research has a much focused 2-class problem, to be very precise it is a class and a complimentary class problem, where class refers to a healthy normal infant cry class and a complimentary-class refers to an unhealthy abnormal infant cry class. In this study, the authors have made use of Discrete Wavelet Transform (DWT), Mel Frequency Cepstral Coefficients (MFCC), Vector Quantization (VQ), and Euclidean Distance measure. The nearest match of the test infant cry sample is identified by correlating it with the infant cry samples present in the database and then it is classified as normal or abnormal (pathological) infant cry. The proposed method used 100 normal and 100 abnormal samples for training. The algorithm has been tested on the test dataset consisting of 25 normal and abnormal samples and the efficiency is found to be 96%.

A Review of Comparison of Various Linear Phase FIR Filter Algorithms to Design an Optimum Filter

Volume 5 Issue 1 January - March 2017

Review Paper

A Review of Comparison of Various Linear Phase FIR Filter Algorithms to Design an Optimum Filter

Mayank Sharma*, Akanksha Khedkar**, Akanksha Jangde***, Bhanu Pratap Patel****, Dharmendra Singh*****

*-**** UG Scholar, Department of Electronics and Telecommunication Engineering, Shri Shankaracharya Institute of Professional Management and Technology, Raipur (C.G), India.

***** Assistant Professor, Department of Electronics and Telecommunication Engineering, Shri Shankaracharya Institute of Professional Management and Technology, Raipur (C.G), India.

Sharma, M., Khedkar, A., Jangde, A., Patel, B, P., Dharmendra. (2017). A Review of Comparison of Various Linear Phase FIR Filter Algorithms to Design An Optimum Filter. i-manager’s Journal on Digital Signal Processing, 5(1), 39-45.https://doi.org/10.26634/jdp.5.1.13531

Abstract

For better design of the FIR filter, it is necessary for the designer to know the drawbacks of all the design methods. In this paper, the authors have compared two algorithms to get a better solution to design an optimum FIR filter. A lot of works have been already done in the design of FIR filter. So the methods and analysis of the algorithms help in the design of the filter, which at least removes all the drawbacks of both the filter design algorithms, which has been discussed in this paper. The papers based on the Parks McClellan algorithm, Particle Swarm Optimization method (PSO), Dynamic and Adjustable Particle Swarm Optimization (DAPSO), Particle Swarm Optimization with Variable Acceleration Factor (PSOVAF) in Linear Phase Digital Low Pass FIR Filter, planned Hybrid algorithm are quick and economical evolutionary algorithms, Artificial Bee Colony (ABC), Particle Swarm Optimization (PSO) and Differential Evolution (DE) based algorithms are used to compare them to obtain a solution. Therefore an effective and efficient optimized FIR filter can be designed.

ECG Feature Extraction and Parameter Evaluation for Detection of Heart Arrhythmias

Volume 5 Issue 1 January - March 2017

Research Paper

ECG Feature Extraction and Parameter Evaluation for Detection of Heart Arrhythmias

Gandham Sreedevi*, B. Anuradha**

* Research Scholar, Department of Electronics and Communication Engineering, Sri Venkateswara University, Tirupati, India.

** Professor, Department of Electronics and Communication Engineering, Sri Venkateswara University, Tirupati, India.

Sreedevi, G., Anuradha, B. (2017). ECG Feature Extraction and Parameter Evaluation for Detection of Heart Arrhythmias. i manager’s Journal on Digital Signal Processing, 5(1), 29-38. https://doi.org/10.26634/jdp.5.1.13530

Abstract

ECG analysis continues to play a vital role in the primary diagnosis and prognosis of cardiac ailments. This paper presents a new approach to classification of ECG signals based on feature extraction and Artificial Neural Network (ANN) using Discrete Wavelet Transform (DWT). Nineteen ECG signals from MIT-BIH database were used to test the performance of proposed method. A 97.12% of sensitivity and 94.37% of positive predictivity were reported in this test for QRS complex detection. Arrhythmias detected were bradycardia, tachycardia, premature ventricular contraction, supraventricular tachycardia, and myocardial infarction.

SIR Based Dual-Band Bandpass Filter Using Source-Load Coupling

Volume 5 Issue 1 January - March 2017

Research Paper

SIR Based Dual-Band Bandpass Filter Using Source-Load Coupling

Farheen*, Sudhanshu Verma**

* PG Scholar, Department of Electronics and Communication Engineering, Madan Mohan Malaviya University of Technology, Gorakhpur, India.

** Assistant Professor, Department of Electronics and Communication Engineering, Madan Mohan Malaviya University of Technology, Gorakhpur, India.

Farheen., Verma, S. (2017). SIR Based Dual-Band Bandpass Filter Using Source-Load Coupling. i-manager’s Journal on Digital Signal Processing, 5(1), 24-28. https://doi.org/10.26634/jdp.5.1.13529

Abstract

This paper presents the designing of Dual-Band Bandpass Filter (DBBPF). The designed filter consists of two Stepped Impedance Resonators (SIR) of different impedance ratios. These two resonators are coupled parallel to source and load end. Due to source load coupling, several transmission zeros are generated. These transmission zeros will help in the suppression of harmonic responses. Location of transmission zeros can be adjusted by varying the gap between the input-output feed lines. The center frequencies of designed filter are 3.05 GHz and 6.2 GHz, with 3 dB bandwidth of 370 MHz and 1.9 GHz, respectively. In the designed filter, the fractional bandwidth and the center frequencies of both the pass bands can be controlled individually by controlling the impedance ratio. The filter shows better isolation between the pass bands with better out of band rejection without the spurious response. The designed filter has a very simple structure.

Denoising of EEG Signal Using FrFT based Barlett Window

Volume 5 Issue 1 January - March 2017

Research Paper

Denoising of EEG Signal Using FrFT based Barlett Window

Jayalaxmi Anem*, G. Sateesh Kumar**

* Senior Assistant Professor, Department of Electronics and Communication Engineering, Aditya Institute of Technology and Management, Andhra Pradesh, India.

** Professor & Head, Department of Electronics and Communication Engineering, Aditya Institute of Technology and Management, Andhra Pradesh, India.

Anem, J., Kumar, S. G. (2017). Denoising of EEG Signal Using FrFT based Barlett Window. i-manager’s Journal on Digital Signal Processing, 5(1), 18-23. https://doi.org/10.26634/jdp.5.1.13528

Abstract

Electroencephalography (EEG) is an electrophysiological monitoring method to record electrical activity of the brain. EEG recording is highly susceptible to various forms and sources of noise, which present significant difficulties and challenges in analysis and interpretation of EEG data. Noise sources may consist of power line interference, base line noise, random body movements or respiration. A number of strategies are available to deal with noise effectively both at the time of EEG recording as well as during pre-processing of recorded data [8]. In this work, the authors have proposed FrFT based Barlett window to enhance the quality of EEG signal and the fidelity parameters like Signal to Noise Ratio (SNR), MSE, LSE, and sensitivity have to be computed and analyzed in a Matlab environment.

Spectrum Analysis of Nearest Neighbor Signal Using Wigner-Ville Distribution

Volume 5 Issue 1 January - March 2017

Research Paper

Spectrum Analysis of Nearest Neighbor Signal Using Wigner-Ville Distribution

C. Mounika*

PG Scholar, Department of Electronics and Communication Engineering, Sree Vidyanikethan Engineering College, Tirupati, India.

Mounika C. (2017). Spectrum Analysis of Nearest Neighbor Signal Using Wigner-Ville Distribution. i-manager’s Journal on Digital Signal Processing, 5(1), 13-17. https://doi.org/10.26634/jdp.5.1.13527

Abstract

One of the most powerful tools for the representation of the classical signal processing methods is algebraic structure. Algebraic Signal Processing (ASP) provides a whole frame for representing the classical signal processing concepts. In ASP, the signal model is defined as a triple (A, M, Φ), where A is a chosen algebra filters, M is an associated A -module of signals, and Φ generalizes the idea of a Z-transform. By using Nearest-Neighbor shift, a new signal model can be developed, i.e., Nearest-Neighbor signal model. The main aim is to represent the Nearest-Neighbor signal in timefrequency domain and to analyze the spectrum. For studying the stationary signals, that is their properties are statistically invariant over time, Fourier analysis can be used, but for a non-stationary signal, it requires both time-frequency representation of the signal for complete analysis. In the context of signal analysis, Wigner-Ville distribution can be used effectively to analyze the time-frequency structures of a non-stationary signal. So Wigner-Ville distribution is used to represent the Nearest-Neighbor signal in time-frequency domain. At last, Wigner-Ville distribution of Nearest-Neighbor signal is simulated and absolute and Relative errors of Nearest-Neighbor signal and Wigner-Ville distribution of Nearest- Neighbor signal are calculated.

Wavelet Feature and SVM for Detection and Classification of Microcalcifications

Volume 5 Issue 1 January - March 2017

Research Paper

Wavelet Feature and SVM for Detection and Classification of Microcalcifications

Jhansi J*, Kalpana M**, Shobha M***

*-*** Assistant Professor, Department of Electronics and Communication Engineering, Siddharth institute of Engineering & Technology, Andhra Pradesh, India.

Jhansi J., Kalpana M., Shobha M. (2017). Wavelet Feature and SVM for Detection and Classification of Microcalcifications. i-manager’s Journal on Digital Signal Processing, 5(1), 7-12. https://doi.org/10.26634/jdp.5.1.13526

Abstract

The objective of this paper is to detect the microcalcifications from the digitized mammograms using support vector machine, based on effective wavelet feature analysis. Microcalcifications are tiny deposits of calcium in the breast tissue which are potential indicators for early detection of breast cancer. The identification of cancer tissue is prohibited by the poor contrast level of mammograms. In this paper, new approach helps to identify cancer tissue with better accuracy. Microcalcifications are extracted by using wavelet based feature extraction and compared with other feature extraction like Gabor filter based extraction. The results from the feature extraction are classified using support vector machine classifier that provides better performance than other classifiers on wavelet based feature extraction.