The current issue of the IETE Journal of Research (Vol. 60, No. 6, November–December 2014) contains 7 articles. These articles present on-going research and developments taking place in the broad areas of Electromagnetics, Communication Engineering, Electrical Engineering, Biomedical Engineering and Computer Science and Engineering.
The paper on ‘Microstrip-fed Circularly Polarized Antenna Array using Semi-fractal Cells for Implicational Band’, presents details of a novel 1 x 4 semi-fractal radiator patch element antenna array with dual quarter wavelength transformed feed for WiMAX and WLAN-Band applications. The design employs a tree shaped semi-fractal patch with modified feed-network. The measured 3-dB axial-ratio (AR) of the antenna array extends from 4.85 to 5.82 GHz, and impedance bandwidth is between 3.4 and 4 GHz in band I and between 5.1 and 6.7 GHz in band II for VSWR < 2.
The next paper on ‘Improved Plucked String Synthesis Model Considering Tuning and Playing Style’, describes a plucked string synthesis model including a simplified tuning system and a right-hand playing style that is based on digital waveguide synthesis and applied to the Gayageum, a traditional Korean plucked string instrument. The Gayageum has 12 twisted silk strings, each supported by a movable bridge called Anjok, which is placed on the body and is moved sideways along a string to tune it. The simplified tuning system is formulated with parameters related to the tension of the string and location of the Anjok which determine the pitch of a specific string. The unique right-hand playing style is proposed that is implemented by a simple feed-forward filter.
In the paper ‘VCMIPWM Scheme to Reduce the Neutral-Point Voltage Variations in Three-Level NPC Inverter’ the authors presents variable common mode injection pulse width modulation (VCMIPWM) technique to reduce the neutral-point voltage variations in three-level neutral-point-clamped (NPC) inverter. In the proposed scheme, various common mode voltages are injected into the sinusoidal signals and zero vectors are equalized. As the neutral-point voltage varies with the modulation index, proper common mode voltages are calculated for various modulation index regions and are carefully injected with the reference phase voltages. The authors base their calculations on equalizing the switching times of the zero/redundant vectors in a switching cycle for various modulation index regions. Without increasing the size of the capacitor bank and without the use of any additional hardware requirements, neutral-point voltage variations are reduced using VCMIPWM thereby proving efficiency of the proposed PWM scheme.
The paper entitled ‘Hyperspectral Image Classification Based on Quadratic Fisher's Discriminant Analysis and Multi-class Support Vector Machine’ presents a Hyperspectral image classification method based on Quadratic Fisher's Discriminant Analysis (QFDA) and Multi-class Support Vector Machine (M-SVM). The authors have used QFDA for feature extraction and dimensionality reduction of a high dimensional Hyperspectral image and M-SVM for classification purpose. The theoretical and experimental results show that the proposed QFDA-based method is better than the PCA-based method. It has been shown that the proposed method approaches to 94% of accuracy as the training samples increase.
In the next paper ‘Semantic Feature Clustering for Sentiment Analysis of English Reviews “, the authors have proposed a new feature extraction method named clustering features. The proposed feature extraction technique focuses on alleviating the data sparsity faced by supervised sentiment analysis by clustering of semantic features. The clustering features are capable of including semantic information and alleviating data sparseness for machine learning algorithm. The experiments performed for classification use support vector machine and Boolean Multinomial Naive Bayes (BMNB) machine learning algorithms.
The paper entitled ‘Empirical Wavelet Transform Based ECG Signal Compression’ presents an Empirical Wavelet Transform (EWT) along with Discrete Wavelet Transform (DWT) for compression and reconstruction of the ECG signals. The authors have used different threshold for different modes obtained by applying EWT. The proposed algorithm has been thoroughly tested on 20 self-acquired records of the ECG signal and 48 recordings of MIT-BIH Arrhythmia database. Improved results for self recorded ECG signals and MIT-BIH Arrhythmia database have been observed using the proposed technique of signal compression.
The paper on ‘A Novel Segmentation Methodology for Cursive Handwritten Documents’ aims to implement a complete framework of segmentation for handwritten documents in various languages. A novel connectivity strength parameter is used to decide groups of the components which belong to the same line. It has been shown that with the help of depth first search approach and iterative use of the connectivity strength function, over-segmentation is removed. The proposed scheme has been implemented and tested with English, Hindi, and Urdu text images taken from the benchmark database. Overall, the average accuracy of the proposed technique is reported as 97:30%.