Universiti Teknikal Malaysia Melaka: UTeM Open Journal System
Not a member yet
2735 research outputs found
Sort by
Estimation of Gait Parameters using EMG Signal with Extreme Learning Machine
No full textIn this paper, an algorithm to estimate the gait parameters based upon EMG signal is proposed. The algorithm is developed using extreme learning machine (ELM). Experiments were conducted to acquire the gait parameters from 18 healthy human subjects. EMG signals from Tibialis Anterior (TA) and Gastrocnemius Lateral (GL) muscles were obtained during the gait cycle. The target temporal gait parameters are gait speed and stance/swing phase which were measured using inertia sensor and camera system. The ELM algorithm was developed using a single hidden layer feedforward network architecture where the weights from the input layer to the hidden layer are randomized and not updated during the run. Results obtained from ELM were compared with artificial neural network (ANN) model with the same architecture as the ELM algorithm. In ELM, the mean estimation errors of gait speed, stance percentage, and swing percentage were 11.86%, 7.62%, and 6.07% respectively. This was compared to the errors of 12.92%, 11.75% and 9.56% using ANN. Besides that, ELM achieved shorter training and testing time. The robustness of ELM algorithm demonstrated the capability of real-time computation due to superior computing performance compared to conventional ANN models
An Improved Sauvola Approach on Document Images Binarization
No full textDocument image binarization is one important processing step, especially for data analysis. A variable background, non-uniform illumination, and blur give a big challenging task in order to detect the text. In this paper, a new binarization based on local thresholding technique ‘WAN’ was presented. The proposed algorithm is known as ‘WAN’ after the first name of the author of this paper. WAN has been inspired by the Sauvola’s binarization method and exhibits its robustness and effectiveness when evaluated on low quality document images. Sauvola method failed to segment if the contrast between the foreground and background is small or if the text is in thin pen stroke text. The objective of the WAN method is to improve the Sauvola method and achieved a better binarization result. The results of the numerical simulation indicate that the WAN method is the most effective and efficient (f-measure 72.274 and NRM = 0.093) compared to the Sauvola method, Local Adaptive method, Niblack method, Feng Method, and Bernsen method
Recommendations Related to Wheeze Sound Data Acquisition
No full textIn the field of computerized respiratory sounds, a reliable data set with a sufficient number of subjects is required for the development of wheeze detection algorithm or for further analysis. Validated and accurate data is a critical issue in the field of research. In this study, the protocol related to wheeze sound data acquisition is discussed. Previously, most articles focused on wheeze detection or its parametric analysis, but no consideration was given to data acquisition. Second major purpose of this study is to exhibit particulars of our dataset which was attained for future analysis. We compile a database with a sufficient and reliable number of cases with all essential details, in contrast to commercially available wheeze sound data used for research, freely available online data on websites and data used to train medical students for auscultation
Analysis and Classification of Multiple Hand Gestures using MMG Signals
No full textThis research aimed to find out whether the MMG signal is useful in recognition of multiple hand gesture. The following hand gestures are Hand closing, wrist flexion, wrist extension, opening, pointing. MMG is reflects the intrinsic mechanical activity of muscle from the lateral oscillations of fibers during contraction. However, external mechanical noise sources such as movement artifact are known to cause considerable interference to MMG compromising the classification accuracy. First aim to develop various feature extraction algorithms software that can identify multiple hand gesture using MMG signal. The main purpose of this work is to identify the hand gestures that are predefined using the artificial neural network, which is particularly useful for classification purpose. The MMG patterns are extracted from the signals for each movement, the features extracted from the signals are given to the neural network for training and classification since it is the good technique for classifying the bio signals. The features like mean absolute value, root mean square, variance, standard deviation and root mean square are chosen to train the neural network
A Wideband Circularly-Polarized Spiral Antenna for CubeSat Application
No full textIn this paper, a wideband circularly-polarized spiral antenna is proposed to be used for CubeSat application. The antenna covers two of CubeSat frequencies; the S-band (2.2 GHz) and X-band (8 GHz). The proposed antenna belongs to the wideband and frequency-independent antenna category, which are known to provide a constant radiation pattern, impedance and polarization throughout the whole bandwidth. The performance of the spiral is compared in two different conditions; in free space and above a ground plane with a separation distance of λ/4 at the operating frequency of 2.2 GHz and 8 GHz. The spiral above the ground plane exhibits a unidirectional radiation and higher gain with an increase of 26.6% (2.2 GHz) to 34.6% (8 GHz) than the free space spiral. Moreover, the return loss also managed to stay within the ideal limit of S11<-10dB (-19.528 dB at 2.2 GHz, and -21.92 dB at 8 GHz) and exhibits circularly-polarized radiation with low axial ratio<3 dB (0.78 dB at 2.2 GHz, and 0.89 dB at 8 GHz)
Remote Spectrum Analyzer based on Web Software Defined Radio for Use in Telecommunication Engineering Remote Laboratory
Get PDFSoftware defined radio (SDR) is a new paradigm in the design of wireless communication devices. Currently SDR technology is widely used in the field of telecommunications such as mobile phones and is very popularly used on amateur radio. SDR technology not only can be used for radio transceiver system, but SDR can also be used as a spectrum analyzer. The spectrum analyzer is an instrumentation that is needed by researchers in the field of telecommunications and amateur radio activists. The problem of researchers and amateur radio activists is how to use spectrum analyzers that can be accessed remotely, so that they can observe the spectrum of the radio frequency that produced by their communication devices remotely when they experiment with the radio transmitters or antenna. From this background, this research proposes the development of a remote spectrum analyzer system based on web software defined radio. The remote spectrum analyzer will be integrated with remote laboratories that can be accessed by the wider community either by researchers, amateur radio activists, and students at educational institutions, by accessing it through the internet network
Fundamental Shape Discrimination of Underground Metal Object Through One-Axis Ground Penetrating Radar (GPR) Scan
Get PDFGround Penetrating Radar (GPR) was used in this research to detect or recognize the buried objects underground. Hyperbolic signals formed by datagram of GPR after detection the buried objects which quite similar to each other in term of metal shapes. The research was tested on the metal cube and metal cylinder by using the A-scan of GPR. There are steps in this signal processing step which are pre-processing step, feature extraction, and classification process. The segmentation process hyperbolic signals were segmented one by one and normalize from the negative to positive signals. The hyperbole from the metal cylinder and metal cube that had been buried in the ground is differentiated using four features of their respective A-scans which are found the maximum value of amplitude signal graph, the number of peaks in the signals graph, skewness, and standard deviation values. Finally, the classification process used learning algorithm of Multi-Layer Perceptron (MLP) was a test on Bayesian Regulation Backpropagation (BR) was given the highest accuracy, 98.70% as a classifier to classify the metal shapes which are a metal cube and metal cylinder
Influence of growth duration to the Zinc Oxide (ZnO) nanorods on single-mode silica fiber
Get PDFSynthesizing Zinc Oxide (ZnO) nanorods by varying the growth duration will give effect to the morphology of ZnO nanorods which were grown by using microwave assisted hydrothermal method. The effect of different growth duration from 4 to 10 hours is investigated on the surface of bare silica optical fiber. The buffer coating of a silica fiber is stripped off for 5 cm to expose the area for the growth of ZnO nanorods. The ZnO nanorods were grown on the fiber by dipping the fiber into the prepared growth solution in a 90˚C microwave assisted hydrothermal synthesised. The physical characterization through field emission scanning electron microscopy (FESEM) results show the diameter of ZnO nanorods is about 75.0 - 93.8 nm and the length of ZnO nanorods ranges from 666.2 nm to 978.6 nm according to their growth duration. The uniform growth of 43.33 nanorods/(µm)2 reveals that the 10 hours of growth duration performs the highest density growth of ZnO nanorods. The amplified spontaneous emission (ASE) peaking at 1550 nm is used to investigate the effect on the light intensity in the optical fiber with coated and uncoated ZnO for optical characterization. The ASE spectrum shows that the light intensity decreases with the growth durations due to the light scattering effect
FLC-based Renewable Wind Energy Viability Assessment in Central Luzon Philippines
Get PDFIn this paper, the proponent designed and developed a fuzzy logic based control system for renewable wind energy viability analysis specifically in six selected sites in Central Luzon Philippines. The membership functions (MFs) were constructed for three classified parameters: wind resource density, other wind resource parameters and site-related criteria with 5%, 5% and 90% factor rates. The quantitative site assessment made by Karen Conover of Global Energy Concepts (GECs) was used for constructing the MFs and for establishing the 216 fuzzy rules using Sugeno-style of fuzzy inference system. The average score of selected sites were used for fuzzy logic assessment and it was realized that Sual Pangasinan and Carranglan Nueva Ecija are best suited for possible wind farm development considering that the linguistic assessment was “Very Good” while the rest were just “Good”. The results obtained using fuzzy logic and the reviewed literature was compared and a favourable correlation was observed. This means that the FLC-based wind viability assessment could be used as an effective computational tool for Central Luzon and even for other wind energy viability assessments inside and outside the country
A Novel Method for Tuning PID Controller
Get PDFThis research proposed a new tuning technique to search efficiently Proportional Integral Derivative (PID) parameters, by locating near-optimal tuning solutions, which compensate for delay time. The purpose is that to minimize response time by optimized PID gains Kp, Ki, Kd within a deferent order model. Related to survey, numerous existing papers propose to optimize proportional gains by introducing various methods. Most of these works cannot achieve to find the best solution for optimization of different orders system. By using both proposed tuning with improved selective switching, it is possible to obtain a maximum optimization for any order system. Proposed tuning was applied by using 17 steps with less than 39 generation loops; each generation includes four loops calculation. Response time is measured and compared with previous times until reached to optimal gains, then fixed Kp,Ki,Kd. The results show decreasing rise time to 0.0165s in the second order, and 0.119s in the third order with zero overshoot. Results prove that this method leads to more precise, effective, robust, optimization with less iteration and applicable to various plants. Furthermore, it is a quick, simple, powerful and more practical methodology, compared with PID toolbox tune