Final Program

The total number of students, who graduate from all the colleges of National Institute of Technology (NIT) Japan, is over 10,000 every year. The college graduates are highly evaluated from the industrial world for over 50 years. In order for the students to acquire high skills, the curriculum in NIT should be combined lectures with experiments and/or practical trainings. However the skills of experiments and practical trainings are implicit know-how up to now. Therefore, we make the skills be explicit knowledge for improvement of skills. In this paper, we propose a method to visualize the skills, and we introduce an approach to decide the level of achievement of experiments and practical training skills. This approach is almost based on original lab manuals, but in order to evaluate of ICT and Software skills, we attempt to use "i Competency Dictionary" that is formulated for human resource development by Information-technology Promotion Agency (IPA) Japan. IPA was established as a Specially-Approved Corporation (presently Incorporated Administrative Agency, since 2004) based on the Law on Promotion of Information Processing, Business outlines of IPA are IT Security, Improving Reliability of Information Processing Systems, and IT Human Resources Development. In order to propose our methods, we checked and evaluated over 300 lab manuals. The experimental skill-sheets and the evaluation-sheets that are the vital part of our methods, have been developed using the learning items, which derived from those manuals in manually. It is possible to visualize each student’s skills by defining evaluation standard with the sheets. The evaluation standard is one of the tools to guide students to their final target effectively and clearly. Using our methods, the level of achievement and the motivation of students’ skills will be improved drastically in the experimental subjects and practical trainings.

Program comprehension is an important task in the software maintenance process. One of the challenges faced by Java developers is the inability to determine the correct number of class dependencies. The ability to recover class dependencies would help developers to understand the design of an existing system prior to modifying it. Many Java dependency analysis tools for this purpose have been proposed, but few are able to analyze the dependency types associated with Java bytecode. In this paper, we propose a reverse engineering tool to extract the dependencies from a compiled Java program. The tool provides a visualization of the recovered dependencies in a form that facilitates the developer’s ability to examine the classes and class relationships in the software system. The resulting dependency extraction capability will also enhance software maintenance and evolution. The results of experiments conducted with the intent of evaluating the proposed tool demonstrate both its accuracy and a few of its limitations.

The recently introduced ISO/IEC 29110 standard Lifecycle profile for Very Small Entities (VSE) has been adopted and practiced in many small and medium software companies, including in Thailand’s software industry. Many Thai companies complete their software process improvement (SPI) initiative program and have been certified. There are, however, a number of participants fail to success. This study was concerned with the factors that influence the accomplishment of the standard implementation in various VSE characteristics. In order to achieve this goal, exploring and extracting critical factors from prior studies were carried out and then the obtained factors were validated by the standard experts. Data analysis of comments and recommendations was performed using a qualitative content analysis method. This paper presents the initial set of influence factors in both positive and negative impact the ISO/IEC 29110 implementation with an aim at helping such SPI practitioners with some considerations to manage appropriate adoption approach in order to achieve its implementation.

The development of software products is currently being expanded to be based on web applications. Due to the time constraints, software design and development for web applications are driven. It leads the software or applications become more complex. Identifying impacts of changes or relationships takes a lot of time and effort. One challenge is to enable traceability on software artefacts created during the development of web applications. Moreover, it is believed that analysis of data obtained from traceability relations is another way to make software process more efficient. However, it is absolutely not an easy task due to lack of guidelines, support, and tools. This research thus aims to enable software traceability on the development of web applications which involve many types of documents.

In this paper, we propose a series of applications that represent a system family for processing Research Data. The whole system is a model of complete data flow of "research", from data capturing, processing, until intelligent data analysis. Data engineering, Software Engineering, Domain Engineering and Ontology Engineering are used for the development of the system. We also described tools and technologies to be used in building the system. The prototype of a small scale system has been built as a proof of concept of this family of system, for the domain of "Research" in Indonesia context. The aim of building this prototype are: (a) to provide an architectural description prototype of a knowledge repository where Indonesian Context Research as a central domain (b) to comprehend a software engineering for software system family.

The post enrolment based course timetabling problem (PECTP) is one type of university course timetabling problem which a set of events has to be assigned into time slots and suitable rooms according to students’ enrolment data. This problem is classified as a combinatorial optimization problem and it is very hard to solve the problem efficiently because solving the problem is to find an optimal timetable which it must satisfy all hard constraints and should satisfy soft constraints as much as possible. Moreover, this problem is technically complicated and highly time-consuming and it is known to be in the NP-complete class. In this paper we have developed a genetic algorithm hybridized with a local search technique and a tabu search heuristic for solving the PECTP. The algorithm takes advantage of the exploitation ability of a local search technique and a tabu search heuristic to improve the results obtained in the exploration phase of the genetic algorithm. In addition, the proposed hybrid approach was tested on a set of standard benchmark problem in comparison with other methods from the literature, and experimental results show that the proposed hybrid approach was able to find promising solutions for solving the PECTP.

This paper proposes a comparative study of commonly-used global optimization methods to improve training performance of back-propagation neural networks. The optimization methods adopted herein include Simulated annealing, Direct search, and Genetic algorithm. These methods are used to optimize neural networks’ weights and biases before using back-propagation algorithm in order to prevent the networks from local minima. Four benchmark datasets of prediction (regression) task were used to evaluate the established models. The experimental results indicated that optimizing neural network’s parameters is a complicated problem due to its high dimension of variables to be optimized. And only genetic algorithm was able to solve this difficult optimization problem. In addition, this paper also applied this success method to predict monthly rainfall time series data in the northeast region of Thailand. The results indicated that using of genetic algorithm with back-propagation neural network is a recommended combination

Nowadays, the classification problems has become more challenging due to the variety types of data set. Some data is appropriated for machine learning techniques and some data is appropriated for statistical leaning techniques. This works proposed a new hybrid ensemble of machine and statistical learning models using confidence-based boosting. The proposed method which uses variants of based classifiers can solve classification problems in variant data set. Moreover, combining the confidence value to the current boosting method can improve the performance of classification. The performance of proposed method is compared to the ensemble of decision trees and MRN created by Adaboost.M1 on data sets from UCI. The experimental results show that the proposed method can improve the accuracy in both binary and multiclass classification problems.

Surface Electromyography (sEMG) signal analysis is a challenging task in neuroscience. The signal is associated with an activity of muscles in Human body. It is a part of how human can control the robotic arm for helping people with disabilities. In this paper, we propose a new method based on Singular Value Decomposition (SVD) and SMO algorithm for classifying sEMG signals into six basic hand movements. By this proposed method, SVD is adopted for feature extraction and SMO classifier is used for classifying sEMG signals into six classes of basic hand movements in five subjects. In preliminary experiment, we investigates the number of features that can yield the best performance in the classification and it found that the optimal number of features is 50. For performance evaluation, five classifiers including Decision Tree, K-nearest neighbor, Naïve Bayes, RBF, and SMO, with 10 fold cross-validation technique are adopted. The experimental results have shown that SMO algorithm with V2M-SVD feature extraction can achieve the best performance for the classification of basic hand movements.

This paper proposes a hybrid differential evolution algorithm with grey wolf optimizer for solving continuous global optimization problems. The proposed algorithm introduces a new improved mutation schemes. In this algorithm, the control parameters are self-adapted by learning from previous evolutionary search. Beside, the grey wolf optimizer algorithm is used to enhance the crossover strategy. The performance of the proposed algorithm was evaluated on nine well-known benchmark functions and it was compared to particle swarm optimization, the traditional differential evolution algorithm and the self-adaptive differential evolution algorithm (jDE). The experimental results suggested that the proposed algorithm performed effectively to solving complex optimization problems.

The research on video has been developing along with the research of the digital image processing and technological advances. The development of the internet has led to the increased production of negative images and video content. There are many challenges faced in creating filtering systems for negative content, especially on video. Most researches on the negative content filtering have been based on the skin segmentation of image. In this study, the method of negative content detection on video (porn video) based on the skin segmentation in the video composer frame was developed. By combining the two color spaces namely RGB and YCbCr color space as the skin detection algorithms to improve the accuracy of the class determination on the video. The sampling method used in this research was the 8 bytes key-frame extraction or about 256 frames with a certain distance based on the total video frame. Based on the number of frames extracted, the porn percentage value was calculated. The data were 100 videos with duration ranging between 2-15 minutes each. The datasets were divided into 60 data, namely 30 porn and 30 non-porn. The other 40 data were used for accuracy testing. The determination of classes was limited by a porn percentage threshold value (pornographic %). Based on the research, from the result yielded, the video threshold was classified into porn class when the value of porn percentage threshold ≥ 70, and non-porn class when the value of porn percentage threshold < 70. The result of the 40 video tests showed the accuracy of about 90%.

Many researches have worked for the development of automotive technologies, so as to realize a convenient and safety automotive society such as Advanced Safety Vehicle (ASV) and Traffic Monitoring System (TMS). As a result, traffic accidents have decreased as the years pass, and the number of the dead and injuries have been reduced. However, autonomous driving is not only required to detect bicycles around vehicles, but also expected to understand the behaviors of bicycles. In addition, since the proportion of bicycle accident shows a high percentage in the total traffic accident, it has been predicted an increase of the accident of the bicycle in the future. Therefore we propose a multi-directional bicycle recognition system using the distance information obtained by the stereo camera. In the proposed system, we divide the moving direction of the bicycle into three directions. In addition, we propose a robust feature extraction method on rotation in order to cope with the bicycle between each direction. Finally,we show the experimental results and verify the effectiveness of the proposed method.

In recent years, there are many studies being performed in order to solve the problem of social robot, it is predicted that social robot will co-exist with people to expand the robot market in the future. The user-friendly robot which would be able to express emotion recognition and facial expression by the image sensing technology. The robot has much added value functions, i.e. conversation, facial recognition, facial expression recognition and emotion. By utilizing these function of the robot, the communication becomes convenient between human and robot.

In this paper, we propose a rapid motion retrieval technique using Dynamic Time Warping. Frames of motions are represented by feature vectors whose elements are integer values. The number of the feature vector dimension is reduced by using the principal component analysis method and values of elements in the vector are quantized into a couple of bits. The similarity matrix giving distances between frames of motions is generated for Dynamic Time Warping. Preliminary experiments are conducted to find the optimum parameter values by evaluating the motion retrieval performance. One important feature of the proposed method is that once the bit length for frame representations is fixed, the distance between any two frames in any two motions can be found as an element of the similarity matrix without changing its size,which can achieve rapid motion retrieval using Dynamic Time Warping. Comparative experiments with existing methods have proved that our proposed technique can complete retrieval tasks more than six times faster than the traditional Dynamic Time Warping method and shown almost the same level of accuracy and calculation time as the method described in [1] using the k-d tree algorithm. By using simple and discrete representations of frames, possibilities of achieving rapid retrieval retaining high retrieval accuracy are explored.

Quadcopter, a popular Unmanned Aerial Vehicle (UAV), is able to land, take off, hover, and move on 3D trajectory. The ability requires accurate control of the rotors velocity based on input from its sensors. One of the control mechanisms is the altitude control. This paper presents a new algorithm to identify altitude change of a quadcopter based on image processing techniques. The algorithm is designed to be simple and efficient in terms of computation and memory usage. The algorithm identifies altitude change by calculating correlation function of 10 sampled rows of pixels. This paper also presents some experiments conducted to investigate the performance of the algorithm. The results indicated that the algorithm is able to properly identify altitude change with accuracy of more than 96%.

Currently, VANET has been developed for communication be-tween vehicles (V2V). There is MI-VANET protocol that is developed for packet forwarding by using only buses. It has high successful packet delivery rates. However, its performance degrades when the number of buses is low. Therefore, this paper proposes a new hybrid method called Hybrid VANET. The method finds routes to forward packets considering both buses and cars. Packets are mainly transmitted by buses. If buses are not available, cars are used to assist packet forwarding. The method combines with two algorithms which are routing algorithm and forwarding algorithm. In the performance evaluation, the experimental results show that our propose method has higher successful packet delivery ratio and less packet delays than MI-VANET.

This paper presents a compact dual wideband asymmetric cross patch antenna fed by cross strip line for WLAN and WiMAX applications over the frequency band of (WLAN: 2.400-2.484 GHz, 5.15-5.35 GHz and 5.725-5.825 GHz) and (WiMAX: 2.5-2.69, 3.3-3.7, and 5.25-5.85GHz). The antenna designed on an FR4 substrate with a thickness of 1.6 mm and the relative permittivity of 4.4, is fed by a 50Ω transmission line with SMA connector, and is of size with 25mm×35mm in dimensions. The measured results show that the compact antenna achieves a broad operating bandwidth of 2.38-2.79 GHz and 4.51-5.87 GHz for |S11|< -10 dB and omnidirectional radiation pattern. The measured gains of the antenna at 2.4 GHz and 5.5 GHz frequencies are 1.55 dBi and 3.11 dBi, respectively.

This paper studies the feasible solution of power control for cognitive radio networks in the presence of primary user. The primary user allows the secondary users to spatially reuse the spectrum as long as they do not disrupt the communication of the former user. In addition, the primary user is willing to adjust its transmit power to meet the Signal to Interference Ratio (SIR) target, thus allowing the secondary users to achieve their QoS requirement. A feasible solution of power control can be achieved if there is a non-negative transmit power vector P* under constraint of the SIR target of SUs and PU. The simulation results show that the proposed power control scheme based on the feasible solution effectively controls the transmit power of PU and SU transmitters so that each receiver meets the SIR target.

Angular skew arising from imperfect biasing of the transmitter can cause a significant performance degradation in high speed optical QAM systems. A simple approach to detect QAM signal under the effect of angular skew is Gram-Schmidt orthogonalization procedure (GSOP). However, this approach has poor performance when the size of QAM constellation exceeds 256. To solve this problem, a new detection algorithm is proposed. From the results, the proposed technique outperforms the GSOP. It provides the skew tolerance of up to 25-degrees at BER of 10-5 for the SNR penalty of 1 dB.

In this paper we study the structure of reversible cyclic codes of both even and odd lengths over the ring R = F_4 + uF_4, u^2 = 0. We present necessary and sufficient conditions for a cyclic code over R to be reversible-complement. We also study cyclic DNA codes over nucleotide base pairs S_{D_16}, where S_{D_16} = {AA,AT,AG,AC,TT,TA,TG,TC,GG,GA,GC,GT,CC,CA,CG,CT}. For this we first establish a one-to-one correspondence between R and S_{D_16}, and then cyclic DNA codes are constructed over SD16 as the images of reversible-complement cyclic codes over R.

This paper presents the realization scheme of an electronically tunable first-order voltage-mode allpass filter using a single voltage differencing gain amplifier (VDGA) as an active component, together with one floating capacitor and one grounded resistor as passive components. The proposed circuit is efficient of providing an independent electronic control of the pole frequency (o) and the passband gain (HAP) through the transconductance gains of the VDGA. PSPICE simulation results, including frequency response and transient analysis, are incorporated to verify the theoretical analysis.

A simple circuit design technique for the realization of compact low-voltage low-power CMOS four-quadrant analog current multiplier circuit has been suggested. It is based on the use of the square-law characteristic in the NMOS current squaring function circuit operating in the saturation region. The suggested four-quadrant current multiplier circuit is designed for implementing in TSMC 0.25-um CMOS technology with a low supply voltage of +-0.75V. To evaluate the circuit performance, the circuit has been simulated by PSPICE program. The simulation results show that the circuit has a linearity error of about 1%, a THD of 1.07% at 100 kHz, the total power consumption of 87.6 uW and -3dB bandwidth of 1.15 GHz.

This study presents the resistorless floating inductance simulator circuit based on employing voltage differencing buffered amplifier (VDBA) as new active component. The proposed floating inductance simulator circuit uses two VDBAs and only one grounded capacitor, which is suitable for integrated circuit design. The resulting equivalent inductance value of the proposed simulator can be tuned electronically through the transconductance parameter of the VDBA. As application example, the second-order RLC bandpass filter has been simulated using the proposed tunable floating inductance simulator. The simulation results using standard 0.35 µm BiCMOS process model are included to verify the theoretical analysis.

An alternative topology of voltage-mode sinusoidal quadrature oscillator based on VDGA (voltage differencing gain amplifier) is proposed. The proposed quadrature oscillator consisting of two VDGAs and two grounded capacitors with the absence of the external passive resistors can generate two sinusoidal output voltages with 90 phase difference. The oscillation condition and oscillation frequency (wo) of the circuit are controllable independently through the transconductance gains of the VDGAs. Moreover, the proposed circuit configuration requires minimum of active and passive components, and also exhibits low sensitivity performance. Computer simulation results using PSPICE are performed to confirm the theory.

This paper presents a novel three-input single-output current-mode universal biquadratic filter using only one voltage differencing current conveyor (VDCC) and three grounded passive components. The proposed circuit can realize low-pass (LP), band-pass (BP), high-pass (HP), band-stop (BS), and all-pass (AP) biquadratic functions. The natural angular frequency (Wo) and the quality factor (Q) can be orthogonally tuned by varying the circuit components. The proposed configuration has a low component count, low active and passive sensitivities and suitability to be integrated circuit implementation. PSPICE simulation results using TSMC 0.18 μm CMOS process parameters are included, which show good agreement with the theoretical predictions.

Ontology is one of semantic web implementation and knowledge representation that can be implemented and consumed by any kind of application. Therefore, it will face the interoperability issue, as there are lots of application platform nowadays, and developers can’t restrict usage of ontology only in the application with same platform. As ontology is part of semantic web, the most suitable method to solve the interoperability issue is using web environment. One of solution that can be used is serving ontology through HTTP protocol and accessing the ontology by SPARQL. This paper explain one alternative of SPARQL implementation over HTTP protocol by using SPARQL endpoint to store and serve the ontology.

It is well known that it is not easy for disabled persons such as handicapped persons or blind persons to access common information on their personal computer (PC), although there have been many efforts, especially systems of information retrieval by speech. Unfortunately, current systems may not support for every language. Therefore, this work aims to present a prototype system of retrieving Thai document by speech, where it may help handicapped persons or blind persons to access information in their PC. The proposed system will accept speech queries, and then it provides search result on a personal computer screen. Finally, it will return the results as a text-to-speech output.

In response to natural disaster, refugee tracking system is intended to find the refugees. Dealing with the chaos condition during disaster, speedy disaster response and accuracy according to the type of resources and services that are delivered and their quantities, is of importance tasks. One of critical concerns in humanitarian is the separation of family members because of high people movement in natural disaster evacuation. Restoring the family member links is more important for people than receiving relief assistance. Community based crowdsourcing is seen a major breakthrough in information sharing and data collection of refugee tracking through techniques known as voluntary geographic information. Through refugee tracking system, refugees can be searched, reunited, and treated well. Designing a refugee tracking system integrated with family identity number is such an important thing. The objective of this paper is to track the refugees, including the position, the condition of treated refugee, and the needs of refugee. As a result, a simulation of refugee tracking integrated with demographic data will be able to track refugees and decrease redundancy data of refugees.

In this paper we present a preliminary research with proposed framework for automatic Entity Relationship (ER) Diagram Grading System. The proposed framework use ER Diagram in XML file format (XMI file) as an input to the system. There are two proposed approaches. First approach is using Tree Edit algorithm to assess ER Diagram similarity between input and the answer. The outputs are similarity score and feedback for inputted ER Diagram. Second approach is using machine learning algorithm to build classifier that grade ER Diagram automatically. These two approaches will be implemented and evaluated in the next phase of research to see the result.

Word Sense Disambiguation (WSD) has become a popular method for solving the ambiguous meaning of the words in Information Retrieval (IR) field area. Under the Natural Language Processing (NLP) community, WSD has been described as the task which able to select the appropriate meaning among the ambiguous meanings to a given word. Among three approaches, supervised based, unsupervised based and knowledge based approaches to WSD, this paper focuses on both supervised based and knowledge based approaches by proposing new Jaccard coefficient-based WSD algorithm to overcome the vocabulary miss match problem. WordNet and corpus external knowledge resources are utilized as the sense repositories by linking up with the new WSD algorithm to consider additional semantic for WSD. According to sample testing, IR system with new WSD algorithm attains more about 20 percentage of total accuracy rate than traditional IR system.

The sequence control is used in many industrial products, and it is important for the students who are studying the industrial field. Currently, active learning has been introduced in various countries. However, learning the sequence control is an acquisition of knowledge type. Learning of the sequence control is not suitable to active and creative learning. Therefore, we developed the sequence control learning kit that corresponds to the PBL. Learning contents are from the basis of the sequence to the PLC sequence.

The students learn the relay sequence as basic knowledge of the sequence. Using the Basic FA kit which is sequence control learning kit that donated to NCT from OMRON co., ltd, the fact that the relay sequence is the basic of the sequence can be learned. Using this kit, it is also possible to learn about matters such as the input/output device, control device, the circuit. But, it difficult to learn the PLC sequence in this kit. The student needs to advance learning kit. Therefore, we developed a new kit that the students can learn the PLC sequence. We selected the positioning control as a subject matter of the PLC sequence. the positioning control is a very important control method. The positioning control is used at various places. For example, elevators and automatic doors, factory production line. In this way, positioning control is indispensable to our life.

Developed the sequence control learning kit has close relationship with game characteristics. The game is a pinball game which is catching the ball using the positioning control. This kit can be learned in a variety of conditions. For example, such as various sensors, changes in the speed of the ball, the movement position of the positioning control. The students can be set various combinations themselves.

In addition, we developed Web contents. The contents which is each degree of difficulty support learning of the sequence control of the students. The students can learn at any time using a PC and tablet if they have Internet environment. Additionally, we developed a self-check test of quiz format. So the students need to be check their level of understanding. In fact, we have assumed a learning method using the sequence learning kit and web content which is indicated below.

First, the students need to acquire basic knowledge. They repeated learning acquisition knowledge and experiment. Then, they learn the positioning control using the pinball game as PBL. They make a circuit out of the box in the rule for each group. Lastly, Students compete the point of pinball game in the rival group. The purpose is to improve the learning motivation by elements of competition.

We are considering the learning method for more efficient PBL learning. So we decide and learning procedures and rules of the game. We repeat the improvement of the sequence control learning kit. We aim to develop efficient PBL learning environment.

In modern society, learning sequence control techniques is essential for engineers since apparatus using sequence control is used in various fields, for example, factory production lines, vending machines and so on. Therefore, we developed some teaching materials of sequence control for international students and students in our college. Relay sequential control is the subject matter in these materials for beginners, because it is the foundation of sequence control. The purpose of manufacturing the materials is to enable students to understand the foundations of sequential circuits, such as logic circuit and self-hold circuit, using these teaching materials.
Next, as an advanced step-up course of fundamental teaching materials, we developed new teaching materials whose subject is inverted-pendulum system. We selected inverted-pendulum system, because inverted-pendulum system is unstable and it is possible to check the pendulum visually. Our system has three sections: a PLC as the controller, the actuator section consisting of a potentiometer as an angle sensor and a servo motor, and a cart with a pendulum made of an aluminum square bar with ball-screw-stage as the controlled object. The motor is controlled by the sequence teaching materials for beginners mentioned above and the PLC sequence controller. The controlled object is controlled based on the position information from the potentiometer, and the pendulum is kept upright as the cart moves by the electrical signal the controller generates. Therefore, the objective of this inverted-pendulum system is to keep pendulum standing upright. The purpose of developing this teaching material is to enable students to understand sequence control using PLC-sequence control. This advanced teaching material will be used for lectures in our college and visiting lectures in overseas partner universities.
At present, we are testing the inverted pendulum system, but we can only keep it standing for one second. The reasons why we cannot keep pendulum standing may be: moving velocity of the ball-screw-stage is too slow and oscillation period of the pendulum is too short. We will consider how to solve these problems.

In a field of education, we need to introduce of International Education and Active Learning. We suggest education material and theme as the Global Engineering PBL education. So we developed the learning kit of sequence control. That is simple positioning control system which consist of PLC, sensors and some balls. It looks like game board, but that can change difficulty of the problem that the angle is changed and the number of the sensor is changed. Students cooperate to solve the problem that many balls are caught. This will be active learning and be PBL of the Problem Based Learning. We will implement of Global Engineering PBL between the NIT Sendai College and KMITL. We investigate how student’s Generic Skill grows by implementation of Global PBL. We’ll report on implementation of Global Engineering PBL and student’s skill investigation in 3 years from now on.

In this paper, a novel model-based method for induction motor with stator inter-turn short-circuit fault detection is presented. The proposed technique is based on the whiteness of innovation sequence developed by the standard extended Kalman filter. Nonlinear Generalized Likelihood Ratio method is applied to identify the faulty phase along with its severity. This technique just requires current sensors which are available in most induction motor drive systems to provide good controllability, and induction motor design details are not necessary. Computer simulations are carried out for a 3kW squirrel cage induction motor using MATLAB environment. The results show the superiority of the proposed method as it provides better estimates for stator interturn fault detection.

The finest design of the proportional-integral derivative (PID) controller plays an important role in achieving a satisfactory response of any rotational electro-mechanical system. This paper presents an optimal design of the PID controller in the DC motor by using the Taguchi method. The proportional gain, the integral gain, the derivative gain constitute the search space for the optimization problem. The objective of PID optimization is to minimize the integral squared error of the step response. The predicted optimum values of the control variables are determined by the Taguchi method using analysis of means. Analysis of variance (ANOVA) is used to select the most significant control parameters. Computer simulation shows that the performance of fine-tuned PID controller using Taguchi method is better than that of traditional open loop Ziegler-Nichols technique.

Direct Current (DC) high voltage transformers are one of the most important equipments in High Voltage Direct Current (HVDC) transmission systems. On the basis of the application of DC high voltage transformers from home and abroad, in this paper the main principles and characteristics of DC voltage transformers are summarized. The key issues of resistance/resistance-capacitance voltage dividers are analyzed in practical use. Considering the problems generated when using optical sensing principle for DC voltage measurement, this paper also presents some researches on this issue.

Recently, a new failure mode has been discovered in kraft paper, which is associated with corrosive sulfur, due to Dibenzyl Disulfide (DBDS) contamination. Simple tests using cylinder-to-plane electrode models were devised to highlight the presence of copper sulphide on paper surface. The purpose of this paper is to investigate the effect of copper sulphide through Partial Discharge (PD) measurement having behavior specifically associated with this type of pollution. The tendency of increasing of copper sulphide deposition on kraft paper insulation was found to be inversely proportional to the Partial Discharge Inception Voltage (PDIV). Breakdown Voltage (BDV) and Dielectric analysis were also considered to achieve a complete understanding of the influence of copper sulphide deposition on paper.

In typical medium voltage feeder wireless current monitoring systems, the current measuring units send computed RMS values to a receiver unit. But these RMS values cannot be further processed by advanced algorithm and more information about faults cannot be extracted. To have more powerful computing capabilities, the measuring unit will have to use more powerful CPU which in turn require higher power consumption. This is a limitation because the measuring unit has to be self-powered by current induction from the feeder line, and oftentimes there is inadequate current available. In this paper, we design the monitoring system so that the measuring unit wirelessly sends all the raw data of the sinusoidal waveforms to a receiving unit using high sampling rate (256 samples per cycle). The receiving unit is powered by low voltage distribution lines therefore removing the constraint of insulations and power consumption. We can then put the powerful CPU at the receiving unit end to process and analyse received raw data. Our wireless current monitoring system also use the split-core type current transformer (CT) for easier installation than conventional CT. We studied and presented the output results of the monitoring system under influence of 3 fault scenarios.

A fault tolerant single phase five level inverter is proposed in this paper for islanded photovoltaic (PV) generation system. The topology has the capability of maintaining same output voltage magnitude in case of switch open circuit fault and/or source open or short circuit fault with slightly reduced number of voltage levels. This helps in supplying uninterruptable power to essential loads even under fault condition. The topology also has the major advantage of energy balancing between two batteries using redundant switching states. This helps in reducing difference in state of charge (SOC) of batteries during partial shading or hotspots on PV panels. The fault tolerant single phase five-level inverter is simulated using MATLAB/SIMULINK and results are verified with laboratory prototype

This paper deals with optimal power flow considering several contingency states. Initially, contingency screening is conducted to make a sort list of contingencies based on impact on operation cost. Then, some severe contingencies are incorporated into standard optimal power flow problem. All considered states, normal and contingency states, are simulated simultaneously. Thus, if contingency occurs, it is sure that contingency states will satisfy all constraints i.e. generation limit, transmission limit and ramp rate. To solve the problem, quadratic programming is applied. IEEE 30 bus system is used as test system to show the ability of the proposed approach.

This paper presents a method to obtain electric power consumption of charging station based on the Monte Carlo simulation. The state of charge (SOC) of battery, charging time and arrival time to charging station will be considered. After obtaining the charging profile, the voltage profile will be analyzed by DIgSIENT Powerfactory. Subsequently, an analysis of the impact of electric vehicle charging stations to the voltage profile of the Provincial Electricity Authority distribution system is carried out. The simulation results indicate that charging station affects changes of voltage profile and voltage drop in the distribution system. However, if the power of the charging station or distribution system changes, an approach presented in this research can also be applied to analyze the impact on voltage.

Oral Potential Malignant Disease (OPMD) is a growing and significant health problem all over the world. OPMDs have the potential to lead to oral cancer. The detection and diagnosis of these OPMDs in their early phases is a challenging task for clinicians. This paper focuses on the identification of OPMDs in Visually Enhanced Lesion scope (VELscope) images using the contrast stretching technique. The parameters of the contrast stretching technique have been set automatically based on a statistical analysis of a set of images. The results show a better differentiation of OPMDs from the normal region. The proposed technique has shown a higher (greater than 90%) accuracy, sensitivity and specificity.

Falls are major problems that could have happen to elderly, and could cause paralysis, hip fractures, or could lead to disabilities or accidental deaths. An algorithm for accurately detecting the falls is necessary in order to decrease the rate of disabilities or accidental deaths. In this paper, a new algorithm to detect the falls from the acceleration signal using the wavelet transform and multilayer perceptron neural network is proposed. In our experiments, 5 volunteers who were healthy with the ages between 21 to 25 year old were asked to attach a tri-axial accelerometer at the right side of their waists. The orientation of the accelerometer was vertical direction. Next, the volunteers were asked to perform 5 daily-life activities: 1) walking 2) standing up from a chair 3) sitting down on a chair 4) lying down on a bed and 5) getting up from a bed; and 5 falling activities: 1) falling forward 2) falling backward 3) falling to the right side 4) falling to the left side and 5) falling while standing up. The experimental results showed that the proposed algorithm outperforms the current falling detection algorithms.

Differential box-counting (DBC) is one of the commonly used methods to estimate fractal dimension (FD) for gray scale images. It has been successfully applied in many applications such as image segmentation, pattern recognition, texture analysis and medical signal analysis. However, the accuracy improvement of FD estimation is still a grand challenge. This paper proposes a modified differential box-counting method using weighted triangle-box partition (MDBC) to reduce the estimation error caused by an undercounting problem. The proposed method is derived from two assumptions: (i) increasing the precision of box-counts by using unequally triangle box partition, and (ii) weighting the box-counts in proportion to the size of triangle-box partition. Based on these assumptions, on each grid a square box is divided into four asymmetric triangle-box patterns. Each pattern is calculated the box-counts by a weighted box-counting technique. The maximum number of box-counts represents the better estimation. In this way, the experimental results show that MDBC outperforms the baseline methods in terms of fitting error. Furthermore, the proposed method applies to finger-knuckle-print recognition in order to test its efficiency. The results illustrate that it significantly enhances the recognition rate when compared with the conventional differential box-counting (DBC) and improved triangle box-counting in combination with DBC (ITBC-DBC) methods.

In collaborative videoconferencing systems, the lack of gaze awareness between remotely located participants degrades the effectiveness of visual communication. This is because the position of camera and remotely located partners on the screen are not relative to produce the communicative gaze direction. Various elaborate researches have been proposed to solve this problem with the increase in difficulty for hardware implementation or software analysis. We have overcome these expensive alternatives by developing the image layout and camera-human positioning scheme that identify the position of camera, participants and remotely located partners on the screen relative to each other. The proposed approach allows the conversations to flow more natural as the visual attention of another person to any others can be recognized, providing a sense of more visually being together in collaborative activity.

This paper proposes the distance measurement for robot using 3D Stereoscopic Technique, which is a new approach for solving the problem of measuring objects in a 3D image system. This can effectively measure the images precisely. Therefore, this paper focuses on the detection system and measuring an object using image processing techniques and using 3D stereoscopic for robot eyes. We propose the new equations to calculate the distance by looking through 2 cameras. In order to calculate the distance by using an equation and watching the virtual image in the real place. We have tested by setting the camera focus at 2 meters. Then, perform a test measuring system that generated on error that lower than 10 percent when compared to the actual measuring.

In this paper, we present an interactive multimedia artwork, which awakens motionless images to interactive animations. This animation simulates the colour-flow movement from painting images. Particle movement and interaction present the rhythm of the brushstrokes. This artwork conveys lively image feelings. Human colour perception is the main idea of this work. We use image-processing techniques to extract colour distinction from painting images. After that, we render numerous particles from distinct colour areas information. Steering behaviour directs particle movement to flow smoothly on a touch-screen monitor. This artwork attracts audiences to admire painting in the new aspect that audiences play with it rather than just watching it. This works designed for arbitrary painting images, unlike previous work which ties to a particular image.

In this paper, we present a simple saturation-based image fusion technique for static scenes, which is aimed at generating high-dynamic range images. A set of images of the same stationary scene with different exposure levels, such as overexposed and underexposed images is blended together into one. The weighting factors for blending the images are determined by the saturation values of the images. Higher saturation indicates more vivid colors. Thus, we assign a heavier weight to the pixels with higher saturation, while lighter weights are used for those pixels whose saturations are lower. Our experimental results show that the proposed method can produce satisfactory results both visually and numerically at a low computational cost without adjusting any parameters. Moreover, the output images are fairly free from the halo effect problem.

A robust hand gesture recognition algorithm for remote human-machine interaction is proposed that has been optimized for implementation on an embedded platform. Hue-saturation-value (HSV) thresholding and unit-gradient vector (UGV) background subtraction methods are employed to overcome common issues related to variations in lighting conditions. Top-hat transformation is used to detect fingers and hand gestures, which are translated to command inputs for remotely controlling a media player. Experimental results demonstrate that the algorithm performs efficiently and accurately on an embedded board with an average computation cost of 1.67 second per gesture and is robust to changes in illumination.

This paper proposes methods to detect hemorrhages which are known as a kind of lesions in diabetic retinopathy. To detect the symptom, eye fundus structures (blood vessels and fovea) as well as microaneuysms need to be discriminated to filter out only the hemorrhages. Five processing steps are proposed based analysis on fundus images. First, preprocessing step is processed to improve the quality of the image. Then all red features are filtered out. They include blood vessels, fovea, microaneurysms and hemorrhages. After that, morphology operation and compactness measurement are applied to eliminate the fovea, and blood vessels. Finally, hemorrhages can be classified by using area method to remove microaneurysms and some small noise. 579 fundus images from Bhumibol Adulyadej Hospital were tested. The results were analysis by ophthalmologist in order to define system accuracy and preciseness. According to results of comparison, we found that the accuracy is 90 %.

Although offline handwritten signature recognition has been continually researched, it still requires an improvement of recognition rate. Most of existing techniques focus on feature extraction to improve their performance. This paper proposes an alternative way to increase the recognition rate by analyzing an important characteristic of input information, namely variability of signatures. The proposed method is based on the hypothesis; reducing the variability of signatures leads to boost up the recognition rate. Therefore, the variance reduction technique is applied to normalize offline handwritten signatures by means of an adaptive dilation operator. Then the variability of signatures is analyzed in terms of coefficient of variation (CV). The optimal CV is obtained and used to be a threshold limit value for the acceptable variance reduction. Based on 5,739 signature samples with 150 classes, the experimental results show that the adaptive variance reduction procedure helps improve the recognition rate when compared to the traditional schemes without adaptive variance reduction, including histogram of gradient (HOG) and pyramid histogram of gradient (PHOG) techniques.

Removal of rain from still images is a complex and a challenging task. The rain drops affects only on a very small region of an image, and hence, leading to a confusion to determine which region should be considered and which should not. In this paper, a new technique has been implemented which effectively uses the L0 gradient minimization approach to remove the rain pixels. The minimization technique can globally control how many non-zero gradients are resulted in the image. The method is independent of local features, but instead locates important edges globally. These salient edges are preserved and low amplitude and insignificant details are diminished. The rain pixels are removed in this manner. Finally the rain removed images are enhanced in intensity using histogram adjustment technique to get better contrast images. Experimental results show that the proposed algorithm is highly efficient as it removes rain effectively even under heavy rain conditions, while preserving the details of the image.

Batik cloth is Indonesia’s national heritage. Across the archipelago, there are numerous patterns and motifs of batik, each having its own meaning and cultural significance. In this paper, we present the results of our investigation of various combinations of SIFT features moments used in automatic classification of batik motifs. The classification method used in this paper is the k-Nearest Neighbor. Our experiments show that the best performance of the system is obtained using feature vectors of length 7, yielding a classification accuracy rate of 31.43% for 7 classes of batik motifs with no batik motif classes having zero classification accuracy rate. Furthermore, our experiments suggest that the feature moment that seems to be the best for the classification process is the m_c, while the feature moment that seems to hinder the classification process is the s^2_c.

Nowadays, cloud computing has been widely used. The Virtual Machines (VMs) are created on servers in cloud computing. The VM scheduling on servers for energy saving in the cloud computing has been studied. The Virtual Machine Scheduling Algorithm is proposed to schedule VMs in cluster environment clouds. However, it is not effective and has high computational complexity. In this paper, we propose the new scheduling method called Energy-aware Virtual Machine Placement (EVP) method to schedule VMs that reduce power consumption. In addition, the EVP method has lower computation complexity. We also formulate power consumption model to evaluate the performance of the EVP method. Finally, we evaluated the EVP method by simulation. The experimental results show that the EVP method has better performance.

Currently IEEE 802.11n is widely adopted standard with a large number of users in each service area. The large number of users can significantly degrade the overall system performance and availability of the system. In this paper, the usage time limiting technique for supporting a large number of users in Wi-Fi network has been proposed. The simulation results reveal that with the suitable threshold value for cutting off certain users who utilize the network for long period, the system can support more incoming users by scarifying low amount of usage time of the heavy users.

Network technology is being developed as a next-generation network (NGN) which is a software-defined networking. The network management can be done directly from the control unit separated from the transmission unit, call as decoupling controller and forwarder, which makes network management efficient. In the near future, the next generation network will be used to replace a traditional network. This paper presents a framework of Quality of Service (QoS) for improvement video streaming service on Mesh Topology. We used the optimal dynamic routing to distribute the flow of data. Our framework has enhanced quality for transferring video streaming when many clients request at the same time. In simulation results, which show that our proposed the QoS dynamic routing framework can achieve significant improvement in overall quality of streaming more than none QoS scenarios. In additional, the performance to maintain the server is good for clients even the system reaches the critical point of the network.

Flood is one of the natural hazards and it is critical to be controlled through proper management. Severe flood events in Thailand (2011) caused damage to both life and property. Moreover, heavy flood or drought affected life, existence and the country’s economy. Therefore flood control is important. This study proposes methodology for analyzing drainage paths in Chao Phraya river basin by graph theory and simulation connection between floodgate and water gauging stations to find out capability of stations. After that capability and efficiency of station were compared to determine which stations can receive water more than maximum performance and to find out solutions for stations.

In intermittently connected ad hoc networks, a reliable path between communication endpoints may not exist because mobile nodes move randomly during their operation. Furthermore, routing protocols have to work efficiently under constraints of limited bandwidth and energy. These challenge researchers to develop a suitable protocol for these networks. In this paper, we present a new routing protocol, which limit the number of message replication. In addition, node should spread the message with lower probability if the message is currently spreading in its local neighborhood. This increases the opportunity for a node to spread messages in to other network areas. Simulation results show that our work can increase packet delivery ratio while maintaining low end-to-end delay and message exchange overhead.

The increase in services that utilize ultra-high-speed networks have caused an increase in network traffic problems. Existing networks require new technologies because the management of these networks is difficult and the traffic problems cannot be easily solved. These challenges have prompted the development of software-defined networks (SDNs). SDNs divide existing network structures into data planes, which manage data transmission, and control planes, which manage network controls. These planes enable easy, flexible network management or expansion through centralized structures. Specifically, SDNs efficiently control networks by monitoring network traffic or utilizing the state information of the connected switches. However, due to the characteristics of SDNs that centralize network management, specifically switch performance, controller overhead may increase, which may lead to problems throughout the entire network. This paper discusses these problems and proposes an algorithm that promotes efficient network management when the flows are replaced at the flow tables.

Wireless communication systems are widely used in our daily lives, thus the power density of ambient radio frequency (RF) energy highly increased accordingly. Therefore, the energy harvesting system can be applied to recycle the energy to use with small devices such as sensor devices which consume only a small amount of energy. Furthermore, the application can be modified to use with wireless chargers and power detectors. The system consists of a rectifier circuit and an antenna. The RF energy is harvested by the antenna and the received RF signal is converted to be direct current (DC) power by rectifier circuit, thus the performance of the system will depend on the quality of the antenna and the efficiency of the rectifier circuit. The high efficiency rectifier can be achieved by the low power dissipation in the diode and the optimal impedance of the matching between the receiving antenna and the rectifier circuit. This paper proposes the source-pull simulation to determine the optimal impedance matching for the rectifier that can provide 35.53% of the maximum efficiency with 1.09 V of the DC output voltages and 3.16 mA of the DC load currents at 10 dBm of the input power by 2.42 GHz of continuous wave signal. At 2.1 GHz, the circuit can provide 21.41% of the efficiency with 0.939 V of the DC output voltages and 2.75 mA of the DC load currents at 10 dBm of the input power. For the broadband performance the circuit can provide the efficiency higher than 27% from 2.4 GHz to 2.5 GHz of the operation frequency and the efficiency more than 25% for the frequency from 2.1 GHz to 2.2 GHz at 10 dBm of the input power.

This paper describes an analysis of a two-phase interleaved step-up converter and its application in maximum power point tracking (MPPT) for photovoltaic (PV) panel. The transient as well as steady-state characteristics of the interleaved converter were analyzed by using state-space averaging technique. A two‑phase interleaved step‑up converter was then constructed and implemented to keep maximizing the PV output power in spite of irradiance changes. Compared to the single-phase topology the input current ripple was reduced by about 35% and the load voltage ripple was reduced by about 50%.

This paper presents an interfacing circuit for capacitive sensor using charge amplifier formed a capacitance-to-voltage converter. The proposed circuit is suitable for the front-end analog-to-digital converter (ADC) and wide variable range of sensing capacitance. The achieved output voltage provides a linear transfer characteristic and fast response. The circuit configuration is implemented using only commercially available devices. The performances of proposed circuit are confirmed by experimental results.

This paper presents a real-time sound synthesis of a musical instrument, named "Khim", one of popular kinds of traditional string instruments in Thailand. An approach of spectral modeling for synthesizing the Khim sound was implemented with a TMS320C6713 DSP board. Due to resource limitation of the DSP board, such as limited memory and only 4 external interrupts available, but up to 21 sounds of Khim pitches from vibration of 42 strings were needed to be synthesized, challenge occurred for the design of the real-time synthesizer. By using approximation of dominant frequency components and applying digital IIR resonators implemented with the DSP processor, the Khim sounds were successfully synthesized in real-time without significant difference from the original. An algorithm for controlling the key selection, implemented with another microprocessor, allowed the synthesizer to have enough input keys. The constructed synthesizer was fast responsive and since it had two input keys for a pitch, it allowed fast and continuously-striking sounds to be generated with unnoticeable latency.

The objective of the Ethernet-based coaxial switching system for an amateur radio station with multiple antenna was to create a system that could choose the appropriate antenna for a radio station in order to reduce the mismatch problem. That is essential because each antenna has specific operating frequency and specific radiation pattern. To control the switching mechanism, the microcontroller was an essential part of this embedded system. The system was implemented with high quality relays having good response in the amateur radio frequency range. System control could be done through personal computer or Android mobile phone. In addition, this system was integrated with a software-defined receiver. The implemented system was capable of selecting appropriate antenna as designed. The attenuation loss and signal reflection were acceptable by conventional amateur radio standards. This system has a user-friendly interface via mobile phone and personal computer. In addition to regular amateur radio setup, this work was tested with a software-defined receiver and it could operate in the similar designed conditions.

Acting as the gateway between the "real world" analog signal and digital signal, data converters have become a critical element of modern electronic devices. High-performance applications have put a particular emphasis on high-speed data conversion converters. A variety of converter architectures are being used to reach these higher speeds, each with special advantages. In modern life, the improvements of technologies and design methods have allowed to implement Successive Approximation Register (SAR) analog-to-digital converter (ADC) for higher performance. However, this converter’s has "N" time conversion steps required to digitize a sample for "N" bit due to the nature of successive approximation algorithm. To solve this situation, the high performance architecture is created based on SAR ADC with charge redistribution DAC. In this paper, the proposed system modifies the SAR function to get high speed by reducing the number of bit cycles. Therefore, the proposed architecture can obtain better speed than the traditional architecture.

The network monitoring system (NMS) monitors network service and system, resource capacity plan, statistics and accounting, fault management and performance; such as through- put, latency and round trip time. Audit log is a importance data source that are used for network behaviour analysis. There are a terabyte of traffic and event log that needed any database system to store and reply any query. This paper study the traditional log database (text file), MySQL and MongoDB to be implemented in network monitoring system. The storage growth and execution time are the performance metrics of this experimental. The result show that MongoDB consume storage more that other database system. However, the storage grown as stepwise function. In the execution time perspective, MongoDB process and reply query result within μs that less than MySQL and Text File.Because MongoDB file management is B-Tree, it result in searching time grown as O(log(n)). Whereas big O of searching in MySQL without indexing is O(n).

This paper presents the methodology employed and findings of the mobile technologies performance for Internet services in Bangkok, Thailand. The measurements involved the latency, user data rates, and speed tests. The performance comparisons were conducted among Thai major mobile network operators in the two frequency bands 850/900MHz and 2100MHz in Thailand. The findings indicated that not only Thai mobile network operators have already occupied more than one frequency band; but also in need for the QoS perceived by the customers. Hence, 2100MHz offered better performance when compared with 850/900MHz. The results revealed that TrueMoveH offered better performance in terms of the user data rates. While AIS had achieved the best results in the latency tests; the speed test performance were spitted between DTAC and TrueMoveH.

In this paper, multi-source single-relay power allocation in cooperative wireless networks is considered. We model the interaction among sources and relay as a Stackelberg bargaining game. A new price function is defined which set a different price for different source, then a relay power allocation scheme based on a distributed algorithm is derived. Simulation results show the proposed solution could effectively achieve fairness in multi-source and optimize system performance.

The proliferation of social networking services (SNS) Social Networking Services has highlighted the need of providing reliable but efficient mechanism to report information from mobile users to the data center. In this paper, a mechanism to reduce periodic update of position and sensor data for activities recognition combined with resource awareness data sensing from mobile device is proposed. The sensing mechanism need to be adjusted according to the remaining battery. To improve the reporting mechanism, variance analysis from sensor data captured in sliding windows is applied. A prototype of SNS has been developed and integrated with the improved reporting protocol using the proposed. Experiments to verify the performances of the proposed approach in term of bandwidth efficiency and energy saving have been conducted with promising results.

Intrusion Detection System (IDS) is a tool for anomaly detection in network that can help to protect network security. At present, intrusion are developing attack methods and difficult to detect accurately. In this paper, we concentrate on ensemble learning for detecting network intrusion data, which are difficult to detect. In addition, correlation-based algorithm is used for reducing some redundant features. Adaboost algorithm is adopted to create the ensemble of weak learners in order to create the model that can protect the security and improve the performance of classifiers. The U2R and R2L attacks in KDD Cup’99 intrusion detection dataset is used to train and test the ensemble classifiers. The experimental results show that reducing features can improve efficiency in attack detection of classifiers in many weak leaners.

This paper explains the AR.Drone position control scheme using Fuzzy Logic Controller (FLC) in a 3 dimensional coordinate. This control scheme uses two FLC block, for X-Y position and Z position. The inputs of FLC block for X-Y position are distance and angle, while the output is pitch and yawrate. Z-position will be controlled by another FLC block, which has two inputs, namely setpoint of z and real position of z, while the output is vertical rate. To compensate the sideward speed of the drone, roll compensation is used. The implementation results show that the AR.Drone can achieve the desired position with settling time for x, y position approximately 6 seconds, while z position around 10 seconds. Response x has the oscillation of approximately 5% around the set point. The implementation result are also compared with other fuzzy control for the same setpoint position

The aim of this paper is to present a simplified vector control implementation with no current sensors in order to reduce cost of overall system, design complexity, and improving by using fuzzy-logic control (FLC) for high performance in wide-speed operation. The voltage source inverter uses space vector pulse width modulation (SVPWM) method that is good voltage utilization and low total harmonic distortion (THD), the voltage command to inverter is derived from indirect vector control (IVC) theory and dynamic motor equation. The complete vector control scheme of the IM drive is experimentally implemented using dsPIC4011 board for ½-hp squirrel cage induction motor (IM). The scheme is compared to the conventional proportional-integral (PI) control and validated by MATLAB/Simulink and experiment. Both control techniques are tested at different dynamic conditions such as sudden speed change and speed tracking. The comparative experimental results show that the PI-fuzzy vector control is more robust for this application.

Object classification has many method such as sound signal or object compression. However, sound signal comes with a high noise level. Therefore, force response of knocking object become more interesting method. In this paper, we proposed the method with combination of force response and knocking sound for improve the analysis. The master-slave robot based on bilateral control system is used to knock the objects. Then the force response is estimated by disturbance observer instead of force sensor which has a limitation. The results of the experiment can show the different of each objects clearly.

As one of the most successful game kit, Kinect has been used widely outside gaming. Kinect provides easy and cheap solutions for stereo camera. Its openness in the software architecture provides an easy platform to many applications. This paper provides a platform that can be used to connect Kinect, arduino, MATLAB and a 4-DOF (RRRR 3D) manipulator. The manipulator is designed as a Wayang (Javanesee puppet) character that can be played by the operator called dalang. Using a simple vector manipulation, the skeleton data from Kinect is translated into joint angles, which are sent to the manipulator. Different Kinect parameters are investigated to find the optimal control settings. Experimental results show that the platform can effectively control the manipulator although only using feedforward control scheme. Two wayang characteristics have been played well using the platform. The platform can also be used as a robotics teaching kit.

Inspection of petroleum pipelines is very crucial for oil and gas companies. Using a ROV (Remotely Operated Vehicle) and a ROV support vessel for pipeline inspection requires extensive expenses due to heavy equipment and a large number of crew members for operations. In order to save the inspection cost, xxx university and yyy have cooperated to conduct a research and development of a specialized AUV (Autonomous Underwater Vehicle). This paper describes a design and implementation of our lab-scale AUV. The AUV is controlled by using a 6-degree-of-freedom PID controller. Sensors used on the AUV include a sonar, a DVL, a pressure sensor, and an altimeter. A software structure of our AUV is based on ROS (Robot Operation System).

Due to the recent epidemic outbreak of the coconut black-headed caterpillars in the southern provinces of Thailand, this research aims to develop a tree-specific pest control equipment using robotic and computer simulation technology. The climbing device allows the spraying head to reach the foliage of the coconut regardless of the height of the tree. The spraying should access the area of the foliage that infested with the coconut black-headed caterpillars. This research proposed the new techniques of modeling spray deposition on to the coconut leaves using image processing and probabilistic model. By spraying the substance directly to the foliage from an optimum position, the proposed method shows that it can largely reduce the excessive amount of chemical pesticide. The coconut climbing robot was developed as a wheel robot with a series elastic actuator. There are two motors on the robot, the drive motor drives the robot up and down the tree while the other motor is attached to a spring mechanism which keeps the holding force between robot and the tree.

Post insulators are exposed to pollution which can influence the performance of the insulation. To overcome this problem, post insulators were designed and made with different type and materials. One of the methods to observe the insulation characteristic of the post insulators is by performing withstands voltage test and measuring the leakage current. This test objective is to perform withstand voltage test and leakage current measurement under alternating voltage and direct voltage on 245 kV post insulators of different materials (ceramic, silicon-coated and hybrid) and analyze the behaviors and performance of the 245 kV post insulators under electrical and environmental stress.

Internet using behavior of teenagers is very vulnerable to threats, because internet information has both positive and negative aspects in which teenagers try to imitate what they see, but they cannot judge for themselves what is appropriate or not. Teenagers are curious, and always want to know more about new thing, but they lack the experience in decision-making, and since this phase of life is known as the transition period of life full of physical, mental, emotional changes which affect the social relationships; therefore, they need attention and proper advice and guidance from their parents which is very important to determine the role or behavior from their children. From the above mentioned, we investigated the relation between the upbringing of their parents, internet using behavior and experience of teenagers. We use data mining technique to group parents and teenagers into clusters, and then use the association rule technique to investigate the relationship of upbringing of parents that affect internet using behavior of teenagers.

Ontologies play a very important role in information exchange and sharing, and are typically constructed by human experts. However, this process is very costly in both time and effort. Given this, there is a need for automated ontology construction from various knowledge (such as text files). A key challenge of automated ontology learning from text is to extract key concepts, which are relevant to the domain, from the documents. Existing approaches typically require large set of training data with prior domain-specific knowledge. However, it is not always possible to provide such knowledge and trained data sets. To overcome this issue, we present a method to obtain key concepts from unstructured texts by using word association measure and statistical knowledge. To demonstrate the efficiency of our method, we compared its performance with that of a state-of-the-art method through extensive experiments, using two real-world datasets and show that in case of not having domain-specific knowledge, our method significantly outperforms the state of the art.

Personal information becomes an important asset for service providers in the age of data-driven. They repetitiously request and collect large amount of personal information from their consumers recently. The personal information collection activities raise privacy concern to their customers. Sometimes service provider collects personal information which consumers feel uncomfortable to disclose with no intention. Several recent studies showed that consumers might reject the collection activities, if they feel their privacy has been invaded or uncomfortable. Service providers have to take precautions to avoid that situation. However, each consumer has different ideas and judgment about the value of each personal attribute. This reason makes this issue is still complicated. This study aimed to estimate a disclosure value of each personal attribute to support service provider’s decision. We used a questionnaires result that collected consumers’ feeling when they have to disclose each personal attribute. We used a probability technique to find a relation of disclosure between personal attributes and graph mining techniques to construct a tree graph. Then, we proposed a method to estimate the value of personal attribute disclosure. Moreover, we showed a case study which resulted from our proposed method. The results showed the differences in the value of personal attribute disclosure between different groups of consumers.

This paper proposes a new framework to identify and rank Twitter accounts, of which short messages or tweets may influence a specific financial stock price. In this paper, we start by mainly focusing on the first step of our framework, selecting potential influencers based on their association with a particular stock market. With numerous limitations of Twitter service to access and acquire Twitters data, a new methodology is also proposed to use a List feature to select potential financial influencers. It requires only Lists of accounts that contain an official Twitter account of the specific financial market, provided by Twitter’s users, through Twitter API. Our methodology is designed to use only a small amount of data, by which it can be practically used under the limitation of Twitter API’s crawling rate. Experimental results show that most of the potential influencers returned by our methodology are similar and related to the specific financial markets, which are companies listed on S&P 500 in this experiment. Most of the returned influencers are the official accounts of company or organization from the same sector and news media with a special emphasis on the same industry. Comparing to Twitter’s user recommendation service (Who-To-Follow) and a crowdsourcing search for topic experts (Cognos), our methodology returns more related accounts in both percentage and the number.

Educational data mining is a widely interesting issue in data mining research field. One of topics is feature selection method that use to reduce a feature set. The main purpose of this study is to compare feature selection methods that improve the efficiency of student performance prediction. In this research, we proposed 4 feature selection methods: greedy algorithm, Information gain ratio, chi-square, and mRMR that combine with 4 classification models. The example data were 6,884 engineering students in Rajamangala University of Technology Thanyaburi, Thailand from year 2004 to 2010. In the Experiments demonstrate the effectiveness of the feature selection method in classification of student performance prediction. The result show that greedy forward selection with neural network classification model is the best efficiency couple with 91.16% accuracy.

Student performance classification is a challenging task for teacher and stakeholder for better academic planning and management. Data mining can be used to find knowledge from student data to improve the performance of classifying model. Before applying a classification model, feature selection method is proposed in data preprocessing process to find out the most significant and intrinsic features. In this research, we propose a comparison of four feature selection methods: genetic algorithms, support vector machine, information gain, and minimum redundancy and maximum relevance with four supervised classifiers: naïve bays, decision tree, k-nearest neighbor, and neural network. The experimental results show that the minimum redundancy and maximum relevance feature selection method with 10 feature selected give the best result on 91.12% accuracy with a k-nearest neighbor classifier. The result of the present study show that the advantage of future selection to find a minimum and significant of feature is more effective to classify the student performance.

A real-time tracking-based change detection using FAST features and a background feature model are proposed as a base system for detection of human entrance and exit.A speedy FAST feature extraction and tracking has to tradeoff its accuracy, which sometimes causes a failure in human entrance/exit detection. Many video sizes are therefore tested in the system to examine the trade-off effects on the accuracy of feature extraction, tracking, and entry/exit detection.Tracking parameters are also investigated to determine the optimal values for each video resolution, such that stable tracking and detection are achieved. Experimental results show that the higher the video resolution is the more the error is likely to happen. Instability of feature extraction and position which increases in higher resolution is proved to be the main reason of failure. Increasing the number of previous images used in the update of background feature model, proportional to the resolution of video, takes into account the feature uncertainty. As the result, the proposed method is robust to changes in video resolution and runs at 30 fps without a miss of human entrance/exit detection and false alarm.

The World Diabetes Foundation has predicted that more than 439 million people in 2030 will suffer from diabetes. Long-term diabetics can lead to the damage of retinal blood vessels, known as diabetic retinopathy, the leading cause of blindness in developing countries. One of the clinical features of diabetic retinopathy is exudate. Exudates have similar characteristic with optic disc. Therefore, in this research work, removal of optic disc is conducted to reduce false positive of exudates detection. The optic disc detection is done by finding the small area of the optic disc which is enlarged to obtain its total area. Green channel that contains useful information for exudates detection is filtered based on high pass filter. Afterwards, segmentation of exudates is conducted by using thresholding and morphological operations. Final result of exudates is validated with ground truth images by measuring accuracy, sensitivity and specificity. The results show that proposed approach for exudates detection achieves accuracy, sensitivity and specificity of 99.99%, 90.15% and 99.99%, respectively. This result indicates that the proposed method successfully detects exudates and is useful to assist the ophthalmologists in analysing retinal fundus image especially for exudates detection to diagnose diabetic retinopathy.

Recently, more attention is given to automatic detection of cancer. However, the multitude of kind of cancer, lung, breast, brain, skin … etc, complicate the detection of this disease with common approach. Adaptive method for each cancer is the alone response to achieve this aim. The segmentation of interest region is the first main step to differentiate between the suspicious and non suspicious part in the image. In this specific work we will focus in segmentation step using Total Variation methods. In this paper, we contribute in theory part of generalization of Chan and Vese (CV) model and its application in specific skin cancer images.

Principal component analysis (PCA) is one of the successful techniques for applying to face recognition, but its challenge still remains for solving in illumination effect condition. This paper proposes an improved 2DPCA (I-2DPCA) for overwhelming the face recognition in illumination effects. The proposed method is based on two assumptions. The first assumption is to create the covariance matrix that can effectively decompose the components of illumination effects from the eigenfaces. This avoids the illumination effect problem. The second assumption is to select the suitable eigenvectors that can significantly improve the recognition rate. Based on the Extended Yale Face Database B+ containing 60 illumination conditions, the experimental results show that not only does the proposed method decrease the computing time, but it also improves the recognition rate up to 95.93%.

This paper presents an approach method to detect vehicle color and to classify multi vehicle from video data. The multi vehicle is classified by blob analysis and feature-based. The proposed method uses a video file recorded by traffic surveillance camera as input. This technique applied RGB (Red, Green and Blue) to detect color of vehicle image. The vehicle is separated from background by using optical flow. Blob analysis and feature-based are performed the type of vehicles. Feature-based is extracted by localized color clusters. This method can classify vehicle images into 3 types of car, pickup, and truck. The K-Nearest Neighbor algorithm is used to detect color possible. The proposed vehicle recognition method can be applied to spot target vehicles which match the input (type and color).

Breast cancer screening used some methods, including by ultarasound examination. The reading of the ultrasound results are subjective, depending on the radiologist. Therefore, we need a help of tools that can help the physicians to make decision. System on the device should be able to diagnose breast cancer malignancies of various parameters including the breast lesion form parameter. Image classification of breast lesion form begins with input image processing by filtering with a median filter then performs segmentation with chan-vese active contour, extraction of characteristics with Zernike moments, then classification by support vector machine (SVM). At the Zernike moment, accuracy obtained using SVM classifier by 72,73%, while using the MLP classifier by 66,67%. At the invariant moment, accuracy obtained using SVM classifier by 69,69 %, while using the MLP classifier by 75,76%.

Routing path quality is an important parameter, which characterizes the reliability of data transmissions in multi-hop wireless networks. An accurate method to predict path quality can provide a valuable insight into the routing process design for supporting reliable transmissions in various types of wireless network implementation. Existing studies describe the data transmission process based only on the information from Network layer. However, Data link layer also affects the outcomes of the data transmissions too. In this paper, a theoretical mathematical model for predicting path quality in term of Packet Reception Ratio (PRR) is presented. The proposed mathematical model is based on the cross-layer information from both the network layer and the data link layer. The proposed model is validated with the results from NS2 simulator and the results from existing model. The results from the comparisons show that the proposed model can provide the estimated PRR values that are much closer to the simulation results than the existing model.

We considered an energy-saving scheme for smart-grid systems that utilizes the indoor location information of persons derived by an accurate localization scheme. To perform an accurate location estimation, we composed a localization scheme using ultra-wideband (UWB) signals that achieves low complexity and high accuracy in both line-of-sight (LOS) and non-line-of-sight (NLOS) environments. Energy saving is achieved by an adaptive limited illumination around persons with on-off switching control of lights, utilizing the location information. This can reduce energy consumption by approximately 2/3 or 1/2 without lighting inconvenience. However, in the previous study, only on-off switching was considered, and dimming control based on the illumination intensity at the person was not used. Therefore, in this paper, we propose a new energy-saving scheme that controls the illumination intensity in an office, and shows that energy saving is achieved by satisfying the intensity conditions through computer simulations. In addition, we propose an improved lighting control scheme with improved localization information in which prior information is considered in the location estimation of a mobile target.

Flash flood is a disaster that requires fast detection in order to prevent loss of life. We proposed a low-cost system, which is developed on Android phone and Arduino board. The system is about to measure water level by an echolocation method, collect information by its sensors, and transmit information to the server via mobile network or long-range WIFI. Low-level features, such as color, are analyzed at the phone as a decision trigger for network connection in order to reduce network cost. Other high-level features are extracted at the server in order to determine whether it is a flash flood.

Visible light communication (VLC) has recently become popular due to its benefit of VLC. VLC can provide not only the data communication via light, but also illumination. Indoor localization using VLC is a solution to locate objects or people inside a building using visible light wave by reading the identification data from visible light tubes. However, the identification data cannot be read in the light overlapping area. Time division multiplexing (TDM) is used to deal with this issue. In this work, the proximity technique is used for localization. The LED spotlights at fixed locations are used as transmitters that transmit their unique IDs. While the photodiode is used as a target receiver. By using the proximity technique, the rough location of the target is identified using the received one ID or more.

In this paper, a pilot-aided frequency synchronization scheme for orthogonal frequency division multiplexing (OFDM) system is proposed. A maximum likelihood (ML) estimator based on Double Dwell Synchronization (DDS) is employed for offset estimation and correction. The proposed DDS scheme provides an accurate estimation over a wide estimation range with reduced computational complexity for both AWGN and Rayleigh fading channels. The number of pilot subcarriers as well as the type of used pilot sequence are investigated to improve the system performance. Simulation results show that the proposed DDS provides accurate estimation that approaches the Cramer-Rao Bound (CRB) for higher number of subcarriers.

Visible Light Communication (VLC) Technology is the short range optical wireless communication technology using LEDs for communication and illumination. In this paper, we propose a repeater system for visible light communication to extend the coverage. Our proposed repeater system is composed of hardware and software components. The transmitter converts text data sending from the user into light signal data by using the on-off keying modulation (OOK) technique and transmits this signal via LED. The repeater receives light signal data by photodiode then transmits it out again. With this method, light signal is amplified at each repeater. The receiver converts light signal data back into text data and displays it on user interface for further use. This multi-hop communication gives more flexibility in communication because it helps to increase the distance between transmitter terminal and receiver terminal. Moreover, it also helps to avoid obstacle objects, such as walls, because VLC is a line-of-sight technology.

Design-for-testability (DFT) reduces the test complexity of sequential register-transfer-level (RTL) circuits. Only enhanced scan technique from the scan based approaches guarantee two-pattern testability with a large area and test time overhead. This paper proposes a path delay DFT technique for functional RTL circuits. Data paths are modified into hierarchical single-port-change (SPC) two-pattern testable (TPT) paths. The state register of the controller is transformed into a parallel-scan register. A snooping mechanism for the control, status and the not clear control lines to register and multiplexer is presented. Control lines considered as the segment of the RTL data path, not clear control signals and status lines are snooped to test without affecting the functionality of the RTL circuit. Two observation multiplexers are inserted to support the testing of control lines, status lines, and the state register. The proposed approach is based on the path delay fault model and supports the hierarchical test generation. The results show that for the given circuit, the area overhead of the proposed method rapidly decreases with the increase in bit width of the circuit data path. The proposed technique performs at-speed testing with small test application time and can obtain the fault coverage as achieved with the enhanced scan method.

A new mixed-mode quadrature oscillator using a single differential difference current conveyor transconductance amplifier (DDCCTA), two grounded resistors, and two grounded capacitors is presented. The proposed circuit provides both quadrature voltage and quadrature current outputs simultaneously. The condition of oscillation (CO) and frequency of oscillation (FO) of the circuit are independently controllable. The circuit is beneficial to monolithic integrated circuit implementation by using all grounded passive components. Moreover, the active and passive sensitivities of the proposed configuration are low. The validity of the proposed oscillator has been established by PSPICE simulations using MIETEC 0.5 μm CMOS technology.

In this paper, a hybrid OTA-C structure which consists of an ordinary differential OTA and a common drain is constructed as a fully differential notch filter for 50 Hz signal attenuation. The power consumption and linear range of this notch filter is 0.525 nW and 300 mVpp at 0.1% THD. This design only requires the total capacitance at 6.66 pF for the second order notch filter.

This paper presents a TDC-based low cost and high quality on-chip delay measurement with adjacency testable scan design. Adjacency test is useful for on-chip delay measurement with Time to Digital Converter (TDC) because it can generate arbitrary 1-bit transition to arbitrary input with smaller number of seed vectors. However the area overhead is high because it requires an extra shift register whose length is the same as the number of registers to store seed vectors. The proposed adjacency testable scan design does not require the extra shift register for its Chiba scan-based architecture. Therefore the area overhead is lower. The evaluation shows that the number of sensitizable paths is 7.1 times of that of LOS-based measurement and it is 3.5 times of that of LOC-based measurement. The number of vectors is 56.2% of that of enhanced scan design on average. The area overhead is 49.3% on average, which is the same order of that of enhanced scan design-based measurement.

This paper presents a fast scan-based on-chip delay measurement with variable clock generator using extra latches and multiple asynchronous transfer scan chains. Usual scanbased on-chip delay measurement requires continuous scan-in operation for assigning the identical test vectors and continuous scan-out operations for transferring the identical test responses, both of which result in long measurement time. The proposed delay measurement system reduces the measurement time with accelerating the time for transferring the identical test responses using the proposed asynchronous multiple scan chains as well as using the extra latches. The pulse width modification circuits are inserted to the asynchronous transfer scan path for robust asynchronous transfer. The experimental results show that the measurement time of the proposed method is 30.8 % of the conventional one under the condition that the length of scan chains is 64. I

In this paper, a fast geometric Type2 Fuzzy developed for altitude stabilization with barometric sensor as an input. MS5611-01BA03 as a barometric sensor to measure altitude of QuadRotor. The barometric sensor outputs are pressure value and temperature. This two kind of outputs should be converted to altitude value. Altitude which comes out from MS5611-01BA03 cannot be really steady like as a sonar sensor output for measure the altitude. This is the challenge in our research. With sonar sensor, it is produce a steady measurement but it’s have a limitations of height measurement below 300 cm only. In the other side, barometric sensor can measure any altitude but an outputs has a random noise for one measurement in QuadRotor applications. As a data information to controller, that random noise will be have a big effect if processed by simple controller, so the QuadRotor cannot steady in one of desired altitude. For that reason, this paper propose the method to reducing effect of random noise of MS5611-01BA03 outputs for altitude stabilization using a complex controller Type2 Fuzzy.

The signal-to-noise ratio (SNR) is unknown to the receiver in most wireless communication applications, and its estimation is often required. This paper deals with the estimation of SNR in a wireless communication system employing quadrature amplitude modulation (QAM) signals in complex additive white Gaussian noise (AWGN) channel. The estimator has been designed using the maximum likelihood approach for data-aided scenario. The Cramer-Rao lower bound (CRLB) has also been derived for the estimator. The results have been observed for different square and cross QAM constellations, and for different packet lengths. The obtained results confirm the efficacy of the ML estimator with respect to CRLB.

This paper presents a development of circular ring antenna for mobile broadband systems. The antenna is fed by a 50 ohm with two-port micro-strip line elements. The dimension size of antenna is 38 mm x 80 mm with low-cost FR4 substrate. The simulated and experimental result achieves the average gain about 3 dBi that covers the frequency range 3.1GHz-10.6GHz. The antenna has correlation coefficient average less than 0.1. For far field radiation patterns is omnidirectional in XZ-plane and bi-directional in YZ-plane. The experimental results are in the same trend with the simulated ones. This antenna is suitable for Multiple-Input Multiple-Output (MIMO) covering a UWB applications.

Free Space Optics (FSO) are going to be popular in the communications system, including commercial, military, and also in deep space communications, due to the higher bandwidth and data rate compared with the microwave communication. There are many factors to make the FSO system more efficiency. One of them is the alignment of the transponders. In this paper, the transponder alignment algorithm which is one of step track algorithm using in radar and satellites tracking antenna is employed. The most used step track algorithm for the transponder alignment is the gradient algorithm. The simulation results show that the gradient algorithm can be used for transponder alignment of the optical transponder.

Importance of power control is related to the interference problem between users on the cognitive radio network. Power control also keeps the battery device remains durable. Completion method uses an algorithm feasible solution to the centralized power control by looking at the value of the power vector is feasible. Feasible solution can be achieved if the value of the power user is non-negative, which means SINR target can be achieved and the system can be implemented. This study focused on the multi-channel that applied to multi-user. It can be concluded that the addition of the channels will increase SINR of user and the average power used by the user will decrease. The smaller the size of the user group, the SIR that can be achieved will be higher. The larger the size of the channel group, the greater the probability of interference affected to user.

A CPW antenna is designed at resonant frequency 2.3 GHz on low cost FR4 substrate by using simulation software. The simple rectangular slot loop antenna fed by coplanar waveguide is referred as the prototype antenna to show the design technique for bandwidth enhancement. To enhance the first band, the rectangular slot is extended from the top of slot loop into inner conductor and more wideband by inserting triangular conductor at the bottom of rectangular slot. Finally, the technique to enhance wide bandwidth covers mobile 4G, WLAN, WiMAX, and UWB by inserting U-shaped strip into inner rectangular slot. The simulated impedance bandwidth of our propose wideband antenna is 8.52 GHz from frequency range 2.22 GHz-10.74 GHz coverage mobile 4G (2.3 GHz-2.5 GHz), WLAN standard of IEEE 802.11b/g/j/a/n (2.4 GHz-2.4835 GHz, 4.9 GHz-5.091 GHz, 5.15 GHz-5.35 GHz, and 5.725 GHz-5.825 GHz), WiMAX (2.5 GHz, 3.5 GHz, 5.8 GHz), and UWB (3.1 GHz-10.6 GHz). The measurements of this real antenna confirm that our proposed techniques can achieve wide frequency range of 2.26 GHz-11.14 GHz, which is suitable for low-frequency of mobile 4G and cover up high-frequency of UWB. Therefore, this antenna supports all standard frequency of applications in WLAN, WiMAX, and UWB. By this performance, the proposed antenna can be used in wideband applications.

This paper presents model-based control that can be used to simultaneously track periodic reference signal and reject periodic disturbance. Two control methods named as Dual Repetitive Control (DRC) and Model Reference Repetitive Control (MRRC) are studied. The control methods are designed based on known model of the plant. Another assumption in the design is that the reference/disturbance period is also known. A simulation on Servo motor is carried out to analyze the performance of both DRC and MRRC. Transient response, tracking accuracy, and robustness under period variation are discussed.

This paper proposes gain scheduled (GS) control for active magnetic bearing (AMB) system. The system levitates and supports a rotor without contact. The rotor is a solid of revolution. AMB is unstable and strongly nonlinear due to characteristics of the magnetic levitation. Furthermore, gyroscopic effect corresponding to the rotational speed and the moment of inertia of the rotor occurs. Thus, AMB system tends to be unstable by the gyroscopic effect. The rotational speed varies in operation. It is treated as a time-varying parameter. The moment of inertia is different by rotor shape. It is treated as an uncertain parameter. The robust stability for the rotational speed and the moment of inertia is guaranteed by using polytopic representation. Linear fractional transformation (LFT) is applied to design the GS controller via parameter dependent Lyapunov function. The problem of the GS controller can be formulated as solving a finite set of linear matrix inequality (LMI) conditions. The effectiveness of the proposed method is illustrated by simulations.

This paper presents robust control design for Control
Moment Gyroscope (CMG). There are practical difficulties
in the control design of CMG. First, friction occurs on the
rotation axis of CMG. Second, CMG is a nonlinear system.
Third, there are uncertainties in the mathematical model of
CMG. In this study, those practical difficulties are solved. The
controller is designed to compensate the friction. The controller
has the integrator for the error between state and reference to
eliminate steady-state error. There are trigonometric functions
of a state variable in the mathematical model of CMG. In this
study, the trigonometric function is approximated more exactly
by not first-order but third-order Taylor series expansion. The
mathematical model involving those high order terms is represented
as the equivalent first-order system by using descriptor
representation and linear fractional transformation (LFT). The
robust stability for the system with those trigonometric functions
is guaranteed by using polytopic representation based on Linear
Matrix Inequalities (LMIs). The moment of inertia has one of
the uncertainties in the mathematical model. It is treated as the
uncertain parameter in this study. The robust stability for the
system with the uncertainty is guaranteed by using polytopic
representation based on LMIs. The effectiveness of the proposed
controller is illustrated by simulations.

Nowadays, energy saving solutions have become an interesting topic in all fields of technology, including Android mobile devices. Data polling and data pushing techniques are the two methods used by Android mobile devices to gather and retrieve information from the Internet. Due to the limited amount of battery, there is a need for a way that can help reduce the battery consumption. In order to achieve the objective, this paper proposes that data compression known as GZip is applied to the data before it is sent to the Android mobile device. The results show that the amount of battery used to send data is reduced by approximately 65% and 30% for the data polling and pushing techniques respectively.

Industries have now leveraged commodity hardware on their big data products. Big-data technology enables large scale processes to be executed with a huge amount of data being processed and optionally stored in persistent storage. One of leading innovation in alternative NoSQL and big data which enables low cost commodity hardware plays role is Redis, an in-memory database technology. Instead of disk, memory used in Redis is mainly to avoid latency during I/O processes. Big data processes require tons of I/O process on storage. This makes in-memory database run faster than conventional disk-optimized database. This research is an attempt to expose latency and improve performance by experimenting multi-thread approach during short message (SMS) delivery in single-thread Redis environment in combination with Rapidpro, a mass SMS management platform.

The main objective of this research is to integrate honeypot and IDS, which can generate and activate snort rule automatically based on the data sending by honeypot server. The new technic is present in this paper, honeypot will collect the data, send the data to IDS, and then IDS will evaluate and generate the rules automatically. Rule that has been made will be active to filter packets sent by the user on the network. We compare rule generated automatically with default rule in snort system for the same pattern. The performance of the proposed technique was evaluated by measuring the effectiveness of IDS server from the attacking.

Nowadays, Twitter data has been becoming more interesting in social science study since it can effectively reflect a nature of human behavior. Unfortunately, a process to analyze Twitter data is very complicated, and the results of existing tools are still limited and not suitable for the study in this domain. In this paper, we present a system that was tailored to analyze Twitter data for the social science research. The system comprises four main functions including: (i) interest group customization, (ii) case study management, (iii) user/keyword search, and (iv) user-friendly visualization. Also, three kinds of measures which are connectivity, reciprocity, and mentioning measures were proposed to support the analysis process. Then the experiments were conducted on more than two millions Twitter activities related to the political situation in Thailand during May-June 2014. With the system abilities providing scenario-based analysis and capturing interactions among user groups, significant parameters and useful knowledge about the network can be extracted. The results showed that our proposed measures can effectively reveal meaningful knowledge in Twitter social group with the aid of the system.

Learning is the important one for learner. Every learner must learn, but how to learn for the most effective? The goal of this paper is to design framework for e-Learning recommendation base on Index of Learning Styles (ILS) model. The framework divided into four modules. 1) Data Module is the module that storages data of student’s surveying by using questionnaires. 2) Rule based Module is the module that uses technique of data mining to find their learning styles of students. 3) LMS Module is the module for learning and teaching. 4) Content Module is the module that stores Science’s content that was is designed by Index of Learning Styles. This paper adapts ILS with e-learning system. Besides that it can forecast the best learning style for learner automatically by using data mining. The evaluate result from experts by using the interview without structure has found that experts has some related comments of design framework are appropriate with the average score equal to 3.87 (S.D.=1.21).

This study aims to investigate the factors that influence the Thai’s elderly intention to use LINE application for Quality of Life. To this end, this study is developed by integrating the Unified Theory of Acceptance and Use of Technology (UTAUT). This research model is empirically using survey data from 434 participants of the elderly in Thailand. All of six research hypotheses were positively significant supported.

It is commonly believed that the development of ICT contributes in various fields. Accordingly, the Ministry of Internal Affairs and Communications have promoted "ICT community development business". In local area where does not have enough public transport facilities, it is hard for elderly person especially who cannot drive a car to go out. Additionally, it is supposed that the time national disasters like earthquakes happen all of transport would be stopped. To solve these problems is the one of goal of the business, and we research and develop a vehicle called city commuter that has convenience ICT functions. Its design is based on three-wheeled electric bicycle. This paper shows the way we decided the design of the city commuter. After that, it explains the detail of the method to create each ICT system. At the end, it describes the conclusion and the future prospects of this study.

This study aims to build a model of BI adoption in logistics service providers industry in order to investigate factors influencing BI adoption. This model is based on Technology-Organization-Environment (TOE) framework generated by Tornatzky and Fleisher. The research model consists of eight factors categorized in technological context (relative advantage, compatibility, and complexity), organizational context (organization size, organization readiness, and top management support), and environmental context (competitive pressure, and government support). Quantitative approach is used in this study in order to validate the research model. Questionnaire survey is conducted with 168 logistics service companies in Thailand for data gathering to test the relationships among three contexts in the research model by applying structural equation model (SEM). The results show that compatibility and organizational readiness have a direct effect on BI adoption in Thailand logistics. In addition results and implications in this study provide contribution to extend understanding of the determinants affecting BI adoption in logistics service companies.

Cloud computing is applauded as a promising IT solution for SMEs which are generally constrained by limited time, expertise and resources. However cloud implementation is a complex and evolutionary process. It is further complicated by the cloud technological characteristics, organizational contexts and concerns outside an organizational boundary. How can SMEs approach their cloud implementation in a sensible manner and achieve expected results? This article recounts the experience of two SMEs and proposes a cloud implementation framework that can help small and medium SaaS providers to better leverage cloud infrastructure to support their business goals and eventually benefit from their cloud infrastructure.

Traditional approaches to ranking documents in Information Retrieval (IR) are under the assumption that the representation of information needs is clear and well-defined. This representation, which is usually in the form of a search query, is arguably considered ambiguous or underspecified. To deal with this uncertainty, much recent research has focused on creating IR systems that diversify search results so as to satisfy the multiple possible information needs underlying the query. To validate these IR systems, many new evaluation measures have been proposed to quantify their effectiveness in terms of diversity and redundancy. Among these, a new diversity-based measure, called normalized Coverage Frequency (nCF), has lately been proposed to quantify diversity in a ranking. When a new measure is proposed, its reliability needs to be validated. This paper conducts an empirical experiment to compares and contrast state-of-the-art diversity and redundancy-based measures, in term of discriminative power and stability of system rankings. Our experiment shows that the nCF is rated the best among all the studied measures. Moreover, this finding is confirmed by when nCF is interpolated with other redundancy-based measures (i.e., ERR-SA and A-nDCG). the nCF is considered more relatively robust than another diversity measure, subtopic-recall.

This paper presents the improved algorithm for the Hybrid Approach of Neural network and Level-2 Fuzzy set (HANN-L2F). The main structure is including 2 parts. The first part is Neuro-Fuzzy system, including the MLP Neural network with the combination of the level-2 Fuzzy system. The second part is using k-nearest neighbor to classify the output from Neuro-fuzzy. The HANN-L2F is an algorithm with high classification performance. However, the process to determine the number of membership functions in HANN-L2F is take time and sometimes results the high number of clusters, this make the high complexity of the overall process. This paper, including the data adjustment in HANN-L2F by using the determination process and implement k-means algorithm to find the appropriate number of clusters. Several types of standard datasets from UCI repository machine learning are used to verify the performance of the propose algorithm. The additional data adjustment process can improve the classification performance of HANN-L2F, furthermore, can reduce the classification time as displayed in the experimental results.

A tropical cyclone disaster is one of the most destructive natural hazards on earth and the main cause of death or injury to humans as well as damage or loss of valuable goods or properties, such as buildings, communication systems, agricultural land, etc. To mitigate severe impacts, track and intensity forecasting is a world-widely adopted process. With accurate forecasting, proactive measures can be appropriately applied on time to reduce both human and property losses. However, Thailand has insufficient meteorological data to apply the NWP model. In fact, the forecasting is done manually in Thailand. This makes the forecasting unreliable and time consuming, which leaves not enough time to prepare a good warning bulletin. To address these problems, this paper proposes an integrated short-range tropical cyclone track and intensity forecasting system by using only 11 features which were extracted from satellite images with improvement of the traditional statistical methods. The performance of the model is satisfactory, giving an average of 3.72 degrees of 6 hour, 12 hour and 24 hour track forecasting errors from best track data and the average errors is lower than traditional techniques by 35.86% on Mercator projection map and the average intensity forecasting errors of 6 hour, 12 hour and 24 hour is lower than traditional techniques by 33.56%.

This paper presents a new clustering algorithm, called Cell-MST-Based Method that is a combination of Cell-based method and Minimum Spanning Tree based (MST-based) methods. The algorithm is dedicated for Big Dataset on limited memory computer, especially for thin big datasets which have a small number of attributes but a very large number of instances. Firstly, the Cell-based method converts a big dataset to a small grid of cells in such a way that required memory to store an edge-weighted graph created from the grid is less than the available memory of a computer. Then the MST-based methods obtain an optimal threshold, estimate the number of clusters and determine the initial centroids. The proposed new combination of Cell-MST-based methods can reduce more than 99% of the required memory of the previous similarity-based and MST-based cluster number estimation methods. Moreover, this new Cell-MST-based method also outperforms the quantization error modeling method in terms of executing time and estimated accurate level.

This paper proposes a new methodology that automatically generates English mnemonic keywords to support the learning of basic Japanese vocabulary. A new phonetic algorithm, called JemSoundex, is also introduced for phonetically transliterating the Japanese and English languages for phonetic matching. The effective mnemonic keywords are selected and ranked by considering their phonetic, orthographic and semantic similarities, as well as psycholinguistic power. A system-oriented evaluation is conducted to evaluate the proposed methodology, and in particular an approach on the basis of the JemSoundex algorithm. The experimental results show that the JemSoundex outperforms other comparative approaches, i.e., IPA, the original Soundex and Metaphone.