KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 10 of 16

Some of the metrics are blocked by your
consent settings
Machine Learning-Based Multi-Room Indoor Localization Using Fingerprint Technique
(2023-01-01)
Adiyatma, Farid Yuli Martin
;
Suroso, Dwi Joko
;
Cherntanomwong, Panarat
Nowadays, developing Wi-Fi-based indoor localization systems has become an attractive research topic due to the growing need for pervasive location determination. The fingerprint technique offers higher positioning accuracy in indoor localization than the distance-based technique. Fingerprint-based techniques via machine learning have been proposed for many years to provide high-accuracy indoor localization services. These works attempt to establish the optimal correlation between the user fingerprint and a pre-defined set of grid points on a radio map. In this paper, a comparative analysis of selected machine learning algorithms is conducted within the context of online phase fingerprint techniques for localization, focusing on implementation in a multi-room case. The experiment involves measurements using a Wi-Fi module in a laboratory, an aisle, a lobby, and a typical classroom, resulting in a small-sized fingerprint database covering a total area of 573.71 m2. The results reveal that Naïve Bayes (NB) obtains the highest localization accuracy in the laboratory and classroom. Meanwhile, Support Vector Machine (SVM) outperforms other algorithms in the aisle, while K-Nearest Neighbor (KNN) delivers the best accuracy in the lobby. In summary, NB, KNN, and SVM are suitable pattern-matching algorithms for multi-room indoor localization and relatively small fingerprint databases.
Some of the metrics are blocked by your
consent settings
Intravascular Ultrasound image Registration based on Geometric Spatial Energy
(2023-01-01)
Yookwan, Watcharaphong
;
Sangrueng, Jiranun
;
Onpans, Janya
;
Srikamdee, Supawadee
;
Chinnasarn, Krisana
This paper introduces a novel method for intravascular ultrasound (IVUS) image registration based on geometric spatial energy. By combining geometric vessel features with spatial energy computations, the proposed approach achieves accurate image alignment, even in the presence of vessel deformations. Key geometric features serve as reference points, guiding the optimization process to minimize spatial energy and achieve optimal image correspondence. Extensive experiments on diverse IVUS image pairs demonstrate the method's superior performance over traditional techniques, promising enhanced clinical applications in vessel analysis and treatment planning. The performance of the proposed method are 10.33 of MSE, 10.55 of HD, and 0.15 of TRE, respectively.
Some of the metrics are blocked by your
consent settings
Thai Voice Recognition for Controlling Electrical appliances Using Long Short-Term Memory
(2020-04-01)
Saheaw, Wuttichai
;
Jaiyen, Saichon
;
Hanskunatai, Anantaporn
Human speech possesses characteristics in each of the word that can be recognized and learned by computers. In this research, It is being proposed the use of the Deep Learning Model to predict speech turn-on and turn-off various electrical appliances, by using the sound conversion method that has been through the process to get the value of sound waves and applied toward training process in different ways. As the sound has more than 1 syllable and having characteristics of similar words that might difficult to predict. This research is based on Convolutional Neural Network (CNN) for comparison with the use of Long Short-Term Memory (LSTM), which is part of the Recurrent Neural Network (RNN) and Thai language Speech Dataset turn-on and turn-off by the 7 types of electrical appliances, the process of reducing noise and silence of the front and back of the audio files by 14 classes in total. The experimental results signify that the proposed Long Short-Term Memory can achieve the best accuracy.
Some of the metrics are blocked by your
consent settings
Artificial magneto-dielectric metamaterial with microstrip antenna for wireless applications
(2019-07-01)
Pinsakul, Atcharaporn
;
Promwong, Sathaporn
This paper refers to a detailed analysis on the design and implementation of artificial magneto dielectric substrate and apply to use with microstrip antenna for wireless application. As the results are show the advantage of metal dielectric above conventional antenna. The key advantage of using the metamaterial with microstirp patch antenna is to size reduction of microstrip patch antennas when compare with original antenna. The metamaterial antenna proposed in this paper achieves 11.5% for downsizing in the area of the patch at the same frequency. The modified quadrilateral loop unit cell is used in design. The modified quadrilateral loop unit cell is used in design which the value of permittivity (ϵ<inf>eff</inf>) of 4.8 and relative permeability (μ<inf>eff</inf>) of 2.2. The result found that at the center frequency is 2.45 GHz, the artificial magneto dielectric metasubstrate antenna radiation has the efficiency of 89% and the gain of 6.9 dBi while the conversional antenna has the efficiency of 83.27% and the gain of 5.23 dBi. The performance of the antenna can be improved by controlling the dielectric constant and the magnetic permeability.
Some of the metrics are blocked by your
consent settings
Design of Repetitive Controller Using Optimization in Frequency Domain with Maximum Gain Constraints
(2018-09-18)
Prasitmeeboon, Pitcha
Spacecraft inherently have rotating parts such as the momentum wheels, reaction wheels or CMGs. Slight imbalances can produce vibrations that disturb fine pointing equipment. Repetitive control (RC) can be applied on an active vibration isolation to eliminate the effect of such periodic disturbances. An effective RC design method based on optimization in frequency domain was presented. In situations that a system needs large amount of correction, a compensator with high gains can generate a command that exceeds the capacity of the equipment. The error continues to grow and the system will become unstable. A method to reduce the magnitude of the compensator gains available in literature requires a weighting function that is difficult to design. This paper presents an alternative method to design an RC compensator that can limit the maximum magnitude of compensator gains. Simulation results show that the proposed method is more effective than the existing method.
Some of the metrics are blocked by your
consent settings
Human Identification Technique Based Behavioural Standing Still on Force Plates
(2018-09-11)
Sirisilpa, Pattraporn
;
Sinchai, Sakkarin
;
Wardkein, Paramote
;
Sangpisit, Wipa
This article proposes the human identification technique based behavioural standing still on force plates. Since the body sway that occurs during a quiet stand is caused by the movement of the center of mass and the center of pressure, thus the transfer of weight on the left and right feet of each person has a unique characteristic. This characteristic is then applied to identify individuals. In addition, the signals obtained from eyes open and eyes close condition are studied. The processing results are obtained by MATLAB program. The results of the proposed method show that the behavioural standing still signal in eyes open condition can be applied for human identification. However, the further study and improvement of the proposed principle is needed to acquire better in false rate of identification.
Some of the metrics are blocked by your
consent settings
Distance Detection Technique Using Enhancing Inverse Perspective Mapping
(2018-09-11)
Wongsaree, Preaw
;
Sinchai, Sakkarin
;
Wardkein, Paramote
;
Koseeyaporn, Jeerasuda
This paper presents a distance determination technique using an image from the single forward camera. Since too dark or too bright of the image and non-linear relation between height of the object and distance from the camera have effect on the performance of the detection process. Therefore, automatic brightness adjustment and inverse perspective mapping (IMP) is applied in the proposed scheme. In addition, region of interest (ROI) determination is used to decrease the processing time. The experimental results confirm that the proposed technique can detect distance of the object in front of the car where the error is 7.96%.
Some of the metrics are blocked by your
consent settings
Distance Measurement Technique Using Phase Difference of Two Reflected Ultrasonic Frequencies
(2018-09-11)
Chongchamsai, Munchuphon
;
Sinchai, Sakkarin
;
Wardkein, Paramote
;
Boonjun, Surapol
This paper proposes a simple method of distance measurement utilizing phase difference of two reflected ultrasonic frequencies. To obtain the phase difference, an alternating signal which contains two sinusoidal signals having different frequencies is fed to an ultrasonic transmitter. When the transmitted signal is reflected and received by the ultrasonic receiver, the received signal is first squared. Then, a time shift property of Fourier transform is applied to the squared signal to acquire the phase difference. The phase difference is converted to a time-of-flight (TOF) value and turned to the distance. To experiment the proposed technique, a DSP TMS620C6713 board is employed to generate the alternating signal as well as to receive the reflected signal, and to process the received signal. In the experiment, two sinusoidal signals having frequencies of 39.85 kHz and 40.6 kHz are used. The results show that the obtained distance by the presented approach has fairly low error only 0.15 centimeters compared to the actual distance.
Some of the metrics are blocked by your
consent settings
Frequency Estimation of Short Range Radar Using Adaptive FIR Notch Filter and SAGC for Velocity Detection
(2018-09-11)
Chinakaew, Pannipa
;
Sinchai, Sakkarin
;
Wardkein, Paramote
;
Koseeyaporn, Jeerasuda
This research proposes the velocity detection for short range radar using an adaptive FIR notch filter and a sinusoidal automatic gain control scheme (SAGC). A HB 100 module with amplifying gain of 20 times produces Doppler signal and this signal is fed to a laptop through a sound card with sampling frequency of 8000 Hertz. The signal is processed by the SAGC to amplify the amplitude of the signal to be unity. To detect the frequency of the signal, an algorithm of indirect frequency estimation based on adaptive FIR notch filter is performed. After that, the resulting frequency is converted to velocity. All procedures in this work are processed by MATLAB program. The results of the proposed method have better signal to noise ratio compared with existing method. In addition, the presented work consumes low computational time.
Some of the metrics are blocked by your
consent settings
Effect of Entropy in Icons and Background on Selection Accuracy
(2018-09-11)
Satcharoen, Kleddao
The purpose of this research is to study the effect of icon and background entropy on selection accuracy under differing cognitive conditions. A series of two experiments are reported here. Both experiments used a 4x2 design (very high/ high/ medium/ low entropy background x high/ low entropy icon), measuring selection accuracy under two different conditions (untimed/3 second timed). The sample included undergraduate students at a university in Thailand (n = 100). Results were analysed using chi-square. The analysis showed that under untimed conditions, there were no significant differences in selection accuracy between different background and icon entropy levels. However, under timed conditions, very high entropy background/high entropy icon selection was significantly lower than other conditions, while low entropy background/low entropy icon selection was higher. This implies that while under normal conditions, visual entropy in field or figure does not impede distinguishing different figures, under conditions of cognitive stress entropy could reduce ability to distinguish such images.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results