KMITL

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

Browse

Search Results

Now showing 1 - 10 of 12

Some of the metrics are blocked by your
consent settings
Performance Characterisation and Predictive Modelling of WireGuard VPN on Kubernetes: Efficient Resource Management Towards Auto-scaling
(2026-01-01)
Sridee, Pachara Threerapat
;
Khawne, Amnach
This study investigates WireGuard Virtual Private Network (WG VPN) performance on Kubernetes (K8s), analysing resource use and developing regression models. Metrics (Throughput, CPU, Memory, Jitter) were analysed across vCPU and physical core allocations. We found that over-provisioning vCPUs degraded performance due to increased overhead, with optimal throughput achieved, especially with two vCPUs and two physical cores. Among these optimal matched configurations, one demonstrated superior cost-effectiveness (high throughput/core, low CPU) and minimal jitter, which is vital for real-time applications. Pod-level CPU lacked traffic correlation due to the host kernel reliance on WGVPN, making Node-level Horizontal Scaling more suitable than Pod-level Vertical. Regression models for Node-level CPU (best: Linear Regression, with MAE decreasing by 16.74% compared to Multi-Layer Perceptron Model) and Traffic Demands (best: Instance-Based k-Nearest Neighbours, with MAE decreasing by 9.02% compared to Multi-Layer Perceptron Model) were developed. Findings offer insights for optimising WGVPN on K8s and future auto-scaling.
Some of the metrics are blocked by your
consent settings
Blockchain as a Notarization System for Military Data Sharing
(2025-01-01)
Chamnikul, Chatchawan
;
Khawne, Amnach
This paper presents a blockchain-based notarization system designed to secure and authenticate military data sharing. In the context of national defense, where the integrity and confidentiality of sensitive military data are paramount, traditional centralized systems often fail to meet the required security standards. Our proposed system leverages a permissioned blockchain platform, specifically Hyperledger Fabric, combined with smart contracts to create an immutable and transparent ledger. This ensures that once military data is recorded, it cannot be altered or tampered with, thereby enhancing data integrity and auditability. The system is further integrated with existing military information infrastructures through secure APIs and middleware, enabling seamless data exchange. Experimental evaluations conducted in a simulated environment demonstrate that the system achieves high transaction throughput, low latency, and robust error management. Although challenges remain in scalability and fine-grained access control, the results confirm that blockchain technology has the potential to significantly improve the security and transparency of military data sharing. Future work will focus on addressing these challenges and refining the system for real-world deployment in defense operations.
Some of the metrics are blocked by your
consent settings
Improving Water Salinity Forecasting in Bang Pakong River with Attention Mechanism
(2024-12-02)
Saksopit, Thongthai
;
Khawne, Amnach
Seawater intrusion in the Bang Pakong River estuary poses a significant threat to freshwater resources used for agriculture, municipal consumption, and industrial applications. Accurate prediction of salinity fluctuations is crucial for effective water management strategies. This study proposes an enhanced univariate salinity prediction method utilizing a Long Short-Term Memory (LSTM) model augmented with an Attention Mechanism. The Attention Mechanism empowers the LSTM to selectively focus on crucial information within extended historical salinity data sequences. The optimal input sequence length for the model is determined through a training process, aiming for the most accurate predictions. Here, the model forecasts salinity values 24 hours ahead and is evaluated against actual measurements. Performance metrics demonstrate that the Attention-LSTM model achieves the lowest error (MAE: 0.007834, MSE: 0.000094, RMSE: 0.009697, MAPE: 0.048736) and the highest accuracy (R<sup>2</sup>: 0.782927) at an input sequence length of 504 hours. These findings highlight the potential of the Attention-LSTM model for improved salinity prediction in the Bang Pakong River estuary, aiding water resource management strategies.
Some of the metrics are blocked by your
consent settings
Predictive quality assurance of a linear accelerator based on the machine performance check application using statistical process control and ARIMA forecast modeling
(2020-08-01)
Puyati, Wayo
;
Khawne, Amnach
;
Barnes, Michael
;
Zwan, Benjamin
;
Greer, Peter
Purpose: A predictive linac quality assurance system based on the output of the Machine Performance Check (MPC) application was developed using statistical process control and autoregressive integrated moving average forecast modeling. The aim of this study is to demonstrate the feasibility of predictive quality assurance based on MPC tests that allow proactive preventative maintenance procedures to be carried out to better ensure optimal linac performance and minimize downtime. Method and Materials: Daily MPC data were acquired for a total of 490 measurements. The initial 85% of data were used in prediction model learning with the autoregressive integrated moving average technique and in calculating upper and lower control limits for statistical process control analysis. The remaining 15% of data were used in testing the accuracy of the predictions of the proposed system. Two types of prediction were studied, namely, one-step-ahead values for predicting the next day's quality assurance results and six-step-ahead values for predicting up to a week ahead. Results that fall within the upper and lower control limits indicate a normal stage of machine performance, while the tolerance, determined from AAPM TG-142, is the clinically required performance. The gap between the control limits and the clinical tolerances (as the warning stage) provides a window of opportunity for rectifying linac performance issues before they become clinically significant. The accuracy of the predictive model was tested using the root-mean-square error, absolute error, and average accuracy rate for all MPC test parameters. Results: The accuracy of the predictive model is considered high (average root-mean-square error and absolute error for all parameters of less than 0.05). The average accuracy rate for indicating the normal/warning stages was higher than 85.00%. Conclusion: Predictive quality assurance with the MPC will allow preventative maintenance, which could lead to improved linac performance and a reduction in unscheduled linac downtime.
Some of the metrics are blocked by your
consent settings
The implementation of spectrum sensing and spectrum allocation on cognitive radio
(2017-03-29)
Rungsawang, Pitcha
;
Khawne, Amnach
Cognitive Radio is a system designed to use radio spectrum in the most efficient way possible - as demands for radio spectrum have been heavily increased nowadays. The system works by allowing unlicensed users to use radio channel on the licensed users' spectrum without interfering with the licensed users. The system needs spectrum sensing to verify the usage of each specific spectrum and in ever-shifting environment, Reinforcement Learning is also used for establishing spectrum allocation as well.
Some of the metrics are blocked by your
consent settings
Optimum watermark detection of ultrasonic echo medical images
(2015-03-01)
Khawne, Amnach
;
Attachoo, Boonwat
;
Hamamoto, Kazuhiko
To prevent unauthorized access to a person's medical images, it is widely acknowledged that the security features of confidentiality, availability and integrity should be in place. This paper proposes the watermarking of ultrasonic echo images together with optimal watermark detection, in which pseudorandom noise is added to the images for integrity. The optimum watermark detection is the integration of the generalized Gaussian distribution (ρ-GGD) and the Cauchy distribution. The results show that the proposed method gives good detection performance. The proposed method not merely achieves optimum detection using the Rao test but also leads to the highest detection probability with JPEG2000 compression. Compared with other detection methods, our proposed method exhibits better watermark detection performance even when the watermark-to-document ratio (WDR) is -50 dB.
Some of the metrics are blocked by your
consent settings
Image watermarking based on adaptive models of human visual perception
(2010-01-01)
Khawne, Amnach
;
Hamamoto, Kazuhiko
;
Chitsobhuk, Orachat
This paper proposes a digital image watermarking based on adaptive models of human visual perception. The algorithm exploits the local activities estimated from wavelet coefficients of each subband to adaptively control the luminance masking. The adaptive luminance is thus delicately combined with the contrast masking and edge detection and adopted as a visibility threshold. With the proposed combination of adaptive visual sensitivity parameters, the proposed perceptual model can be more appropriate to the different characteristics of various images. The weighting function is chosen such that the fidelity, imperceptibility and robustness could be preserved without making any perceptual difference to the image quality. © 2010 The Institute of Electrical Engineers of Japan.
Some of the metrics are blocked by your
consent settings
Ultrasonic echo image adaptive watermarking using the just-noticeable difference estimation
(2009-12-01)
Khawne, Amnach
;
Hamamoto, Kazuhiko
;
Chitsobhuk, Orachat
Most of the image watermarking methods, using the properties of the human visual system (HVS), have been proposed in literature. The component of the visual threshold is usually related to either the spatial contrast sensitivity function (CSF) or the visual masking. Especially on the contrast masking, most methods have not mention to the effect near to the edge region. Since the HVS is sensitive what happens on the edge area. This paper proposes ultrasound image watermarking using the visual threshold corresponding to the HVS in which the coefficients in a DCT-block have been classified based on the texture, edge, and plain area. This classification method enables not only useful for imperceptibility when the watermark is insert into an image but also achievable a robustness of watermark detection. A comparison of the proposed method with other methods has been carried out which shown that the proposed method robusts to blockwise memoryless manipulations, and also robust against noise addition.
Some of the metrics are blocked by your
consent settings
A study of using L1-norm with image watermarking on SVD domain
(2007-12-01)
Khawne, Amnach
;
Chitsobhuk, Orachat
;
Nakamiya, Toshiyuki
This paper presents a study of image watermarking using robust singular value decomposition in LI-norm subspace. The watermarked image attacked by noise is greatly degraded. This results in the effects of transparency and robustness of the watermarked image. Although the water-marking in SVD domain is sensitive to noise and outliers, incorporating L1-norm regression to the watermarking algorithm can help handling the missing data caused by noise and help increasing the robustness of the proposed algorithm. Experimental results show that the proposed algorithm can not only excellently reduce the bit error rates of the recovered watermark but also retain the transparency property of the watermarked image.
Some of the metrics are blocked by your
consent settings
A quick scene search with constructed mapped charts for TV sport programs
(2006-12-01)
Fuangrod, Todsaporn
;
Khawne, Amnach
This paper presents the fast scene search algorithm using fixed-image regions of the superimposed-text in a video. Our scheme focuses on the fixed-image region of the superimposed-text which is often an intuitive clue for an end-user for specifying a certain frame in a sport video. The video scene, consisting of superimposed-text detection using MFI algorithm, and Mapped Chart (MC) which constructed the index of a pattern matching, will be searched by the user. Our method found that the searching method with MC was faster (reduced the time consuming more than 50%) and more robust. ©2006 IEEE.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results