University of Technology Malaysia
Universiti Teknologi Malaysia Institutional RepositoryNot a member yet
70456 research outputs found
Sort by
Issues and challenges of technology-enhanced learning during the Covid-19 era: A case study
No full textThe conventional teaching and learning (T&L) approach involving face-to-face (F2F) classes and activities had to be adapted to the new norm of on-line teaching and learning, when the world faced the Covid-19 pandemic in March 2020. At the University of Technology Malaysia (Universiti Technologi Malaysia) (UTM), Johor Bahru, Malaysia, on-line T&L began soon after the onset of the pandemic and has continued to the present day. In this article, the authors focus on the challenges experienced by UTM electrical engineering students and instructors, including network connectivity, the learning experience and teaching transferable skills in engineering. The motivation to conduct this research was due to the drastic changes from F2F to on-line T&L. It was found that the mean Internet connectivity rate was more than acceptable for smooth on-line T&L throughout the country. However, the large standard deviation values show that there were differences in the students’ Internet experience and accessibility. The results also showed that students had difficulty mastering the engineering skills when they were learned through an on-line teaching method
Understanding the working primate: An ethogram of jon, a Southern Pig-Tailed Macaque - Macaca Nemestrina
Get PDFA study of Southern Pig Tailed Macaque in human community was conducted in Kijal, Malaysia. The study aimed to document how this species responds to commands of performing one of the riskiest and most break-necking tasks, plucking coconut. An ethogram of Macaca Nemestina was constructed by characterising and defining the behavioral patterns of this species dealing with the task given. A series of observations amounting to 28 visits with a total of 52 hours (3120 minutes) was made. There were four phases of observation conducted starting from 15 August 2019 to 3 December 2019 between 0800 and 1700 hours. It was found that the monkey is a smart animal that managed to duly perform its duty based on few rudiment utterances of sound system/vocabulary and some other repetitious forms of non-verbal communication. The ethogram of the pig-tail macaque and its coconut plucking activity is presented here for the first time
Performance analysis of ISFET biosensor with different structures
Get PDFIon-sensitive field-effect transistor (ISFET) biosensor has gained popularity in clinical research field for biomolecules detection due to its high detection sensitivity, mass-production capability, and low manufacturing cost. Proteins are needed by all living cells for structural and functional purposes. However, some proteins such as Bovine Serum Albumin (BSA) can cause allergic reactions in human body. Besides, instability of Liposome may result in medication leaks that could harm cells. The first objective of this project is to study the effect of different proteins on the performance of different structures of ISFET biosensor. There are several limitations of ISFET biosensor such as lack of good solid-state electrodes, parasitic sensitivity to temperature and light and time dependent instability of sensor parameters. Enhancement on ISFET structures could be a good approach to improve the ISFET performance. The second objective of this research is to compare the performance of ISFET biosensors of different nanoelectronics structure in terms of settling time, sensitivity, and selectivity. In this project, an open-source software, nanohub BioSensorLab was used to simulate the settling time, sensitivity, and selectivity of the biosensors. The target proteins are Liposome and Bovine serum albumin (BSA) while the bioreceptor is collagen. Different structures of ISFET biosensor were simulated and analysed, which are planar ISFET, cylindrical nanowire, nanosphere and double-gate FET biosensors. In this research, the impact of proteins’ diffusion coefficient, structures of biosensor and analyte concentration towards the settling time was analysed. Next, the sensitivity relied on the structures of biosensor. Therefore, the analysis carried out independently for different structures of biosensor. Selectivity is determined by the size of receptor molecules and parasitic molecules, concentration of target molecules and parasitic molecules. From the simulation results, the settling time decreased when analyte concentration and protein diffusion coefficient increased. In addition, the settling time increased from planar and double-gate, cylindrical nanowire to nanosphere biosensor. To obtain high sensitivity in planar ISFET biosensor, the width and the electrolyte concentration are increased while the oxide thickness and the length are decreased. For cylindrical nanowire biosensor, increasing the radius, oxide thickness and reducing the buffer ion concentration can improve the sensitivity. To obtain high sensitivity in double-gate FET, the width, back oxide thickness, and silicon body thickness need to increase while the length and top oxide thickness need to decrease. The selectivity is the same regardless of the ISFET structures. High selectivity can be obtained by increasing the size of receptor and parasitic molecules, concentration of target molecules, and decreasing the concentration of parasitic molecules. Diffusion coefficient of proteins have no impact towards the sensitivity and selectivity of ISFET biosensors
Evaluation of graphene electrode in capturing the changes of muscle activity during flexion and extension
Get PDFThis study was performed to determine and evaluate the performance of the graphene-based electromyogram sensor as an alternative to silver/silver chloride (Ag/AgCl) electrode for data acquisition. The main advantage of using graphene electrode is that it is more comfortable for sensitive skin, reusable and convenient. An experiment to capture the changes of biceps brachii and triceps brachii during flexion and extension was conducted to test the performance of graphene electrode over Ag/AgCl electrode. The test measurement was carried out using a portable surface electromyography (sEMG). MATLAB software was used to process the acquired signals and analysed its signal-to-noise ratio (SNR) of both electrodes. In addition, SPSS software was used to determine the significant different of both electrodes and to measure the agreement level of both acquired data by using Bland Altman (BA) analysis plot. At 1kg load, graphene electrode’s SNR = 27.081 dB from biceps brachii, SNR = 23.709 dB from triceps brachii. Meanwhile, Ag/AgCl electrode get SNR = 24.932 dB from biceps brachii, SNR = 24.348 dB from triceps brachii. Besides that, it shows agreement of 100 % between both electrode’s performance when using BA analysis. This shows that both electrodes are statistically comparable in terms of SNR, and the graphene electrode are able to perceive 100 % of Ag/AgCl electrode to the standard reading of EMG. The solution presented in this study could be a key factor for continuous innovation to increase user comfort and save earth from electrode waste
Modelling and simulation of Heteromaterial Dual-Gate Dopingless TFET (HTDGDL-TFET) and its application as digital inverter
Get PDFTunnel Field -Effect Transistor (TFET) has been known as one of the promising devices which will be replacing Conventional Metal Oxide Semiconductor Field-Effect Transistor (MOSFET) as a future low-power and high-speed logic application. This is because as the size of MOSFET reduce decade by decade, to achieve better speed and lower power, and currently moving towards the nanometer regime, has leads to the limitation of the performance of MOSFET. These few bottlenecks such as increasing of leakage current, Short Channel Effects (SCEs) and complexity in device fabrication have been faced while scaling down the size of MOSFET. Therefore, TFET which work on principle of tunnelling phenomenon has been proposed as one of the devices to replace MOSFET which work on the principle of thermionic emission that limits the device’s sub-threshold swing to 60mV/decade. TFET has various of features such as immunity from most of the Short Channel Effects, lower leakage current, lower sub-threshold swing which is below 60mV/dec, lower threshold voltage and higher OFF current over ON current ratio. However, there are also some drawbacks for TFET such as complexity of fabrication process in doped TFET which cause various defects. These can be overcome by using dopingless technique. This technique helps in producing defects-less and more economical devices. Another drawback would be TFET exhibits lower ON state current. Heteromaterial TFET can be used to solve the low Ion issue. To have a better controllability of heteromaterial TFET channel, dual gate is proposed. Sub-threshold swing (SS) is one of the important parameters to determine a device performance. By lowering the SS, the device performance will be better in term of lower leakage current, better Ion/Ioff ratio and lesser energy. There are 3 objectives for this project: To model and simulate Heteromaterial Dual-gate Dopingless TFET (HTDGDL-TFET). To compare the performance of TFET between Ge, Si and GaAs as Source region material. To apply the HTDGDL-TFET as a Digital Inverter. This project will be simulated using Silvaco TCAD tool. Single-Gate and Double-Gate HTDL-TFET has been successfully modelled. 4 simulation test cases have been done for this project to select the best structure of proposed TFET. Several important parameters such as Vth, SS, Ion, Ioff and Ion/Ioff ratio are used to measure the performance of TFET. Among all of the 4 test cases, the best TFET structure is with Ge as source region material, source and drain region carrier concentration of 1×1019 cm−3and channelcarrier concentration of 1×1017 cm−3and dopingless. This is because the deviceshows Vth value of 0.97V, SS value of 15mV/dec, and Ion/Ioff ratio of 7×1011. Thepropagation delay for designed TFET inverter is 75 times shorter than the inverter from [21]and is 29 times shorter than the market inverter [SN74AUC1G14DBVR]. Somefuture works also have been suggested in this thesis
Tomato ripeness detection using deep learning
Get PDFTomatoes are fruits with high nutrition and high in fibre; packed with vitamin C, vitamin K1, vitamin B9 and minerals. The global tomato processing market has reached 43.4 million tons in 2021. It is important to determine maturity level of the crops before harvesting to optimize yield. However, manual inspection of ripe tomatoes required huge labour resources and is time consuming. The amount of labour force for fruit harvesting has increased over the years due to increasing demand. Recently, some studies have attempted to evaluate the feasibility of smart agriculture involving machine learning for harvest ripeness detection. However, these works typically used smaller data size, simple dataset with no background or leaves or explored limited machine learning model. Hence, this thesis aimed to identify tomato ripeness detection using two machine learning networks such as Mask RCNN and YOLOv5. Both models were compared based on minimum average precision. The results of these algorithms were benchmarked with previous works in terms of precision and recall. The dataset for this work consisted of 1000 high resolution images (3024 x 4032) with a total of 9063 tomatoes consisting of unripe, half ripe and ripe tomatoes with leafy background to emulate actual environment in a tomato field. The images were annotated with bounding box in VGG image annotator prior to training and testing with the Mask R-CNN, YOLOv5 networks. After that, these images were divided to training, validation and testing set with 80:10:10 ratio and trained using TensorFlow. Parameters such as epochs, step per epochs, learning rate, batch size were tuned to improve training accuracy and reduce training loss. Minimum average precision achieved for Mask R-CNN was 0.903 and YOLOv5 was 0.927. Precision and recall for Mask RCNN was 89.94% and 87.14% respectively. YOLOv5 achieved better precision and recall of 92.72% and 90.87% respectively, which were better compared to Mask RCNN
A case study of industrial MBR process for poultry slaughterhouse wastewater treatment
No full textThe wastewater discharged from the poultry slaughterhouse always contains high levels of chemical oxygen demand (COD) and biochemical oxygen demand (BOD) and thus, it requires proper treatment to minimize its negative impacts on the receiving water bodies. In this work, we presented a local case study of the full-scale implementation of membrane bioreactor (MBR) process with capacity of 144 m3/day to treat the poultry slaughterhouse wastewater. Over the 6-month monitoring period, our results showed that the permeate flow rate of the MBR process was relatively stable and only suffered from approximately 16% flux decline for the entire period with 8-h operation daily. Such flux deterioration is acceptable given the membrane was not subjected to any cleaning process. With respect to the separation efficiencies, the MBR process showed a very promising performance by meeting almost all of the parameters' limit of the National Water Quality Standards (Class IIB Limit), except for the dissolved oxygen (DO) that displayed slightly higher value than the maximum limit. A chemical cleaning process using sodium hydrochloride as agent was found to be effective to retrieve the permeate flow rate of the fouled membrane by 99%, indicating the deposited organic foulants were mainly reversible ones. The findings from this case study clearly demonstrated the potential of MBR process for treatment of poultry slaughterhouse wastewater and played an important role to minimize the negative impacts of discharged effluents on the environment
WHTE: Weighted Hoeffding Tree Ensemble for Network Attack Detection at Fog-IoMT
No full textThe fog-based attack detection systems can surpass cloud-based detection models due to their fast response and closeness to IoT devices. However, current fog-based detection systems are not lightweight to be compatible with ever-increasing IoMT big data and fog devices. To this end, a lightweight fog-based attack detection system is proposed in this study. Initially, a fog-based architecture is proposed for an IoMT system. Then the detection system is proposed which uses incremental ensemble learning, namely Weighted Hoeffding Tree Ensemble (WHTE), to detect multiple attacks in the network traffic of industrial IoMT system. The proposed model is compared to six incremental learning classifiers. Results of binary and multi-class classifications showed that the proposed system is lightweight enough to be used for the edge and fog devices in the IoMT system. The ensemble WHTE took trade-off between high accuracy and low complexity while maintained a high accuracy, low CPU time, and low memory usage
A conceptual framework for democratization of Big Data Analytics (BDA)
No full textOrganizations implement Big Data Analytics (BDA) to help the organization to improve the efficiency and effectiveness of the business. However, the successful of the BDA implementation is not achieved because of lack communication and collaboration between the BDA stakeholders. As we know stakeholders for example a data scientist played an important role in BDA implementation. Therefore, democratization is introduced to address these issues. This paper introduces the concept of democratization in big data analytics (BDA) by applying the Technology Organization Environment Model (TOE) in a conceptual framework that is used in the process of BDA implementation
Construction 4.0 intervention in quality assessment system in construction’s implementation
No full textThe quality of construction projects (buildings and infrastructure) is an important parameter in determining the maturity level of construction industry. The issues related to quality in construction projects is not relatively new and has been around for decades. To address this issue, Construction Industry Development Board (CIDB) Malaysia has introduced Quality Assessment System in Construction (QLASSIC) or also known as Construction Industry Standard 7 (CIS7) back in 2007 as a benchmark yardstick to rate the quality level of building projects specifically in the workmanship area. The association of Industry 4.0 concept with construction delivery process has resulted the Construction 4.0 concept. Through the adoption of Construction 4.0, the construction industry is expected to experience major changes in the entire project delivery process. These changes require further update on the existing CIS7 to suit the measurement methodology with the upcoming emerging technologies and new construction methods. The clarity of gaps between existing measurement parameters stated in CIS7 and the requirement of emerging technologies must be made clear and well-understood. This will be the foundation for CIDB Malaysia as sole construction leader in Malaysia to improvise CIS7 to a higher standard and ready for Industry 4.0 era. As there are limited study and research conducted in understanding the integration of Construction 4.0 in QLASSIC and limitation of CIS7, this study will contribute to the enrichment of academic studies documentation related to Construction 4.0 generally and QLASSIC in specific