Proceeding of the Electrical Engineering Computer Science and Informatics
Not a member yet
649 research outputs found
Sort by
Obtaining Reference's Topic Congruity in Indonesian Publications using Machine Learning Approach
Get PDFThere are some criteria on how an article is categorized as a good article for publications. It could depend on some aspect like formatting and clarity, but mainly it depends on how the content of the article is constructed. The consistency of the topic that the article was written could show us how the authors construct the main idea in the article content. One indication that shows this consistency is congruity in the article’s topic and the topic of literature or reference cited in the document listed in the bibliography. This works attempting to automate the topic detection on the article’s references then obtain the congruity to the article title’s topic through metadata extraction and text classification. This is done by extracting metadata of an article file to obtain all possible reference title using GROBID than classify the topic using a supervised classification model. We found that some refinements in the whole approach should be considered in the next step of this work
Privacy Control In Social Networks By Trust Aware Link Prediction
Get PDFSocial networks are exceedingly common in today’s society. A social network site is an online platform where people build social relations with others and share information. For the last two decades, rapid growth in the number of users and applications with these social networking sites, make the security as the most challenging issue. In this virtual environment, some greedy people intentionally perform illegal activities by accessing others’ private information. This paper proposes a novel approach to detect the illegal access of a particular’s information by using trustaware link prediction. The facebook dataset is used for experiments and the results justify the robustness andtrustworthiness of the proposed model
River Water Pollution Monitoring using Multiple Sensor System of WSNs
Get PDFThe river is a natural phenomenon thatcommonly available in the tropical region because of rainintensity. Many peoples and community like to live along theriverside for a few decades ago. The river used by thecommunity for transportation and daily activities uses riverwater. In this research objective to design and develop a newsystem with multiple sensors system to monitor river waterpollution because most of the people use it. Wireless SensorNetworks (WSNs) used in this design and developmentbecause of advantages WSNs system, multiple sensor nodesinstalled for detection of water pollution such as watertemperature, pH, electrical conductivity (EC) and dissolvedoxygen (DO). The system designed to be able to monitorriver water pollution parameters and send the information tothe data center (backend system). Arduino microcontrollerused to process and filtering the data before sending to thebackend system, only valid and valuable information tocollect and keep in the database. Results show system be ableto detect polluted water with indicating parameters andshows in a graph. Based on analysis can be concluded thatpolluted water indicator mostly from residence waste andindustry. Furthermore, WSNs sensors will deploy in somearea then compare the results each other
Optimizing Design of Core-clad Width for Single Mode Fiber with Zero Dispersion Shift
Get PDFFiber optics have become a vital role in telecommunication technologies with many benefits, i.g. high speed transmission, non-electromagnetic interference and low energy consumption. An excellent single mode fiber (SMF) must provide a low attenuation and dispersion which occurs at same wavelength, i.e. 1550 nm. But, in silica based SMF, this property cannot be achieved in a bulk form. Meanwhile, the direct experiment is really not the best choice. Therefore, a simulation fiber design take a crucial role into account for obtaining zero dispersion shift. We design SMF geometry with zero dispersion by resizing the width of core and cladding. This design consists of inner and outer core-clad profile. We also provide several width boundaries for matching the lowest dispersion to the lowest attenuation in silica fiber-based. Moreover, the results shows that dispersion property of the design is suitable for long-haul optical communication systems
Detection of EEG Signal Post-Stroke Using FFT and Convolutional Neural Network
Get PDFStroke is a condition that occurs when the blood supply to the brain is disrupted or reduced. It may be caused by a blockage (ischemic stroke) or rupture of a blood vessel (hemorrhagic stroke) so that it can cause disability. Therefore patients need to undergo rehabilitation. One of the procedures of monitoring of the recovery of stroke patients using the National Institutes of Health Stroke Scale (NIHSS) method, but sometimes subjectively. Electroencephalogram (EEG) is an instrument that can measure electrical activity in the brain, including abnormalities caused by stroke. This study investigates EEG signal detection in post-stroke patients using Fast Fourier Transform (FFT) and 1D Convolutional Neural Network (1D CNN). Fast Fourier Transform (FFT) extraction can increase accuracy from 60% to 80.3% from the use of Adam's optimization model. Meanwhile, the AdaDelta model gave 20% accuracy without FFT. And its condition increased to 79.9% with FFT extraction. Therefore, Adam's stability has the advantage of remembering to use hyper-parameter. On the other hand, FFT is beneficial for directing information used for the use of 1D CNN, thus increasing accuracy. The results showed that using of Fast Fourier Transform (FFT) in identification could increase accuracy by 45-80% compared to identification using only 1D CNN. Meanwhile, the results of the study show that the relative weight correction model using Adaptive Moment Estimation (Adam) provided higher accuracy compared to the Adaptive learning rate (AdaDelta)
Performance improvement of MO surge arrester using high gradient arrester block against VFTOs
Get PDFIt is well known that the metal oxide surge arrester is inoperative against very fast transients overvoltages (VFTOs) because of its strong stray capacitive effect. This stray effect causes a time lag between the peak of residual voltage and peak of the current surge and so there is a delay its response. In order to reduce the stray effect, high gradient material is used for preparing metal oxide arrester blocks with different compositions. For simulation study, the required electrical parameters of high gradient arrester blocks are calculated with estimated height of arrester. This model is simulated using Electromagnetic transient program (EMTP) for different arrester ratings against switching, lightning, steep and very fast transients. The simulated value of residual voltages are compared with experimental values. From the observed results, it is perceived that the newly developed high gradient arrester decreases the delay and so the dynamic performance of the arrester is improved especially against very fast transients
Low-Frequency Oscillation Mitigation usin an Optimal Coordination of CES and PSS based on BA
Get PDFSmall signal stability represents the reliability of generator for transferring electrical energy to the consumers. The stress of the generator increases proportionally with the increasing number of load demand as well as the uncertainty characteristic of the load demand. This condition makes the small signal stability performance of power system become vulnerable. This problem can be handled using power system stabilizer (PSS) which is installed in the excitation system. However, PSS alone is not enough to deal with the uncertainty of load issue because PSS supplies only an additional signal without providing extra active power to the grid. Hence, utilizing capacitor energy storage (CES) may solve the load demand and uncertainty issues. This paper proposes a coordination between CES and PSS to mitigate oscillatory behavior of the power system. Moreover, bat algorithm is used as an optimization method for designing the coordinated controller between CES and PSS. In order to assess the proposed method, a multi-machine two-area power system is applied as the test system. Eigenvalue, damping ratio, and time domain simulations are performed to examine the significant results of the proposed method. From the simulation, it is found that the present proposal is able to mitigate the oscillatory behavior of the power system by increasing damping performance from 4.9% to 59.9%
Middleware for Network Interoperability in IoT
Get PDFOne solution for interoperability issue in IoT is a middleware which is competent on resolving the problems of syntactical, semantic, and network interoperability. In previous study, a middleware capable of addressing semantic and syntactical interoperability challenges has been developed, yet has not responded to network interoperability matter. In this paper we continue our previous research by adding BLE and 6LoWPAN features to the middleware's communication media, so it may communicate with various devices. Interoperability test results show that the middleware is capable of responding to network interoperability challenges and able to receive data from multiple nodes simultaneously
Determination of Router Location for Optimizing Computer Network Using Dominating Set Methods
Get PDFThe aims of this research is to develop a system to determine an optimal router location in a computer network. The router location is optimal if the computer network need the minimal number of router without losing the network connectivity. The methods used in order to optimize the router location are dominating set and Greedy algorithm. The data is the router location of computer network in University of Jember. The result showed that the optimal router needed in the networks is 4. These routers were found by using the determination system that implement Greedy algorithm to find the minimum number of dominating set
Investment Analysis of Smart Connected Motorbike in Machine to Machine Application in Indonesia
Get PDFAt present the cellular operator is faced with difficult conditions, where the customer growth rate is at its peak. In terms of income, there was also a decline in growth because legacy services (voice & SMS) tended to decline due to threats from OTT. Therefore, a strategy is needed to survive this condition, one of which is to look at the digital market. The digital service business development strategy is carried out to be able to increase the company's revenue through the implementation of Machine to Machine (M2M) services given the promising potential and support of the digital ecosystem. But in reality, the implementation of digital services has not been able to have a significant impact on the company's revenue growth so that a new strategy is needed with a service innovation. This study aims to model M2M digital services into the Smart Connected Motorbike (SCM) service, as well as analyze the service delivery business model in the PT. XYZ so that it is expected to become a new source of income for cellular operators of PT. XYZ. Furthermore, this research will do investment analysis in machine to machine application. Moreover, this research also indicated that this implementation is also feasible to increase the company's earnings