Indonesian Journal of Electrical Engineering and Computer Science
Not a member yet
9109 research outputs found
Sort by
Marrying deep learning within blockchain technology for credit card fraud prevention
Get PDFOver the last decade or so, an excessive turnout on e-financial transactions by companies and customers results in a pinnacle growth of credit card fraudulent acts, leading them to lose frequently huge amounts of money. In their trial to find the key to this issue, specialists and experts have founded a bunch of fraud detection and prevention models relied on data mining, machine learning and deep learning. Yet, the outcomes were not effective nor optimal. Thereupon, to foster these prototypes’ function, Blockchain -a safe, decentralized and unchangeable database- was deployed to ban any sort of anomaly or data alteration after storing. For identifying malicious financial behaviours, our work managed to intermingle a pre-designed deep learning prototype with Blockchain. That is to say -for the sake of preventing fraudulence that concerns credit card- we applied the former prototype in Blockchain system. Still, Blockchain showed impotence in terms of using off-chain data, which embeds deep learning pattern, specifically through smart contract. Hence, we activated chainlink boosting our model to surmount this obstacle
Modified-LSTM and feed forward neural network enabled resource allocation for 6G wireless networks
Get PDFThe 6G wireless networks utilize terahertz (THz) frequency and intended to tremendously dynamic and diverse applications with deep learning enabled network, harvested significant attention and able to solve complex problems. Efficient resource allocation is a key requirement of next generation wireless networks. This research focuses on the resource allocation optimization challenge which includes storage, computing power, bandwidth and memory in the milieu of 6G wireless networks with device-to-device (D2D) communication enabled. The proposed model uses modified long short-term memory (mLSTM) and feed forward neural network to allocate resources to various tasks as per requirement such as information access, audio/video streaming, information access and productivity activity applications. The proposed work focuses on network parameters like channel noise, signal to noise ratio (SNR), distance from base station and includes D2D communication decisions to improve network performance. This research gives a novelty learning based solution for resource allocation for 6G wireless networks which contributes to the enhancement of next generation wireless communication networks. The lowest computing power utilized is 1%, Bandwidth utilized is 3% of total bandwidth and 2% storage
Performance analysis of wireless optical link with RIS aided and the implications of using Gaussian Q-function
Get PDFThis paper is concerned performance analysis of a wireless optical link with reconfigurable intelligent surfaces (RISs) aided and the implications of using the Gaussian Q-function. The Gaussian Q-function plays a critical role in the performance analysis of communication systems, particularly in scenarios involving noise and fading, including optical links and those aided by technologies like RISs. The analytical expression for performance is derived from the average of the symbol error rate (ASER) conditioned on the signal-to-noise ratio (SNR), incorporating the effects of noise and fading. Depending on the modulation scheme and fading model, the specific expressions for ASER can vary. The Gaussian Q-function significantly impacts the performance of systems; its mathematical properties and relationships with RISs and quadrature amplitude modulation (QAM) provide essential insights into system behavior under various conditions. The upper and lower bounds are commonly used to approximate the Gaussian Q-function, since the Gaussian Q-function doesn’t have a closed-form solution, bounds and approximations are often used
Improving farming by quickly detecting muskmelon plant diseases using advanced ensemble learning and capsule networks
Get PDFIn modern agriculture, ensuring plant health is essential for high crop yields and quality. Plant diseases pose risks to economies, communities, and the environment, making early and accurate diagnosis crucial. The internet of things (IoT) has revolutionized farming by enabling real-time crop monitoring and using drones and cameras for early disease detection. This technology helps farmers address challenges with precision and sustainability. This research propose an ensemble learning model incorporating multi-class capsule networks (MCCN) and other pre-trained model with majority voting system is implemented to predict plant diseases and pests early. The research aims to develop a robust MCCN-based ensemble prediction model for timely disease identification. To evaluate the performance of the ensemble model, various key metrics, including accuracy, and loss value, are assessed. Furthermore, a comparative analysis is conducted, benchmarking the MCCN model against other well-known pre-trained models such as residual network-101 (ResNet101), visual geometry group-19 (VGG19), and GoogleNet. This research signifies a substantial stride towards the realization of IoT-driven precision agriculture, where advanced technology and machine learning contribute to the early detection and mitigation of plant diseases, ultimately enhancing crop yield and environmental sustainability
Distributed formation control with obstacle and collision avoidance for humanoid robot
Get PDFFormation control has become a popular research topic in recent years. A common challenge in formation control is ensuring that robots can avoid obstacles and maintain a safe distance from one another to prevent collisions while forming a formation. In this research, a distributed formation control approach for a multi-robot system (MRS) with obstacle and collision avoidance is presented. The distributed formation control architecture is based on a consensus algorithm and consists of four layers: consensus tracking, consensus-based formation control, behavior, and physical robot layers. The system was implemented and evaluated through both simulations and experiments. Humanoid robots were used as the platform for these implementations. The result of the simulations and experiments show that the distributed formation control system successfully guided the robots into desired formation while also avoiding obstacles and preventing collisions with other robots
Region based lossless compression for digital images using entropy coding
Get PDFImage compression is a method for reducing video and image storage space. Moreover, enhancing the performance of the transmission and storage processes is important. The region based coding technique is important for compressing and sending medical images. In the medical field, lossless compression can help telemedicine applications achieve high efficiency. It affects image quality and takes a long time to encode. As a result, this study proposes region-based lossless compression for digital images using entropy coding. The best performance is achieved by segmenting these areas. In this case, an integer wavelet transform (IWT) is utilized after the ROI of the image was manually generated. The IWT compression method is helpful for reversibly reconstructing the original image to the required quality. For enhancing the quality of compression, entropy coding is utilized. By passing images of varying sizes and formats, various quantitative metrics can be determined. The simulation results demonstrate that the region based lossless compression technique utilizing range blocks and iterations resulted in reduced encoding time and improved quality
End-user software engineering approach: improve spreadsheets capabilities using Python-based user-defined functions
Get PDFEnd-user computing enables non-developers to handle data and applications, boosting collaboration and productivity. Spreadsheets are a key example of end-user programming environments that are extensively utilized in business for data analysis. However, the functionalities of Excel have limitations compared to specialized programming languages. This study aims to address this shortcoming by proposing a prototype that integrates Python's features into Excel via standalone desktop Python-based user-defined functions (UDFs). This method mitigates potential latency concerns linked to cloud-based solutions. This study employs Excel-DNA (dynamic network access) and IronPython; Excel-DNA facilitates the creation of custom Python functions that integrate smoothly with Excel's calculation engine, while IronPython allows these Python UDFs to run directly within Excel. Core components include C# and visual studio tools office (VSTO) add-ins, which enable communication between Python and Excel. This approach grants users the chance to execute Python UDFs for various tasks such as mathematical computations and predictions — all within the familiar Excel environment. The prototype showcases seamless integration, enabling users to invoke Python-based UDFs just like built in Excel functions. This study enhances the capabilities of spreadsheets by harnessing Python's strengths within Excel. Future work may focus on expanding the Python UDF library and examining user experiences with this innovative approach to data analysis
Hierarchical enhanced deep encoder-decoder for intrusion detection and classification in cloud IoT networks
Get PDFSecuring cloud-based internet of things (IoT) networks against intrusions and attacks is a significant challenge due to their complexity, scale, and the diverse nature of connected devices. IoT networks consist of billions of devices, computer servers, data transmission networks, and application computers, all communicating vast amounts of data that must adhere to various protocols. This study introduces a novel approach, termed hierarchical enhanced deep encoder-decoder with adaptive frequency decomposition (HED-EDFD), and is designed to address these challenges within cloud-based IoT environments. The HED-EDFD methodology integrates adaptive frequency decomposition, specifically adaptive frequency decomposition, with a deep encoder-decoder model. This integration allows for the extraction and utilization of frequency domain features from time-sequence IoT data. By decomposing data into multiresolution wavelet coefficients, the model captures both high-frequency transient changes and low-frequency trends, essential for detecting potential intrusions. The deep encoder-decoder model, enhanced with deep contextual attention mechanisms, processes these features to identify complex patterns indicative of malicious activities. The hierarchical structure of the approach includes a hierarchical wavelet-based attention mechanism, which enhances the accuracy and robustness of feature extraction and classification. To address the issue of imbalanced intrusion data, a cosine-based SoftMax classifier is employed, ensuring effective recognition of minority class samples
Enhancing privacy in document-oriented databases using searchable encryption and fully homomorphic encryption
Get PDFIn cloud-based not only SQL (NoSQL) databases, maintaining data privacy and the integrity are critically challenged by the risks of unauthorized external access and potential threats from malicious insiders. This paper presents a proxy-based solution that provides privacy-preserving by combining searchable encryption and brakerski-fan-vercauteren (BFV) fully homomorphic encryption (FHE) to facilitate secure search and aggregate query execution on encrypted data. Through extensive performance evaluations and security analyses, we show that our approach offers a robust solution for privacy-preserving data operations, with performance overhead introduced by the use of FHE. This solution gives an opportunity for a robust framework for secure data management and querying in NoSQL databases, with promising implications for practical deployment and future research. This work represents a significant advancement in the secure handling of data in NoSQL oriented databases, supplying a practical solution for privacy-conscious organizations
Blue light therapy device for wound healing
Get PDFCuts, diabetic ulcers, and pressure sores are examples of chronic skin wounds that pose a serious healthcare danger because of their delayed healing rates. This problem emphasizes the necessity of creating noninvasive, economical, and successful wound treatment plans. Conventional treatments, such as skin grafting, negative pressure wound therapy, and hyperbaric oxygen therapy, have demonstrated effectiveness; nevertheless, they are frequently costly, intrusive, and have possible side effects. On the other hand, blue light treatment has become a viable substitute due to its antimicrobial characteristics and capacity to encourage cellular restoration. However, there is a crucial gap in the development of a portable, noninvasive, and cost-effective photobiomodulation device for wound treatment and monitoring, despite its demonstrated potential in wound healing. This work aims to address this gap by creating a novel blue light therapy tool specifically suited for wound healing. The gadget allows for controlled blue light exposure and real-time temperature monitoring to minimize overheating. It has a portable arm housing with integrated blue light strips, a temperature sensor, and an integrated fan. An STM 32 microcontroller powers the system’s pulse width modulation (PWM) technology, which modifies light intensity and therapy duration in response to conditions unique to each wound. This novel strategy seeks to improve the effectiveness of wound healing, lower the likelihood of adverse effects, and offer patients and healthcare providers a workable alternative that is noninvasive, inexpensive, and easy to use