Journal of Electronics Technology Exploration

Implementation of Retrieval-Augmented Generation (RAG) and Large Language Models (LLM) for a Document and Tabular-Based Chatbot System

Imam Chalish Rafidhul Haque — 2025-07-10

The challenge of accessing information from disparate sources—unstructured documents and structured tabular data—hinders efficiency in enterprise information systems. This study addresses this challenge by presenting the design, implementation, and validation of a unified chatbot system powered by Retrieval-Augmented Generation (RAG) and Large Language Models (LLM). For unstructured documents, the system implements a RAG pipeline utilizing ChromaDB for vector indexing and OpenAI embeddings. Meanwhile, for structured data, it leverages a Text-to-SQL engine to translate natural language queries into SQL commands, with results visualized via QuickChart. The architecture is built upon a modular FastAPI backend with role-based access control and was rigorously validated through blackbox functional testing. Results demonstrate 100% functional success across all endpoints, confirming the architecture's reliability. This study confirms the viability of a unified RAG and Text-to-SQL architecture, offering a practical blueprint for creating more intelligent and integrated data interaction systems in enterprise environments.

Improving V2V Communication Reliability in Dynamic Vehicular Networks: A Software-Defined Radio-Based Approach

Mohammad Yanuar Hariyawan — 2025-07-03

Smart Transportation Systems (STS) leverage Vehicle-to-Vehicle (V2V) communication to enhance road safety, traffic efficiency, and urban mobility. However, ensuring reliable V2V communication remains challenging due to signal power instability, environmental interference, and scalability limitations. This study explores the optimization of V2V communication using Software Defined Radio (SDR) technology, which offers a cost-effective and adaptable approach for real-time signal processing. An SDR-based V2V communication system was developed using GNU Radio and HackRF One, with signal power calibration conducted through comparative measurements involving a Spectrum Analyzer across varying distances (3-15 meters) and environmental conditions. Performance evaluation focused on Bit Error Rate (BER) and Signal-to-Noise Ratio (SNR) under different vehicle speeds (20-40 km/h). Results indicate that increasing distance leads to signal degradation, with BER reaching 36.83% and SNR dropping to -3.17 dB, emphasizing the need for adaptive signal optimization techniques. While SDR-enabled calibration provided accuracy in signal measurements, environmental factors such as multipath interference and atmospheric attenuation significantly impacted communication reliability. Despite its flexibility, the system exhibited high BER and limited communication range, necessitating further enhancements through adaptive modulation schemes, machine learning-based power control, and hybrid 5G-DSRC integration. The study highlights SDR's potential for improving V2V communication while addressing key limitations in urban mobility networks. Future research should focus on enhancing scalability, security, and energy efficiency through advanced signal processing techniques. This study contributes to developing next-generation STS by providing empirical insights into SDR-based V2V communication optimization, supporting safer and more efficient transportation systems.

Programming the 8031 Minimum System in Proteus Simulator using the C: Issues and Solutions

putut son maria — 2025-07-10

An essential required course in electrical engineering, computer science, and informatics is the microprocessor. Students may consider using Proteus software in cases wherein microprocessor trainers are unavailable. Yet, the simulation of the 8031 microprocessor-based minimum system circuit that Proteus executes fails to operate correctly, despite the fact that the source code and circuit wiring comply to programming and circuit theory standards. This is in contrast to other microcontroller-based minimum system circuits that it can be simulated successfully and as intended. This research aims to get hints in programming the 8031 minimum system circuit simulated using Proteus. The problem was investigated and analyzed by observing the parameters that become the properties of each element in the circuit, especially the RAM, then comparing them with the specifications of the microprocessor. The experimental results showed that some adjustments on the program code were necessary either written using assembly language or C program code.

Performance Analysis of Long Short-term Memory (LSTM) Model for Remaining Useful Life Prediction on Turbofan Engine

Themy Sabri Syuhada — 2025-07-10

Accurate Remaining Useful Life (RUL) prediction is critical for the predictive maintenance and operational safety of aircraft turbofan engines. This research develops and evaluates a stacked Long Short-Term Memory (LSTM) network for RUL prediction using the NASA C-MAPSS FD001 dataset as a fundamental case study. A systematic data preprocessing pipeline was employed, including sensor selection, RUL value clipping at 130 cycles, and feature normalization to prepare the data for modeling. The LSTM model was trained with regularization techniques and an EarlyStopping callback to ensure robustness and prevent overfitting. Evaluation results on the unseen test data show the final model achieved a solid and competitive performance with a Root Mean Squared Error (RMSE) of 15.22 and a PHM08 Score of 311.20. These results demonstrate that a well-configured LSTM architecture provides a reliable baseline for engine prognostic tasks, exhibiting strong generalization capabilities on new data.

Ev Battery Controller Tuning For Efficient Thermal Management Based On Grasshopper Algorithm And Particle Swarm Optimization Algorithm

Allif Nazmie — 2025-10-18

Electric Vehicles (EVs) offer low emissions and reduced fossil fuel dependence but require efficient battery thermal management to ensure performance and safety. This research aims for tuning proportional-derivative(PD), proportional-integral(PI) and proportional-integral-derivative (PID) controller for Electrical Vehicle (EV) Thermal Management System using Particle Swarm Optimization (PSO) and Grasshopper Optimization Algorithm method (GOA) method to optimize the compressor power consumption to contribute to the development of better EV battery thermal management systems. By minimizing and maximizing the factors involved in the challenges, optimization is the process of identifying the best way to make something as useful and effective as feasible. Simulation results show that GOA outperforms PSO for all controllers. Objective function values for GOA are lower, 1.6783 (PD), 0.8517 (PI), and 0.8114 (PID), compared to PSO, 1.7578, 0.8665, and 0.8254, respectively. Improvement percentages of GOA over PSO are 4.73% (PD), 1.70% (PI), and 1.65% (PID). The PID controller achieved the best performance overall, showing 51.65% improvement over PD and 4.91% over PI. The findings confirm that GOA is more effective than PSO in optimizing controller performance, and that PID is the most suitable for stable and efficient EV battery thermal management.