Journal of information and
communication convergence engineering
(J. Inf. Commun. Converg. Eng.)

Vol.22 No.4, December 31, 2024 Article TypeRegular paper

• Regular paper December 31, 2024

  0 49 11

  MMSE-Based Power Control for Amplify-and-Forward Relay Systems in Time-Varying Channels

  Han-Gyeol Lee , Duckdong Hwang , and Jingon Joung

  Journal of information and communication convergence engineering 2024; 22(4): 267-272 https://doi.org/10.56977/jicce.2024.22.4.267

  Abstract

  Full Text Cite Share

  Abstract : This study proposes a novel power control method for amplify-and-forward (AF) relay systems operating in time-varying channels. Transmit power control between the source and relay nodes significantly enhances the performance of the AF relay system, and the improvements are proportional to the number of antennas. However, these enhancements are restricted by the presence of time-varying channels, and this limitation increases in severity with increasing number of antennas. To address the challenges posed by these channel variations, the proposed method adaptively optimizes the power-scaling factors to minimize the mean-squared errors under a power-inequality constraint. Numerical results demonstrate that the proposed method effectively mitigates the bit-error-rate performance degradation caused by channel variations, maintaining robust performance even in systems with a large number of antennas. This approach offers a promising solution for improving the reliability and efficiency of AF relay systems under dynamic channel conditions.

• Regular paper December 31, 2024

  0 41 6

  Mid-span Spectral Inversion System with Random Triangular Dispersion Maps

  Jae-Phil Chung and Seong-Real Lee , Member, KIICE

  Journal of information and communication convergence engineering 2024; 22(4): 273-279 https://doi.org/10.56977/jicce.2024.22.4.273

  Abstract

  Full Text Cite Share

  Abstract : In dispersion-managed links with mid-span spectral inversion (MSSI) systems, the symmetry of the cumulative dispersion profile is important for compensating the distorted wavelength division multiplexing (WDM) signals. The triangular dispersion map has antipodal symmetry with respect to the midway optical phase conjugator (OPC). A triangular dispersion map is proposed in this paper that has a singularity in which the lengths of the optical fibers that contribute to forming the cumulative dispersion profile are determined randomly. We analyzed three types for randomly determining and deploying the length of optical fibers to form the triangular dispersion map. We confirmed that triangular dispersion maps combined with MSSI systems are more advantageous for distorted WDM channel compensation than traditional uniform dispersion maps are. In particular, the dispersion map created via the “random-inverse” scheme, which randomly arges the optical fiber lengths of each span before the midway OPC while reversing the arrangement of the optical fiber lengths in the sections after the midway OPC, results in the best compensation.

• Regular paper December 31, 2024

  0 54 9

  Automatic Classification of Scientific and Technical Papers Using Large Language Models and Retrieval-Augmented Generation

  Jaehan Jeong and Dongsup Jin , Member, KIICE

  Journal of information and communication convergence engineering 2024; 22(4): 280-287 https://doi.org/10.56977/jicce.2024.22.4.280

  Abstract

  Full Text Cite Share

  Abstract : This study proposes artificial intelligence (AI) technology for the automatic classification of Korean scientific and technical papers, aiming to achieve high accuracy even with a small amount of labeled data. Unlike existing BERT-based Korean document classification models that perform supervised learning based on a large amount of accurately labeled data, this study proposes a structure that utilize large language models (LLMs) and retrieval-augmented generation (RAG) technology. The proposed method experimentally demonstrates that it can achieve higher accuracy than existing technologies across all cases using various amounts of labeled data. Furthermore, a qualitative comparison between manually-generated labels, and recognized as correct answers and those produced by LLM responses confirmed that the LLM responses were more accurate. The findings of this study, while limited to Korean scientific documents, provide evidence that a system utilizing LLM and RAG for document classification can easily be extended to other domains with diverse document datasets, owing to its effectiveness even with limited labels.

• Regular paper December 31, 2024

  0 89 17

  Enhancing Transformer-based Cooking Recipe Generation Models from Text Ingredients

  Khang Nhut Lam , My-Khanh Thi Nguyen , Huu Trong Nguyen , Vi Trieu Huynh , Van Lam Le , and Jugal Kalita

  Journal of information and communication convergence engineering 2024; 22(4): 288-295 https://doi.org/10.56977/jicce.2024.22.4.288

  Abstract

  Full Text Cite Share

  Abstract : Recipe generation is an important task in both research and real life. In this study, we explore several pretrained language models that generate recipes from a list of text-based ingredients. Our recipe-generation models use a standard self-attention mechanism in Transformer and integrate a re-attention mechanism in Vision Transformer. The models were trained using a common paradigm based on cross-entropy loss and the BRIO paradigm combining contrastive and cross-entropy losses to achieve the best performance faster and eliminate exposure bias. Specifically, we utilize a generation model to produce N recipe candidates from ingredients. These initial candidates are used to train a BRIO-based recipe-generation model to produce N new candidates, which are used for iteratively fine-tuning the model to enhance the recipe quality. We experimentally evaluated our models using the RecipeNLG and CookingVN-recipe datasets in English and Vietnamese, respectively. Our best model, which leverages BART with re-attention and is trained using BRIO, outperforms the existing models.

• Regular paper December 31, 2024

  0 44 5

  Blockchain and IPFS-based IoT Massive Data-Management Model

  Ting Chain , Am-Suk Oh , and Seung-Soo Shin , Member, KIICE

  Journal of information and communication convergence engineering 2024; 22(4): 296-302 https://doi.org/10.56977/jicce.2024.22.4.296

  Abstract

  Full Text Cite Share

  Abstract : The exponential growth of the Internet of Things has resulted in a substantial surge in the volume of data produced by numerous sensors and microcontroller devices. The objective of this study is to create a comprehensive data-management framework using blockchain and InterPlanetary File System with the aim of enhancing the security, dependability, and decentralized processing of large-scale data storage. This study utilizes a hyperledger blockchain and an interstellar file system to transmit data via ZigBee and Wi-Fi networks. In addition, edge devices are employed for pre-processing to guarantee the security of data storage and exchange. The performance of the model in terms of data management is verified and analyzed using simulations. The results demonstrate a significant enhancement in both the data-storage access speed and overall system efficiency owing to the implementation of the model. This study aims to offer efficient methods for managing large amounts of data in the Internet of Things context.

• Regular paper December 31, 2024

  0 39 5

  Simple 1D CNN Model for Accurate Classification of Gait Patterns Using GRF Data

  Ansary Shafew , Dongwan Kim , and Daehee Kim

  Journal of information and communication convergence engineering 2024; 22(4): 303-309 https://doi.org/10.56977/jicce.2024.22.4.303

  Abstract

  Full Text Cite Share

  Abstract : Gait analysis plays a pivotal role in clinical diagnostics and aids in the detection and evaluation of various disorders and disabilities. Traditional methods often rely on intricate video systems or pressure mats to assess gait. Previous studies have demonstrated the potential of artificial intelligence (AI) in gait analysis using techniques, such as convolutional neural networks (CNN) and long short-term memory (LSTM) networks. However, these methods often encounter challenges related to high dimensionality, temporal dependencies, and variability in gait patterns, making accurate and efficient classification difficult. To address these challenges, this study introduces a simple one-dimensional (1D) CNN model designed to analyze ground reaction force (GRF) patterns and classify individuals as healthy or suffering from gait disorders. The model achieved a remarkable classification accuracy of 98.65% in distinguishing healthy individuals from those with gait disorders, demonstrating significant improvements over the existing models. This performance is bolstered by the attention mechanism and standardization techniques that enhance robustness and accuracy.

• Regular paper December 31, 2024

  0 47 10

  Utilization of XGBoost for Behavior Analysis of Lottery Purchasers

  Esther Kim , Yunjun Park , Gwanghyun Jo , and Seong-Yoon Shin

  Journal of information and communication convergence engineering 2024; 22(4): 310-315 https://doi.org/10.56977/jicce.2024.22.4.310

  Abstract

  Full Text Cite Share

  Abstract : In this study, we conducted a data-driven analysis of lottery purchase behavior by using the XGBoost algorithm to predict future lottery purchase amounts based on purchase patterns of the previous four weeks. We began by judiciously defining key features including the weekly average purchase amount and variance in purchase amount. Subsequently, we evaluated the proposed method’s performance, finding the predicted future purchase amounts to match the actual purchase amounts. A key strength of this study was the interpretability of feature variables. Through the feature importance score from XGBoost, we found that features that capture impulsive patterns in purchases (e.g., variability in purchase amount) are strongly correlated with future spending, which agrees with conventional behavior analysis. Our study can be extended to the development of early warning systems designed to identify at-risk and potentially addicted purchasers on online lottery platforms.

• Regular paper December 31, 2024

  0 40 8

  Mode Decomposition and Long-Term Solution Prediction for the Vlasov-Poisson Equation

  Donghyun Kim

  Journal of information and communication convergence engineering 2024; 22(4): 316-321 https://doi.org/10.56977/jicce.2024.22.4.316

  Abstract

  Full Text Cite Share

  Abstract : The Vlasov-Poisson (VP) equation plays an important role in plasma physics. Most numerical methods for the VP equation are based on the finite difference method (FDM) or finite element method (FEM), where the computational costs are high. However, this study focuses on the efficient reconstruction of solutions to the VP equation. We begin by generating short-term solutions to the VP equation using an FDM-type algorithm. Among various versions of FDM schemes, we employ backward semi-Lagrangian-based methods with weighted, essentially non-oscillatory schemes for interpolation. Subsequently, a stable dataset without spurious oscillations is obtained. The spatiotemporal patterns within these snapshot solutions are then analyzed via dynamic mode decomposition (DMD). By projecting solution spaces onto the DMD modes, we efficiently extend the solution to unobserved future time steps. Experimental results indicate that the time cost for the DMD prediction is within one second, showing the efficiency of the proposed algorithm.

• Regular paper December 31, 2024

  0 31 6

  The Impact of Backchannel in Medical-Questionnaire Robot on User Perception

  Da-Young Kim , Min-Gyu Kim , and Sang-Seok Yun , Member, KICCE

  Journal of information and communication convergence engineering 2024; 22(4): 322-329 https://doi.org/10.56977/jicce.2024.22.4.322

  Abstract

  Full Text Cite Share

  Abstract : Robot backchanneling is implemented using verbal expressions, such as reiterating after remembering the response of the user or chiming in with the user. We performed experiments to investigate how the backchannel behavior of the robot affects the user's perception of the robot's empathy and attentiveness toward their reported pain and sleep concerns. The results indicate that users prefer the robot that provides backchannel and perceive it as more intelligent. Where the robot backchanneled, the number of conversational turns between the user and the robot decreased compared to situations where the robot did not backchannel. The refusal rate of the robot repeating the same question to the user decreased. When a robot is conducting a medical questionnaire and offers backchannels to the user, the overall efficiency of the conversation improves. Users also perceive the robot as understanding and empathetic towards their pain and discomfort, leading to a positive evaluation of the interaction.

• Regular paper December 31, 2024

  0 61 5

  Development of a HUD-based Integrated Navigation Information System User Interface for Effective Autonomous Ship Navigation

  Younggeun Choi

  Journal of information and communication convergence engineering 2024; 22(4): 330-335 https://doi.org/10.56977/jicce.2024.22.4.330

  Abstract

  Full Text Cite Share

  Abstract : Various AI-based digital navigation technologies, such as autonomous navigation, are being developed in the maritime shipping industry. It is necessary to effectively integrate the information provided individually by various navigation information systems installed on a ship and efficiently deliver it to navigators to realize the first stage of autonomous navigation, known as an onboard decision support ship. In this study, we developed a user interface for an HUD-based integrated navigation information system that supports navigators in performing navigation tasks safely and efficiently. Through a literature review and expert interviews, the detailed procedures of navigation tasks and pain points from the perspective of practitioners were investigated, and three types of user interfaces (lookout-focused, information-focused, and nighttime-supportive) that can be selectively used depending on the context were proposed. The user interface proposed in this study is expected to be useful in the development of navigation information systems for autonomous ships.

• Regular paper December 31, 2024

  0 37 12

  Computer-Vision-Based Mobile Application for Translating Sundanese Scripts to Modern Indonesian Language With Gamification Strategies

  Wanda Gusdya Purnama , Handoko Supeno , Anggoro Ari Nurcahyo , Ayi Purbasari , Aria Bisma Wahyutama , and Mintae Hwang

  Journal of information and communication convergence engineering 2024; 22(4): 336-343 https://doi.org/10.56977/jicce.2024.22.4.336

  Abstract

  Full Text Cite Share

  Abstract : In the digital era, teaching endangered local languages and scripts to children has become challenging owing to the scarcity of learning media and materials. The present study addresses this problem through the development of a mobile application that classifies and automatically converts Sundanese scripts into Latin using computer vision algorithms. The proposed method represents an innovative solution for capturing children's interest using gamification strategies. We discuss the development, implementation, and evaluation of YOLOv8, a deep learning technology for computer vision in mobile applications. A pilot study conducted on children aged 7-12 years revealed significant improvements in their interest and knowledge of Sundanese scripts, as the children were able to memorize, identify, and write 5-8 words in Sundanese characters out of 10 randomly selected words. Furthermore, the model achieved 80% accuracy for almost all Sundanese-scripted words, indicating satisfactory results. This study combines computer vision with gamification to facilitate the learning of Sundanese scripts, thereby paving the way for future innovation.