J.H. Kim

ORCID: 0000-0001-9484-7541
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Particle Detector Development and Performance
  • Radiation Detection and Scintillator Technologies
  • Water Systems and Optimization
  • Financial Markets and Investment Strategies
  • Stock Market Forecasting Methods
  • Water resources management and optimization
  • Particle physics theoretical and experimental studies
  • Photonic and Optical Devices
  • Urban Stormwater Management Solutions
  • Semiconductor Lasers and Optical Devices
  • Video Coding and Compression Technologies
  • 3D Printing in Biomedical Research
  • Stochastic processes and financial applications
  • Diamond and Carbon-based Materials Research
  • Multimedia Communication and Technology
  • Market Dynamics and Volatility
  • Semiconductor materials and devices
  • Advanced Data Compression Techniques
  • Nanofabrication and Lithography Techniques
  • Anodic Oxide Films and Nanostructures
  • Geotechnical Engineering and Underground Structures
  • Infrastructure Maintenance and Monitoring
  • Muon and positron interactions and applications
  • Advanced Fiber Optic Sensors
  • Advanced Surface Polishing Techniques

Korea University
2004-2024

Hankuk University of Foreign Studies
2024

Bank of Korea
2024

Imperial College London
2024

Jeonbuk National University
2002

Abstract In recent decades, evolutionary optimisation algorithms have been used successfully for a wide variety of water resources engineering problems and their applications are still increasing. this research work, hybrid harmony search algorithm, ‘Non‐dominated Sorting Harmony Search’ algorithm is developed compared with two state‐of‐the‐art multi‐objective – the non‐dominated sorting genetic ( NSGA ) ‐II particle swarm MOPSO assigning optimal rehabilitation plans sewer pipe networks. The...

10.1111/jfr3.12143 article EN Journal of Flood Risk Management 2015-01-20

A low-loss, thermally stable TE-mode selective optical waveguide was fabricated using a photosensitive fluorinated polyimide. The polymer undergoes photocrosslinking under UV exposure, thus changing its refractive index. photocrosslinking-induced index change utilized to form channel waveguides. propagation losses of the polyimide waveguides were less than 0.3 and 0.5 dB/cm for TE polarization at wavelengths 1.3 1.55 μm, respectively. measured extinction ratio higher 29 28 dB film remains...

10.1109/lpt.2002.801088 article EN IEEE Photonics Technology Letters 2002-09-01

Abstract The importance of the effective distribution runoff in sewer network throughout a detention facility has recently been recognized. This study aims to determine location and capacity facilities urban network. optimal can be determined by numerous iterations an optimization model. In this paper, interfacing Visual Basic (VB) programming tool with SWMM (Storm Water Management Model) DLL for hydraulic hydrological computation completes routine. A multi-objective genetic algorithm (MOGA)...

10.1016/j.proeng.2014.02.115 article EN Procedia Engineering 2014-01-01

This study proposes an optimal scheduling model for rehabilitation based on the deterioration prediction of existing pipes by using survey method a water distribution system. The divides degree each pipe into 5 degrees Probabilistic Neural Networks (PNN). Furthermore, maximum residual service time is estimated calculated and diameter. integer programming (IP), shortest path, can calculate cost maintenance, rehabilitation, replacement. Consequently, proposed be utilized as quantitative management

10.2166/ws.2005.0020 article EN Water Science & Technology Water Supply 2005-09-01

Utilizing market forecasts is pivotal in optimizing portfolio selection strategies. We introduce DeepClair, a novel framework for selection. DeepClair leverages transformer-based time-series forecasting model to predict trends, facilitating more informed and adaptable decisions. To integrate the into deep reinforcement learning-driven framework, we introduced two-step strategy: first, pre-training on data, followed by fine-tuning architecture using this model. Additionally, investigated...

10.48550/arxiv.2407.13427 preprint EN arXiv (Cornell University) 2024-07-18

Utilizing market forecasts is pivotal in optimizing portfolio selection strategies.We introduce DeepClair, a novel framework for selection.DeepClair leverages transformer-based timeseries forecasting model to predict trends, facilitating more informed and adaptable decisions.To integrate the into deep reinforcement learning-driven framework, we introduced two-step strategy: first, pre-training time-series on data, followed by fine-tuning architecture using this model.Additionally,...

10.1145/3627673.3680008 article EN 2024-10-20

We have developed the BSAC audio decoder for T-DMB (terrestrial DMB) receiver. Here, we introduce and present an efficient implementation of which consists fast computation algorithm filterbank, effective memory management DSP, auto sampling frequency detection, simple lip sync method A/V synchronization, error recovery circuit against some contaminating data compact SDRAM hardware interface to receive parsed MPEG-2 TS (transport stream). Because this operates with low power consumption in...

10.1109/icce.2006.1598408 article EN 2006-03-21

We made various patterns on the PMMA surfaces and observed cellular responses them. Nanoimprinting using AAO (anodized aluminum oxide) template was used for topology control of polymeric culture surfaces. has numerous regularly-spaced nanopores. The pore diameter center-to-center distance can be easily controlled by altering anodizing etching conditions. One main advantages as a in nanoimprinting is that nanometer-scale structure templates larger than 25 cm/sup 2/ are easily, economically...

10.1109/mmb.2005.1548422 article EN 2005-12-10

10.1016/j.ultrasmedbio.2006.02.1080 article EN Ultrasound in Medicine & Biology 2006-05-01

Multi-asset portfolio selection is an asset allocation strategy involving a variety of assets. Adaptive investment strategies which consider the dynamic market characteristics individual assets and classes are vital for maximizing returns minimizing risks. We introduce HADAPS, novel computational method multi-asset utilizes Soft-Actor-Critic (SAC) framework enhanced with Hierarchical Policy Network. Contrary to previous approaches that have relied on heuristics constructing allocations,...

10.1109/access.2023.3285613 article EN cc-by-nc-nd IEEE Access 2023-01-01

This paper introduces the digital filter architecture that reduces power and chip-size. The implementation is largely divided into two methods. One software based on DSP, other hardware filter. disadvantage it's difficult to design a high sampling rate change types. In order solve problems, we propose new structure of programmable filter, which includes memory for controls types sampling-rates.

10.1109/sice.2002.1195573 article EN 2004-01-24
Coming Soon ...