OVERVIEW

AI PxM Semiconductor

AI semiconductors are currently focusing on memory integration technology to solve data bottlenecks and memory access speed problems, and are also researching new memory integration technologies that can solve these problems because integrating memory and logic causes reliability, thermal problems, and mechanical defects.
Depending on the location and size of the operation unit in 3D coordinates, there is a complex statistical correlation between device-circuit input/output variables. As a result, factors that deteriorate memory reliability at low voltages, such as 1) thermal stress due to processor operation, 2) structural stress, and 3) external electromagnetic wave stress, change. The PxM structure design is an optimal 3D memory structure which minimizing the complexity arising from the optimization of each part of the process-device-circuit of memory semiconductors by applying AI simulation big data-based Design Technology Co-Optimization (DTCO) to increase design efficiency.

Research in semiconductor device design focuses on achieving low voltage operation, minimizing leakage current, ensuring high reliability, and pursuing device miniaturization. To address these objectives, our analysis employs advanced methodologies, including artificial intelligence (AI) and Monte Carlo simulations. AI techniques facilitate the optimization of design parameters, enhancing performance and efficiency.
Experimental analysis of semiconductor device and circuit : DCIV reliability analysis, charge pumping analysis, sensing margin analysis, write/read failure analysis.

Analysis of Radiation and 3D mechanical stress semiconductor device : AI-optimized element design considering the radiation effect and mechanical stress in three-dimensional structure.
Design semiconductors by considering reliability issues, which is an important problem in AI semiconductors.

Cutting-edge research spans low-power AI and autonomous vehicle circuits, alongside innovative ultra low-power sensor circuits and co-optimized designs for enhanced performance and reliability.
Improving circuit reliability through statistical analysis using Monte Carlo simulation method : Enhancing sense amplifier reliability design considering semiconductor device performance variability and improving sensing fail rate.

Fast switching performance for high frequency and wide bandwidth, device miniaturization for space efficiency and easy integration, low power design for energy efficiency, SOI platform design for reduced latch-up, higher packing density, low leakage current, interference elimination and so on.
Silicon based power device design for high breakdown, low on-resistance power semiconductor, low leakage current design for low power devices. High reliability and ruggedness which contains electrostatic discharge(ESD) robustness, electromagnetic compatibility(EMC), unclamped inductive switching(UIS), thermal management performance are importantly considered for electrical autonomous vehicle(EAV) application.

Our research revolves around High Voltage Direct Current(HVDC) circuit breakers, crucial for efficient high-power direct current transmission. While HVDC is promising for high-power demand, abrupt disconnection poses challenges, leading us to explore the innovative Dual-Pole DC Circuit Breaker(DCCB).
We are also researching the High step-up converter. Microorganisms are extracted using wastewater, and through this, fuel cells are developed and supplied as voltage sources. We study boosting the 0.4V DC voltage of this fuel cell to 100V and charging it to ESS (Energy Storage System) or outputting it to the system through an inverter.