187037 research outputs found
Sort by
Bandgap-Engineered Graphene Quantum Dot Photosensitizers for Tunable Light Spectrum-Activated NO2 Sensors
No full textVisible-light activation is highly desirable for gas sensors due to its energy-efficient operation and broad accessibility. Photocatalysis offers a promising strategy for visible-light activation; however, a limited understanding of the band engineering-mediated activation process restricts the rational design of photocatalysts for gas sensors. In this work, we systematically investigate the impact of band tuning in photocatalysts on the nitrogen dioxide (NO2) sensing performance of In2O3-based sensors, employing graphene quantum dots (GQDs) as photosensitizers. By controlling the sp2 carbon core size in GQDs, the bandgaps are tuned from 3.3 to 1.9 eV, enabling precise band engineering. It modulates the carrier transfer dynamics between GQDs and In2O3 layers, while surface functional groups of GQDs facilitate gas adsorption through their catalytic effects. By integrating sensitization effects, 7 nm GQDs optimize the photocarrier efficiency under visible light (blue light), leading to enhanced NO2 sensing performance in the GQD-decorated In2O3 system (R g/R a = 97.1 toward 1 ppm) with a fast response/recovery time (T 90/T 10 = 136/100 s). The bandgap tuning of GQDs highlights the critical role of band engineering in light-assisted gas sensing, enabling the photocatalyst-based sensor system construction for visible-light activation.
On the Conservation Property of Density Correction Methods in Smoothed Particle Hydrodynamics
No full text금속과 실리콘 접합 구조를 적용한 자기장 센서
No full textA magnetic sensor includes a silicon substrate, a cross-shaped metal pattern formed on the silicon substrate and directly contacting the silicon substrate, and an insulating layer covering the cross-shaped metal pattern
DEVICE OF CHANNEL PRUNING FOR EXPLAINABLE AI MODEL AND METHOD OF THEROF
No full text본 발명의 일 실시예에 따른 설명 가능한 인공지능 모델 채널 프루닝 장치 는 복수의 채널 및 상기 복수의 채널에 대응되는 필터 행렬(filter matrix)을 포함하는 설명 가능한 인공지능(XAI) 모델의 학습용 데이터 및 상기 필터 행렬을 획득하는 입력부; 및 상기 학습용 데이터에 대한 상기 필터 행렬의 손실 함수의 변화량(gradient) 크기에 기초하여, 채널 민감도를 연산하고, 상기 연산된 채널 민감도에 기초하여 프루닝될 채널을 결정하고, 상기 결정된 채널을 마스킹하여 프루닝하는 제어부를 포함하되, 상기 설명 가능한 인공지능 모델은 기 학습되지 않은 것일 수 있다
METHOD AND APPARATUS FOR PERFORMING COMMUNICATION BASED ON BACK SCATERRING IN WIRELESS COMMUNICATION SYSTEM
No full text본 실시예에 의한 통신 장치는: 제1 주파수에서 제2 주파수까지 변화하는 단 위 처프 신호(unit chirp signal)들이 연속되는 심문 신호(interrogation signal) 를 출력하는 심문기(interrogator) 및 상기 심문 신호를 제공받고 주파수 변조하여 택 신호를 형성하고 제공하는 후방산란 택(backscatter tag)을 포함하며, 상기 심 문기는, 상기 택 신호를 제공받고 상기 택 신호를 복조한다
Apparatus and method for recommending images for diagnosis and treatment of developmental disorders
No full text챗봇과의 대화를 통해 발달장애 진단 및 치료를 위한 영상을 제공하는 장치 및 방법이 개시된다. 일 실시예에 따른 영상 추천 장치는 데이터 입출력을 위한 인터페이스부; 및 인터페이스부와 연결된 제어부를 포함하며, 제어부는 인터페이스부를 통해 채팅을 위한 화면을 출력하며, 인터페이스부를 통해 수신한 사용자 입력 데이터를 기초로 발달장애와 관련된 분석을 수행하며, 분석된 결과를 기초로 하나 이상의 추천 영상을 검색하여 인터페이스부를 통해 출력할 수 있다
Bio-Electro-Hybrid System for Direct CO2 Conversion into L-Lactic Acid via Metabolically Engineered Escherichia coli
No full textDeepDDI2: Predicting Interactions between Paxlovidand Prescription Drugs Using Computational Methods
No full text
Darwin: A DRAM-Based Multi-Level Processing-in-Memory Architecture for Column-Oriented Database
No full textWe propose Darwin, a practical LRDIMM-based multi-level Processing-in-memory (PIM) architecture for data analytics, which exploits the internal bandwidth of DRAM using the bank-, bank group-, chip-, and rank-level parallelisms. Considering the properties of data analytics operators and DRAM's area constraints, Darwin maximizes the internal bandwidth by placing the PIM processing units, buffers, and control circuits across the hierarchy of DRAM. Darwin supports a novel PIM instruction architecture that concatenates instructions for multiple thread executions on bank group processing entities, addressing the command bottleneck by enabling separate control of up to 512 different in-memory processing units simultaneously. We build a cycle-accurate simulation framework to evaluate Darwin with various DRAM configurations, optimization schemes and workloads. Darwin achieves up to 14.7x speedup over the non-optimized version, leveraging many optimization schemes. Darwin architecture achieves 4.0 x -43.9x higher throughput and reduces energy consumption by 85.7% than the baseline CPU system (Intel Xeon Gold 6226 + 4 channels of DDR4-2933) for essential data analytics operators. Compared to the state-of-the-art PIM, Darwin achieves up to 7.5x and 7.1x in the basic query operators and TPC-H queries, respectively. Darwin in GDDR6 configuration requires only 5.6% area overhead, suggesting a promising PIM solution for the future main memory system.