Results 51 to 60 of about 62,963 (220)

Feature Disentangling and Combination Implemented by Spin–Orbit Torque Magnetic Tunnel Junctions

Advanced Intelligent Systems, EarlyView.
Spin–orbit torque magnetic tunnel junctions (SOT‐MTJs) enable efficient feature disentangling and integration in image data. A proposed algorithm leverages SOT‐MTJs as true random number generators to disentangle and recombine features in real time, with experimental validation on emoji and facial datasets.
Xiaohan Li, Yingqian Xu, Caihua Wan, Ran Zhang, Meike Hindenberg, Raik Hoffmann, Shiqiang Liu, Tianyi Zhang, Shilong Xiong, Dehao Kong, Jihao Xia, Fangshuo Gao, Zhuyang Nie, Thomas Kämpfe, Guoqiang Yu, Xiufeng Han   +15 more
wiley   +1 more source

Fast Bitmap Fit: A CPU Cache Line friendly memory allocator for single\n object allocations [PDF]

, 2021
Dhruv Mátáni, Gaurav Menghani
openalex   +2 more sources

An ultra low-power hardware accelerator for automatic speech recognition [PDF]

, 2016
Automatic Speech Recognition (ASR) is becoming increasingly ubiquitous, especially in the mobile segment. Fast and accurate ASR comes at a high energy cost which is not affordable for the tiny power budget of mobile devices.
Arnau Montañés, José María, González Colás, Antonio María, Segura Salvador, Albert, Yazdani Aminabadi, Reza   +3 more
core   +1 more source

Ultralow‐Power Real‐Time On‐Chip Thermal Prediction via Finite Element Method–Machine Learning Codesign and Field‐Programmable Gate Array Deployment

Advanced Intelligent Systems, EarlyView.
A lightweight machine learning (ML)‐based thermal prediction framework is demonstrated and implemented on a field‐programmable gate array (FPGA). Using measured temperature data from a real chiplet, the approach enables real‐time, die‐level heat‐map inference with low power consumption, validating practical on‐chip thermal monitoring for advanced ...
Jun Ho Lee, Jae Gyu Kim, Sung Jin Kim, Jae Yong Song, Byoung Don Kong   +4 more
wiley   +1 more source

ControlIt: A Universal Framework for Translational, Adaptive, and Online Brain–Computer Interfaces

Advanced Intelligent Systems, EarlyView.
Brain–computer interfaces (BCIs) lack a unified platform that works across signals and algorithms. ControlIt, an open‐source, modular ROS2‐based BCI framework supporting electroencephalography (EEG), electrocorticography (ECoG), and spike‐based decoding across both classification and regression tasks is presented.
Wanlin Yang, Chenyang Li, Xinxiu Xu, Xindong Wang, Yongcheng Guo, Xiaoke Chai, Tianqin Cao, Jiuxiang Song, Nan Wang, Xueming Zhang, Yi Yang, Cong Zheng, He Cui   +12 more
wiley   +1 more source

Implementation of RSA Signatures on GPU and CPU Architectures

IEEE Access, 2020
This paper reports a constant-time CPU and GPU software implementation of the RSA exponentiation by using algorithms that offer a first-line defense against timing and cache attacks.
Eduardo Ochoa-Jimenez, Luis Rivera-Zamarripa, Nareli Cruz-Cortes, Francisco Rodriguez-Henriquez   +3 more
doaj   +1 more source

Accelerating CPU-Based Sparse General Matrix Multiplication With Binary Row Merging

IEEE Access, 2022
Sparse general matrix multiplication (SpGEMM) is a fundamental building block for many real-world applications. Since SpGEMM is a well-known memory-bounded application with vast and irregular memory accesses, considering the memory access efficiency is ...
Zhaoyang Du, Yijin Guan, Tianchan Guan, Dimin Niu, Hongzhong Zheng, Yuan Xie   +5 more
doaj   +1 more source

A Survey of Methods For Analyzing and Improving GPU Energy Efficiency

, 2014
Recent years have witnessed a phenomenal growth in the computational capabilities and applications of GPUs. However, this trend has also led to dramatic increase in their power consumption.
Mittal, Sparsh, Vetter, Jeffrey S.
core   +1 more source

Performance Assessment of the Immersed Boundary Method in the Framework of the Moment Representation Lattice Boltzmann Method

International Journal for Numerical Methods in Fluids, EarlyView.
The immersed boundary method (IBM) was coupled with the moment representation lattice Boltzmann method (MR‐LBM), reducing bandwidth requirements compared to population‐based LBM formulations. A systematic assessment of IBM parameters was conducted to quantify their effect on computational performance.
Marco A. Ferrari, Ricardo de Souza, Admilson T. Franco   +2 more
wiley   +1 more source

Design and Implementation of Multi-channel Data Acquisition Controller Based on FPGA [PDF]

Jisuanji gongcheng, 2017
In order to improve the real-time performance of Data Acquisition Controller(DAC) based on general PC,a multi-channel real-time data acquisition controller is designed.By using the Field Programmable Gate Array(FPGA) hardware technology,the data ...
ZHAO Tao,GUO Meng,GU Yaliu,ZHANG Yang
doaj   +1 more source