- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Finite-Difference Time-Domain Modeling for Electromagnetic Wave Analysis of Human Voxel Model at Millimeter-Wave Frequencies
摘要： The finite-difference time-domain (FDTD) modeling of a human voxel model at millimeter-wave (mmWave) frequencies is presented. It is very important to develop the proper geometrical and electrical modeling of a human voxel model suitable for accurate electromagnetic (EM) analysis. Although there are many human phantom models available, their voxel resolution is too poor to use for the FDTD study of EM wave interaction with human tissues. In this paper, we develop a proper human voxel model suitable for mmWave FDTD analysis using the voxel resolution enhancement technique and the image smoothing technique. The former can improve the resolution of the human voxel model and the latter can alleviate staircasing boundaries of the human voxel model. Quadratic complex rational function is employed for the electrical modeling of human tissues in the frequency range of 6–100 GHz. Massage passing interface-based parallel processing is also applied to dramatically speed up FDTD calculations. Numerical examples are used to illustrate the validity of the mmWave FDTD simulator developed here for bio electromagnetics studies.
关键词： human tissue,electromagnetic wave,Doppler radar,dispersion model,bioelectromagnetics,Finite-difference time-domain (FDTD) method,parallel processing
更新于2025-09-23 15:23:52
-
[IEEE 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Chicago, IL, USA (2019.6.16-2019.6.21)] 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Implementation of Temporal Parallelization for Rapid Quasi-Static Time-Series (QSTS) Simulations
摘要： Quasi-static time-series (QSTS) analysis of distribution systems can provide critical information about the potential impacts of high penetrations of distributed and renewable resources, like solar photovoltaic systems. However, running high-resolution yearlong QSTS simulations of large distribution feeders can be prohibitively burdensome due to long computation times. Temporal parallelization of QSTS simulations is one possible solution to overcome this obstacle. QSTS simulations can be divided into multiple sections, e.g. into four equal parts of the year, and solved simultaneously with parallel computing. The challenge is that each time the simulation is divided, error is introduced. This paper presents various initialization methods for reducing the error associated with temporal parallelization of QSTS simulations and characterizes performance across multiple distribution circuits and several different computers with varying architectures.
关键词： PV grid integration,parallel processing,quasi-static time-series,distribution system modeling
更新于2025-09-23 15:19:57
-
Parallel information processing using a reservoir computing system based on mutually coupled semiconductor lasers
摘要： Via the nonlinear channel equalization and the Santa-Fe time series prediction, the parallel processing capability of a reservoir computing (RC) system based on two mutually coupled semiconductor lasers is demonstrated numerically. The results show that, for parallel processing the prediction tasks of two Santa-Fe time series with rates of 0.25 GSa/s, the minimum prediction errors are 3.8 × 10?5 and 4.4 × 10?5, respectively. For parallel processing two nonlinear channel equalization tasks, the minimum symbol error rates (SERs) are 3.3 × 10?4 for both tasks. For parallel processing a nonlinear channel equalization and a Santa-Fe time series prediction, the minimum SER is 6.7 × 10?4 for nonlinear channel equalization, and the minimum prediction error is 4.6 × 10?5 for Santa-Fe time series prediction.
关键词： Santa-Fe time series prediction,parallel processing,reservoir computing,nonlinear channel equalization,semiconductor lasers
更新于2025-09-19 17:13:59
-
Multi-Core DSP Based Parallel Architecture for FMCW SAR Real-Time Imaging
摘要： This paper presents an e?cient parallel processing architecture using multi-core Digital Signal Processor (DSP) to improve the capability of real-time imaging for Frequency Modulated Continuous Wave Synthetic Aperture Radar (FMCW SAR). With the application of the proposed processing architecture, the imaging algorithm is modularized, and each module is e?ciently realized by the proposed processing architecture. In each module, the data processing of di?erent cores is executed in parallel, also the data transmission and data processing of each core are synchronously carried out, so that the processing time for SAR imaging is reduced signi?cantly. Speci?cally, the time of corner turning operation, which is very time-consuming, is ignored under computationally intensive case. The proposed parallel architecture is applied to a compact Ku-band FMCW SAR prototype to achieve real-time imageries with 34 cm × 51 cm (range × azimuth) resolution.
关键词： real-time imaging,FMCW SAR,Parallel processing,multi-core DSP
更新于2025-09-04 15:30:14