Files in this item

FilesDescriptionFormat

application/pdf

application/pdfECE499-Sp2019-he-Xinwei .pdf (2MB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Optimization MEAT-COMMS for higher throughput
Author(s):He, Xinwei
Contributor(s):Singer, Andrew
Subject(s):MEAT-COMMS
code optimization
OpenMP
CUDA
Abstract:By using signal processing and acoustic communication methods, Professor Singer’s group has designed and implemented a novel system to communicate with biomedical implanted devices. The system, named MEAT-COMMS (Mbps Experimental Acoustic Through-Tissue Communication), sends communication signals through biological tissues at very high data rates using ultrasound signals. Our current goal is to build a transmission system that could run with 1-3 Megabits/second, which will allow real-time video streaming. Currently, the model achieved these high data rates using offline data processing. We hope to upgrade our model to be able to not only transmit but also process data very fast so that we are able to display video in real time. To achieve this goal, we solved this problem in two steps: First, transcribing our model (currently implemented in MATLAB) to a lower-level programming language (i.e. C/C++) to ensure efficiency. Then, using general code optimization techniques and parallel programming techniques (CPU and GPU) to speed up our implementation by fully exploiting our computational resources.
Issue Date:2019-05
Genre:Other
Type:Text
Language:English
URI:http://hdl.handle.net/2142/104015
Date Available in IDEALS:2019-06-14


This item appears in the following Collection(s)

Item Statistics