1 / 10

Vyacheslav Yushin Igor Derzhavets Winter 2007

Implementation of neuronetwork system on FPGA (characterization presentation) supervisor: Karina Odinaev. Vyacheslav Yushin Igor Derzhavets Winter 2007. Project definition. Goal: create FPGA based system for string/pattern matching with high code optimization for FPGA structure.

Download Presentation

Vyacheslav Yushin Igor Derzhavets Winter 2007

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Implementation of neuronetwork system on FPGA(characterization presentation)supervisor: Karina Odinaev Vyacheslav Yushin Igor Derzhavets Winter 2007

  2. Project definition. • Goal: create FPGA based system for string/pattern matching with high code optimization for FPGA structure. • The system will have 3 main hardware elements: • XUPV2P FPGA board. • DLP-USB245M USB Adapter • PC with USB port.

  3. Data Flow Definition • Running special API with testing file as one of the arguments. • The structure of input file is strings of different length. • API will inject the string to the FPGA through the USB adapter. • FPGA will process the data and update the status bit according to processing result. • Status bit from FPGA will be transferred through the USB adapter to API. • API will write the status to the output file

  4. Data Flow Definition (cont.) input file API output file API

  5. FPGA processing. • FPGA programmed according previous project with small tunings of result calculation, allowed by modular design. • FPGA will be synchronized with input device (no dummy clocks allowed) .

  6. Neuron network processing way. • The algorithm of networks processing based on the CLA (constructive learning algorithm) • Reference: http://ieeexplore.ieee.org/iel2/3013/8556/00374191.pdf?arnumber=374191 and http://citeseer.ist.psu.edu/16290.html

  7. Timelines • 27/11/06 - VHDL code ready. • 1/12/06 - Finish API code writing. • 5/12/06 - synthesize the whole project and debug the API -> FPGA input interface. • 10/12/06 – finish debug whole project. • 15/12/06 – make placement optimization (if possible) and run full benchmark for performance validation.

  8. Thank you

  9. BACKUP

  10. BACKUP Input file data Board: http://www.xilinx.com/univ/xupv2p.html USB adapter

More Related