Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

[SOLVED] Neural Network with Matlab in FPGA Spartan 3E, ADC Problem

Status
Not open for further replies.

Allahyarzadeh

Newbie level 4
Joined
Mar 1, 2012
Messages
7
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Iran,Tehran
Activity points
1,347
Hi, thanks in advance

I want to implement a trained Neural Netw in Fpga(Spartan 3E), I know it's not the first time but my question lies in hardware level.
you know to train a Neural netw I need some sample data Input,but the system after implementing is going to use the out put of ADC of board as the neural network Input( :?: )
I use MATLAB neural network toolbox's pattern recognition tool ( nprtool ) which is a wizard like tool to train the neural network.(No I don't use command line to train network)
the problem is the output of ADC is unknown to me,the board use a LTC1407 ADC which is a Serial 12-Bit, 3Msps ADC with Two Simultaneous Differential Inputs and has two channel Inputs and 1.5Msps Throughput per Channel,and SPI output(serial).so I should some how model this kind of converting in matlab to train my network with the exact output of adc.
I've trained my Neural network with 4096 point fft wave data like output of adc(input of network),but it didn't work.
so how I should train my network to fit its input when it's implemented ?
if my question is unclear let me know:smile:
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top