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.

Register Log in

What causes integration error of sinusoidal signals?

Status
Not open for further replies.
I

ILv2Xlr8

Newbie level 6
Joined
Mar 4, 2009
Messages
12
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,357
Integration error

What causes integration error of sinusoidal signals?

I am using the Euler integration method to integrate sinusiodal signals, but get magnitude error that increases exponentially as the frequency of the sinusoid increases.

What causes this error and how can I remove it?

This was noticed per disucssions **broken link removed**

Any advice would be greatly appreciated.
 
Last edited by a moderator:

Integration error

why cant U use help "quad" function in matlab it does numerical integration.
 

Re: Integration error

ILv2Xlr8 said:
What causes integration error of sinusoidal signals?
I am using the Euler integration method to integrate sinusiodal signals, but get magnitude error that increases exponentially as the frequency of the sinusoid increases.
What causes this error and how can I remove it?
Click to expand...

Do you know that the EULER method is just an approximation to the ideal integration procedure ? You cannot "remove" it. Instead, you should use time steps as small as possible.
 

Re: Integration error

rramya said:
why cant U use help "quad" function in matlab it does numerical integration.
Click to expand...

I am hand coding these function in C, mexing them, and using them as S-fcns in simulink. I can not use the canned matlab functions other than to compare results with and validate what I've coded in C.

If some one has the source code for the quad function, I suppose I could adapt that to C code...

Yes I know that the Euler method is an approximation, but I was not aware of its magnitude in error. I am using the smallest time step I for data sampled at 1kHz (1/1024) Would you sugguest another numerical method to code for integration?

Thanks for any advice.
 

Re: Integration error

get magnitude error that increases exponentially as the frequency of the sinusoid increases
Click to expand...
exponentially isn't what you should expect from an error calculation. I would start with an exact calculation of expectable error and compare with your results. It may be a coding error as well. I understand that you are using 1024 steps per period. The integration error should be in a 0.1-0.5 % range, I think.

Generally, if you are processing sampled empirical data without a functional model, there's basically no better integration method.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top