WCDMA Simulink error due to the Root Raised cosine filter

[amitdegada]

wcdma raised cosine filter

Hi!
I am working on WCDMA simulation in Simulink.

I have spreaded and scrambled the data as per the WCDMA standard and as per the standard i have to pass this complex data to Root-Raised cosine filter having roll off factor 0.22.

I dont have idea how to adjust group delay parameter in it as i have not found or perhaps i dont know how to adjust it.

Kindly help me to solve this error.

 

[Communications_Engineer]

WCDMA

First my question is that how did you know to put
2.828427125 as the Linear amplitude filter gain value

 

[amitdegada]

Re: WCDMA

Ya!

Actually the demo of "WCDMA spreading and Modulation" is given in MATLAB R2008a. The same thing i want to develop without using WCDMA library attached to that demo..In short i want to develop design without using any preavailable library.

In that demo they have set the parameter of linear amplitude filter gain as sqrt of upsampling factor( that i have chosen to be 8(eight).

See i dont have much information regarding the parameters other than Root-raised cosine filter. By trial and error method I have make the filter response of my filter as same as of filter given in demo.

Eventhaugh if i adjust the roll of factor 0.22 and keep other parameter default steal i am receiving same error, that you may see after running my model.

So please do needfull things.....

Thanking you for your appreciation.

 

[Communications_Engineer]

WCDMA

Ah... so you have an error. I downloaded you model and tried it out my self and had an error. I have managed to remove that error. Its in a different PC so I'll upload it later. But I couldn't find the answer to the relation between roll-off and group-delay although I understand both terms clearly

I think you should as this in the Digital Signal Processing group at edaboard.com also there is a usenet group on google group, I think its com.soft.dsp, some over there may know.

BE ADVISED: people at the google group are extreme;y arrogant and usually reply harshly so use a dummy name before asking them otherwise your name will be all over the internet with fun being made of you. I read a post where they ridiculed a guy studying at MIT

 

[amitdegada]

Re: WCDMA

Thanks a lot!

Really thanking you for your support. i am waiting for your model...

 

[amitdegada]

Re: WCDMA

Hi Communication_Engineer!

Would you upload that model as early as possible.

Thanking YOu

 

[Communications_Engineer]

Re: WCDMA

Here is your corrected model. I have made changes in the "configure parameters" which you can see from Simulition option in the menu.

However, since your sampling time too small i.e. 1/(3.84 Mcps) the simulation will take a long time to complete. But we can't do anything about it, right, the standard says it.

By the way please answer my question that why do we mulitply the Quadrature component of our signal by 'j' before scramling?


------------------------------------------------------------------------------------------
groups.yahoo.com/group/telecom_research

 

[amitdegada]

Re: WCDMA

HI Communication Engineer!

Thanking you for the help!

In WCDMA Two successive symbols are at first converted from serial to parallel, and mapped to I and Q branches. Even-number-sequenced symbols are mapped to I phase, whereas odd number-sequenced symbols are mapped to Q phase.

Subsequently, both I and Q branches are spread up to the chip rate by the same channelization code. The two realvalued chip sequences of I and Q branches are treated as one complex-valued sequence, and is randomized by the complex-valued scrambling code Sdl,n .According to best of my knowledge When sequence is divided in two parts we will get bitperiod double and hence processing time we get at receiver will be double.

it is done to make overall sequence complex, so at the receiver when it is again divided in two parts I and Q we will get double time for each bit to process.

By the may the model that you have uploaded it is taking to much time to simulate. If you change the configuration parameter will it make any change in output?

Thanking you!

 

[Communications_Engineer]

WCDMA

Actually, I said that it has changed it step size to that of 1/3.84Mcps so it will take long time, basically the simulation points are a count from 0 to what ever simulation end time you have chosen with a step of 1/3.84e6, that is a down side.

But for expert knowledge you can contact mathworks.com or ask them at their google group. Usually people from mathworks answer queries there.

 

[amitdegada]

Re: WCDMA

Hi!

Ok i will pit my question there.....

If get proper answer then will put on this Forum.