Receiver system design - RF Cafe Forums

 Post subject: Receiver system design

Posted: Wed Apr 12, 2006 8:38 am 

Hi guys. I'm getting ready to embark on one of my 1st receiver design projects (2nd year out of engineering school - MSEE). They taught us about receiver budgets, but most of those classes were taught by TAs that had never really designed anything in real life. I' going to need some software to help out. Anybody care to suggest something other than the $10k programs like MDS et al? I'd like to buy this myself so I can use it at home.

Here's the kind of parameters I'm looking to plan:
Gain
Noise Figure
Intercept Point
1 dB Compression Point
Downconversion
Filtering
BER (maybe but not)

Can anyone suggest a good book on the subject that convers strategy?

Thanks.


Eric

 Post subject:

Posted: Thu Apr 13, 2006 11:10 am 

Take a look at some of the free vendor software available. I've used Agilent AppCAD and Cascade32 from Spectrum Microwave, both of which I believe are linked through RF Cafe. I've also used an Excel spreadsheet passed down through co-workers with changes I've made. It's created from the basic cascaded noise figure and intermod equations. Also, I believe Kirt has a spreadsheet for a small fee. Doing your own spread sheet would be very instructive. Use one of the other programs mentioned as a double check.

Keep in mind that a mixer will add out of band noise unless that is filtered out of the applied signal.

Filters can be handled by allowing for a bandwidth term in the spreadsheet. There will be some manual tweaking of the equations but it will assist you in seeing the effects of a 5 MHz BW filter and a 20 MHz BW filter.

None of the mentioned software does BER, however, that is dependent upon the SNR and SFDR you have, so you should be able to get close.

Also, be aware that the simulations don't typically take into account component to component variations and also rely on ACCURATE data at all frequencies involved. This means a lot of bench testing of simple amplifiers. (You haven't lived until you've taken intermod data on a mixer over temperature and LO drive for ten units and then find there is no pattern to the distortion products or losses that can easily be quantified or described in any equation.)

Temperature drift and stage to stage mis-match need to be taken into account also.

To sum up, your best bet is to get one of these software packages and start using them with component manufacturer's data, building small functional blocks, always comparing simulation with measured data.

GOOD LUCK!!

_________________
Jim Beckford
RF Engineer

 Post subject:

Posted: Tue Jul 10, 2007 11:00 pm 

Just a follow up to the previous post:

Another program which can be used is SystemView by Elanix. It has proven to be pretty accurate for system level design in the few projects I have used it for. Another program you can use is the Simulink portion of Matlab. Although it does cost some money to get licenses for these, you can install these on your PC at home and they won't set you back $10k. It will definitely take some getting used to and some trial and error before you are good with these programs, but in addition to the aformentioned software, these programs are definitely starting to provide good systems with decent examples in the nominal libraries.

Hope this helps.

-J

 Post subject:

Posted: Wed Jul 11, 2007 9:03 pm 

I've written a bunch of receiver and transmitter spreadsheets over the years, and you can certainly calculate most of the parameters you want in a spreadsheet as a starting point. I'd recommend Dr. Egan's Book "Practical RF System Design" as an excellent starting point. It's the only book I've come across that covers spreadsheet analysis.

Dave