Designer’s Guide Community

jcpu2006 Junior Member Offline Posts: 19 Hsinchu	EVM simulation Nov 11^th, 2006, 12:13am Hello, Need reference about how to simulate Error vector (EV) for transmitters of UWB or 802.11a. Any advices / suggestion is welcome. Foe example, can I somehow correlate EV to IP3 index?? Thanks in advance,
Back to top	IP Logged

pancho_hideboo

Senior Fellow

Offline

Posts: 1424
Real Homeless

Re: EVM simulation
Reply #1 - Nov 12^th, 2006, 4:50am

Hi.

See http://www.designers-guide.org/Forum/YaBB.pl?num=1129102926/0#0

It is difficult and not so helpful to relate IP3 to EVM analytically
since EVM is affected by not only AM/AM(IP3) but also AM/PM... etc.

I think you should build system model to survey various effects.
But this is dependent on your availability of EDA tools.

I use Agilent ADS Ptolemy and WTB in both ADS_Native and RFDE for this purpose.
http://eesof.tm.agilent.com/pdf/wlan.pdf

If you can prepare UWB or 802.11a signal and measurement for Cadence Environment,
you can do same system design study also in Cadence Environment.

For this purpose in Cadence Environemt, modeling candidates for signal and measurements are Verilog-AMS, MATLAB, SPW etc. in AMS Designer.

If you will do only simple system characteristics such as ACPR and EVM, Spectre with Verilog-A might be helpful.
But for full system characteristic study, I think it is very difficult and laborious to bulid models using only Verilog-A.

See Cadence RF Design Methodology Kit
http://www.cadence.com/products/kits/RF_Design/index.aspx

Back to top

« Last Edit: Nov 12^th, 2006, 6:47am by pancho_hideboo »

Kita━━━━━━(ﾟ∀ﾟ)━━━━━━ !!!!!
http://www7.plala.or.jp/ungeromeppa/flash/kita.html
http://www.youtube.com/watch?v=mjIxGh55bMM&feature=related

IP Logged

jcpu2006 Junior Member Offline Posts: 19 Hsinchu	Re: EVM simulation Reply #2 - Nov 18^th, 2006, 4:48am Thank you sir: I went through them. The problem wasn't really solved. Let me re-phrase it: EVM estimation by real OFDM signal would be very time consuming. If we design a circuit, say power amplifier, with such simulation. I don't think we have much chance before the time is up. Best Regards, Chewnpu Jou
Back to top	IP Logged

pancho_hideboo

Senior Fellow

Offline

Posts: 1424
Real Homeless

Re: EVM simulation
Reply #3 - Nov 18^th, 2006, 7:11am

Hi.

Using behavioral models, you can relate IP3 to EVM, etc.

This task doesn't take much time, at most half day or one day.

http://www.designers-guide.org/Modeling/modeling-rf-systems.pdf

When I determine specifications, I always do system simualtion using behavioral models, where phase noise, IP3, NF, AM/PM, etc are involved to study ACPR, EVM,.. etc.

Back to top

Kita━━━━━━(ﾟ∀ﾟ)━━━━━━ !!!!!
http://www7.plala.or.jp/ungeromeppa/flash/kita.html
http://www.youtube.com/watch?v=mjIxGh55bMM&feature=related

IP Logged

RFICDUDE

Community Fellow

Offline

Posts: 323

Re: EVM simulation
Reply #4 - Nov 24^th, 2006, 4:43am

If you are specifically interested in OFDM then there were some papers several years ago relating the out of band distortion of CDMA (multi user basestation) case to the IP3 and IP5 of an amplifier. There is a copy of the paper and a Mathcad sheet based on it.

A OFDM signal is fairly close to Gaussian (this is the main assumption of the method), so you could use the method directly to calculate the in-band distortion and the signal to distortion ratio. The in-band signal to distortion ratio is directly related to evm by

EVM=1/sqrt(SNR)

You would have to do a little work to figure out the inband distortion since they mainly focus on ACPR.

http://www.circuitsage.com/lnapa/acpr.pdf
http://www.circuitsage.com/lnapa.html

Also there was a paper on calculating EVM for OFDM signals directly from AM-AM AM-PM simulations/measurements of a PA or transmitter. This technique can be implemented in Matlab or even in simulator like ADS.

Yamanouchi, S.; Kunihiro, K.; Hida, H., "An efficient algorithm for simulating error vector magnitude in nonlinear OFDM amplifiers," Custom Integrated Circuits Conference, 2004. Proceedings of the IEEE 2004 , vol., no.pp. 129- 132, 3-6 Oct. 2004

Hope this helps
RFICDUDE

Back to top

IP Logged