Calculation of Eb/N0 in OFDMA
24 visualizzazioni (ultimi 30 giorni)
Mostra commenti meno recenti
Hello,
I'm working on OFDMA performances, i'm a little bit confused between SNR & Eb/N0 because of the presence of many definitions in the Internet,in some books,and in Matlab help!
In my simulation,i generate an OFDMA signal X,i use the Matlab function awgn to add AWGN noise to my signal X:
awgn(X,snr,'measured');
SNR: [0:1:20] [dB]
Then i demodulate my OFDMA signal & calculate BER vs SNR.
My next task was to try to get BER vs Eb/N0. I searched how is it calculated,i found many definitions.I found that:
1) In AWGN channels,Eb/N0 is part of the calculation of the variance (sigma^2) of the additional noise n.Linear equation:
sigma = sqrt(2)*sqrt(1./(2*Rm*Rc*EbN0));
Rm = log2M: Modulation rate
Rc: Code rate
2) I found then SNR[dB] is calculated from Eb/N0[dB]! found many equations in dB format:
a) From Matlab help:
snr = EbNo + 10*log10(Rc) + 10*log10(Rm); [in dB]
b) From Matlab help:
snr = Eb/N0 + 10*log10(Rc) + 10*log10(Rm) - 10*log10(Tsymb/Tsampl);
I'm really confused.My questions is:
1) What's difference between Eb/N0 & SNR? Which is important on performances basis?
2) Do i have to program my own awgn function to simulate Eb/N0 or is it possible to use awgn Matlab function knowing it requires snr in entry not eb/b0?
3) Equation 2)a) & 2)b) are different knowing it's both on matlab help,wich one is adequate? and how using them?
I will really appreciate your answers.Thanks.
Regards,
0 Commenti
Risposta accettata
Neels
il 25 Lug 2011
1. Eb/No is normalized SNR. ie, if you consider SNR for individual bit Eb/No. If you generally consider FEC during transmission or for comparison of various modulation schemes Eb/No is used. In your case SNR appears to be best(my opinion).
2. I am not sure what you are trying to do. if you post your entire code, it will be clear. But if possible, its better you write your own AWGN
3. 2 a looks fine, i hope you have the parameters
Più risposte (1)
Drew
il 20 Ott 2021
1) ) Eb/No is energy per bit to noise power ratio, while SNR is signal to noise power ratio.
The two are related as:
SNR = EbNo+ 10*log10(Rc) + 10*log10(Rm) -10*log10(Tsymb/Tsamp) (dB)
Either can be used for the performance comparison.
2) awgn can be used with SNR as an input, adjusted as per the relation above.
3) 2a assumes Tsymb/Tsamp =1, so either can be used based on the link parameters.
0 Commenti
Vedere anche
Categorie
Scopri di più su Propagation and Channel Models in Help Center e File Exchange
Prodotti
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!