1
$\begingroup$

I'm using the Rayleigh Channel in Simulink but I can't calculate the BER. I always get values near to 0.5 which doesn't make sense (even using high EbNo values). [In this simulink model][1] I found a trick to calculate the BER but I also would like to use delays and attenuation vectors.

Any idea of how to calculate the BER? I have tried also with a Matlab script comparing bit to bit but I also get values near 0.5.

Edit1: I can calculate the BER but it is always around 0.5, regardless EbNo values. I'm using this simulink model.

Edit2: I'm using delay and attenuation vector so I have problems with the array dimensions:

enter image description here

Improve this question
$\endgroup$
2
  • $\begingroup$ Welcome to DSP.SE! Why can't you calculate the BER? What is stopping you? If you "can't" calculate it, where did you get 0.5 from? Can you edit your question and show or explain how you are currently calculating the BER? $\endgroup$
    – Peter K. ♦
    Aug 31, 2015 at 19:30
  • 1
    $\begingroup$ The theoretical BER on a Rayleigh fading channel is indeed close to $\frac 12$ even at high (average) SNR. This is because the large number of errors made when the channel is in a deep fade (and so the effective SNR is very small) swamp the good performance when the channel is not faded and the effective SNR is high. In an AWGN channel, the SNR decays exponentially with increasing SNR; on a Rayleigh fading channel, only inversely with increasing average SNR. $\endgroup$
    – Dilip Sarwate
    Sep 1, 2015 at 3:07

2 Answers 2

Reset to default
1
$\begingroup$

After the AWGN block you need to divide with the channel coefficient value before you apply the BPSK demodulator.

If you click on the Rayleigh block it gives the option to enable a "Complex Gain Path Port". You can use this output to accomplish the division mentioned above.

Instead of division you can also multiply the output of the AWGN block with the complex-conjugate of the channel gain.

enter image description here

Improve this answer
$\endgroup$
1
  • $\begingroup$ I'm using delay vector and attenuation vector so I have problems with the matrix dimensiones (see Edit2 on the post). I don't really understand what to do. $\endgroup$
    – alrevuelta
    Sep 1, 2015 at 14:26
1
$\begingroup$

When you pass your signal into a Rayleigh Fading Channel, it distorts your signal, causing Intersymbol Interference (ISI). You need to Equalize your signal at the receiver in order to perform the detection correctly. You can do it in various ways. I suggest you to search about zero-forcing and MSE equalization.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Also see Dilip's comments on the question. $\endgroup$
    – Peter K. ♦
    Sep 1, 2015 at 13:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.