Hello,
In wirelss In-Band Full-Duplex systems, I want to know how much Self Interference Cancellation (SIC in dB) a scheme I am simulating can achieve. I am running link-level simulations in MATLAB. As per my understanding, the SNR for which the BER floors should give an idea about the residual interference. But is there any specific simulation technique I can use to determine the amount of cancellation? For example, if I need to plot the amount of SIC (dB) versus the transmit power or Tx-Rx distance etc, I'll need to quantify this. How is this done?
Dear Abhay,
welcome,
I think the first comment was so so useful that we could model our signal.
If the noise is same then you can subtract the two received signals with interference cancellation and with out interference cancellation.So you have to build your system to keep the noise the same.
You can also get out the residual of the interference by minimizing the received signal which is subtracted by an attenuated sent signal. The attenuation factor can be adjusted to minimize the received signal. This could be considered an interference cancellation method.
Best wishes
Dear Abhay,
welcome!
You have the sent signal s(t) which is filtered through the frequency duplexing band pass filter of the receiver hr(t) then the received signal will be hr(t)s(t)= ri(t) which is the received self interference signal.
Assume that the intended received signal is rr(t)
Then you can get the signal to interference signal by power of the received signal pr= average rr^2 and pi= average ri^2
And then the S/I= pr/pi
Best wishes
Abdelhalim abdelnaby Zekry , thank you. But I wanted to determine the amount of cancellation of a particular scheme, not the amount of SI itself. That is, We pass
y(t) = ri(t) + rr(t) + n(t).
through a Self Interference Canceller to get y'(t). This signal now is:
y'(t) = ri'(t) + rr(t) + n(t)
where ri'(t) is the residual self interference.
I wanted to know how to get an estimate of ri'(t) /ri(t), which is the amount of self interference that was cancelled. Can we do this by subtracting the power of y'(t) from y(t)?
Dear Abhay,
welcome,
I think the first comment was so so useful that we could model our signal.
If the noise is same then you can subtract the two received signals with interference cancellation and with out interference cancellation.So you have to build your system to keep the noise the same.
You can also get out the residual of the interference by minimizing the received signal which is subtracted by an attenuated sent signal. The attenuation factor can be adjusted to minimize the received signal. This could be considered an interference cancellation method.
Best wishes
- Badges
- Science topic