Hi Mostafa,
On 12/28/2014 03:54 PM, Mostafa
Alizadeh wrote:
Hi Martin,
Thank you for response. I don't want Eb/N0, because there
is not bit stream but complex symbols.
A symbol represents information you want to transmit. The unit for
information is bit.
As you said, there are two possible solutions:
1 If one uses a specific modulation, so by sending a long
known bit sequence, the BER (bit error rate) can be found
accordingly then the SNR can be obtained from the theoretical
relationship between BER and SNR (here SNR is Eb/N0). I think
this is not practical due to the fact that we use the analytic
relationship while in practice we want to, for instance, prove
that relationship is true.
2 Measuring SNR in analog is, in fact, a hard task.
Because as you said it depends on both receiver hardware
parameters (such as phase noise, frequency offset, timing
offset, quantization, ... ) and channel characteristics (such
as small and large scale fadings). This method also may need a
sort of channel sounding to extract the channel response
corresponding to the measurement environment.
The third way I could add here is: using a kind a
calibration between Tx/Rx with sending a sine wave pilot.
Suppose a sine wave (in base band) with frequency 'fm' is sent
via a Tx with the power of 'A' dB. At Rx we take FFT of the
signal which has a shape like this:
I assumed that the power level out of sine wave is about
10dBm, call it 'noise_floor', and the received power level of
the sine is 0dBm, call it 'peak'. By using the following
formula the SNR can be calculated:
SNR = peak  noise_floor  10 log (N_fft/2).
N_fft is the number of fft points. Once the SNR value is
calculated for 'A' dB Tx power, it seems reasonable to have
different SNRs with changing just the Tx power.
I'm not sure whether this approach is accurate as we need
or not! Is it true? I want to know your recommendations.
Assuming you're noise is a) purely additive, b) your channel is
absolutely flat, and so is your noise (white), c) everything is
linear (memoryless channel) and c) you're able to reliably measure
the power of a infinitely narrow peak (which implies you can produce
a infinitely narrow peak), this applies. As you can see in your
picture, the peak is *not* infinitely narrow (obviously, the FFT has
limited length, so frequency resolution can't be arbitrarily fine).
It's but a basic estimate. Giving you the rx power for a specific
frequency and estimating noise power by averaging for a limited
amount of time, you get something that, depending on how your system
behaves, might or might not represent actual SNR well.
I have to stress this: SNR is *signal* to noise ratio. Signal is
what your application defines to be signal!
Greetings,
Marcus
