My SNR is better with DC Free transmission. Why?


1.) Datasignal with DC:

Datastream consisting of 0V and 1V.

2.) Datasignal without DC:

Datastream consisting of -0.5V and 0.5V.

The variance is the same at both examples.

So I don't get, why the SNR is better without DC.

  • \$\begingroup\$ It's not as a rule, unless you are sensitive to 1/f noise, show your assumptions or real issue \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Mar 19 '19 at 23:08

Noise performance depends on the difference between levels. Since both your signals have a difference of 1 volt, they have the same resistance to noise.

However, they have different energy. Let's say that your data signal has a period of 1 second. Then, the pulses you transmit have the following energy:

  • The 0V pulse has energy 0 J.
  • The 1V pulse has energy 1 J.

Then, on average your signal requires 0.5 J per pulse.

The second signal uses pulses of the following energy:

  • The 0.5 V pulse has energy 0.25 J.
  • The -0.5 V pulse has energy 0.25 J.

So, on average the energy in this signal is 0.25 J per pulse.

So, removing DC results in a signal with the same error performance but less energy. This is equivalent to saying that you have a larger SNR.

(Note that I have made an assumption that is common in communications, that the load resistance is \$R_L = 1\,\Omega\$. In any case, when comparing the SNR of the two signals, \$R_L\$ cancels out, so its actual value is not important).

| improve this answer | |

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.