What is the best technique in detecting signal in very low S/N ratio?

Answer 1

If you are talking about detecting a signal in the presence of noise in a communications receiver, then there are several techniques.

In digital communication receivers, for example, there is a mathematical technique known as 'cross correlation'. In this method the received signal which is corrupted by noise during transmission is 'cross correlated' with a known good copy of the signal stored in the memory within the receiver.

Mathematically, cross correlation is the sum of the product of received signal samples and known signal samples shifted in time (delayed) appropriately.

If the output of the correlator shows a peak above a pre-determined threshold then the known signal is said to have been detected.

If you are an electrical engineering student or a physics/math student then you can read up more on the theory of signal detection using correlation method and several other advance techniques in a number of engineering textbooks or even 'Google' it.

Good luck

Answer 2

Typically you would integrate repeated occurrences because only the signal part would reinforce while the random noise would tend to cancel. That is the technique used to get the very weak GPS satellite signal to show up in a GPS receiver (even with a small antenna and the interior of a building).

Answer 3

That depends on what equipment that you have at hand. There are meters, or transmission line test sets that could sort out the desired signal from the noise. You could also use a spectrum analyzer with the capability of retrieving the signal desired and sending it on to other circuits for further processing. A frequency selective voltmeter might also be useful.

Answer 4

AGC=Automatic Frequency Control with Double Diversity receiver method in single sideband and PCM is one of best technique in very low S/N ratio.