Assuming we sample the noise, we get a per-sample variance of Averaging a random variable leads to the following variance: Since noise variance is constant : demonstrating that averaging realizations of the same, uncorrelated noise reduces noise power by a factor of.
Signal power for sampled signals
Considering vectors of signal samples of length : the power of such a vector simply is Again, averaging the vectors, yields the following averaged vector In the case where, we see that reaches a maximum of In this case, the ratio of signal to noise also reaches a maximum, This is the oversampling case, where the observed signal is correlated.
Time-locked signals
Averaging is applied to enhance a time-locked signal component in noisy measurements; time-locking implies that the signal is observation-periodic, so we end up in the maximum case above.
A specific way of obtaining replicates is to average all the odd and even trials in separate buffers. This has the advantage of allowing for comparison of even and odd results from interleaved trials. An average of odd and even averages generates the completed averaged result, while the difference between the odd and even averages constitutes an estimate of the noise.
Algorithmic implementation
The following is a MATLAB simulation of the averaging process: % Create matrix % Fill the matrix with noise sz = 256; NOISE_TRIALS = randn; % Create signal with a sine wave % Divide the array SZ by sz/2 SZ = 1:sz; SZ = SZ/; S = sin; for i = 1:sz; NOISE_TRIALS = NOISE_TRIALS + S; end; % Create the average average = sum/sz; odd_average = sum / ; even_average = sum / ; noise_estimate = odd_average - even_average; % Create plot figure hold on plot plot plot plot hold off xlabel ylabel title legend
The averaging process above, and in general, results in an estimate of the signal. When compared with the raw trace, the averaged noise component is reduced with every averaged trial. When averaging real signals, the underlying component may not always be as clear, resulting in repeated averages in a search for consistent components in two or three replicates. It is unlikely that two or more consistent results will be produced by chance alone.
Correlated noise
Signal averaging typically relies heavily on the assumption that the noise component of a signal is random, having zero mean, and being unrelated to the signal. However, there are instances in which the noise is not uncorrelated. A common example of correlated noise is a hum. To apply the signal averaging technique, a few critical adaptations must be made, and the problem can be avoided by:
