Hello wizards, Is there a "best" known way for comparing two finite-same-sized digital signals? I want a good way to measure the degree of similarity of the *shape* of the two signals. This is for comparing magnetic signatures of vehicles. Is the best way just to do a scaled cross correlation between the two signals? --Shafik

# Methods for digital signal comparison?

Started by ●December 13, 2004

Reply by ●December 14, 20042004-12-14

Errr, find the correlation? If I remember correctly, remove the DC, multiply and sum the terms of series A and B.. ie sum = a1*b1 + a2+b2 +..... The sum will be proportional to the correlation between the two signals. Or if they are very similar, finding the mean squared error between the two signals ie sum= (a1-b1)^2 + (a2-b2)^2 . Failing that, neural networks that can be trained to accept or reject? The magnetic signals from cars - are these cars are moving - have you factored in the different car velocities ? Ie the same car might have different signatures at different speeds etc.

Reply by ●December 14, 20042004-12-14

Errr, find the correlation? If I remember correctly, remove the DC, multiply and sum the terms of series A and B.. ie sum = a1*b1 + a2+b2 +..... The sum will be proportional to the correlation between the two signals. Or if they are very similar, finding the mean squared error between the two signals ie sum= (a1-b1)^2 + (a2-b2)^2 . Failing that, neural networks that can be trained to accept or reject? The magnetic signals from cars - these cars are moving - have you factored in the different car velocities ? Ie the same car might have different signatures at different speeds etc.