How to interpret the correlation functions?

[engineer_eda]

Hi All,

Cross-correlation is used between two signals to see if there is a similarity between them, and do so lets say that we have noisy signals x(t) and y(t) which are received from the channel at the receiver.

By performing the cross-correlation we see that there is a peak in the plot diagram of the correlation function in time domain.
My question here is that how do we interpret this peak in the plot diagram?


Please could you also explain me the followings?

What do we get from the plot diagram of the correlation function in time domain?
How to interpret the spectrum of the correlation function?


Thanks in advance!

 

[engineer_eda]

I would appreciate if someone please explain this?

Thanks a lot!

 

[rkt_167]

There are few thing u need to understand first you already mentioned...
1) Cross-correlation is used between two signals to see if there is any similarity between them or rather I would say how much similarity between two signals.
2) Generally, correlation plots are between Energy of the signal Vs time. i.e |x(t)|^2 Vs time.

so, higher peak shows that signals are largely correlated with each other and hence we are finding correlation between two signals by comparing their energy at each and every point in time domain.
This is why we obtain perfect Gaussian distribution in auto correlation.

 

[engineer_eda]

Thank you rkt_167.

I am also trying to understand how we use correlation in OFDM system. And how to correlate the signals at the OFDM system receiver?
Cyclic Prefix added OFDM symbol is received at the receiver, and in order to get each OFDM symbol we want to know where the OFDM symbol starts in time and then decision is made to capture one OFDM symbol. So here the correlation is used but at this point could you please also explain how the correlation works here? Is it cross or auto correlation?

Thanks In advance!