I'm trying to model some time series data, and I've been reading about tapped delay line and sliding window to transform the input data.
In my understanding a sliding window with windows size 1 shifts the inputs by one at each time step and feeds them back to a feed forward neural network.
target| input
2| 1 0
3| 2 1
4| 3 2
5| 4 3
6| 5 4
7| 6 5
In my understanding, a tapped delay line with delay 2 takes the two most recent inputs and outputs and feeds them back to a recurrent neural network
y(t) = f(x(t-1),y(t-1))
Is there a difference? Why are there different names? Please correct me if I'm wrong.