 |
The error for each parameter
after compression was considered relative to the parameter from the raw
signal before compression.
relative error = ( X - Xtrue) / Xtrue
For each group of signals, both the mean relative error, and the standard
deviation of relative errors were calculated at each compression ratio.
So, for each of the eight target entropies, there is a calculation of the
two errors for each signal group.
A * indicates the mean relative error for
the group.
A o indicates the standard deviation of
relative error for the group. |
