All the tasks below should be applied to the following
The project should be reported in a concise but high
standard research-like article.
The report should be returned by the 15th of January.
Characterisation of the frequency content
- Specify the frequency content of the data using
- Filter out the 60 Hz power line and the base-line
- Extend the analysis into time-frequency representation
both in Fourier and Wavelet bases. Justify your choice of
Wavelet basis. Explain the reasons for the differences
between the two approaches.
Segment the data in portions having:
You may choose any segmentation method from the
book. Justify your choice.
Modelling and time-series prediction
Model the data using a suitable
linear auto-regressor (AR).
- Find a good balance between model complexity (the
number of time lags) and predictive power.
- Check for stationarity in the data by computing the
innovation process (i.e. the residual between the data
and the prediction of the model)
- Predict 10 next values for the data.
Thursday, 02-Dec-2004 16:30:34 EET