next up previous contents
Next: Scaling Up: Dynamic Range Previous: Weighting Functions   Contents

Optimization of Dynamic Range

The dynamic range of a filter is dependent on the filter transfer function and the realization of that transfer function. One method to improve the dynamic range of a filter is increasing the total capacitance that is used to realize the filter. If for instance the capacitance is enlarged by a factor two, the conductances of the filter are also to be enlarged by a factor two, if the transfer function of the filter is to remain unchanged by this action. This implies a reduction of the noise output of the filter by a factor two, and thus an increase of the dynamic range by 3dB.

The total capacitance that can be used to realize the filter is limited, usually by the chip area, sometimes also by the supply power. In the following we assume that the capacitance is limited to a total value of Ctot .

In that case the dynamic range is still dependent on the maximal signal levels in the filter. These levels are usually limited at the inputs or outputs of the integrators, and are dependent on the realization of the integrators. We assume that these maximal signal levels are known, and are (for all the integrators) equal to Vmax .

There are three ways to further increase the dynamic range: optimization of the capacitance distribution, scaling and network transformation.



Subsections
next up previous contents
Next: Scaling Up: Dynamic Range Previous: Weighting Functions   Contents
2009-06-03