We report the key results of a study of the optimization of capacitive
sensing for LISA. This work was a component of an ESA study into drag
-free satellite control. We discuss the problems associated with the c
apacitive sensing and control of a proof mass which is electrically is
olated and develop models for candidate non-resonant and resonant dete
ction schemes. We calculate acceleration noise assuming an ideal contr
ol system whose only noise sources are the capacitive sensors and actu
ators. On the basis of this model it appears that the design goal of a
cceleration noise for the LISA mission is achievable using both resona
nt and non-resonant detection schemes. The residual acceleration noise
on the proof mass is dominated by oscillator noise which produces for
ce and displacement noise due to asymmetries in the electrode capacita
nce values and bridge components.