Improved TMR budget through a frequency shaped servo design

This paper introduces a method of establishing disturbances and measurement noise power spectra from the position error signal (PES) and presents a discrete-time linear-quadratic Gaussian/loop-transfer recovery (LQG/LTR) design technique for a disk drive track following servo system. Having the knowledge of these noise power spectra, a controller can be designed to minimize the energy corresponding to the actual read/write (R/W) head position.