Noise control in the active headrest based on estimated residual signals at virtual microphones

A system known as the active headrest, generating the so-called zones of quiet at the ears of a person sitting in a chair is considered. Shape and dimensions of the headrest have been designed to mimic a real headrest. An adaptive algorithm using the idea of virtual microphones has been designed to shift the zones of highest attenuation to the user's ears. It is based on the assumption that the primary noise source is far enough from the headrest for the difference between its contribution to field response at locations of the errors and respective virtual microphones to be negligible. This assumption allows estimating residual signals at the virtual microphones. Then, mean square values of these signals are minimised in a multivariable adaptive control system implemented on a DSP board. The areas of attenuation are presented in the form of surfacial distributions of zones of quiet. This algorithm is compared with another virtual microphone - based algorithm designed for stationary noise and with classical adaptive Internal Model Control algorithms employing two and four secondary sources.