Identification input design for simultaneous estimation of head-related transfer functions

In our previous work, we proposed a method for creating input signals for a multi-input single-output (MISO) system identification. The identification using our previous method is numerically stable, however, the estimation error increases under some condition. This paper proposes a new method for designing input signals for MISO system identification, which enables estimation not only to be numerically stable, but also to be as accurate as measuring one by one. The effectiveness of the proposed method is demonstrated through simultaneous estimation experiments of head-related transfer functions.