This paper presents a verification procedure for finite-difference time-domain-simulated head-related transfer functions (HRTFs) from a simplified model of a human head, a sphere. The analytic solution required by the code verification is computed with the multipole reexpansion technique and used to estimate convergence rates. A solution verification process follows in which asymptotic predictions are computed. For the HRTFs considered and employed grids, results show that the convergence rates attain the expected first-order accuracy at lower frequencies, after which scattered estimates are observed. Results also reveal that the asymptotic predictions are accurate up to 10 kHz, after which bias is observed.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1121/10.0011736 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!