A frequency-domain multipoint equalization algorithm that combines fractional-octave smoothing of measured impulse responses (IRs) in multiple locations and the extraction of a representative prototype is presented. The proposed approach is evaluated considering different methods to combine the IRs for the prototype extraction and to obtain an inverse filter for the equalization, using sets of IRs measured in realistic environments. With reference to previous publications the influence on the equalization performance of the number of positions considered and of the size of the equalization zone is investigated. Also a comparison with the single-point equalization approach is reported. Finally the robustness of multipoint equalization is also evaluated for positions that differ from those used for the equalizer estimation.
