Floating-point to logarithmic encoder error analysis | IEEE Journals & Magazine | IEEE Xplore