FPGA Implementation of Base-N Logarithm

Abstract

In this work, we present an area optimized FPGA implementation of an IP core to compute the base-N logarithm. Nevertheless, we also discuss the area, speed and precision trade-offs. We selected an algorithm that could be implemented on any FPGA avoiding vendor specific features like block RAMs, embedded multipliers, etc. We report the implementation results of a fixed point version of the algorithm using various common configurations on Xilinx and Actel devices. This implementation achieved the required area goals providing a very good speed-area ratio.