Low cost logarithmic techniques for high-precision computations

In this paper, we propose a novel high-precision logarithmic conversion method that relies on linear approximation to reduce the hardware cost and non-linear approximation to increase the conversion precision. It has been shown that for an 8-bit number conversion, the proposed method can achieve a precision of 10 fractional bits with only 24 Look-Up Table (LUT) entries, thereby justifying the approach for incorporation into low-cost and high-precision applications such as an environment modeling strategy for dynamic path planning.