I. Introduction
Unary functions, such as trigonometric functions and logarithms, are extensively used in computer graphics, digital signal processing, communication systems, robotics, astrophysics, fluid physics, etc. For these high-speed applications, software solutions are in many cases not sufficient and a hardware implementation is therefore needed. Implementing a numerical function , by a single look-up table is simple and fast which is strait forward for low-precision computations of , i.e., when only has a few bits. However, when performing high-precision computations a single look-up table implementation is impractical due to the huge table size and the long execution time. Approximations only using polynomials have the advantage of being ROM-less, but they can impose large computational complexities and delays [1]. By introducing table-based methods to the polynomials methods the computational complexity can be reduced and the delays can also be decreased to some extent [1]. The CORDIC (COordinate Rotation DIgital Computer) algorithm [2] [3] has been used for these applications since it is faster than a software approach. CORDIC is an iterative method and therefore slow which makes the method insufficient for this kind of applications.