Abstract
CORDIC algorithm is used for low-cost hardware implementation to calculate transcendental functions. This paper proposes a low-latency high-precision architecture for the computation of hyperbolic functions sinhx and coshx based on an improved CORDIC algorithm, that is, the QH-CORDIC. The principle, structure, and range of convergence of the QH-CORDIC are discussed, and the hardware circuit architecture of functions sinhx and coshx using the QH-CORDIC is plotted in this paper. The proposed architecture is implemented using an FPGA device, showing that it has 75% and 50% latency overhead over the two latest prior works. In the synthesis using TSMC 65 nm standard cell library, ASIC implementation results show that the proposed architecture is also superior to the two latest prior works in terms of total time (latency × period), ATP (area × total time), total energy (power × total time), energy efficiency (total energy/efficient bits), and area efficiency (efficient bits/area/total time). Comparison of related works indicates that it is much more favorable for the proposed architecture to perform high-precision floating-point computations on functions sinhx and coshx than the LUT method, stochastic computing, and other CORDIC algorithms.
Highlights
Scientific computing has penetrated almost all scientific and engineering computing and is widely used in energy survey, game rendering, meteorology and oceanography, finance and insurance, computer-aided design, etc
coordinate rotation digital computer (CORDIC) algorithm, this paper proposes a novel QH-CORDIC architecture
The hardware architecture of basic CORDIC and QH-CORDIC is presented in Figure 1a,b, respectively
Summary
Scientific computing has penetrated almost all scientific and engineering computing and is widely used in energy survey, game rendering, meteorology and oceanography, finance and insurance, computer-aided design, etc. Its properties, which are simple arithmetic units [17], fault tolerance, and allowance for high clock rates [18], result in extremely low hardware cost and power consumption, but its disadvantages, including degradation of accuracy and long latency [19], cannot be ignored in some cases Overall, this technique is likely to find more applications in massively parallel computation driven by the very low-cost hardware [20]. A novel architecture based on the CORDIC prototype is proposed to fill in this gap This architecture, called quadruple-step-ahead hyperbolic CORDIC (QH-CORDIC), is demonstrated to be well suited to calculate hyperbolic functions sinhx and coshx in high-precision FP format with low latency. It is coded in Verilog Hardware Description Language (Verilog HDL) to implement the two functions.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have