Low-resource low-latency hybrid adaptive CORDIC with floating-point precision

Abstract

Despite being proposed since more than 50 years ago, COordinate Rotation DIgital Computer (CORDIC) is still one of the most effective algorithms for elementary function calculation so far. Original CORDIC, however, suffers high latency due to its nature of unvarying number of rotations. As a result, a low-latency hybrid adaptive (HA) CORDIC is proposed in this paper. Firstly, adaptive angle selection decreases total iterations up to 50% with respect to higher accuracy of results. Secondly, hybrid architecture including fixed-point input and floating-point output reduces the total hardware utilization and enhances the dynamic range of final results. Lastly, parallel and pipeline processing together with resource sharing technique allow the design to operate fully at 175.7 MHz with low resource consumption -1,139 LUTs and 489 registers.

Minimum adder-delay architecture of 8/16/32-point DCT based on fixed-rotation adaptive CORDIC

“…The ultimate goal of the ACor method [15] is to reduce the iterations number by choosing only several selected micro-rotations instead of all of the angles as in the conventional algorithm. By doing so, the latency and the resources cost can be reduced sharply.…”

“…By doing so, the latency and the resources cost can be reduced sharply. For accuracy performances, the ACor architecture was proven to have the equivalent or even better accuracy results than those of the conventional CORDIC [15]. i = j = 0; x 0 = 1; y 0 = 0; z 0 = input residual angle; K = 1; while |z j | > c 15 and The pseudo-code of the algorithm is given by Fig.…”

“…The key idea of AARC algorithm is to choose only a few selected micro-rotations to perform instead of all micro-rotations as in the conventional method. The latest development of AARC is the Adaptive CORDIC (ACor) algorithm which was proposed by Hong-Thu Nguyen et al [15] in 2015. A proposed ACor system was also implemented and proven to have the low-latency, low-resources, and high-accuracy experimental results [15].…”

“…In this paper, the proposed DCT architecture was developed based on the ACor algorithm [15] and the extension of the previous work [16]. Six different versions were implemented based on the number of DCT-point, i.e., 8-point (8p), 16-point (16p), and 32-point (32p), and the number of ACor stages, i.e., 2-Stage (2S) and 3-Stage (3S).…”

Design of a High-Speed High-Accuracy 2048-Point FFT Using Single-Precision Floating-Point Adaptive Cordic on Fpga

“…Therefore, the system will be faster with an approximation accuracy result. Based on the previous works such as [11,12], in this study, Adaptive CORDIC has been optimized and implemented for a Radix-2 Multi-path Delay Commutator FFT system in hardware as an application. In this research, the experimental results of Adaptive CORDIC and 2048-point Radix-2 Multi-path Delay Commutator FFT designs is built and verified based on three kinds of Look-up Table that cost 16, 8 and 4 constant angles.…”