关键词: CORDIC, 相位补偿, 最佳一致逼近, FPGA

Abstract: An amplitude and phase compensation algorithm based on the best uniform approximation method is proposed. It overcomes the limitations of the traditional CORDIC when used in high-precision calculation of the amplitude and phase, such as too many iterations, long delay time, and slow phase convergence. By utilizing the vector information obtained from several iterations of traditional CORDIC, sectionalized first-order polynomial of best uniform approximation compensating for the amplitude and phase results is constructed, thus efficiently improving the computation accuracy. Simulation and test results show that, by using the proposed algorithm with 4 iterations of traditional CORDIC, the relative error of amplitude can reach 10^-5 level, and the absolute error of phase can reach 10^-5 degree level. At the same time, the maximum delay time is no more than 100 ns. And with the use of some dedicated multipliers, the registers and LUTs are reduced by 42.5% and 15.5% respectively. Moreover, the phase precision can be increased approximately one order with one more iteration. Hence, compared to conventional CORDIC algorithm, the proposed algorithm improves in iterations and computation precision, and is suitable for high-precision computation applications.

Key words: coordinate rotation digital computer (CORDIC), phase compensation, best uniform approximation, FPGA