Method for achieving FPGA digital signal processing based on segmented lookup trigonometric function

Abstract

The invention provides a method for achieving FPGA digital signal processing based on a segmented lookup trigonometric function. The method for achieving the FPGA digital signal processing based on the segmented lookup trigonometric function comprises the following steps: dividing FPGA into 3 parts including an index value calculation module, a trigonometric function lookup table, and a linear interpolation module, wherein the trigonometric function lookup table includes correspondences among angles and trigonometric functions and is fixed in ROM storage resources of the FPGA; calculating an index position in the corresponding trigonometric function lookup table by the index value calculation module based on the inputted angle value; applying a linear interpolation to the trigonometric function value obtained from looking up the table by the linear interpolation module for obtaining a trigonometric function calculation value having a higher precision. In the method for achieving the FPGA digital signal processing based on the segmented lookup trigonometric function, the contradiction of calculating the trigonometric function value precision and the lookup table depth by the FPGA is solved by the segmented lookup, thereby saving the limited storage space of the FPGA.

Description

technical field [0001] The invention relates to the field of FPGA digital signal processing, in particular to a method for realizing FPGA digital signal processing based on trigonometric functions. Background technique [0002] Trigonometric functions are often used in digital signal processing using FPGAs. Common FPGA trigonometric function implementation methods include look-up table method and CORDIC module method. Among them, the look-up table method has the characteristics of small delay, but it will occupy more storage resources of the FPGA. Using the CORDIC module method occupies less storage resources, but has a longer calculation delay. In the patent CN 102542336, the use of the CORDIC module requires about 20 clock cycles to calculate the delay. The present invention adopts the method based on the look-up table to calculate the trigonometric function, and the conventional look-up table method needs larger internal memory and is unfavorable for realizing in FPGA, ...

AI Technical Summary

Problems solved by technology

Using the CORDIC module method has less storage resources, but there will be a longer calculation delay. In the patent CN 102542336, the use of the CORDIC module requires about 20 clock cycles to calculate the delay.

Images