GPU-based digital NCO high-precision parallel implementation method

Abstract

The invention discloses a GPU-based digital NCO high-precision parallel implementation method, belongs to the technical field of communication, and can design and implement an efficient and high-precision numerical control oscillator NCO. The method specifically comprises the following steps that a sine lookup table LUT is established, and the sine lookup table LUT is stored by utilizing a texturecache of a graphics processing unit GPU; the GPU receives the input data and performs segmentation processing; In each segment, after the initial phase of the current segment is determined, the phasevalue of each data point in the segment is calculated according to the phase difference of the sampling points; integer-cycle elimination operation is carried out on the phase value of the last datapoint in the segment, and a sampling point phase difference value is added to serve as the initial phase of the next segment; and therefore, an accumulated phase value of each data point of the inputdata is obtained. A lookup table index value index (iL) is constructed by utilizing the accumulated phase values of the data points, and the sine lookup table LUT in the texture cache is searched to obtain sine and cosine orthogonal numerical control oscillator NCO signals.

Description

technical field [0001] The invention relates to the technical field of communication, in particular to a GPU-based digital NCO high-precision parallel implementation method. Background technique [0002] Digital Down Converter (DDC) system is an important subsystem of modern aerospace measurement and control system. In a typical DDC system, the digital local oscillator is the most complex core device. [0003] In a traditional DDC system, the digital local oscillator is mainly realized by a numerical control oscillator (Numerical Control Oscillator, NCO). A typical NCO consists of an N-bit phase accumulator, a phase register, and an M-bit sine look-up table. Increasing the number of N and M bits can effectively improve the frequency and phase resolution, but the phase resolution accuracy of the lookup table is restricted by the storage space on the FPGA chip and cannot be effectively improved. In recent years, with the increase of on-chip storage space, the look-up table ...

AI Technical Summary

Problems solved by technology

[0005] Although the GPU-based NCO implementation method has the advantages of flexibility and high efficiency, the GPU-based floating-point number calculation method will accumulate errors due to rounding errors in the phase accumulation process of floating-point numbers.

Images