Data processing method and device based on DFT ( Discrete Fourier Transform)

Abstract

The invention discloses a data processing method and a device based on DFT( Discrete Fourier Transform). The method comprises the following steps of: receiving input data; carrying out the serial DFT base computation on the input data and the serial DFT being consistent to the DFT base in number according to the number of the discrete Fourier transform (DFT) base; outputting the data after the DFT base computation. In the embodiment of the invention, the serial DFT base computation is carried out on the input dataso that the high-efficiency stream treatment is realized; all data is processed in series without restoring, thus the computation delay is not generated; in addition, the data is unnecessary to call at different moments, thereby the realization complexity of the DFT processing structure is reduced.

Description

technical field [0001] The present application relates to the field of communication technologies, and in particular to a DFT-based data processing method and device. Background technique [0002] DFT (Discrete Fourier Transform, Discrete Fourier Transform) is an important transformation in signal analysis and processing, and is widely used in the field of communication. With the development of communication technology, the requirements for the implementation architecture of DFT are gradually increasing, and the DFT implementation architecture with a single or few processing points has been difficult to meet the communication requirements. LET technology is the trend of future communication development. In the LET (Long-Term-Evolution, long-term evolution) protocol, the points processed by DFT are required to be within 6 to 1296, satisfying the number of points N=2 α 3 β 5 γ (wherein, α, β, and γ are integers). [0003] According to N is within 6 to 1296 and satisfies N=...

AI Technical Summary

Problems solved by technology

[0004] The inventor finds in the process of realizing the present invention, because only one Radix-2DFT module, Radix-3DFT module, Radix-5DFT module are respectively included in the existing DFT processing structure as computing unit, so each module needs to wait for the previous data After the packet processing is completed and the computing unit is released, the next data packet can be input and processed, resulting in the inability to achieve efficient pipeline processing; because the data after each operation of the Radix-2DFT module, Radix-3DFT module, and Radix-5DFT module are It needs to be stored back, and the next operation of the Radix-2DFT module, Radix-3DFT module, and Radix-5DFT module can only be performed after waiting for all the data to be stored back, so the operation delay is increased; and, due to the need for the Radix-2DFT module, The Radix-3DFT module and the Radix-5DFT module are called in time-sharing, so the complexity of implementing the increased DFT processing structure

Images