Model-driven Turbo code deep learning decoding method

Abstract

The invention discloses a model-driven Turbo code deep learning decoding method, which comprises the following steps of: firstly, unfolding a Turbo code iterative decoding structure into a tiled structure, and replacing each iteration with a DNN decoding unit to form a network Turbo Net for Turbo code decoding; then constructing a graphic structure of a traditional Max-Log-MAP algorithm and carrying out parameterization of the graphic structure to obtain a deep neural network based on the Max-Log-MAP algorithm as a sub-network in a TurboNet decoding unit, so that an SISO decoder in a traditional decoding structure and calculation of external information obtained through output of the SISO decoder are replaced; training a TurboNet composed of M DNN units to obtain model parameters; and finally, normalizing an output value of the Turbo Net by using a sigmoid function, and performing hard decision on a normalization result to obtain an estimated value of the real information sequence u todecode the Turbo code. According to the invention, Max-can be improved; log Compared with a model driven by pure data, the error rate performance of the Max-Log-MAP algorithm is improved, the numberof parameters is reduced by two orders of magnitudes, and the time delay is greatly reduced.

Description

technical field [0001] The invention relates to a model-driven Turbo code deep learning decoding method, which belongs to the technical field of wireless communication. Background technique [0002] As an important channel coding technology, Turbo codes have been widely used in the third and fourth generation mobile communication systems due to their excellent performance close to the Shannon limit. In deep space communication, Turbo codes are also written into CCSDS recommendations as standard codes. However, the traditional iterative decoding algorithm of Turbo code can only be processed serially and requires multiple iterations, which is difficult to meet the extremely low delay characteristic proposed by the fifth generation mobile communication system. [0003] In recent years, deep learning technology has achieved great success in the fields of image and voice, making applications such as unmanned driving and smart home possible. Deep learning is an important branch ...

AI Technical Summary

Problems solved by technology

However, whether this model-driven decoding method based on BP algorithm can be applied to sequential codes (such as Turbo codes) to improve the performance of decoding is still unknown.

Images