Method for constructing Polar-LDPC concatenated codes

Abstract

The invention provides a method for constructing Polar-LDPC concatenated codes. The method serially concentrates Polar codes as outer codes and LDPC codes as inner codes and is characterized by comprising the following steps of (1) carrying out coding operation of an information sequence (img file='DDA000402432330000011.TIF' wi='63' he='64' / ) through a generated matrix according to encoding rules of Polar codes to obtain an information sequence (img file='DDA000402432330000012.TIF' wi='86' he='81' / ); (2) encoding an information sequence (img file='DDA000402432330000013.TIF' wi='66' he='84' / ) after Polar code encoding according to LDPC codes to obtain an information sequence (img file='DDA000402432330000014.TIF' wi='88' he='64' / ) to be sent to a channel for information transmission; (3) receiving an information sequence (img file='DDA000402432330000015.TIF' wi='105' he='80' / ) from the channel and sending the information sequence to LDPC code decoding, and decoding the information sequence using a BP iterative decoding algorithm to obtain an information sequence (img file='DDA000402432330000016.TIF' wi='95' he='84' / ); and (4) sending an information sequence (img file='DDA000402432330000017.TIF' wi='64' he='67' / ) obtained after LDPC code decoding and sending the information sequence to a Polar code decoder to obtain an information sequence (img file='DDA000402432330000018.TIF' wi='96' he='72' / ) using a continuous deletion decoding algorithm. The novel concatenated codes have better decoding performances and fewer error floors, and can be applied to fields of deep space communication, image transmission and the like.

Description

technical field [0001] The invention relates to the field of image coding, in particular to concatenated codes. Background technique [0002] Polar code is a new channel coding method proposed in recent years. It passes the same and independent N=2 n Channels W are first merged and then separated, so that the symmetrical capacity I(W) and Barthel's parameter Z(W) of each channel W are polarized. According to the channel polarization phenomenon, the advantages and disadvantages of the channel are screened, and the information bits are only transmitted on the channel whose channel capacity is close to 1 through encoding, and only the redundant bits agreed by the communication parties are transmitted on the channel whose channel capacity is close to 0 , this redundant bit is called a dormant bit in the Polar code. [0003] The encoding of the Polar code can be done by generating a matrix, and the encoding formula is where G N to generate the matrix. G N The calculation ...

AI Technical Summary

Problems solved by technology

Therefore, long codes are an effective way to improve the performance of channel coding and decoding, but the increase in code length will inevitably lead to a sharp increase in the complexity and calculation of the decoder, especially when using soft-decision decoding algorithms for decoding. As the code length increases exponentially, when hardware simulation is performed on the decoding algorithm, the consumption of hardware and the complexity of implementation are even more impractical

Images