Method and system for iterative decoding of full integer turbo codes

Abstract

The present disclosure provides a method and system for iteratively decoding a received information sequence with a full integer Turbo code. The method includes: receiving the k-th bit of the received information sequence, the a priori information of the current iteration about the k-th bit of check information. In the current iteration, it is obtained based on the received k-th bit of check information a priori information. The external component and channel component of the branch metric Mk(s',s) will convert the obtained external component into the integer part to be continued iteration and the common reduced part that will be eliminated in the continued iteration, and the channel component will be converted into the integer part to be continued iteration. The integer part and the common reduced part that will be eliminated in continued iterations; add the external component and the integer part of the channel component; use a full integer memory to store only the external component and the integer part of the channel component and as the branch metric Mk(s',s) of the current iteration for the next iteration; the integer decoding result of the k-th bit is obtained based on the branch metric formed by multiple iterations.

Description

technical field [0001] The present invention relates to the field of Turbo code decoding, and more particularly relates to a method and system for iterative decoding of full-integer Turbo codes. Background technique [0002] Turbo codes are also called parallel concatenated convolutional codes (Parallel Concatenated Convolutional Code, PCCC). During encoding, two simple component codes are concatenated in parallel using a pseudo-random interleaver to obtain a long code with pseudo-random characteristics, and during decoding, two soft-in / soft-out (Soft In Soft Out, SISO) Multiple iterations are performed between the decoders to realize pseudo-random decoding, and the performance close to the theoretical limit of Shannon coding is achieved. [0003] The coding and decoding technology of Turbo codes has been greatly developed at present, and has been widely used in various application fields such as coding, modulation, and signal detection, and has achieved good decoding perfo...

AI Technical Summary

Problems solved by technology

[0004] As we all know, in the decoding process of Turbo code, since the memory only stores integers, if floating-point operations are to be performed, it is generally considered to only store integers. In this case, decimals need to be rounded. Therefore, when using rounding After the stored number is iterated, the accuracy of the obtained storage and iterative operation will gradually decrease. If you want to store a few digits after the decimal point, it will take up a lot of storage space

Images