LT code decoding method for short code length system

Abstract

The invention discloses an LT code decoding method for a short code length system, and the method is an improved decoding method based on a BP (Belief Propagation) algorithm. The method comprises thefollowing steps: 1, performing system LT code encoding on an information source bit information sequence and performing binary phase shift keying (Binary Phase Shift Keying) on the information sourcebit information sequence; wirelessly transmitting the modulation symbol sequence to a receiver, enabling a receiver demodulator to adopt a traditional demodulation method to obtain channel log-likelihood ratio information of a system LT code coding bit, enabling a receiver LT decoder to adopt a BP decoding algorithm to decode, and after the BP algorithm is completed, enabling the receiver LT decoder to decode the channel log-likelihood ratio information of the system LT code coding bit. The receiver calculates the soft information of the coded bits, resorts and judges the soft information, andfinds out a minimum weighted Hamming distance by performing bit flipping on a bit sequence obtained by judgment, thereby obtaining a finally received information sequence. Compared with a traditionalBP decoding algorithm, the improved decoding method provided by the invention has the advantages that the soft information of the coded bits is fully utilized, so that the frame error rate (FER) canbe further reduced, and the purpose of reliable transmission is achieved.

Description

technical field [0001] The invention relates to a short code length system LT code decoding method suitable for a wireless noise channel in a wireless communication system, and belongs to the technical field of communication coding. Background technique [0002] Fountain code is a new type of rateless error correction code technology, and its typical applications include multicast and broadcast services, distributed network storage, etc. LT (Luby Transform) codes and Raptor codes are currently the two most important types of fountain codes. Among them, Raptor codes are composed of high bit rate precoding and LT code cascading. Fountain codes were originally used in erasure channels to resist packet loss. In recent years, academic circles have proved that fountain codes also have excellent performance in wireless noise channels (such as AWGN channels, fading channels), (see "Fixed-Rate Raptor Codes Over Rician Fading Channels”, IEEE Transactions on Vehicular Technology, Vol....

AI Technical Summary

Problems solved by technology

The performance of the traditional Belief Propagation (BP) decoding algorithm is close to the maximum likelihood decoding performance and has an acceptable decoding algorithm complexity when the LT code length is long

However, when the code length is short (such as the code word length is less than 500), the decoding performance deteriorates seriously, so it is necessary to study a decoding algorithm with better performance when the code length is short

Images