Based on the analysis of the constellation mapping in bit interleaved coded modulation and iterative decoding(BICM-ID),design rules are proposed in iterative demodulation and decoding.

The new codes perform well using the sum-product algorithm for iterative decoding.

In addition,it gives input vs output characteristics curve of Viterbi decoder, and puts forward error correct threshold of Viterbi decoder.

This paper applied duration information into Viterbi decoding to overcome these errors.

Recurrent neural networks(RNN) convolution decoder can perform close to the conventional Viterbi decoder,and the complexity increases linearly with the constraint length while that of Viterbi s increases exponentially.

Recurrent Neural Networks(RNN) convolution decoder was proposed as an applicable decoder performing close to the conventional Viterbi decoder,and the RNN decoder′s complexity increases linearly with the constraint length while Viterbi′s increases exponentially.

Viterbi decoder,as a kind of optimal decoding for convolutional codes based on Maximum-Likelihood(ML) criterion,is widely used in many digital communication systems,but brings challenges to the chip design because of its high algorithm complexity.

Based on the factor graph representation of maximum a posteriori probabilities decoding of the user s information bits and the sum product algorithm, an iterative multiuser decoding based on parallel partial interference cancellation detection for coded CDMA system was proposed.