中文摘要：

为了逼近多址接入信道的容量限，对功率与低密度校验阵（LDPC）码的度分布进行了优化．基于互信息最大化准则设计了功率分配算法，并得出分配方案，利用高斯近似来近似取得因子图上消息的概率密度函数，得到功率分配条件下优化的LDPC码的度分布．采用迭代的检测与译码算法进行多址干扰的消除．仿真结果表明，经功率和LDPC码优化后的误码性能在误码率为10^-5时比等功率只优化LDPC码度分布时的性能提升了1．6dB．

英文摘要：

A study on approach multiple access channel capacity by means of power allocation and low density parity check （LDPC） codes optimization is presented. The power allocation scheme is designed based on the maximization of mutual information and then the power profile is obtained. The probability density function of message on factor graph is obtained by Gaussian approximation. The optimized LDPC code degree profiles are obtained by using the power profile. Iterative detection and decoding algorithm is also applied to combat multiple access interference. Simulation showes that the bit error probability performance can be improved by 1.6 dB at 10^-5 with the proposed method compared with that of only optimization of LDPC code.