基于LDPC码的BICM系统中的迭代盲信道估计

Iterative blind channel estimation in LDPC coded BICM system

针对传统基于导频辅助的信道估计方法频带利用率低的不足,提出一种块衰落信道下的迭代盲信道估计算法,该算法基于最大似然估计(maximum likelihood,ML),将译码产生的信息经判决、交织和映射后的符号作为发端的符号估计进行联合迭代信道估计和译码,随着迭代次数的增加,译码输出值更加精确,信道估计值也越来越精确。将之应用到基于低密度校验(low-density parity-check,LDPC)码的比特交织编码调制(bit-interleaved coded modulation,BICM)系统中,仿真结果表明,在对迭代信道估计算法性能的影响上,信噪比与衰落信道的块长之间存在着折中关系,即在低信噪比区域,块长较大时的性能要好于块长较小时的性能,而在高信噪比区域,情况正好相反。

Aiming at the shortcomings of the low spectrum efficiency of the classical pilot symbol assisted channel estimation methods, an ML-based iterative blind channel estimation algorithm over block fading channels is proposed, which regards the symbols obtained by judging, interleaving and mapping the information provided by the decoder as the estimated symbols to perform the channel estimation and decode iteratively. With the increasing iterative numbers, the value of channel estimation is more and more accurate because the information provided by the decoder is more and more accurate. The proposed algorithm is applied in LDPC coded BICM system. Simulation results show that there is a tradeoff between SNR（signal noise ratio） and the block length of block fading channels which determine the performance of the proposed algorithm, that is, the performance under long block is better than that under short block in low SNR but it is opposite in high SNR.