双向中继信道中物理层网络编码的检测  

Detection for physical-layer network coding in two-way relay channels

作　　者：颜伟[1] 蔡跃明 潘成康 

机构地区：[1]解放军理工大学通信工程学院,江苏南京210007 [2]中国移动研究院,北京100053

出　　处：《通信学报》2012年第2期82-86,共5页Journal on Communications

基　　金：国家自然科学基金资助项目(60972051;61001107);国家科技重大专项基金资助项目(2010ZX03006-002-04);江苏省自然科学基金资助项目(BK2010101)~~

摘　　要：对双向中继信道中物理层网络编码的检测进行了研究,最大似然检测性能好但是实现复杂度高。因此,在信源节点未知信道状态信息情况下,提出了2种检测方案:基于似然比函数的似然比检测和基于最大后验概率准则的最大后验概率检测。同时,针对信源节点已知信道状态信息的特殊情形,进行了同样的推导。分析和仿真结果表明,相比于分别检测出2个信源信息的最大似然检测,似然比检测的BER性能更优,但似然比检测需要知道额外的噪声方差信息,最大后验概率检测与最大似然检测等价,而且最大后验概率检测在实现复杂度上相对较低。The detection for physical-layer network coding （PNC） in two-way relay channels was investigated. Maxi- mum-likelihood （ML） detection not only had great performance but also had high computational complexity. Therefore, without channel state information （CSI） at source nodes, two detection schemes were proposed. The first scheme was likelihood ratio （LR） detection based on LR function, and the other was maximum a posteriori probability （MAP） detec- tion based on MAP rule. Meanwhile, It carried out similar analysis if CSI was available at source nodes. Compared to ML detection which separately detects two source messages, numerical and simulation results show that LR detection outper- forms it in terms of BER metric with the requirement for additional noise variance information. And MAP detection is equivalent to ML detection, and it has lower computational complexity.

关 键 词：物理层网络编码 最大似然 似然比 最大后验概率 检测 

分 类 号：TN911.2[电子电信—通信与信息系统]