公开(公告)号：KR1020080034050A

公开(公告)日：2008-04-18

申请号：KR1020060099327

申请日：2006-10-12

Applicant: 삼성전자주식회사 ,  한국과학기술원

Inventor： 이주현 ,  윤상보 ,  권영훈 ,  황성수 ,  김종인 ,  박현철 ,  김현국 ,  전기환

IPC: H04B7/04 ,  H04L27/26

CPC classification number: H04L1/0045 ,  H04L1/0656

Abstract: An apparatus and a method for detecting a signal in a communication system using a multi-antenna are provided to estimate a reception signal by considerably reducing the amount of calculation compared with estimation of a reception signal by using an existing QRD(QR Decomposition)-based M algorithm. A receiving end generates a tree structure(502), and determines the first threshold value(504). The receiving end removes a branch with a metric value smaller than the first threshold value(506) and extends each survived candidate symbol as candidate symbols by the number corresponding to a modulation order(508). The receiving end determines a candidate symbol having a minimum accumulated branch metric value as a reception symbol(512), and extends the determined candidate symbol by the number corresponding to the modulation order(514). The receiving end determines a minimum accumulated branch metric value, among the extended candidate symbols, as the second threshold value(516). The receiving end excludes a candidate symbol having a value larger than the second threshold value, among the M number of candidate symbols(518), and extends the surviving candidate symbols by the number corresponding to a modulation order(520). The receiving end determines a candidate symbol corresponding to a minimum accumulated branch metric value as a reception symbol at the third stage(522).

Abstract translation: 提供一种用于使用多天线的通信系统中的信号检测的装置和方法，用于通过使用现有的QRD（QR分解）为基础大大减少与接收信号的估计相比计算量的接收信号 M算法。 接收端生成树结构（502），并确定第一阈值（504）。 接收端删除具有小于第一阈值（506）的度量值的分支，并且将每个存活的候选符号作为候选符号延伸与调制阶数相对应的数目（508）。 接收端确定具有最小累积分支度量值的候选符号作为接收符号（512），并且将所确定的候选符号扩展为与调制阶数相对应的数字（514）。 接收端确定扩展候选符号之中的最小累积分支量度值作为第二阈值（516）。 在M个候选符号（518）中，接收端排除具有大于第二阈值的候选符号，并将剩余候选符号扩展为与调制阶数相对应的数字（520）。 接收端在第三级（522）确定对应于最小累积分支度量值的候选符号作为接收符号。

公开(公告)号：KR1020080023032A

公开(公告)日：2008-03-12

申请号：KR1020060087006

申请日：2006-09-08

Applicant: 삼성전자주식회사 ,  한국과학기술원

Inventor： 최영곤 ,  박현철 ,  김응선 ,  정진세 ,  김정호 ,  김카일 ,  이지훈 ,  이성원 ,  노용성 ,  김재홍

IPC: H04B7/14 ,  H04L12/28 ,  H04L1/16 ,  H04L1/00

CPC classification number: H04L1/005 ,  H04B7/15592 ,  H04L1/0066 ,  H04L1/0069 ,  H04L1/0077 ,  H04L2001/0097

Abstract: A data relay method and a system for the same are provided to obtain a relatively big code gain as a source station, a relay station, and a destination station identically use convolutional turbo codes, not cyclic convolutional codes, in relaying data through a relay network. A data relay system(200) comprises the first puncturing unit(210), the second puncturing unit(220) and a data restoration unit(230). The first puncturing unit(210) codes data into a turbo codeword using convolutional turbo codes, primarily punctures the turbo codeword, and transmits the primarily punctured codeword to a relay station and a destination station from a source station. The second puncturing unit(220) decodes the primarily punctured codeword, secondarily punctures the decoded codeword, and transmits the secondarily punctured codeword to the destination station from the relay station. The data restoration unit(230) combines the turbo codewords, respectively received to the destination station from the source station and the relay station, into a turbo codeword, decodes the combined codeword, and restores the data.

Abstract translation: 提供数据中继方法及其系统，以获得作为源站，中继站和目的站的相对较大的代码增益，在通过中继网络中继数据时，相同地使用卷积turbo码而不是循环卷积码 。 数据中继系统（200）包括第一穿孔单元（210），第二穿孔单元（220）和数据恢复单元（230）。 第一穿孔单元（210）使用卷积turbo码将数据编码成turbo码字，主要对所述turbo码字进行穿孔，并且将来自源站的主要穿孔码字发送到中继站和目的地站。 第二删截单元（220）对主要被穿孔的码字进行解码，二次对解码的码字进行穿孔，并从中继站向接收站发送二次删余的码字。 数据恢复单元（230）将分别从源站和中继站接收的目的地站的turbo码字组合成turbo码字，解码组合码字，并恢复数据。