【摘要】：為了提高低密度奇偶校驗(yàn)(LDPC)碼的單最小值最小和(single-minimum Min-Sum,sm MS)算法的誤碼性能,提出了一種基于變量節(jié)點(diǎn)LLR(Log Likelihood Ratio)消息加權(quán)的改進(jìn)最小和(Improved Min Sum algorithm based on weighted message LLR of variable nodes,IMS-WVN)算法。首先,將迭代次數(shù)所確定的次小值的估值參數(shù)與最小值相加后取代次小值,以增強(qiáng)sm MS算法校驗(yàn)節(jié)點(diǎn)的可靠度。然后,將變量節(jié)點(diǎn)輸出LLR消息與迭代前LLR消息進(jìn)行加權(quán)處理,降低變量節(jié)點(diǎn)的振蕩幅度,降低平均譯碼迭代次數(shù)。仿真結(jié)果表明,在信噪比為3.2 d B時(shí),IMS-WVN算法的誤碼性能比VWMS算法提升0.53 d B,當(dāng)誤碼率為10-5時(shí),IMS-WVN算法平均譯碼迭代次數(shù)較MS算法減少58%。
[Abstract]:In order to improve the bit error performance of the single minimum sum (single-minimum Min-Sum,sm MS) algorithm for low density parity check (LDPC) codes, An improved minimum sum (Improved Min Sum algorithm based on weighted message LLR of variable nodes,IMS-WVN) algorithm based on variable node LLR (Log Likelihood Ratio) message weighting is proposed. Firstly, the estimation parameters of the sub-minimum value determined by the number of iterations are added to the minimum value, and then the sub-minimum value is replaced to enhance the reliability of the sm MS algorithm to verify the node reliability. Then, the output LLR message of the variable node and the LLR message before the iteration are weighted to reduce the oscillation amplitude of the variable node and the average number of decoding iterations. The simulation results show that when the SNR is 3.2dB, the error performance of IMS-WVN algorithm is 0.53dB higher than that of VWMS algorithm. When the bit error rate is 10-5, the average decoding iteration times of IMS-WVN algorithm are reduced by 58% compared with MS algorithm.
【作者單位】： 桂林電子科技大學(xué)信息與通信學(xué)院;
【基金】：廣西自然基金項(xiàng)目(2013GXNSFFA019004,2014JJ70068) 廣西教育廳重點(diǎn)項(xiàng)目(ZD2014052)
【分類號(hào)】：TN911.22

【相似文獻(xiàn)】

相關(guān)期刊論文 前2條

1 譚曉衡;張建慧;;基于LLR算法的多姿態(tài)人臉識(shí)別[J];計(jì)算機(jī)應(yīng)用研究;2011年01期

2 ;[J];;年期

，

本文編號(hào)：2469642