【摘要】 为了降低低密度奇偶检验码的误码平底,提出一种基于陷阱集状态检测的两级置信度传播译码算法。该算法分2步进行,首先,按照标准的置信度传播算法进行迭代,直到检测到陷阱集状态的出现;然后,尝试对可能出现错误的比特的初始值进行翻转扩张继续进行迭代,直到译码成功或达到最大尝试次数。仿真结果表明:该算法大大降低了误码平底区域迭代译码所需的最大迭代次数,同时在进行很少次的尝试后,低密度奇偶检验码的误码平底可以降低近2个数量级。更多还原

【Abstract】 A two stage belief-propagation(TSBP) decoding algorithm was developed based on trapping set state(TSS) detection to reduce the error floor of low-density parity-check(LDPC) codes.The algorithm uses belief-propagation algorithm until the TSS is detected,then iterative decoding is attempted to continue through flipping and augmenting the initial values of the possible error bits until the decoding succeeds or the maximum number of attempts is reached.Simulations show that the algorithm significantly reduces the maximum number of iterations for decoding in the error floor region,and with a few additional attempts,the error floor of LDPC codes can be reduced by nearly two orders of magnitude.更多还原