纠正同步错误的级联码研究

发布时间：2018-12-26 07:58

【摘要】：同步错误广泛存在于各类通信系统中,这类错误具有记忆性,会造成灾害性的错误传播,无法用传统的纠错编码方案纠正。在同步错误纠错码方案中,级联码通过内码识别插入删节错误的位置,通过外码纠正由替代和错误识别导致的错误,可有效纠正多个插入、删节和替代错误,但译码复杂度高。反转级联码改变了级联方案的编码器结构,避免了内译码器中稀疏码字似然比的计算,有效降低了级联码的译码复杂度,但译码性能的损失较大。为此,本文设计了一种可改善反转级联码性能的迭代译码算法。具体而言,针对反转级联码的性能损失问题,提出了一种内外译码器之间进行硬判决迭代的译码算法。所设计算法利用内外译码器的串行结构,将外译码器输出的码字硬判决信息反馈给内译码器,内译码器利用该信息重新进行同步。仿真结果表明,该方案可使内译码器确定的同步错误位置更准确,改善了反转级联码的性能。进一步,本文分析了提出的反转级联码迭代译码的复杂度,并与原始级联码、反转级联码进行对比,给出了三种方案中译码器的各个步骤所需的加法、乘法、指数和对数运算的次数。由分析可知,在一定的迭代次数内,反转级联码的硬判决迭代可实现性能与复杂度的良好折中,最优的整体迭代次数可通过仿真确定。综上所述,本文提出了反转级联码的硬判决迭代方案,以增加少量复杂度为代价改善了反转级联码的性能。将提出的反转级联码硬判决迭代方案的复杂度分别与原始级联码和反转级联码方案进行对比。
[Abstract]:Synchronous errors are widely used in all kinds of communication systems. These errors have memory and cause disastrous error propagation, which can not be corrected by traditional error correction coding scheme. In the scheme of synchronous error correction codes, concatenated codes can recognize the position of insert and delete errors by internal codes, and correct errors caused by substitution and error recognition by outer codes. It can effectively correct multiple inserts, deletions and substitution errors, but the decoding complexity is high. The inverted concatenated code changes the encoder structure of the concatenated scheme, avoids the computation of the sparse codeword likelihood ratio in the internal decoder, and effectively reduces the decoding complexity of the concatenated code, but the decoding performance is greatly impaired. Therefore, an iterative decoding algorithm is designed to improve the performance of the inverted concatenated codes. In particular, a decoding algorithm based on hard decision iteration between internal and external decoders is proposed for the performance loss of inverted concatenated codes. Using the serial structure of the internal and external decoder, the algorithm feedback the codeword hard decision information from the external decoder to the internal decoder, and the internal decoder uses the information to resynchronize. Simulation results show that the proposed scheme can make the synchronization error location determined by the internal decoder more accurate and improve the performance of the inverted concatenated code. Furthermore, this paper analyzes the complexity of the iterative decoding of the inverted concatenated code, compares it with the original concatenated code and the inverted concatenated code, and gives the addition and multiplication for each step of the decoder in the three schemes. The number of exponent and logarithmic operations. The analysis shows that the hard decision iteration of the concatenated code can achieve a good compromise between performance and complexity within a certain number of iterations, and the optimal number of global iterations can be determined by simulation. To sum up, this paper proposes a hard decision iterative scheme for inverted concatenated codes, which improves the performance of inverted concatenated codes at the cost of increasing a small amount of complexity. The complexity of the proposed scheme is compared with the original concatenated code and the inverted concatenated code scheme.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.22

【相似文献】