带禁止符号的算术码序列译码算法

如何有效检测错误以及如何构造编码树型结构是算术码抗误码性能研究的两个关键性问题。文章利用多禁止符号实现快速、高效检错，并结合删减编码树型结构分支点的序列译码算法，在降低序列译码复杂度的同时提高了算术码的抗误码性能。仿真结果表明，在同样误包率的条件下，多禁止符号的抗误码性能优于单禁止符号0．5dB。而且在数据包长固定的条件下，禁止符号冗余度的选择和序列译码堆栈空间大小密切相关，联合优化后可以达到性能最优。

Sequential Decoding Algorithm of Arithmetic Code with Forbidden Symbols

How to detect error efficiently and how to construct decoding tree are two key problems in the study of error correction of arithmetic code. In this paper, multiple forbidden symbols are used in arithmetic code to provide quick and efficient error detection capability. Thecombination of sequential decoding with reduced branch points of decoding tree is a significant method to improve the error correction performance of arithmetic code while reducing the complexity of sequential decoding. The simulation result shows that, under the same packet error rate, the multiple forbidden symbols outperforms 0.5 dB than single forbidden symbol in error correction. And under the fixed length of data packet, how to select the redundancy of forbidden symbol is highly related to the size of stack of sequential decoding, so the joint optimization is the best way to achieve the best performance.