本文根据AMR技术在GSM系统中物理层信道卷积编码规范,提出了一种可控制信道差错,降低工程实现难度的译码MATLAB仿真方案。结果表明,通过该仿真方案适合AMR语音业务信道卷积码的译码。

According to channel convolution coding standards in the physical layer of GSM system of AMR tec hnology,this paper proposed a decoding MATLAB simulation program that can control channel error and reduce difficulty of project implementation.The results show that the simulation program is appropriate for the de coding of channel convolution codes of AMR voice service.

介绍了一种长距离无线数字通信中的信道编码设计方案。为改善通信质量,采用了(255,233)RS编码+(8,255)交织+(2,1,7)卷积编码这一信道编码方式。MATLAB仿真结果表明,该级联码能有效降低数字通信系统的误码率,在无线数字通信领域中有着很好的应用前景。

This paper introduces a method of designing the channel coding on long-distance wireless digital communication, in order to enhance the quality of communication,(255,233)RS code and(2,1,7)convolu-tion code are cascaded by (8,255) inter-leaver. The MATLAB simulation result shows that cascaded code will be widely used in wireless digital communication region because of its low error rate.

递归系统卷积码（RSC码）是Turbo码子编码器常采用的编码形式。针对（n,1,m）递归系统卷积码的盲识别问题,给出了完整的编码参数和生成多项式识别方法。首先,采用矩阵分析法先求得卷积码参数,避免后续识别生成多项式时因参数循环估计带来的额外计算量。然后,根据RSC码的特性推导出生成多项式识别模型,给出了模型的具体求解步骤,并对所采用的Walsh-Hadamard变换算法进行了介绍。最后,运用Matlab平台进行仿真验证。仿真结果表明,该方法容错性能明显优于常规方法,在误码率为10-2条件下成功识别概率能达到90%,对于Turbo码的进一步研究具有重要意义。

Recursive system convolutional code( RSC code) is the commonly used encoding form of Turbo sub-encoders. For the blind recognition problem of (n,1,m) RSC code,a method based on Walsh-Had-amard transform is proposed. Firstly,code parameters are estimated by analyzing matrix constructed by the received codeword. Then,the mathematical recognition model is derived according to RSC code’s charac-teristics,the specific implementation steps are provided and the principle of the method is introduced. The simulation results from Matlab show that the performance of the method is obviously more excellent in com-parison with other methods. The recognition probability reaches 90% when bit error rate( BER) is as high as 10-2 ,which indicates the great significance of the proposed method for further research on Turbo code.

针对无人机(unmanned aerial vehicle,UAV)平台数据链机载端机所要求的信道编码算法简单、运算速度快、纠错能力强和实时性高等特点,研究分析了适用于无人机数据链系统的卷积码、Turbo码、LDPC(low density parity check code)码等信道编码的编译码特点,采用基于信道编码的OFDM(orthogonal frequency division multiplexing)传输方案设计了高速无人机数据链系统,并针对不同的编码方案分别进行了在Rayleigh衰落和Rician衰落信道下的性能仿真分析,仿真结果表明,在高速无人机信道下,不论在短码和长码情况下,综合考虑编译码实现复杂度和性能2个方面,LDPC码均表现出了比卷积码和Turbo码较优异的性能,可作为无人机数据链系统中较好的信道编码方案.LDPC码相比卷积码与Turbo码具有更好的误码率性能,是较好的解决方案.

Channel coding is the key technology to improve anti-interference performance of data link system,the channel coding algorithm used for UAV (unmanned aerial vehicle) data system should be simple,quick,reliable and real-time.We analyzed the system of data link of UAV and,we also compared the coding and decoding characteristics of convolutional codes,Turbo and LDPC Codes,finally we designed high-speed data link of UAV based on OFDM and simulated different coding schemes in Rayleigh fading and Rician fading channel respectively.Simulation results show that LDPC codes has better BER performance than Turbo codes and convolutional codes in the high-speed UAV data link channel; whatever is short code or long code,LDPC codes can be the better scheme for future UAV data link system.

在通信系统中，为了保证通信系统的可靠性，克服信道中的噪声和干扰，需要引入信道编码。当利用计算机搜索到大约束长度的卷积码生成多项式，为了剔除搜索到的恶性卷积码生成多项式，必须对搜索到的生成多项式进行检验。判定一个编码器的生成多项式是否是恶性的方法是数学推导和Matlab计算。然而这些方法只适合约束长度较短的卷积码，而当约束长度较大时，运用现有的计算手段很难对编码器生成多项式进行检验。提出了一种基于卷积码自身状态特性的适用于并行就算的检验算法，实验表明该算法可在很大程度上提高检验的效率，有很大的实用价值。

In order to ensure the reliability of communication system and overcome the noise and interference, channel coding is introduced in communication system. With computer to search the convolution code generator polynomial of long constraint length, the searched convolutional code generator polynomial must be tested, thus to remove the malignant polynomial. The mathematical and Matlab methods are used to determine whether the encoder generator polynomial is malicious or not. However, these methods are only suitable for the short constraint length convolutional code, whereas when the constraint length is longer, the existing computing means is hard to test the generator polynomial of the encoder. This paper proposes a parallel computing inspection algorithm based on the features of convolutional code itself. The experiment result shows that the proposed algorithm could greatly improve the inspection efficiency, and is thus of great practical value.