联合信源信道编码技术研究

发布时间：2018-06-16 16:54

本文选题：联合信源信道编码 + 过采样滤波器组　；参考：《河北联合大学》2014年硕士论文

【摘要】：无线信道中有噪声、干扰和衰落的影响，为了达到较低的误码率，就需要较强的纠错编码方法。然而较强纠错编码会有系统复杂度提高、时延加大等问题。联合信源信道编码（JSCC, Joint Source Channel Coding）技术是解决这一矛盾的主要方法。文中搭建了两种联合信源信道编码系统，构建了联合信源信道模型。考虑量化噪声和信道噪声，由量化、给定转移概率的二进制对称信道、反量化构成的GBG（Gaussian Background Noise and Bernoulli Gaussian Impulse Noise）信道模型构成联合信道。首先，研究分析基于Real BCH编码的联合信源信道编码系统的性能。之后，搭建了基于过采样滤波器组（OFBS, Oversampled Filter Banks）的联合信源信道编码系统。研究通信领域中新的基于过采样滤波器组的纠错码。对OFBS进行多相分解，找到以贝叶斯假设检验为理论依据的新的阈值计算方法，研究系统传输的纠错性能。 Matlab仿真结果表明，，信道在不同的转移概率条件下，基于Real BCH编码的联合信源信道编码技术在图像传输方面有较好的传输效果。当信道的转移概率较小信道状况良好时，图像的传输质量很好；当信道情况变差，图像的质量下降平缓。由于Real BCH编码过程中加入了冗余，所以在一定程度会减弱编码效率。OFBS产生的冗余是信号通过滤波器组时产生的，带有原始信号的信息成分。当信道质量较好时，可以在接收端恢复出高质量信号。当信道质量变差时，恢复出来的信号质量也随之降低，但即使在信道质量很恶劣的情况下，恢复的信号质量依然可以接受。基于过采样滤波器组的新阈值的计算简洁有效。
[Abstract]:In order to achieve low bit error rate (BER), strong error-correcting coding method is needed for the influence of noise, interference and fading in wireless channel. However, strong error-correcting coding will have some problems, such as increasing system complexity, increasing delay and so on. JSCC (Joint Source Channel coding) is the main method to solve this problem. In this paper, two kinds of joint source and channel coding systems are built, and the joint source and channel model is constructed. Considering the quantization noise and channel noise, the joint channel is constructed by the GBG Gaussian background noise and Bernoulli Gaussian Impulse Noisemodel, which is composed of binary symmetric channel with quantization, given transition probability and inverse quantization. Firstly, the performance of the joint source channel coding system based on Real BCH coding is studied. Then, a joint source-channel coding system based on OFBS (Oversampled filter Banks) is built. A new error-correcting code based on oversampling filter banks in communication field is studied. The multiphase decomposition of OFBS is carried out, and a new threshold calculation method based on Bayesian hypothesis test is found to study the error-correcting performance of the system transmission. The Matlab simulation results show that the channel is in different transition probability conditions. The joint source and channel coding technology based on Real BCH coding has good transmission effect in image transmission. When the channel transition probability is better than that of the channel, the quality of the image is very good, and when the channel becomes worse, the quality of the image declines slowly. Because redundancy is added in the Real BCH coding process, the redundancy generated by OFBS is generated when the signal passes through the filter bank and has the information component of the original signal. When the channel quality is better, the high quality signal can be recovered at the receiving end. When the channel quality becomes poor, the recovered signal quality also decreases, but even when the channel quality is very bad, the recovered signal quality is still acceptable. The calculation of the new threshold based on oversampling filter banks is simple and effective.
【学位授予单位】：河北联合大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.22

【参考文献】