Study On Blind Recognition Of Linear Block Code And Interleaver

Channel coding, which mainly use the sender information sequence to makes kinds of processing, makes correlation between the originally separate information bits. At the receiving end, this correlation is used to check and correct the errors when the information element is transmitted in the channel transmission. Interleaving which exchange the position of code elements can weaken the relevance and improve the reliability of codes. The blind recognition of channel coding, which use the unknown encoded data, can recognize the encoding method and encoding parameters without the encoding information. The blind recognition analysis of channel coding is widely used in the field of intelligent communication, communication investigation and spread spectrum communication field.Channel coding has unique coding structure not only can achieve its excellent performance but also left mathematical features to identify analysis. Blind identification of channel coding parameters mainly concentrated in the convolution code and convolution interleaving, and the methods of the block code limited to linear matrix method. This thesis will improve the method in chapter Three. The recognition of block interleaving stay on the interleaving length, and the recognition of the position relationship and subsequent coding parameters are not perfect. From a macro sense, there are significant topics for the identification of block code encoding, and there are not many article involving this subject.This thesis focuses on the parameters blind identification of linear block code and block interleaving. The main work of the thesis includes the following aspects:(1) Briefly introduces the basic knowledge of the channel coding, describes the encoding process of BCH code and the principle of block interleaving in detail. And pointing out that the linear block code and block interleaving is the object of this thesis.(2) In order to solve the blind identification problem of the block interleaving based on BCH code, we come up with a blind identification approach based on Binary Column Gaussian Elimination and depth distribution in this thesis. Firstly, the length and synchronization of interleaver are identified by Binary Column Gaussian Elimination, and then the position relationship is determined. Secondly, code length is determined after determining the position relationship. Thirdly, the generator matrix is recognized by the feature of depth distribution, typical generator matrix is produced by using Gauss Elimination. The recognition has good performance at identification the interleaver length, synchronization parameters, position relationship of interleaver, code length and generator matrix.(3) A Binary Iteration Column Elimination and depth distribution algorithm is proposed to the blind estimation problem of the linear block code.Firstly, the length and synchronization of the linear block code are estimated by Binary Column Gaussian Elimination. Secondly, code rate is estimated by the depth distribution. Thirdly, the generator matrix is recognized by the Binary Iteration Column Elimination, typical check matrix is produced by using Gauss Elimination. The recognition has good performance at identification the code length, synchronization parameters and check matrix.(4) In order to solve the blind identification problem of the linear block code, we come up with a blind identification approach based on the logarithmic likelihood ratio in this thesis. First of all, the expression of logarithmic likelihood ratio equation is derived. According to fact that "0" and " 1" occured in the linear block code at the equal probability,we get a posteriori logarithmic likelihood ratio calculation formula. Then we use different check matrix posterior probability mathematical characteristics of logarithmic likelihood ratio to identify the encoding of coding. The method has a better performance to estimate the linear block code encoding, and has a small amount of calculation. The method is more successful to estimate the linear block code encoding, and has a small amount of calculation.This thesis studied the blind identification approach of linear block code, blocks interleaving parameters and carried out the simulations with MATLAB. Research shows that the mixed BCH code parameters of blind identification algorithm is more close to the actual situation of channel transmission, practicality is better. Blind identification algorithm of binary linear block code parameters more precise, the recognition of generator matrix is more stable; Identification of a block code encoding provides the basis of linear block codes parameters.