Research On Blind Identification Algorithms Of Convolutional Code Parameters Under Error Conditions

Recently,with the development of channel coding and the application of intelligent mobile communication systems,the study of blind identification technology of Convolutional code parameters is also becoming increasingly important.Convolutional code is a very promising coding method in channel coding,the blind recognition technology of convolutional code has long become a research hotspot in the intelligent mobile communication system.The purpose of blind identification of convolutional code parameters is to identify the coding parameters of convolutional code used at the sending terminal only based on the intercepted data bit flow containing error codes according to a certain blind identification algorithm without any prior knowledge.This thesis mainly studies the blind identification algorithm of the code length of convolutional code under Binary Symmetric Channel(BSC)and the blind recognition algorithm of the check matrix of convolutional code under Additive White Gaussian Noise(AWGN)channel.The main research content are divided into the following two parts:In the first part,we mainly investigate the blind identification of code length of convolutional codes under BSC channel.To better solve the problem of blind identification of code length in the noise environment,we analyzes the influence of the position of the error bits on the Gauss-Jordan Elimination Through Pivoting(GJETP)algorithm,Combined with the own structural properties of the convolutional code.By performing a GJETP transformation of the convolutional code data matrix,as well as the statistical analysis,a blind identification algorithm for code length based on iterative statistical polarity is proposed.The optimal threshold obtained under the algorithm by histogram analysis is given.Meanwhile,the optimal threshold formula under the proposed algorithm is derived.Finally,through experimental simulations,we show that the proposed recognition algorithm can effectively improve the recognition rate of code length.In the second part,we study the blind identification algorithm of convolutional code parameters under the AWGN channel.In order to improve the parameter recognition performance of the convolutional code under the AWGN channel,we fully makes full use of the reliability information of soft information carrying,further studies and improves the existing GJETP algorithm,and proposes a Gaussian elimination element algorithm based on soft information,which is called the Soft-GJETP algorithm.The proposed algorithm can effectively reduce the propagation of error bits in the Gaussian elimination process,compared with the general GJETP algorithm.Subsequently,a convolutional code parameter recognition method based on the convolutional algorithm and the proposed Soft-GJETP algorithm.Through experimental simulations,we can find that the algorithm presented in this paper can significantly improve the convolutional code parameter recognition performance under the AWGN channel.