Research And Design For Concatenation Scheme Of Polar Codes With Convolutional Codes

Polar codes have attracted widespread attention because of the advantages of definite coding and decoding structure and accessible capacity.At the critical moment of the development of 5G mobile communication era,polar codes have been identified as the coding scheme of control channel in 5G enhanced mobile broadband scenario.However,the performance of polar codes is not close to Shannon limit because of the limitation of code length in practical applications.In order to improve the error-correcting performance of medium and short code length polar codes,a scheme of concatenating convolutional codes with polar codes is proposed in this paper.The main work is as follows:The concatenated coding and decoding scheme of convolutional code and polar code is completed,and the simulation on the platform of MATLAB is carried out.The concatenated codes designed in this paper are interleaved and concatenated.The concatenated decoding algorithm which combines parallel SC decoding and Viterbi decoding is adopted.The concrete scheme is designed.The simulation results are given and the expected goals are achieved.It is proved theoretically that the frame error rate of the concatenated code decreases exponentially with the increase of the length of the concatenated code L,and the error propagation of SC decoding can be effectively eliminated,so as to improve the error correction ability.The simulation results show that the bit error rate is very low when the signal-to-noise ratio is greater than 1.6dB,which indicates that the concatenated code has good asymptotic performance when the channel quality is slightly better.An improved parallel linear SC decoding structure is implemented.Two key components are improved: the normalized storage structure based on SRAM and the two-level system and the update unit.It improves the scalability of code length,reduces the storage area and has scalability.The hardware logic synthesis is completed on Synplify platform,and the function verification is completed on Modelsim platform.The experimental results achieve the expected goal.