Method for decoding sequential algebraic cascade convolutional codes for mobile communication systems

The development of radio communication technologies allows the introduction of the latest electronic communication services that impose strict requirements on the quality of data transmission. To increase the efficiency and reliability of data transmission in these electronic communications systems, it is advisable to use parallel and sequential cascade code structures. It is shown that the construction of sequential cascade code structures for mobile communication systems can be carried out on the basis of various component codes and corresponding decoders. This makes it possible to adjust the characteristics of these codes depending on the existing requirements and parameters of the communication channel. A sequential algebraic cascade coding scheme is proposed, where the outer stage is implemented based on a non-binary Reed-Solomon block code, and the inner stage is implemented using a non-recursive algebraic convolutional code. A block interleaver is used between these stages. A distinctive feature of this scheme is the use of an algebraic convolutional code with the maximum achievable correction capability. This is achieved by constructing this code based on the generator polynomial of the Reed-Solomon code. A method for decoding sequential algebraic cascade convolutional codes based on a combination of two decoders has been developed. At the first stage, the internal code is decoded using ordered statistics. Within this stage, the most reliable basis is initially found based on the reliability of the symbols of the received sequence. Test error vectors of a given Hamming weight are added to the found basis, resulting in a set of test code blocks. The search for the estimate of the transmitted code block is carried out on the basis of minimizing the weighted Hamming weight between the formed code blocks and the accepted estimate of the code block. At the next stage, the accepted vector for the outer code is formed by inversely permuting the found estimate of the code block in the block interleaver. At the final stage, algebraic decoding of the Reed-Solomon code occurs based on the Berlekamp-Massey algorithm.