МОДИФИКАЦИЯ СПОСОБА ОЦЕНИВАНИЯ МЯГКИХ РЕШЕНИЙ СИМВОЛОВ И ЕГО ВЕРИФИКАЦИЯ

Abstract

To implement permutation decoding (PD), the soft symbol decisions (MSDs) are required for each element of the codeword processed by the decoder. Similar solutions are required for both binary codes and non-binary code structures [1-4]. In [5], it was shown that for binary codes it is advisable to have a maximum score equal to 710. It should be noted that the author of [5], when using integer MRS for convolutional codes, did not mean the PD method, in which it is necessary to rank symbols by reliability over the length of the code vector. The paper proves that this practice of forming the MRS turns out to be inefficient for most communication channels with additive white Gaussian noise (AWGN). It is proposed to lower the value of the index of the maximum estimate of the MPC to the value of 310 or 112 . This will halve the symbol ranking time, which improves the performance of the data receiver processors. A rigorous probabilistic estimate of the proposed approach is given.