Neural network to convert from a positional number system to the code polynomial system of residue class

摘要

u043du0435u0439u0440u043eu043du043du0430u00a0 network u0434u043bu00a0 u043fu0440u0435u043eu0431u0440u0430u0437u043eu0432u0430u043du0438u00a0 of positional system u0441u0447u0438u0441u043bu0435u043du0438u00a0 code u043fu043eu043bu0438u043du043eu043cu0438u0430u043bu044cu043du043eu0439 system classes deduction module of p (z) = z4 + z3 + 1, u0441u043eu0434u0435u0440u0436u0430u0449u0430u00a0 input layer neuron s, u043eu0441u0443u0449u0435u0441u0442u0432u043bu00a0u044eu0449u0438u0439 reception and distribution at the entrance of the original u043fu043eu043bu0438u043du043eu043cu0430 a (z) presented in binary form (0 (z), (z), 2 (z).13 (z), 14 ((z)), and the output layer neurons, u043eu0442u043bu0438u0447u0430u044eu0449u0430u00a0u0441u00a0 in that it u0438u0441u043fu043eu043bu044cu0437u0443u0435u0442u0441u00a0 method direct u0441u0443u043cu043cu0438u0440u043eu0432u0430u043du0438u00a0 where conversion of a (z) used in ra u0441u0448u0438u0440u0435u043du043du043eu043c field gf (p), u043fu043eu043bu0438u043du043eu043cu0438u0430u043bu044cu043du0443u044e class u043eu0441u0443u0449u0435u0441u0442u0432u043bu00a0u0435u0442u0441u00a0 deduction system by a set of constants, u00a0u0432u043bu00a0u044eu0449u0438u0445u0441u00a0 equivalent degrees of justification 2i and u043au043eu044du0444u0444u0438u0446u0438u0435u043d in the u0441u0442u0435u043fu0435u043du00a0u0445 grounds u0430i,the system of class reduction, with the first layer of the neural network consists of 15 neurons 1 - 15, and which are connected to the respective u0432u0445u043eu0434u0430u043c 4 neuron s 16 to 19 of the second u0441u043bu043eu00a0 network u043eu0441u0443u0449u0435u0441u0442u0432u043bu00a0u044eu0449u0438u0445 addition module 2 u0440u0430u0437u0440u00a0u0434u043eu0432 values received from the first u0441u043bu043eu00a0 exits neurons in accordance with the matrix u044212, stories m the entrance 16 of a neuron.u043eu0441u0443u0449u0435u0441u0442u0432u043bu00a0u044eu0449u0435u0433u043e calculating zero 0 (z) u0440u0430u0437u0440u00a0u0434u0430 code u043fu043eu043bu0438u043du043eu043cu0438u0430u043bu044cu043du043eu0439 system classes deductions is connected to exits 1, 5, 6, 7, 8, 10, 12 and 13 on the first u0441u043bu043eu00a0 entrance 17 her ron, u043fu043eu043bu0438u043du043eu043cu0438u0430u043bu044cu043du043eu0439 system classes deductions is connected to exits 2, 6, 7, 8, 9, 11, 13 and 14 of the first u0441u043bu043eu00a0 entrance 18 neurons, neuron, u043eu0441u0443u0449u0435u0441u0442u0432u043bu00a0u044eu0449u0435u0433u043e calculation of 2 (z) u0440u00a0u0434u0430 code to exits 3, 7, 8, 9, 10, 1214 and 15 of the 19 neurons, neuron, u043eu0441u0443u0449u0435u0441u0442u0432u043bu00a0u044eu0449u0435u0433u043e calculation third 3 (z) u0440u0430u0437u0440u00a0u0434u0430 code to exits 4, 5, 6, 7, 9, 11, 12 and 15 on the first u0441u043bu043eu00a0 and exits on the 16 - 19 wto whom u0441u043bu043eu00a0 u00a0u0432u043bu00a0u044eu0442u0441u00a0 exits the device u043fu0440u0435u043eu0431u0440u0430u0437u043eu0432u0430u043du0438u00a0.