The article proposes a method that solves the problem of hiding decimal numbers from unauthorized access while simultaneously detecting error packets in them and correcting single errors. The method is based on the encoding of decimal numbers by binary-coded permutations. With this encoding, each digit of a decimal number is first converted into a binary-coded permutation using a special encryption table and, after further mixing with other permutations, is transmitted to the receiver. On the receiving side, it is checked for errors and, after correcting them with the help of a key, it is converted into decimal digits. Since each digit of a decimal number contains 1 of 10 numbers, their encoding requires 10 permutations and, accordingly, at least 4 permutation elements: 0, 1, 2, 3. These elements form a set of 24 permutations, which consists of 10 used and 14 redundant permutations. This redundancy, as well as the natural redundancy of binary-coded permutations, allows them to detect packets of errors and correct single errors.