JPS593912B2 - Data conversion method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a data conversion method for encrypting data or decoding encrypted data, for example in the field of data communication using wide area networks.

For example, in satellite communication systems, loop network systems, branch line systems, etc., a third party is present in the network, and the third party can easily steal and manipulate data.

In recent years, there has been a trend toward wide-area data communications, but the wider the area, the more unsafe data communications become, which hinders the spread of data communications. Therefore, encryption is required to maintain secret communication between users. There are various encryption methods, the most representative of which is DES (Data Encryption), which was started by the U.S. Department of Commerce in July 1977.
tionStandard) method is known.

In this DES method, a message consisting of plain text is divided into blocks of 8 bytes (64 bits) and each 8 byte is encrypted or decrypted.

For example, when creating a ciphertext, as shown in FIG. 1, an 8-byte plaintext is encrypted with a key of 1, and the encryption process is repeated 15 times with different keys. In each session, we will appropriately combine the encryption method using the transfer method and the encryption method using the substitution method. The former is a method of permuting the bit arrangement of a given code according to certain rules, and the latter is a method of converting a code according to a predetermined table (called a substitution table). FIG. 2 is a diagram showing FIG. 1 in more detail.

Note that the number indicated by (d) in the figure represents the number of bytes of data, and the same applies hereinafter. In the DES method, 8-byte plaintext is first subjected to transliteration processing. This transliteration process is the third
Perform according to the table in the figure. How to read this table is as follows. That is, the numbers indicate the bit numbers of the input data; for example, the number 30 indicates the 30th bit of the 64 bits of input data. The array also indicates the order of output data. Therefore, by this transposition process, the 58th bit appears at the beginning, and the 50th and 4th bits thereafter appear at the beginning.
2...2,60,52...4,62,
54...5, 63, 55...7th bit, etc. are converted from the left to the right of the row, from the soil row to the row below. . Unless otherwise specified, the table below should be read in the same way as in Figure 3.
Next, this output 8 bytes are divided into 32 bits each on the right and left.

Hereinafter, these 4 bytes divided into two will be referred to as LO and RO, respectively. Next is R. Extend transliteration processing is performed on . This enlarged transliteration is a transposition method that overlaps some of the bits of a given code and rearranges the bit array so that the number of bits in the converted code is greater than before conversion. The output of character processing is 48 bits. This enlarged transliteration process is performed according to the table in FIG. Next, a 48-bit key K1 is given to this 48-bit output, and an exclusive OR of these is performed.

Next, the 48-bit output is subjected to a reduction/replacement process. This reduction substitution processing is a substitution method that reduces the number of bits of the code after conversion than before conversion, and the output of this reduction substitution processing is 32 bits. This reduction substitution process is performed as follows. That is, the given 48
The bits are sequentially divided into 6 bits and each is converted according to tables S1 to S3 in FIG. For example, the first 6 bits follow the table S1. How to read Tables S1 to S3 is as follows. The numbers indicate decimal numbers for 4 bits of output data; for example, the number 13 is "1101".
This shows the output data. Further, the rows and columns of the array indicate the upper 2 bits and lower 4 bits of input data, respectively. For example, if the 6-bit input data is 101001F", the first line corresponds to the number 13 of 3 nights and 1 tears, "1101".
appears as output data. As described above, each 6 bits is converted to 4 bits, and therefore 48 bits of input data is converted to 32 bits. Next, character transposition processing is performed on this output 32 bits.

This transliteration process is performed according to the table shown in FIG. Next, this output 32 bits and the previous L. Exclusive OR the This output is R1, and the previous R. is referred to as L1 as it is, and hereinafter each is referred to as R. , LO is repeated 15 times by changing the key to obtain Rl6 and Ll6. Next, Rl6 and Ll6 are concatenated to form 8 bytes.
Next, the 8 bytes of output are subjected to transliteration processing, and the 8 bytes of output are used as ciphertext.

This transliteration process is performed according to the table in FIG. In addition, to decrypt the ciphertext, it is sufficient to perform the inverse transformation of the above encryption process,
Since it is easily inferred, the explanation will be omitted here.

Now, let us consider a case where the above processing is performed by a data processing device such as a microprocessor.

The processing unit of the data processing device is a byte unit such as 1 byte or 2 bytes. Therefore, processing of data in which input or output data has a medium number of bits becomes complicated and requires a large amount of time. Therefore, when encrypting or decoding the DES method described above using a data processing device such as a microprocessor, there is a particular problem with enlarged transliteration processing and contracted transliteration processing, which must be repeated 16 times. So the impact is huge. SUMMARY OF THE INVENTION Therefore, it is an object of the present invention to provide a data conversion method that allows encryption or decryption processing using the DES method to be performed in units of one or more bytes.

In this invention, input data is first l (l is a positive number)
It is divided into bits and placed within m (m is a positive number) bytes.

m may be any positive number, such as 1, 2, or 4, which is a processing unit of a normal data processing device. Next, each of the preceding l bits is expanded or contracted to n bits (n is a positive number) and arranged within m bytes.

The expansion process is a process in which the number of bits of data after conversion is greater than that before conversion, and the contraction process is the opposite. The enlargement processing and reduction processing may be accompanied by transliteration or substitution. Each of the previous n bits is then reduced or expanded to l bits and placed within m bytes.

The reduction processing and enlargement processing may also be accompanied by transliteration or substitution. In this invention, the above process is repeated multiple times to convert into a new code.

Note that processing other than the processing described above may be added. FIG. 8 is a diagram for explaining one embodiment of the present invention. That is, in this embodiment, the transliteration process is first performed on 8-byte plain text. Each byte of the output data of this transliteration process is shown in each row of the table of FIG. Next, this 8-byte output is doubled to 16 bytes.

As shown in FIG. 9, this byte expansion process is a method in which input data is sequentially divided into 4 bits each and each bit is packed into the right half of 1 byte of output data. Next, the 16 bytes output from this are divided into 8 bytes each on the right and left sides.

Hereinafter, these 8 bytes divided into two will be referred to as RO and IO, respectively. Next is R. Extend transliteration processing is performed on . As shown in Figure 9, for byte A in 8 bytes of input data, this enlarged transliteration process adds 2 bits from other bytes in the input data to 4 bits in the byte, and outputs This is a method of padding 6 bits to the right side of the byte. Six bits of each byte of output data of this enlarged transliteration process are shown in each row of the table of FIG. On the other hand, the 48-bit (6-byte) key K1 is expanded to an 8-byte key k1.

This byte expansion process is similar to the previous byte expansion process, in which input data is sequentially divided into 6 bits and each bit is packed to the right of 1 byte of output data. Next, the 8-byte key k1 for the 8-byte output of enlarged transliteration processing
are given, and their exclusive OR is taken bit by bit. Next, the 8 bytes of output are subjected to a reduction substitution process. As shown in Figure 9, when we look at a certain byte B in 8 bytes of input data, we obtain 4 bits of corresponding data for 6 bits in the byte and transfer it to the right side of the output byte. This is a method to fill in 4 bits. The 4 bits of each byte of the output data of this abbreviated character substitution process are determined according to S1 to S3 in the table of FIG. Next, the 8 bytes of output are subjected to character transposition processing.

The 4 bits of each byte of the output data of this transliteration process are the 5th
Indicated by each row of the table in the figure. Next, this output 8 bytes and the previous 1.

The exclusive OR is performed for each bit. This output is r1,
The previous R. are set as 11, and each is R. , 1O by changing the key and performing the same process as above for 1O.
Repeat 5 times to obtain R,6 and 116. Next, this Rl6
and 116 are concatenated to make 16 bytes.

Next, this 16-byte output is reduced to 8 bytes. This byte reduction process is a method in which four bits of adjacent bytes of input data are combined to form one byte of output data, as shown in FIG. Next, the 8 bytes of output are subjected to transliteration processing, and the 8 bytes of output are used for the cipher text. Each byte of the output data of this transliteration process is shown in each row of the table of FIG. Furthermore, ciphertext decoding can be done by performing inverse transformation of the above-mentioned encryption process, and since it can be easily inferred, the explanation will be omitted here.

One embodiment of the present invention has been described above. According to the present invention, for example, when performing encryption processing or decryption processing using the DES method, the processing can be performed in units of one or more bytes. This has the effect of simplifying the process and shortening the processing time.

[Brief explanation of drawings]

Figure 1 is a diagram for explaining the encryption process using the DES method, Figure 2 is a diagram showing Figure 1 in more detail, and Figures 3 to 7 are conversion tables used in the process of Figure 2. FIG. 8 is a diagram for explaining encryption processing according to an embodiment of the present invention, and FIG. 9 is a diagram for explaining the functions of each part in FIG.

Claims (1)

[Claims] 1 Divide the input data into l (l is a positive number) bits and arrange each bit in m (m is a positive number) bytes, and then divide each of the l bits into n (n is a positive number) bits. is expanded or contracted into m bytes, and then for each of the n bits, l is
A data conversion method characterized by repeating a process consisting of performing reduction processing or expansion processing to bits and arranging the data within m bytes a plurality of times.