JPH05793B2 - - Google Patents

Description

[Detailed description of the invention]

[Technical Field of the Invention] The present invention relates to an encoding method for digital magnetic recording, and particularly to an encoding method for digital magnetic recording suitable for magnetic disks. [Technical Background of the Invention and Problems Therewith] In magnetic disk devices that magnetically record and reproduce digital information, various self-clockable modulation methods have been developed as recording density increases. Various methods have also been proposed in which original data is encoded according to certain rules and then recorded using NRZI. The following properties are desirable for such an encoding method. (1) Self-clocking is possible. (2) Minimum magnetization reversal should be wide. (3) The signal detection window should be wide. (4) The maximum magnetization reversal interval is small. (5) The amount of encoding/decoding hardware is small. (6) Contain many synchronization pulses within the sync region. (7) It is easy to cue up the data area. The so-called RLL (Run-Length-Limited) (2,
7) Among these encoding methods, the code is one of the best methods as it allows a wide minimum magnetization reversal interval. However, these encoding methods have points that should be improved as will be described later. A conventional example will be described below with reference to the drawings. Table 1 was disclosed in U.S. Patent No. 3,689,899 (2,
7) Code conversion table. Table 2 is a (2,7) code conversion table disclosed in US Pat. No. 4,115,768.

【table】

【table】

[Table] The former frames the original data string and then converts it into code words, while the latter converts the original data string into code words serially and sequentially. , the same data word and code word are used, and although there is a slight difference in the amount of hardware in (5), which is a desired property of the encoding method described above, properties (1) to (4) are completely different. It's the same. However, the properties shown in (6) and (7) each have their advantages and disadvantages. That is, the former includes many synchronization pulses in the sync area, but requires some work to cue the data area, while the latter allows easy cueing of the data area, but the amount of synchronization pulses included in the sync area is small. This point will be explained below. In magnetic disk devices, recording and reproduction are performed using a self-clockable modulation method, and a so-called sync area is provided at the beginning of each sector for generating synchronization pulses by a PLL. Generally, all "0" data is used as the original data for this sink area since no means for outputting a specific data pattern is required. In addition, the boundary between this sink area and the following data area can be identified from the data string of the code word by “1” in the original data.
This is done by detecting the appearance of by some method. Therefore, the desired properties of the code word pattern used in the sync area include as many "1's" as possible in order to quickly form a stable synchronization pulse by the PLL, and the boundary between the sync area and the data area. Another advantage is that it is easy to identify (cue) the divisions of encoded data. First, let us consider the case where the conversion table shown in FIG. 1 is used. Here the original data oak “0”
The data word “000” corresponding to is the code word “100100”
Moreover, since this corresponds to the highest frequency in this encoding, it is superior in that it has the largest synchronization energy. However, according to this method, locating the data area is not always easy. For example, if the original data string is

In the case of [Formula], the area before the first "1" (cue code) is the sink area, and the area after it is the data area.

[Formula] indicates a word break during encoding/decoding. ) The encoded data string is

[Formula], but depending on how it is separated when reading

Interpreted as [expression] Re

It may be decoded as [Formula], and cannot be decoded uniquely. This is because the code word "100100" corresponding to the data word "000" includes two "1"s in its 6 bits. Therefore, in order to avoid such inconvenience, for example, if the original data string is

[Formula] (x is optional), add one more bit of “0” before “1” to ensure that the cue part is

[expression] or

It is necessary to do some tricks such as making it as [Formula]. On the other hand, considering the case where the conversion table shown in FIG. 2 is used, the code word "000100" corresponds to the data word "000", and only one "1" is included in the code word. Therefore, by using this position of "1" to create a timing clock such that the code word is a pair of 2 bits each, such as 00⌒01⌒00⌒, the inconvenience in the above case can be avoided. Recognize. However, this case has the disadvantage that only half the synchronization pulses can be obtained compared to the case where the conversion table shown in FIG. 1 is used. [Object of the Invention] The present invention has been made in view of the above-mentioned drawbacks of conventional encoding methods, and provides an encoding method that allows easy data cueing and has large PLL synchronization energy. It's about doing. [Summary of the Invention] As an encoding rule for a variable word length PLL (2, 7) code, the present invention uses data words used in the sink area as data words divided into 2-bit units, and divides the data words into 2-bit units. is specified to be converted into a code word containing only one "1" in 4 bits. [Examples of the Invention] Hereinafter, the present invention will be explained in detail with reference to Examples. Table 3 shows a conversion table according to an encoding method according to an embodiment of the present invention.

[Table] As can be seen from this conversion table, in this embodiment, the code word "1000" is assigned to the data word "00". Therefore, according to this conversion rule, the code word used for the sink area has 1 "1" in 4 bits.
According to Table 2 mentioned above, 6 will be included.
The sync region contains a large number of synchronization pulses at a higher rate than 1 bit in a bit, and the value is closer to that of 1 bit in 3 bits in the method shown in Table 1. Furthermore, since there is only one "1" in 4 bits, data can be separated by forming a timing clock of 2 bits each like 10⌒00⌒ in the same way as shown in Table 2. Uniqueness is also ensured. In addition, a 4-bit code word is used in the sync area because a 2-bit code word cannot be self-clocked and contains at least two 0's between successive 1's. It can be seen that this method is the best method that satisfies the conditions that the code word contains only one "1" and as many synchronization pulses as possible. As a specific procedure for encoding/decoding using the present invention, it is also possible to sequentially convert each word while framing as shown in US Pat. It is also possible to serially convert 2 bits at a time, as shown in FIG. [Effects of the Invention] As described above, according to the present invention, the drawbacks of the conventional encoding method are eliminated, the cueing of the data area is easy, and the PLL in the sink area is
An encoding method is provided that allows a large number of synchronization pulses to be used. [Other Embodiments of the Invention] Table 4 shows conversion tables according to other embodiments of the invention.

[Table] This embodiment also provides the same effects as the above-mentioned case. Other similar conversion rules can be created within the scope of the claims of the present invention, but all of these are included in the present invention.

Claims (1)

[Claims] 1 A word used when 1 bit of a data word is code-converted at the ratio of 2 bits of a code word, contains at least two or more 0s between 1 and 1 of consecutive code words, and is encoded according to a data pattern. In the encoding method where the length varies, the data used in the sink area is a repetition of data words divided into 2-bit units,
An encoding method for digital magnetic recording characterized by converting this 2-bit data word into a code word containing only one 1 in 4 bits.