Coded data generation or conversion – Digital code to digital code converters – To or from minimum d.c. level codes
Reexamination Certificate
2000-10-24
2003-07-22
Tokar, Michael (Department: 2819)
Coded data generation or conversion
Digital code to digital code converters
To or from minimum d.c. level codes
C341S059000, C341S068000, C341S069000, C375S287000, C375S340000, C714S709000
Reexamination Certificate
active
06597295
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a decoding apparatus and decoding method for decoding an RF signal reproduced and read from a recording medium on which data is recorded in the form of RLL (Run Length Limited) codes, in accordance with at least one comparator level, thereby to output channel-bit data.
To transmit data or to record data on a recording medium such as a magnetic disk, an optical disk or a magneto-optical disk, the data is modulated to data that may be well transmitted or recorded. Known as one method of modulating data is block encoding. In the block encoding, a stream of data is divided into blocks (hereinafter referred to as “data words”), each consisting of m×i bits. The data words are converted to code words in accordance with appropriate encoding rules. It should be noted that each code word is composed of (n×i) bits. The code word has a fixed length if i=1. If i is of maximum value i
max
=r, where r is 2 or more, the code word has a variable length. The blocks, or data words, shall be called variable-length codes (d, k; m, n; r). The value i is called “restraint length” here, and “r” is the maximum restraint length. The value d is the smallest number of “0s” that may exist between two “1s” in a series of codes, and the value k is the largest number of “0s” that may exist between two “1s” in the series of codes.
The modulation applied to compact disks (CDs) will be described as an example of a method of modulating data. EFM (Eight-to-Fourteen Modulation) is used to record data on a CD. More precisely, an 8-bit data word is converted to a 14-bit code word (channel bits), three margin bits are added to the 14-bit code word, thus reducing the DC component in the code word (i.e., EFM word), and the code word is then recorded by means of NRZI modulation. The 8-bit data word is converted to a 14-bit code word and the margin bits are added, such that the smallest number of “0s” and largest number of “0s” may be 2 and 10, respectively. Hence, the parameter of this modulation is (2, 10; 8, 17; 1). The minimum inversion interval T
min
is 3 (=2+1)T, where T is the interval between bits in a channel-bit stream (a series of recorded waves). The maximum inversion interval T
max
is 11 (=10+1)T. The width T
w
of the detection window is (m
)×T, where T is the interval T
data
between data items in a stream of data items, and has the value of 0.47 (=8/17)T.
The minimum length d′ defined by identical symbols in the code word that has been NRZI-modulated is: d′=d+1=2+1=3. On the other hand, the maximum length k′ defined by identical symbols in the code word is: k′=k+1=10+1=11.
With the CD described above, the recording density can be increased if bits are compressed in the linear-speed direction. When the bits are compressed, the minimum bit length that corresponds to the minimum inversion interval T
min
will decrease. If the minimum bit length decreases excessively, it will be difficult to detect the bits, causing an error.
The error rate in the process of reproducing data from a disk will increase if a skew occurs, that is, if the optical pickup is inclined to the recording surface of the disk. The skew is classified into two types in accordance with the direction in which the optical axis of the pickup inclines to the disk. The first type occurs in a tangential direction in a plane parallel to the direction in which the pickup moves and perpendicular to the disk. The second type occurs in a radial direction in a plane parallel to the radial direction of the disk and perpendicular to thereto. The error rate increases in the tangential direction at a relatively early stage of the process of reproducing data. Both types of skews inevitably reduce the design margin of the system.
The distribution of errors in the minimum length defined by identical symbols was checked in the two skew directions. The errors due to the skew in the tangential direction occurred when the minimum length defined by identical symbols was short. That is, the error rate increased because the code word having a length T
min
(d′) was decoded to a data item having a length of T
min
−1 (d′−1). It was found that many errors were made in the EFM system when 3T (i.e., minimum inversion interval T
min
) was changed to 2T, where T is the interval between bits in a series of recorded waves.
The waves reproduced have their form more distorted when the recording density is increased in the linear-speed direction or when a skew of a large angle takes place in the process of reproducing data from the disk. Thus, the error rate increases as the code word having a length T
min
(d′) is decoded to one having a length T
min
−1 (d′−1) and the code word having a length T
min
−1 (d′−2) is decoded to one having a length T
min
−1 (d′−3). In other words, errors occur in the EFM system when 3T (i.e., minimum inversion interval T min), where T is the interval between bits in a series of recorded waves, is decoded to 2T, 2T is then decoded to 1T, and further 1T is decoded to 0T. Here, “0T” means that the output is too small or too large to cross the comparator level, or that the output cannot be detected at all.
RLL (1,7) codes, which are often used in the modulation performed in magneto-optical recording systems will be described. The parameter of an RLL code (1,7) is (1, 7; 2, 3; 2). The minimum inversion interval T
min
is 2 (=1+1)T and the maximum inversion interval T
max
is 8 (=7+1), where T is the interval between bits in a series of recorded waves.
In the process of reproducing data, using the RLL (1, 7) code, an error is made when the recording density is increased in the linear-speed direction or when a skew of a large angle occurs. That is, the minimum inversion interval T
min
, i.e., 2T, is decoded to 1T, and further 1T is decoded to 0T, where T is the interval between bits in a series of recorded waves.
The value 0T, i.e., an error that can no longer be detected, is often obtained in the case where d=1 in the RLL code (1, 7). This is inevitably because it is believed that 2T is more easily decoded to 1T and then to 0T when d=1 than when d=2, though the waves reproduced must be much distorted to be detected as an error when d=2 and 3T is therefore decoded to 2T, thence to 1T, and thence to 0T.
Some asymmetry margin is provided in the manufacture of optical disks. It is therefore necessary to take into consideration a case where the waves reproduced are asymmetric with respect to the center level.
Viterbi decoding may be employed to reduce the error rate in the process of reproducing signals. Viterbi decoding is one of decoding methods in which code errors are minimized, thereby finding the geometrically shortest way possible, discarding other ways of less likelihood. In other words, Viterbi decoding is a decoding method in which the value having the highest likelihood is searched for in a simple manner. An algorithm for compensating for the minimum inversion interval T
min
can be utilized in Viterbi decoding.
Viterbi decoding, however, is disadvantageous in that a complicated circuit of a large scale must be used to perform it. This decoding method cannot decode data without making errors, if the data has been reproduced from a recording medium that has an asymmetry margin like an optical disk. The circuit for performing Viterbi decoding should therefore be designed to cope with the asymmetry margin and is, inevitably, more complicated in structure.
A recording medium, such as an optical disk, may hardly have a sufficient skew margin. The skew margin is inadequate, particularly in the tangential direction.
It is difficult to reproduce data from a recording medium having high recording density, such as an optical disk, at with a stable minimum inversion interval T
min
. Inevitably, the error rate increase
Nguyen Khai
Tokar Michael
LandOfFree
Decoding apparatus and decoding method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Decoding apparatus and decoding method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Decoding apparatus and decoding method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3071366