JPH0744461B2 - Binary code demodulation method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a block conversion type run length limited.
ed) code or a convolution type run-length limited code demodulation method.

[0002]

2. Description of the Related Art In digital transmission and magnetic and optical recording / reproducing systems, normal information is transmitted or recorded as a series of symbols. Such symbols together form an alphabet (often a binary alphabet; code). In the case of a binary code, one symbol, for example, "1" is recorded on a magnetic disk or tape as a transition between two magnetization states by an NRZM (NRZ-mark) code, or whether there is a recess called a pit. As an optical disc. Then, record the other symbol "0" as the absence of such a transition.

By the way, in such transmission and recording, a data bit string is so-called RLL (Run Le).
It is desired to perform modulation to the ngth Limited) code. That is, some systems require self-clocking, which is why a sequence of symbols to be transmitted or recorded is transmitted or recorded as enough transitions to generate a clock signal used for detection or synchronization. I have to do it. Therefore, the maximum transition inversion interval Tmax is restricted. On the other hand, the minimum inversion interval Tmin is also regulated to limit interference between symbols. This minimum inversion interval Tmin is also regulated by the information density of the recording medium. This is because, at a predetermined minimum distance between two adjacent transitions on the recording medium, if the minimum time interval Tmin corresponding to the recorded signal increases, the information density also increases at the same rate. The minimum required bandwidth Bmin is also associated with the minimum distance between transitions (Bmin = 1 / 2Tmin).

From the above, it is desired that the bit string that has been transmitted and the like satisfy the following d rule and k rule. That is, it is preferable to use the RLL code. d rule: Two “1” s are separated by at least d consecutive “0” strings. k rule: The maximum length of a string of consecutive "0" s is k.

On the other hand, when the information channel does not carry direct current, as in the case of a general magnetic recording channel,
When there is a demand for simplifying the demodulation side, it is necessary to make the bit string in the information channel hardly contain a DC component.

From the above requirements, there is a need for a binary code modulation method that suppresses the DC component while limiting the minimum distance Tmin and the maximum distance Tmax between transitions.

One is to add a redundant bit string to the block coding so that the d and k rules are met in the channel bit string, while reducing the DC imbalance as a whole.

That is, in this binary code modulation method, a data bit sequence is sequentially divided into m-bit data words, and these data words are converted into n ₁ and bit information words, respectively. Here, m <n ₁ and 2 ^m RLL (Run) are selected from the combination of 2 ⁿ¹ bits.
Only Length Limited) codes are selected and these are made to correspond to data words as information words.

Of course, if this is left as it is, when the information words are connected, that is, when the information words are connected, d
The rules and k-rule cannot be satisfied. For example, when an information word ends with k consecutive "0" s and a subsequent information word starts with "0", the k rule is no longer satisfied. The same applies to the d rule.

Therefore, a redundant word of n ₂ bits is added between the information words, and the information word and the redundant word form a channel word. Channel like this
For words, the choice of redundant words allows the d and k rules to be satisfied over successive channel words.

Further, in this method, this redundant word is also used for suppressing DC imbalance. That is, a redundant word that satisfies the k rule and the d rule is selected over successive information words, and a redundant word that minimizes the DC imbalance is selected from among the redundant words that can be realized in this way.

In order to suppress the DC imbalance, this method uses digital summation (DSV). This will be described below with reference to FIG.

In FIG. 9, the information word W is used as an example.
I is composed of 14 bits, and the redundant word WR is composed of 3 bits. The information word WI is obtained, for example, by converting an 8-bit data word (not shown) by table lookup. Where the information word WI is d
= 2, k = 10, so as to satisfy the dk rule.

Information word WI _i followed by information word WI
When continuing _{i + 1} , (000), (100), (010) and (001) can be realized as the redundant word WR _i from the dk rule (FIGS. 9B, C, D and E). . The digital sum in each of these cases is represented as the integral of the digital waveform indicated by WF in FIGS. Where the subsequent information word WI _{i + 1}
If the end of is used as the evaluation time point, and (000) is used as the redundant word WR _i , the digital sum becomes +7. Similarly, (100), (010), and (001) are +1, +3, and +5, respectively. Therefore, redundant word W
(100) is used as Ri. Then, the DC imbalance can be minimized at least at each evaluation time point.

For example, when the spectrum of the data bit string is as shown in FIG. 10A, if a redundant word determined with the digital sum as an evaluation reference is added, a spectrum in which the DC component is suppressed is obtained as shown in FIG. 10B.

Of course, the end of the information word WI _{i + 2} may be selected as the evaluation time point of the redundant word WR _i .

By the way, in such a code modulation method, since the digital sum is used as the evaluation reference, only the DC component can be stably suppressed, and stable suppression over a low frequency band is not guaranteed at all. And this leaves some problems unsolved, especially in the case of optical recording.

The first problem relates to stains (fingerprints) and scratches formed on the recording medium. That is, in the optical reproducing apparatus as shown in FIG. 11, the transition is represented by the presence or absence of pits, and the detection signal (FIG. 12A) is obtained from the detector 1 according to the presence or absence of these pits. By the way, recording medium 2
If there is dirt or scratches on the top, the dirt or the like appears in the detection signal as an asymmetric AM component. For example, FIG. 12C
If there is a fingerprint interference as shown in Fig. 1, the detected signal will be
As shown in 2B, the image is distorted and faithful reproduction cannot be performed.

In order to deal with such a situation, it is necessary to avoid the spectrum region due to dirt and scratches and transmit the channel bit string. For example, if the spectrum due to dirt or the like is as shown in FIG.
The spectrum of the channel bit string should be as shown in (b).

The second problem relates to the servomechanism. That is, as shown in FIG. 11, the optical reproducing apparatus controls the servo mirror 3 and the like based on the envelope (FIG. 14A) of the detection signal to perform tracking and keeps the focus of the objective lens 4 within the depth of focus. ing. However, when the spectrum of the detection signal itself is superimposed on the spectrum of such an envelope, the detection signal in the spectrum region of the envelope acts as a disturbance, and as a result, the tracking meanders or the focus deviates from the depth of focus. . Further, a saturation phenomenon occurs in the tracking loop due to the phase compensation, and a track jump occurs.

In FIG. 11, 5 is a servo circuit,
6 is a laser and 7 is a half mirror.

[0022]

In consideration of such circumstances, the present inventor first examined using the digital sum as an evaluation criterion. As described above, the digital summation is obtained as the integral of the digital waveform represented by WF in FIGS. Instantaneous values of integration are shown in FIGS.
Are indicated by broken lines. And the subsequent information word WI
When evaluating the redundant word WR _i at the end of _{i + 1} ,
The determination may be made based on the integrated value at the end of this information word WI _{i + 1} .

By the way, the instantaneous value of the integration shown by the broken line in FIGS. 9B to E is the step response of the digital waveform WF for the transfer function composed of the integral elements as shown in FIG.
In the frequency characteristic of such a transfer function (FIG. 15), the gain of the DC component is sufficiently large, but the gain gradually decreases over the low frequency band and the high frequency band. This means that the DC component can be sufficiently suppressed, but
This means that the degree of suppression gradually decreases in the low frequency band and the high frequency band. In this way, when the evaluation criterion based on the digital sum is associated with the transfer function, the degree of suppression of the spectrum of the modulated signal can be determined from the frequency characteristic of the transfer function. In other words, if a transfer function that defines a frequency band of interest such as a frequency band to be suppressed is obtained in advance, an evaluation criterion that suppresses the frequency band of interest can be obtained.

The present invention has been made from such a point of view, and an object thereof is to provide a binary code demodulation method capable of arbitrarily designing a frequency band to be suppressed by selecting redundant bits.

In order to achieve such an object, the present invention defines a band which covers a band of interest, for example, a direct current component to a spectrum region of fingerprints or scratches on a recording medium or a band necessary for servo. The value of the step response of the function to be evaluated is obtained, and the redundant word is selected so that this value becomes small.

[0026]

According to the present invention, the above-mentioned redundant word is used for code-modulating an m-bit data word into an n-bit channel word consisting of an n _1- bit information word and an n ₂ -bit redundant word. Of the n-bits in the transfer function that defines the frequency band to be suppressed at the end of the information word following, or at the end of the last information word of the predetermined number of channel words that can follow the channel word. Binary code data in which a predetermined frequency band component is removed and code modulation is performed by selecting a redundant word in which a step response value of a digital waveform of a channel word is minimized, and a synchronization signal is added to a predetermined position. In the binary code demodulation method for decoding the m-bit data word from the sequence, the synchronization signal is detected and the detected Based on the synchronization signal, the channel word is extracted from the binary coded data sequence, the n _1- bit information word is separated from the extracted n-bit channel word, and the separated n ₁
It is a binary code demodulation method characterized by converting a bit information word into the m-bit data word.

[0027]

According to this, the step response of the transfer function defining the band of interest is obtained at each of the plurality of digital waveforms WF corresponding to the plurality of redundant words at the time of evaluation, and the step response is minimized. The redundant word is selected so that Therefore, the spectrum of the modulated signal can be suppressed in a desired band simply by changing the transfer function.

[0028]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention relates to a binary code demodulation method. Specifically, block conversion type run length limited (Ru
n Length Limited) code, that is,
A binary code modulation method of a run-length limited code in which an m-bit data word is converted into an n _1- bit information word and these information words are concatenated with an n ₂ -bit redundant word, or a convolution type run Length limited code, that is, after adding a redundant word of n ₂ 'bits to a data word, for example, MFM code or M
^{The present} invention relates to a demodulation method of a binary code modulation method for modulating a run length limited code such as a ² FM code, and in particular, arbitrarily designing a frequency band to be suppressed of a modulated signal by selecting a redundant word of n ₂ or n ₂ ′ It was made possible.

An embodiment in which the present invention is applied to a block conversion type run length limited code will be described below. In this example, the spectrum of the channel bits in the detection signal of the detector (see 1 in FIG. 11) is prevented from becoming a disturbance of the tracking servo system.

First, the frequency characteristic of the servo system will be described with reference to FIG. In FIG.
The broken line shows the characteristics of the servo device, for example, the servo mirror, and the alternate long and short dash line shows the characteristics of the servo circuit. The characteristics of the entire servo system are shown by the solid line.

In this case, if there is a channel bit component within the band of the servo device, the device moves accordingly, and the device does not trace the center of the track as shown by the one-dot chain line in FIG. Usually, the deviation of this trace needs to be suppressed to 0.1 μm or less when the laser wavelength is 0.8 μm and the numerical aperture NA of the lens is about 0.45, and if the channel bit component is large in this band. , Tracking goes off.

Further, in order to guarantee stable servo, it is necessary to provide the phase compensation characteristic shown by the one-dot chain line in FIG. For this reason, the gain is increased in the band indicated by the alternate long and short dash line. Therefore, if there is a channel bit component in this band, a large noise component enters the servo loop, and as a result, the saturation phenomenon of the servo system is likely to occur. Especially when the cutoff of the servo device is low and therefore the gain is large for phase compensation,
This inconvenience is remarkable.

In order to prevent the channel bit spectrum from entering as a disturbance in such a servo system, a transfer function as shown in FIG. 3 is used as an evaluation function. Here, point a in FIG. 3 corresponds to point a in FIG.

The evaluation by the transfer function of FIG. 3 will be described below. The transfer function of FIG. 3 is

[Equation 1] It is represented by. Then, the step response Y of this transfer function is

[Equation 2] And the momentary value y (t) is

[Equation 3] Becomes

[0035] Here, for example, to do the evaluation from time 0 to t ₁₀ for digital waveform WF as shown in FIG. 4, by obtaining at time t ₁₀ the step response of the transfer function of FIG. 3 of the digital waveform WF Good. Then,

[Equation 4] To get However,

[Equation 5] And

Using the above, in order to determine the redundant word WR _i in FIG. 9 at the end of the information word WI _{i + 1} , for each digital waveform of FIGS. Find the value. Then, the redundant word WR _i may be selected so that this step response is minimized.

Of course, the subsequent information words WI _{i + 2} , WI
_At the end of either _{i + 3} , ... Redundant word WR _i (00
0), (100), (010) and (001) may be evaluated to find the optimum redundant word WR _i .

Further, in this example, it is preferable to use a filter such as a transversal filter when performing the evaluation, as shown as a modified example below. For example, as shown in FIG. 5, the delay circuit 11 and the weighting circuit 1
2 and the adder 13 constitute a transversal filter. Here, as shown in FIG. 6, the information word WI
Of the respective end points of time t _-3 , t _-2 , t _-1 ,
It is assumed that the evaluation has already been completed at t ₀ , and the values x _-3 , x _-2 , x _-1 , and x ₀ of the respective step responses are indicated by arrows. Then, at time t ₁ to be evaluated, there are four values x ₁ according to the redundant word WR (shown by broken lines), and it is considered to evaluate these x ₁ .

To carry out the evaluation, first one of the four values of x ₁ is input to the input terminal 14. At this time, the delay circuit 11
By setting the delay time of each to be the evaluation time interval T, each delay circuit 11 outputs x ₀ , x _-1 , x _-2 , x _-3 , and these are added together with the value of x ₁ via the weighting circuit 12. The signals are added by the circuit 13 and output from the output terminal 15 as the output of the transversal filter.

Here, the weighting of the weighting circuit 12
The values shown by the arrows in the impulse response shown in FIG.
It This is assigned as shown in FIG. Then,
Transversal filter output y₀As y₀= A₃x₁+ A₂x₀+ A_-1x_-1+ A_-2x_-2+ A_-3x_-3 To get Generally speaking y_n= Σa_kx_nk To get And the other three y₁Similar operation for
Do three possible y _nGet these y_nIs the minimum
The redundant word is determined as follows.

Then, in this example, the characteristics based on the transfer function of FIG. 3 described above (shown by hatching in FIG. 8).
Further, the characteristics of the filter shown by the solid line are added. In addition, the characteristics of the digital filter make it possible to obtain more preferable characteristics as a whole.

As shown by the broken line in FIG. 5, the adder 13
If the output of the above is fed back to the adder 13 via the delay circuit 11, the transversal filter can be a cyclic type, and a more preferable configuration can be adopted. The output of the transversal filter in this case is y _n = b ₁ y _n-1 + Σa _k x _nk , but it is also possible to add a feedback loop to obtain y _n = Σb _k y _nk + Σa _k x _nk. Of course.

The recording medium on which the signal modulated in this example is recorded is reproduced by the reproducing device as follows. That is, a synchronization pattern (omitted in FIG. 9) in the channel bit is detected, and (n ₁
+ N ₂ ) -bit channel words are obtained by dividing each channel word, and this channel word is further divided into an n _1- bit information word and an n ₂ -bit redundant word, after which the information word is converted into a data word. Good.

As described above, according to the present invention, a plurality of digital waveforms W respectively corresponding to a plurality of redundant words are provided.
The step response of the transfer function that defines the band of interest for each of F is determined at the time of evaluation, and the redundant word is selected so that this step response is minimized. Therefore, the spectrum of the modulated signal can be suppressed in a desired band simply by changing the transfer function.

The present invention is not limited to the above-mentioned embodiments, and various modifications can be made without departing from the scope of the invention. For example, in the above embodiment, the block conversion type run length limited code is taken as an example, but the present invention can also be applied to the convolution type run length limited code. That is, not only the block conversion type run-length limited code but also the convolution type run-length limited code cannot achieve sufficiently satisfactory band suppression by evaluation by digital summation.
The present invention is also effective in the case of such a convolution type.

[0046]

According to the present invention, the step response of the transfer function that defines the band of interest is obtained at each of the plurality of digital waveforms WF corresponding to the plurality of redundant words at the time of evaluation, and the step response is obtained. The redundant word is selected so that Therefore, the spectrum of the modulated signal can be suppressed in a desired band simply by changing the transfer function.

[Brief description of drawings]

FIG. 1 is a graph diagram showing frequency characteristics of a servo system of an optical reproducing apparatus to which an embodiment of the present invention is applied.

FIG. 2 is a diagram for explaining the frequency characteristic of FIG. 1;

FIG. 3 is a graph diagram showing a frequency characteristic of a transfer function used in the above-described embodiment.

FIG. 4 is a diagram of a time chart used for explaining the above-mentioned embodiment.

FIG. 5 is a circuit diagram showing a modified example using a transversal filter.

FIG. 6 is a diagram for explaining the example of FIG. 5;

FIG. 7 is a diagram for explaining the example of FIG. 5;

FIG. 8 is a diagram for explaining the example of FIG. 5;

FIG. 9 is a diagram for explaining an example of a block conversion type run length limited code.

10 is a diagram of a graph used to explain FIG. 9. FIG.

FIG. 11 is a diagram for explaining a defect of the example of FIG. 9;

FIG. 12 is a diagram for explaining a defect of the example of FIG. 9;

FIG. 13 is a diagram for explaining a defect of the example of FIG. 9;

FIG. 14 is a diagram for explaining a defect of the example of FIG. 9;

FIG. 15 is a diagram for explaining a defect of the example of FIG.

Claims (1)

[Claims] 1. When the m-bit data word is code-modulated into an n-bit channel word consisting of an n ₁ -bit information word and an n ₂ -bit redundant word, the end point of the information word following the redundant word. or, the n bits of the transfer function which defines the frequency band to be suppressed at the end time of the last information word of a predetermined number of channel words can follow to the channel word
Binary code data in which a predetermined frequency band component is removed and code modulation is performed by selecting a redundant word in which a step response value of a digital waveform of a channel word is minimized , and a synchronization signal is added to a predetermined position. In the binary code demodulation method for decoding the m-bit data word from the sequence, the synchronization signal is detected, and the channel word is extracted from the binary code data sequence based on the detected synchronization signal, separating the withdrawn was n the n _1-bit information words from the channel words of bits, characterized in that the separated n _1-bit information words into data words of said m-bit binary Code demodulation method.