DE4120270A1 - DATA STORAGE MEDIUM, DATA STORAGE DEVICE AND METHOD FOR CONTROLLING THE DATA STORAGE DEVICE - Google Patents

Description

The invention relates to a data storage device, namely a Magnetic disk storage device and especially a data storage medium, a data storage device and a control process for this, a reading system based on Parameters read from the data storage medium be controlled.

The usual known magnetic disk device contains a system that ge using a magnetic head on a magnetic disk stored data transformed into electrical signals and by comparing these signals to a fixed smolder lenwert (slice level) differentiates whether this electrical Signals are 0 or 1.

The magnetic disk device has cylinders with several Magnetic disks on. Every magnetic disk of a cylinder has again two magnetic heads on each side of the disc. Further is the threshold as a constant value to the magnet heads of the same cylinder set.

Due to electrical or mechanical characteristics of the single magnetic disk and single magnetic head can however, the case will occur that the output level each single magnetic head is not the same for everyone Magnetic heads. As a result, one read can be performed on each magnetic head errors occur because of the threshold for this magnet  head is unsuitable and is the same for every magnetic head. The magnetic head therefore tries when the reading occurs error, the data from the magnetic disk again read. In most cases, however, is due to the faulty readout with the same threshold value the data of the magnetic head. The threshold must therefore set to an optimal value for each magnetic head will. In addition to the threshold, the data reading points System of the magnetic disk device facilities, which the decision that distinguishes whether the electrical Signal is 0 or 1. These facilities have parameters such as B. an additional size of Cosine equivalent circuit, a gain of the amplifier and a window shift amount of the Comparator. Consequently, these parameters must be the same Way like the threshold into an optimal value be set.

The present invention is based on the above showed problems; it sees a data storage medium for Storing data and at least one parameter before, which is placed in a signal processing system if the data is transformed into binary signals, whereby this parameter in advance in a fixed area of the memory Medium is saved.

The data storage device according to the present invention has

• (a) a data storage medium for storing data and a parameter that is in advance within a fixed range the storage medium is stored;
• (b) a reading device for reading data therefrom Data storage medium; and  
• (c) a signal processing device to process the data were read out by said reading device, based on this parameter in binary signals transform.

Furthermore, there is a control method for the data storage device provided that according to the invention comprises the following steps:

• Writing parameters into a data storage medium, around data read out by a reading device be transformed into binary signals, ent speaking of the state of the data storage medium and the Reading device;
• - Setting the parameter to a signal processing device after reading out; and
• - Read out this data based on this parameter.

It can be seen that the signal in this data storage device processing device by means of the reading device Transfor read data on the data storage medium correctly can lubricate by in spite of characteristic irregular of the magnetic medium or the reading device the parameters are set to optimal values.

The invention is explained in more detail below with the aid of the figures explained. It shows:  

Figure 1 is a block diagram showing the structure of a first exemplary embodiment of a magnetic disk device according to the invention.

FIG. 2 shows a plan view of a magnetic disk device according to FIG. 1;

Fig. 3 is a graph of a comparator circuit for a magnetic disk device shown in Fig. 1;

FIG. 4 shows a flow chart by means of a threshold value at the final shipment test of the magnetic storage device is written an operation;

Fig. 5 is a block diagram showing the structure of a second exemplary embodiment from the invention;

FIG. 6 is a flowchart of an operation of the second embodiment shown in FIG. 5; and

Fig. 7 is a flowchart showing a Lesewiederholoperation of the second embodiment.

First embodiment

Fig. 1 is a block diagram showing the construction of a first embodiment of a magnetic storage device according to the invention. A magnetic plate 1 is rotatably mounted on an axis 2 of a spindle motor. A carrier 4 , which is rotatable about a pivot axis 3 , which is parallel to the axis 2 , is fastened near the magnetic disk 1 . A magnetic head 5 , which is attached to the free end of the carrier 4 , transforms data stored on the surface of the magnetic disk 1 into electrical signals. An at the end of the carrier 4 be fixed plunger motor 6 acts on the carrier 4 such that it conveys the magnetic head 5 to the desired cylinder. The magnetic storage device also has a plurality of magnetic disks 1 arranged on the axis 2 , with one magnetic head 2 being arranged on each disk side of all magnetic disks 1 .

The electrical signal emitted by the magnetic head 5 is fed to a magnetic head amplifier 10 , which amplifies it and then feeds it into an AGC amplifier 11 (automatic gain control). The AGC amplifier 11 raises the already amplified signal again to a fixed value and outputs it to an equalizer circuit 12 . The equalizer circuit 12 arranges a waveform of the electrical signal output by the AGC amplifier 11 and feeds it into a filter circuit 13 . The filter circuit 13 filters out interference signals, which contains the electrical signal emitted by the equalizer circuit 12 , and feeds the filtered signal into a differential circuit 14 . The differential circuit 14 eliminates unevenly steep parts of the electrical signal emitted by the filter circuit 13 and then feeds the signal into a comparator circuit 15 . This circuit compares the made signal with a fixed threshold value (criterion value) and outputs the compared result to a pulse circuit 16 . This pulse circuit finally generates a data pulse, ie 0 or 1, based on the compared result.

A CPU 17 (central processing unit) controls each part of the magnetic storage device, in particular it feeds the threshold value into the comparator circuit 15 , and furthermore feeds control data corresponding to the threshold value into a D / A converter 18 , ie a digital / analog converter. The D / A converter 18 converts the control data into an analog signal and then outputs it to a dispersion circuit 19 . The dispersion circuit 19 disperses the analog signal (= control data) to obtain the fixed threshold value and outputs a dispersed analog signal, ie the fixed threshold value, to the comparator circuit 15 .

Incidentally, information specifying the threshold value supplied from the CPU 17 to the comparator circuit 15 is written in the system cylinders (the outermost peripheral cylinder) of the magnetic disk 1 . That is, a minimum threshold value, if the data readout by the magnetic head 5 is possible, and a maximum threshold value, shortly before this data readout is possible, are measured in the production or factory in a final shipping test of the magnetic storage device. Furthermore, an average between the minimum and maximum threshold value is obtained. The mean value is written into each system cylinder of the multiple magnetic disk as the optimal threshold value.

As already mentioned above, it is a special feature of this embodiment that the threshold value corresponding to the characteristic irregularity in each magnetic disk or magnetic head is already written in advance in each system cylinder and the magnetic storage device reads out the threshold value when it is commissioned later by means of its magnetic head and for Comparator circuit 15 sets.

A write operation of the threshold value in the final dispatch test of the magnetic storage device is explained below with reference to FIG. 4.

First, the threshold value of the comparator circuit 15 is set to zero in step SP1. Then, ie in step SP2, the data are read out from the magnetic disk 1 on the basis of the threshold value. Next, it is tested whether a reading error has occurred or not. The error test is carried out on the basis of CRC (cyclical redundancy control), whereby the error display code, which is in the last bit of the ID field, and the data field on the magnetic disk are used. If the result of the test carried out in step SP3 is positive, ie the reading error occurs in step SP3, control passes to step SP4, after which the threshold value is increased by one bit. The control then returns to step SP2, in which the data are read from the magnetic disk 1 , after which the control transfers to step SP3. As already mentioned, steps SP2 and SP4 are carried out continuously until the threshold value has reached an optimal value.

On the other hand, if the result of step SP3 is negative, i.e. when the threshold becomes an optimal value and the data is read exactly, the control proceeds to SP5. At step SP5, the threshold of Step SP3 set as the minimum value. Next up  the threshold at step SP6 by another bit increases and then tested in step SP7 whether a Read error has occurred or not. If the result from step SP7 is negative, i.e. if the reading error if control has not occurred, control returns to step SP6 back, after which the threshold is increased by another bit becomes. Steps SP6 and SP7 are carried out until the The result of step SP7 becomes positive, i.e. until the read error occurs and thus the threshold the optimal Area exceeds. Then, i.e. if the threshold exceeds the optimal range, the result of Step SP7 is positive and control proceeds Step SP8. At step SP8, the threshold, if the read error occurred at step SP7, set as maximum value. Next is the step SP9 the mean (optimal value) by calculating (Maximum value + minimum value) / 2 determined, namely on the basis of the maxi obtained in step SP8 or SP5 paint or minimum value. Then the remedy value in the system cylinder as the optimal threshold entered. This is the aforementioned write operation of the threshold for all magnetic disks.

If, as explained above, the threshold value has already been previously written into the system cylinder in accordance with the characteristic irregularities of the magnetic disk or the magnetic head, z. B. even the output signal of the magnetic head 5 a special wave form - see. Fig. 3 -, which is caused by characteristic irregularities of the magnetic disk 1 or the magnetic head 5 . The above-mentioned threshold value sets the CPU 17 to the comparator circuit 15 so that it is able to compare the level of the converted signal from the differential circuit 14 with the optimal threshold value. The pulse circuit 16 consequently outputs pulse data as a reference signal B - s. Fig. 3 -.

In addition, if the maximum and minimum values of the threshold value - as shown in FIG. 3 - are identified by a reference signal C, the value L is set as a threshold value on the basis of the maximum and minimum values. Furthermore, if a reading error caused by the threshold value L occurs, the CPU 17 can control the reading system, so that after the threshold value L has been changed, the reading of the data is repeated. In the event that the CPU 17 cannot read the threshold from the magnetic disk 1 , the magnetic storage device may have a ROM (or memory) which temporarily stores a threshold L for input to the CPU 17 .

Although in the above embodiment the threshold value is written as a control parameter of the reading system into a special area of the magnetic disk 1 , information which influences a reading level (reading level) can nevertheless be written into the magnetic disk, the following sizes or Conditions must be observed:

• a) a parameter for changing the window mean of VF0 (i.e. voltage controlled oscillator with PLL: Phase locked loop), the pulse signals based on which generates fixed frequency read data;  
• b) an additional value of a cosine equivalent circuit (ie the equalizer circuit 12 );
• c) a write current of the magnetic head;
• d) a pre-compensation value for compensation to a Verification pattern peak shift of the reading pulse wrestle, with pattern peak the peaks of the read signal are and the difference the difference to the ideal Position means;
• e) at least one gain factor that the reading system puts.

Second embodiment

In the following a further embodiment example according to the invention is explained with reference to FIG. 5, in which a reading system is controlled by several parameters. Here, it is omitted the description of portions of FIG. 5 ver which have the same function as the ent speaking parts of the first embodiment.

In this embodiment 12, the cosine equivalent circuit 12 applies ver instead of the equalizer circuit. This circuit has a delay circuit, a reduction circuit, etc. and generates a delay signal by delaying an input signal, after which this and the input signal are put together. Both signals, ie the delay signal and the input signal, are reduced by the reduction circuit when they are assembled. That is, that the cosine equivalent circuit 12 , the delay time and the half value of the input signal by changing the delay amount of the delay circuit and the reduction factor of the reduction circuit changes due to the reduction of the pattern peak shift (time delay) of the input signal and to one Exceptional data structure (data pattern) leading back bit readout. In this embodiment, the reduction factor of the reduction circuit is changed. The output signal of the cosine equivalent circuit 12 is fed via a low-pass filter 13 into a differential circuit 14 , an amplitude display circuit 30 and a gate generator 31 .

The amplitude display circuit 30 displays an amplitude of the signal output by the cosine equivalent circuit 12 and regulates the gain of the AGC amplifier 11 in accordance with this amplitude. Furthermore, a gain regulating signal is output from the CPU 17 to the amplitude display circuit 30 . The result is that the amplitude display circuit 30 increases or decreases the gain of the AGC amplifier 11 in accordance with the gain control signal, despite the signal amplitude output from the cosine equivalent circuit 12 .

The pulse circuit 32 transforms the differentiated signal into pulse data in accordance with a threshold value that is based on the signal output by the gate generator 31 . The gate generator 31 outputs the threshold signal, which is synchronized with the differential signal, which is output accordingly from the output signal of the low-pass filter 13 from the differential circuit 14 . Here, the amount of the threshold value output by the gate generator 31 is adjusted by a control signal from the CPU 17 . The pulse data output by the pulse circuit 32 are fed to a decoder 33 and a phase display circuit 34 . The phase display circuit 34 forms a PLL (phase locked loop) circuit with a variable frequency oscillator 35 . The phase display circuit 34 displays a phase difference between the pulse data that is input to one input terminal and data that is supplied to another input terminal of this circuit. Corresponding to the phase difference, a corresponding voltage is fed into the frequency oscillator 35 on the output side. This frequency generator generates a signal with a frequency corresponding to the respective voltage, which on the one hand is fed back to the latter input terminal of the phase display circuit 34 and, on the other hand, is fed into the decoder 33 as a distinction window signal. The frequency oscillator 35 generates a signal which has a specific frequency in order to set the voltage to zero in accordance with the phase difference. After the frequency of the output signal of the frequency oscillator 35 is halved, this signal is fed back to the phase display circuit 34 .

The decoder 33 generates a data bit from the pulse signal on the basis of the discrimination window signal just mentioned. From this, a control signal is passed from the CPU 17 to the decoder 33 , so that it can move the differentiation window in accordance with the control signal.

Next, a data format controller 36 performs a write operation in which data read out from the magnetic disk by the magnetic head is written into the buffer RAM 37 by means of a write operation in which data to be written into the magnetic disk is provided by a peripheral Device as supplied to a host computer can be written into the buffer RAM 37 based on the control signal of the CPU 17 . The data format controller 36 transforms data stored in the buffer RAM 37 to the data format (data format) (corresponding to the SCSI; small computer system interface) of the magnetic disk. That is, data fed into the host computer and data output from it are stored in the buffer RAM 37 . The data read out by the magnetic head 5 are fed into the host computer via an interface adapter 38 . On the other hand, the data from the host computer is temporarily stored in the buffer RAM 37 by the interface adapter 38 .

Furthermore, the write data from the host computer are fed to an encoder 39 via the data format controller 36 . This encoder transforms the write data into data that matches the magnetic head amplifier 10 . Furthermore, a control signal with write precompensation value is fed into the encoder 39 by the CPU 17 and this value is adjusted in the encoder 39 by the control signal mentioned above. Here, the write precompensation means that when the write data is stored, the encoder 39 compensates for a writing timing, whereby the write data is shifted in the opposite direction (against the peak shifting direction) by the amount of Reduce pattern peak shift. After the Manget head amplifier 10 has amplified the write data to a certain value, it feeds this data into the magnetic head 5 . The magnetic head is driven by the moving coil motor 6 - s. Fig. 2 - beför changed in a certain position on the magnetic disk 1 . The voice coil motor 6 is in turn controlled by a signal from a magnetic head position controller 40 , to which the CPU 17 outputs a control signal. Furthermore, the CPU 17 feeds a compensation signal into the magnetic head position controller 40 , through which the magnetic head 5 is positioned with a fixed compensation value for the original position signal.

A fixed value ROM 41 and a working memory RAM 42 are connected to the CPU 17 . The fixed value ROM 41 stores standard values (initial values) of the parameters required to store or read out data such as a table in the magnetic disk 1 . The RAM RAM 42 stores parameters that regulate the reading and writing of data. It is used as a repetition counter, which - as will be explained later - stores the number of repetitions in which the parameters made a decision. A history memory RAM 43 is also connected to the CPU 17 and stores the development or the process of repeating a parameter decision.

The mode of operation of the parameter decision of the reading system based on the above-mentioned arrangement is explained below with reference to the flow chart according to FIG. 6.

When the magnetic storage device is connected to the power supply, the flow diagram shown in FIG. 6 is carried out as follows:
First, in step SP12, each parameter set in the device is initiated by CPU 17 . Then, in step SP13, a standard value of the R / W parameter is read out of the fixed value ROM 41 and is set for each part of this data reading system, that is to say for the cosine equivalent circuit 12 , for the gate generator 31 and for the decoder 33 . At the conclusion of this, the R / W parameter, which is written into the magnetic disk 1 as the dispatch date, is read out in step SP14. The R / W parameter is written in accordance with the flow chart according to FIG. 4. In the following, ie at step SP15, it is tested whether the R / W parameter is able to read exactly according to the standard value in step SP13 or not, ie whether a reading error occurs or not. If the test result of this step is positive, ie if the reading error occurs in step SP15, control passes to step SP30, in which a read repeat operation takes place. In step SP30, the R / W parameter is changed on the basis of the standard value stored in the fixed value ROM 41 because the central processor unit CPU 17 is unable to add the data to the magnetic disk by means of the R / W parameter set in step SP13 read.

In the following step SP31, it is tested whether or not a number of repetitions stored in the memory RAM 43 is above a certain number. If the result in step SP31 is negative, the control proceeds to step SP14, in which the stored R / W parameter is read out again. However, if the reading error occurs again in step SP15, steps SP30 and SP31 are carried out. As mentioned above, if the reading error occurs even though the R / W parameter is set based on the default value, then while the R / W parameter is gradually changed by executing the step sequence SP14, SP15, SP30 and SP31 who tried to read the data from the magnetic disk 1 . If the result of step SP31 is positive, that is, if the number of repetitions is above the fixed number, then the operation in this flowchart ends when the CPU 17 has indicated that a read error has occurred.

On the other hand, if the result at step SP15 is negative, that is, if it is possible to read out R / W parameters of the magnetic disk 1 based on the standard value, then control goes to step SP16. In this step, the R / W parameter read from the magnetic disk 1 is set to several parts of the data reading system, namely to the cosine equivalent circuit 12 , to the amplitude circuit 30 or to the gate generator 31 . The control then controls step SP17 and waits here for a command from the host computer, ie it is tested whether a command has been issued or not. If the host computer issues a command, the result of this decision is affirmative and control proceeds to step SP18. This step tests whether the command is a read command or not. If the test result is negative, ie if there is no read command, the control proceeds to step SP32, in which an operation according to the command is carried out. Control then returns to step SP17 and waits for the next command to be issued.

On the other hand, if the test result of step SP18 is positive, that is, if the command from the host computer is the read command, the control proceeds to step SP19, in which data is read out from the magnetic disk 1 . In the following step SP20 it is tested whether a reading error occurs or not. If this is the case, that is to say the reading error occurs, the control proceeds to step SP33, within which a read retry operation takes place.

At step SP33, the R / W parameter is changed based on the R / W parameter read from the magnetic disk at step SP14. Control then proceeds to step SP34, after which it is tested whether or not the number of repetitions stored in the memory RAM 43 is higher. If the test result of this step is negative, the control proceeds to step SP19, after which the data are read out from the magnetic disk again. In the subsequent step SP20, it is tested whether a read error has occurred or not. If the reading error occurred again at this step, steps SP33 and SP34 are carried out. As explained above, if the reading error occurs using the R / W parameter even though the parameter of each part is set based on the R / W parameter of the magnetic disk 1 , then the R / W parameter gradually becomes during the R / W parameter is changed by exercising the sequence of steps SP19, SP20, SP33 and SP34, the reading of the data from the magnetic disk 1 is attempted. If the test result at step SP34 is positive, that is, if the number of repetitions is above the specified number, control proceeds to step SP35, at which CPU 17 indicates that a read error has occurred. Control then passes to step SP17 and again waits for a command from the host computer. If, however, the result from step SP20 is negative, that is to say that no reading error has occurred, the control proceeds to step SP21. In this step, the R / W parameter is returned to the initial value read from the magnetic disk 1 in step SP14. This operation means that the reading error occurs on the data written in a special magnetic disk. The reason for this is that the aforementioned R / W parameter is not a value that is valid for the data of all magnetic disks. In the following, the control continues to step SP22, in which an operation for the read data is carried out. Control then returns to step SP17 and the CPU 17 waits again for a command from the host computer.

The following is the repeat read operation in the above described steps SP30 and SP33.

First, the number of repetitions is incremented at step SP50. Then, at step SP51, it is tested whether the count is 1 or not. If it is 1, that is, the reading error occurs only once, the result of step SP51 becomes positive, and control goes to step SP60, that is, the R / W parameter is not changed and the control becomes that in FIG. 6 returns the flowchart shown, ie to step SP31 or SP34.

If, on the other hand, the count, ie the number of repetitions, is not 1, that is to say the reading error occurs more than twice, the result at step SP51 becomes negative and control passes to step SP52, in which it is tested whether the number of repetitions 2 or not. If the result at step SP52 is positive, ie the reading error has occurred only twice, the control proceeds to step SP61. In this step, a control or control signal of the CPU 17 is fed into the decoder 33 , thereby the discrimination window by 5 nsec. shifted forward, after which control continues in the flowchart shown in Fig. 6, that is, with step SP31 or SP34.

However, if the number of repetitions is not 2, control continues with SP53, where it is tested whether the number of repetitions is 3 or not. If the test result of this step is positive, ie if the reading error has occurred only three times, the control proceeds to step SP62. In this step, a control signal of the CPU 17 is fed into the decoder 33 and thereby the discrimination window by 5 nsec. shifted backwards, after which the control continues in the flow diagram shown in FIG. 6, ie with step SP31 or SP34.

If the number of repetitions is not 3, then control passes to step SP54, where it is tested whether the number of repetitions is 4 or not. If the test result is positive, that is to say if the reading error has just occurred four times, the control proceeds to step SP63. In this step, the amount of the discrimination window shifted in step SP61 or SP62 is reduced to 0 and then the threshold value is increased by 50 mV by feeding a control signal into the gate generator 31 . The control then continues according to the flowchart of FIG. 6, ie with step SP31 or SP34. If the number of repetitions is not 4, control proceeds to step SP55, where it is tested whether the number of repetitions is 5 or not. If the result of this step is positive, ie if the reading error has occurred exactly five times, then control transfers to step SP64, in which the threshold value of 50 mV is lowered by feeding a control signal into the gate generator 31 . The control then continues according to the flowchart according to FIG. 6, ie with step SP31 or SP34.

On the other hand, that is, if the number of repetitions is not 5, control proceeds to step SP56, where it is tested whether the number of repetitions is 6 or not. If the test result is positive, ie the reading error has just occurred six times, step SP65 is triggered. In this step, the threshold value changed in step SP63 or SP64 is returned to the previous value. Then, by supplying a control signal to the cosine equivalent circuit 12, the offsetting amount increases by 5%, that is, this operation makes the above-mentioned amount of reduction small. As a result, the peak value of the output signal of the cosine equivalent circuit 12 drops and its half value becomes small. The control then continues according to the flowchart according to FIG. 6, ie with step SP31 or SP32.

If the number of repetitions is not 6, control proceeds to step SP57 and tests whether the number of repetitions is 7 or not. If the test result is positive, that is to say if the reading error occurs just seven times, the control proceeds to step SP66. In this step, by feeding a control signal into the cosine equivalent circuit 12, the compensation amount is reduced by 5%, that is, this operation makes the aforementioned amount of reduction large. The result is that the peak value of the output signal of the cosine equivalence circuit 12 rises and its half value becomes large. Control then returns to the flow chart of FIG. 6, that is, step SP31 or SP32.

On the other hand, if the result at step SP57 is negative, the control proceeds to step SP58 even if the number of repetitions is not 7. At this step SP58, the compensation amount is returned to the previous value by supplying a control signal to the cosine equivalent circuit 12 , after which control returns to the flowchart shown in FIG. 6, that is, to step SP31 or SP32.

Besides, the history of the repetition is stored in the history memory RAM 43 connected to the CPU 17 and, if necessary, displayed or printed out in the above-mentioned error end operation of the step SP35.

In the exemplary embodiments explained above, the block Circuit diagram adapted to the needs of any parameters will. Both embodiments have a magnet disk as the object of data storage; as part of However, this invention can also be an optical or electro static storage medium can be used.

Claims (12)

1. Data storage medium ( 1 ) for storing data that are read out and transformed by a signal processing device ( 10-19; 30-43 ) into binary signals, characterized in that at least one parameter before in a certain area of the data storage medium ( 1 ) stored and, before this data is transformed into binary signals, set to the signal processing device ( 10-19; 30-43 ). 2. Data storage device with a data storage medium ( 1 ) for storing data, a reading device ( 5 ) for reading out the data of this data storage medium and with a signal processing device ( 10-19; 30-43 ) for transforming the data read out by means of the reading device in binary signals, characterized by a signal processing device ( 10-19; 30-43 ) which transforms the data on the basis of at least one parameter which is previously stored in the data storage medium ( 1 ). 3. A method for controlling a data storage device which has a data storage medium ( 1 ) for storing data, a reading device ( 5 ) for reading out the data of the data storage medium and a signal processing device ( 10-19; 30-43 ), to transform the read data read out by the reading device into binary signals, characterized by the following steps:

• - A parameter, which is set in the signal processing device ( 10-19; 30-43 ), is written into the data storage medium ( 1 ) in order in accordance with the conditions of this data storage medium and the reading device ( 5 ) transform the data into binary signals;
• - The parameter is set after reading the signal processing device ( 10-19; 30-43 ); and
• - The data are read out based on this parameter.

4. Data storage device according to claim 2, characterized in that the signal processing device ( 10-19; 30-43 ) has the following features:

• (a) a discriminator ( 15 ) for discriminating whether the parameter is read out exactly or not; and
• (b) storage means ( 42 ) for storing a substitution parameter in place of this parameter if the result of the discrimination is negative.

5. Data storage device according to claim 2, characterized in that the signal processing device ( 10-19; 30-43 )

• (a) an error display device ( 17 ) for distinguishing whether the data have been read out exactly or not; and
• (b) has a control device for compensating this parameter if the error display device ( 17 ) indicates an error.

6. The method for controlling the data storage device according to claim 3, characterized in that the reading step comprises the following individual steps:

• - The data are read out on the basis of the parameters previously stored in the data storage medium ( 1 );
• - the parameter is compensated if an error occurred during the reading step; and
• - The readout step is repeated after the compensation step on the basis of a compensated parameter.

7. Data storage medium ( 1 ) according to claim 1, characterized in that the parameter is used as a threshold when transforming the data into binary signals. 8. Data storage device according to claim 2, 4 or 5, characterized characterized in that the parameter when transforming the Data in binary signals is used as a threshold. 9. Method of controlling the data storage device according to claim 3 or 6, characterized in that the Parameters when transforming the data into binary signals is used as the threshold. 10. Data storage medium ( 1 ) according to claim 1, characterized in that the parameter is an amplification factor of the signal processing device. 11. Data storage device according to claim 2, 4 or 5, characterized characterized in that the parameter is a gain factor the signal processing device. 12. Method of controlling the data storage device according to claim 3 or 6, characterized in that the Parameter a gain factor of signal processing facility is.