JPH04372774A - Digital data reproducing method - Google Patents

Abstract

PURPOSE:To reduce a data decode error rate when a digital data is reproduced. CONSTITUTION:From an input signal and an output data of a digital data decoder 8, a means 9 extracting a noise included in an input signal is provided. The extracted noise is subtracted from the former input signal with a signal synthesizer 6 through a proper filter 5, then the noise is eliminated. An S/N at the data decode time is improved and the data decode error rate is lowered.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for reproducing digital data when transmitting digital data or recording and reproducing digital data on a recording medium.

0002

2. Description of the Related Art Taking the case of recording and reproducing digital data in a magnetic disk device as an example, FIG. 8 shows a block diagram of a conventional data reproducing circuit, and FIG. 9 shows signal waveforms of various parts at this time. NRZI, commonly used in magnetic disks, etc.
In the recording method, the data in FIG. 9(A) is recorded in the waveform shown in FIG. 9(B). The weak signal read from the head is amplified by the head amplifier 2 so that the amplitude of the output signal is constant and at a manageable level. Furthermore, the equalizer 3 compensates for deterioration in high frequency characteristics that occurs during the recording and reproducing process on the recording medium and eliminates intersymbol interference. The output signal of the equalizer 3 at this time has the waveform shown in FIG. 9(C). The synchronization signal generation circuit 4 generates a synchronization clock corresponding to the recording timing from the reproduced signal. The decoder 5 identifies the signal at the timing of the synchronization clock and reproduces data from its amplitude value.

[0003] Also, when transmitting digital data, the basic configuration is the same except that the head amplifier 2 shown in FIG. 8 is replaced with a receiving amplifier, so a description thereof will be omitted.

0004

[Problems to be Solved by the Invention] When high-density recording is performed on a recording medium using the conventional data reproducing circuit as described above, the high frequency characteristics during recording and reproducing are greatly degraded. To compensate for this, a high gain equalizer 3 is required. However, when a high gain equalizer 3 is used, the noise generated in the media 1 and head amplifier 2 is greatly amplified by the equalizer 3, so the equalizer 3 as shown in FIG. The output signal becomes an equalizer 3 output signal as shown in FIG. 9(D) due to noise. As a result, there was a problem in that a decoding error occurred in the bit indicated by the arrow in FIG. 9(D).

Similarly, when performing high-speed data transfer, there is a problem in that noise is amplified and decoding errors increase.

An object of the present invention is to suppress the increase in noise that occurs during high-density recording on a recording medium or high-speed data transfer, and to lower the decoding error rate.

[0007]

[Means for Solving the Problems] In order to achieve the above object, the present invention separates the noise signal contained in the pre-decoding signal from the digital data pre-decoding signal and the digital data decoding data. Means is provided for band-limiting using a filter and subtracting the band-limited noise signal from the signal before digital data decoding.

[0008]

[Operation] By providing the above means, it is possible to efficiently extract the noise signal contained in the digital data pre-reproduction signal. By subtracting this extracted noise signal from the original pre-reproduction signal, a pre-reproduction signal with fewer noise components can be obtained. As a result, errors in decoding digital data are reduced.

[0009]

Embodiment An embodiment in which the present invention is applied to a magnetic disk device will be described below. FIG. 1 is a block diagram of a data reproducing circuit according to the present invention. FIG. 2 is a waveform diagram showing the operation of the circuit shown in FIG. Here, the bipolar coding method (
Hereinafter, it will be referred to as PR (1, -1). ) shall be used for recording and playback. In this PR (1, -1) method, at the synchronization clock point, the output amplitude becomes 0 for the recording code 0, and the output amplitude a or -a is alternately output from the equalizer 3 for the recording code 1. Determine the equalization characteristics as follows. Therefore, recorded data can be reproduced from the amplitude value at the synchronization clock point. However, in reality, due to noise generated during recording and playback,
The actual waveform is, for example, as shown in FIG. 2(A). In this state, correct decoding cannot be performed due to noise components included in the output of the equalizer 3. Therefore, the noise component included in the output signal of the equalizer 3 is extracted using a circuit described later, and this noise component is subtracted from the output signal of the equalizer 3 to create the waveform shown in FIG. 2(B) with less noise component. Perform decryption processing.

In the decoding process, first, the signal shown in FIG. 2(B) is sampled/held (
(hereinafter referred to as S/H) to produce the pulse signal shown in FIG. 2(D). Letting the amplitude of this pulse signal be S, this is shown in Figure 2.
The decoder 8 compares it with the threshold value indicated by the broken line in (D). If S is larger than a/2, data 1 and sign + are output as output +1. Further, when S is less than a/2 and more than -a/2, the output is 0 and data 0 is output. The sign at this time may be either + or -. Also, S is -a
If it is smaller than /2, it is set as -1 and data 1 and sign - are output. Such an operation of the decoder 8 is achieved by making data 1 correspond to decoding output H, data 0 to decoding output L, sign + to sign output H, sign - to sign output L, and so on.
This can be realized by the circuit shown in FIG.

Further, noise components contained in the signal are extracted by the noise detector 9 as follows. When the output of the decoder 8 is +1, the ideal amplitude is a, and the difference from the actually obtained signal amplitude S can be considered as noise. Therefore, by subtracting a from the input signal S of the decoder 8, the noise component contained in the signal can be extracted. Similarly, if the output of the decoder 8 is 0, the input signal S of the decoder 8 can be regarded as noise, and if the output of the decoder 8 is -1, -a is extracted from the input signal S of the decoder 8. It can be said that the subtracted value is the noise component. FIG. 4 shows an example of a noise detector circuit that performs the above operation. In FIG. 4, a multiplier 40 is a circuit that outputs an ideal amplitude value from the output of the decoder 8. Figure 5 is Figure 4
The operation of the multiplier 40 inside is shown. and subtractor 41
Then, the difference between the ideal amplitude and the amplitude S of the S/H output 10 is output as a noise output 12. The waveform of the noise output 12 obtained in this manner is as shown in FIG. 2(F).

Now, the waveform in FIG. 2(F) shows the noise component at each synchronization clock point. It is considered that the low frequency component of the signal in FIG. 2(F) contains approximately the same level of noise even at the next clock point to be decoded. Therefore, out of the signal in FIG. 2(F), only the low frequency component is filtered through the LPF5
By extracting the signal and subtracting it from the equalizer output shown in FIG. 2(A), the signal shown in FIG. 2(B) with less noise components can be obtained. The characteristics of this LPF 5 may be determined so that the error rate of decoded data is minimized.

Furthermore, if a Viterbi decoder is used as the decoder, the decoding error rate can be further reduced. An example in which a Viterbi decoder is applied to the present invention will be described below. FIG. 6 is a block diagram showing an example of a data reproducing circuit of a magnetic disk device to which a Viterbi decoder is applied. Here, the output of the equalizer 3 is converted by an analog-to-digital converter 13, and subsequent processing is configured by a digital circuit. Even if an analog circuit is used here, the basic configuration is the same and the same effect can be obtained.

Various circuit configurations of the Viterbi decoder 15 shown in FIG. 6 have been proposed; for example, "Digital Video Recording Technology", p.
). In this Viterbi decoder 15,
There is a time delay of several clocks from when a signal is input until the corresponding decoded data is output. Therefore, the reproduced signal is temporarily stored in the data buffer 16, and when the corresponding decoded data is output, it is taken out again to absorb the time delay and the noise detector 18 performs noise detection. Noise detector 18
The low frequency component of the output is extracted by the LPF 17. By subtracting the output of the LPF 17 from the A/D 13, a signal with less noise components can be obtained. Since the Viterbi decoder 15 has a lower decoding error rate than the decoder 8 shown in FIG. 3, it is possible to perform noise detection with high accuracy. A normal FIFO memory can be used as the data buffer 16 used at this time.

However, in FIG. 6, it takes several clocks to extract the noise component from the signal input to the decoder 15, so there is a problem that feedback takes time and high frequency noise components cannot be removed. Therefore, a configuration as shown in FIG. 7 can be considered in order to also remove relatively high frequency noise components. In this configuration, a relatively high frequency component among the noise components detected by the noise detector 18 is transferred to the BPF.
25, and after subtracting this from the output signal of the data buffer 23, decoding processing is performed. This data buffer 23
is provided to absorb the time delay in the noise detector and BPF. Furthermore, among the noise components detected by the noise detector 18, low frequency components are extracted by the LPF 26 and
Subtract from the output of D13. With such a configuration, it is possible to remove noise with high precision while taking advantage of the characteristics of the Viterbi decoder 21, and it is possible to further reduce the decoding error rate.

[0016]

[Effects of the Invention] According to the present invention, the digital data reproduction method can efficiently remove noise components contained in transmitted or reproduced digital signals, and reduce the decoding error rate of decoded digital data. be able to.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment of a data reproducing method according to the present invention;

[Fig. 2] Waveform diagrams of various parts according to the data reproduction method according to the present invention,

FIG. 3 is a block diagram of an example internal configuration of a decoder;

FIG. 4 is a block diagram of an example of the internal configuration of a noise detector;

FIG. 5 is an explanatory diagram of the operation of a multiplier,

FIG. 6 is a block diagram of an embodiment of a data reproducing circuit according to the present invention in which a Viterbi decoder is applied;

FIG. 7 is a block diagram of another embodiment in which a Viterbi decoder is applied to the data reproducing circuit according to the present invention;

[Fig. 8] Block diagram of a conventional data reproducing circuit,

FIG. 9 is a waveform diagram of each part in a conventional data reproducing circuit.

[Explanation of symbols]

1...Media, 2...Head amplifier, 3...Equalizer, 4...Synchronization signal generation circuit, 5...Low pass filter, 6...Signal synthesizer, 7
...Sample and hold circuit, 8.Decoder, 9.Noise detector, 10.Sample and hold output signal, 11.Decoder output signal, 12.Noise signal.

Claims (1)

[Claims] 1. A digital data reproducing method, comprising separating a noise signal included in the non-digital data reproduction signal from a non-digital data reproduction signal and a digital data reproduction result, and band-limiting the noise signal using a filter. A digital data reproducing method characterized in that the band-limited noise signal is subtracted from the non-digital data reproducing signal.