JP4006446B2 - Data recording / playback system - Google Patents

Description

  The present invention relates to a data recording / reproducing system, and more particularly, to a data recording / reproducing system having an improved error detection and correction capability at the time of data reproduction retry (reproduction).

  A magneto-optical recording / reproducing apparatus, which is an example of a data recording / reproducing system, is capable of recording code information for computers from recording / reproducing of images and image information due to large capacity, interchangeability, high reliability, etc. This is a field where there are various devices and the market is rapidly expanding. There is an increasing demand for such an optical disk recording apparatus year after year.

  Such an optical disc recording / reproducing apparatus requires a more accurate method for recording and reproducing data as the recording density increases. As a method for recording and reproducing data with high accuracy, for example, the data sequence is rearranged and modulated and recorded and reproduced, such as turbo coding and decoding, and low density parity check (LDPC). Furthermore, a method of repeatedly decoding data at the time of demodulation has been proposed.

  The turbo code is a code having a large coding gain, and is a coding technique that has attracted particular attention in the communication field.

  FIG. 1 is a diagram showing an example of the configuration of an optical disc recording / reproducing apparatus that reproduces data recorded by turbo encoding using iterative decoding. FIG. 1 shows a configuration particularly when applied to an optical disc recording / reproducing apparatus 100 as a data recording / reproducing system. Hereinafter, the operation of the optical disk recording / reproducing apparatus 100 will be described as an example with reference to the drawings.

Optical disc recording and reproducing apparatus 100 of FIG. 1 mainly includes a recording system 110 to record user data _{u k} in the optical disc 122, and the reproduction system 120 to reproduce the user data from the optical disk.

First, the recording system 110 of the optical disc recording / reproducing apparatus 100 of FIG. 1 will be described. The recording system 110 of the optical disk recording / reproducing apparatus shown in FIG. 1 mainly includes an encoder 111, a MUX / puncture unit 112, an interleaver (π) 113, and a laser driving circuit 114. The encoder 111 generates a parity bit string p _k corresponding to user data u _k to be recorded.

Figure 2 is a diagram illustrating a configuration example of an encoder 111 used to perform encoding of the turbo code to be used from the user data u _k to generate a parity sequence of the turbo code. The encoder 111 used for encoding the turbo code shown in FIG. 2 mainly includes delay units 201 and 202 and adders 203 and 204. Each adder includes an exclusive OR operation unit. The user data input to the encoder 111 is added to the outputs of the delay units 201 and 202 by the adder 203, and the addition result is input to the delay unit 201. Outputs of a delay unit 202 of the adder 203 are summed by an adder 204, from the encoder 111 as the parity bit sequence p _k, it is outputted.

The MUX and puncture unit 112 in FIG. 1 combines the user data u _k and the parity bit sequence p _k generated by the encoder 111 according to a predetermined rule, and from the obtained bit sequence, bits according to the predetermined rule. Is thinned out (this is called a puncture function) to generate an encoded data bit string a _i .

The interleaver (π) 113 changes the arrangement of the encoded data bit sequence a _i from the MUX and puncture unit 112 to generate the encoded data bit sequence c _i . The laser drive circuit controls the light emission amount of the laser based on the encoded data bit sequence c _i, writes the signal on the optical disc. In this way, the user data _{u k,} written on the optical disk 122.

  Next, the operation of the playback system 120 of the optical disk recording / playback apparatus 100 of FIG. 1 will be described.

  1 mainly includes a PR channel 121, an analog / digital converter (hereinafter referred to as an A / D converter) 127, a memory unit 128, an iterative decoder 129, and a controller 130. The The PR channel 121 mainly includes an amplifier (amplifier) 123, an AGC (automatic gain controller) 124, a low-pass filter (low-pass filter) 125, and a waveform equalizer (equalizer, EQ) 126. The

The MO reproduction signal mo obtained from the optical head is waveform-shaped by the amplifier 123, AGC 124, low-pass filter 125, and waveform equalizer 126. When the signal is recorded on the optical disc 122 at such a high density that waveform interference occurs, the MO reproduction signal mo from the optical disc 122 is converted into a PR waveform (partial response waveform) at the output of the waveform equalizer 126. Waveform equalization can be achieved. That is, it can be considered that the PR channel 121 has been encoded with the output of the waveform equalizer 126. Thus, the substantial coding function of the recording system 110 (reference numeral 111) in the PR channel 121, turbo encodes user data _{u k,} the optical disk recording and reproducing apparatus 100 can be realized.

The reproduction system 120 includes a memory unit 128 and an iterative decoder (turbo decoder) 129. The signal after waveform equalization by the waveform equalizer 126 as described above is converted into a digital value by the A / D converter 127, and the sampling value y _i sequentially output from the A / D converter 127 is , Stored in the memory unit 128. The sampling value y _i stored in the memory unit 128 is decoded (turbo decoding) by the iterative decoder 129.

  As described above, the iterative decoder 129 has a decoder corresponding to the encoding function in the encoder 111 and the PR channel 121 in the recording system 110 and the reproduction system 120.

FIG. 3 is a diagram illustrating a basic configuration of the iterative decoder 129. The iterative decoder 129 shown in FIG. 3 includes a PR channel decoder 301, a subtractor 302, a deinterleaver (π ⁻¹ ) 303, a DEMUX and depuncture unit 304, a code decoder 305, a MUX and puncture unit 306, and a subtractor 307. , An interleaver (π) 308 and a hard discriminator 309. The PR channel decoder 301 is a decoder corresponding to the above-described encoding function on the PR channel, and performs a posteriori probability decoding (APP).

The a posteriori probability decoding of PR channel decoder 301, first of all, the sampling value _{Y (y 1, y 2,} ..., y n) inputted under the condition is detected, the bit c _i is 1 A log-likelihood ratio L (c _i *), which is the ratio of the probability P (c _i = 1 | Y) and the probability P (c _i = 0 | Y) that it becomes 0, is calculated.

Next, a priori information La (c _i ) based on an output from a code decoder 305 as described later is subtracted by a subtracter 302 from likelihood information L (c _i *) output from the PR channel decoder 301. Thus, external likelihood information Le (c) is obtained. The sequence of external likelihood information Le (c) obtained sequentially in this way is supplied to the DEMUX and depuncture unit 302 after the arrangement order is changed by the deinterleaver (π ⁻¹ ) 303.

The DEMUX and depuncture unit 302 converts the sequence of likelihood information sequentially input into a sequence of likelihood information L (u _i ) corresponding to the data bit u _k and likelihood information L (p _k corresponding to the parity bit p _k. ). Further, at the time of the decomposition, information is added according to a rule corresponding to the thinning (puncture function) rule in the encoding process (this is called a depuncture function).

The code decoder 305 is a decoder corresponding to the encoder 111 of the recording system 110 described above, and performs a posteriori probability decoding (APP). Based on the prior information L (u _k ) that is likelihood information about the data bits and the prior information L (p _k ) that is likelihood information about the parity bits as described above, the code decoder 305 performs the posterior probability about the data bits. A log likelihood ratio L (u *) expressed by (probability of u _k = 1, probability of u _k = 0) and a posteriori probability for a parity bit (probability of p _k = 1, p _k = 0 and Log likelihood ratio L (p *) expressed by

  The sequence of the log likelihood ratio L (u *) and the sequence of the log likelihood ratio L (p *) sequentially output from the code decoder 305 are supplied to the MUX and puncture unit 306. The MUX and puncture unit 306 combines each log likelihood ratio L (u *) column and L (p *) column, and thins out the information in accordance with a predetermined rule (puncture). function). As a result, likelihood information L (c *) is output from the MUX and puncture unit 306.

The prior information Le (c) (before decomposition into L (u _k ) and L (p _k )) supplied to the code decoder 305 by the subtractor 307 is converted into the above-described likelihood information L (c * ) Is subtracted from. As a result, external likelihood information La (c _i ) is obtained. This external likelihood information La (c _i ) is transmitted through the interleaver (π) 308 to the PR described above.
The information is supplied to the decoder 301 as prior information.

As mentioned above, PR
An iterative decoder 129 having a channel decoder 301 and a code decoder 305 performs an iterative decoding process using prior information supplied from the other decoder (this is called iterative decoding).

Next, the hard discriminator 309 is based on the log likelihood ratio L (u *) for the data bit u _k output from the code decoder 305 after the above iterative decoding process is performed a predetermined number of times. Determine whether the data bit is 1 or 0. This determination is made when the log likelihood ratio L (u *)> 0, the decoded data bit U _k = 1, and when the log likelihood ratio L (u *) <0, the decoding is performed. It is determined that the data bit U _k = 0. This determination result is output as decoded data U _k . Finally, the output decoded data is checked by a CRC (cyclic redundancy check) checker 310 to determine whether or not it is necessary to retry. Thus, by decoding, user data can be recorded and reproduced with high reliability.

  However, as the recording density increases, signal quality such as signal-to-noise ratio (Signal to Noise Ratio or SNR) decreases, and therefore, a demodulation system with higher accuracy is always desired. Turbo decoding enables demodulation with higher accuracy. However, on the other hand, since turbo decoding is encoded and recorded as shown in FIG. 1 and FIG. 2, there is a problem that the influence of one noise affects the entire encoded data unit. . For this reason, when errors occur in the reproduced data at the time of reproduction, it is more important than the conventional reproduction method to perform reproduction (retry) again. At the time of retrying, as shown in FIG. 1, a retry signal 140 is sent from the controller 130 to the PR channel, and the same data is reproduced again from the same playback location as that of the optical disc 122 before the retry.

  The present invention has been made in view of the above points, and an object of the present invention is to provide a data recording / reproducing system with improved error detection and correction capability when retrying data reproduction.

In order to achieve this object, the present invention provides:
In a data recording / reproducing system that records and reproduces a recording signal in which data is encoded by a convolutional code through a partial response channel, and decodes the data from the reproduced signal using iterative decoding using likelihood information.
A memory unit for storing the reproduction signal;
Based on the first time reproduction signal by reproducing, and a retry determining section for determining whether it is necessary to play again,
When the retry determination unit determines that the reproduction is necessary again, the amplitude of the reproduction signal by the first reproduction and the amplitude of the reproduction signal reproduced by the reproduction are predetermined values. It is determined whether it has the above value, and based on the determination result, the data among the reproduction signal stored in the memory unit by the first reproduction and the reproduction signal reproduced by the reproduction again the reproduction signal or use both for decoding, either to use one, and determines whether or not to use both.

  For example, in the case of a system using turbo decoding, generally, a memory unit that temporarily stores sample data at the time of reproduction is provided. Originally, since a recording / reproducing system that performs rearrangement such as turbo decoding or LDPC (low density parity check) has a memory unit, it is necessary to use an existing memory unit or to expand the memory unit. Thus, sample data can be stored. Then, according to the present invention, at the time of retrying, both sample data stored in the memory unit and sample data reproduced again at the time of retrying are demodulated as new sample data by using them under certain conditions. Can do.

  According to the present invention, at the time of retry, the SNR (signal to noise ratio) can be improved by 3 dB at the maximum, and higher error detection and correction capability than the conventional method can be obtained.

  Other objects, features and advantages of the present invention will become more apparent upon reading the following detailed description with reference to the accompanying drawings.

  Embodiments for carrying out the present invention will be described below with reference to the drawings.

  First, a first embodiment of the present invention will be described.

  FIG. 4 is a diagram showing an embodiment showing the basic configuration of the present invention. Among the components shown in FIG. 4, components given the same numbers as in FIG. 3 indicate the same components. The difference between the first embodiment of the present invention shown in FIG. 4 and the basic configuration of the conventional iterative decoder shown in FIG. 3 is that in the first embodiment of the present invention, the memory unit 128 is stored in the memory 401. , The memory 402 and the calculation unit 403.

  Next, the operation of the first embodiment of the present invention will be described below.

The PR channel 121 in FIG. 1 reproduces a recorded signal from the optical disk 122 via an optical pickup (not shown), and this reproduced MO signal is supplied to an amplifier 123, an AGC 124, and a low-pass signal. The filter 125 and the waveform equalizer 126 perform waveform equalization to PR characteristics. The signal equalized by the waveform equalizer 126 is converted into digital data y _i by the A / D converter 127. The data y _i subjected to A / D conversion in this way is stored in the memory 401. In the first embodiment, data y _i is also stored in the retry memory 402 at the same time.

The CRC checker 310 shown in FIG. 4 performs CRC check on the detection data U _k determined by the hard discriminator 309. After the CRC check of the detected data U _k , if it is determined that retry is necessary because the data is incorrect, the PR channel 121 is transmitted from the iterative decoder 129 via the controller shown in FIG. Is sent a retry signal 140. When the PR channel 121 receives the retry signal 140, the data that has been found to be an error from the CRC check result is reproduced again from the optical disk 122 and sent to the A / D converter 127 as described above.

The memory unit 128 stores the new sample data y _i _new replayed by this retry and subjected to A / D conversion only in the memory 401, while the memory 402 stores the new sample data replayed again. The data reproduced before the retry stored in the memory 402 is held as it is without being written. And the new data y _i _new reproduced again by the retry, decodes the user data using the data y _i stored reproduced before retrying in memory 402.

FIG. 5 is a diagram showing a first embodiment of the calculation unit 403 of the present invention. The calculation unit 403 according to the first embodiment illustrated in FIG. 5 includes an adder 501 and a divider 502. The calculation unit 403 of the first embodiment adds the data stored in the memory 402 before retrying and the newly reproduced data stored in the memory 401 after retrying, and the addition result is The value divided by 2 by the divider 502 is output. Thus, the calculation unit 403 of the first embodiment calculates the average value of the data y _i stored in the memory 402 and the data y _i _new stored in the memory 401, and calculates this average value as Output to the PR channel decoder 301.

In this way, the average data of the reproduction signal before and after the retry input to the PR channel decoder 301 is decoded by the above-described iterative decoder 129, and finally determined by the hard discriminator 309. , Data U _k is detected.

  Next, a second embodiment of the calculation unit of the present invention will be described.

  FIG. 6 is a diagram showing a second embodiment of the calculation unit 403 of the present invention. The calculation unit 403 of the second embodiment shown in FIG. 6 includes an adder 601, a first spike noise detector 602, a second spike noise detector 603, a divider 604, and a selector 605. Is done.

  In the calculation unit 403 of the second embodiment, the sample data stored in the memories 401 and 402 is read out and read from the memory 401 in the same manner as described in the first embodiment. The data 612, the data 610 read from the memory 402, the data 612 read from the memory 401, and the data 610 read from the memory 402 are added by the adder 601 and divided by 2 by the divider 604. Three types of data 611 are input to the selector 605.

  On the other hand, the data 612 read from the memory 401 is also input to the first spike noise detector 602, and the data 610 read from the memory 402 is also input to the second spike noise detector 603. Is done. Then, the first spike noise detector 602 and the second spike noise detector 603 each of the data 612, the data 611, and the data 610 according to the amplitude values of the data 612 and the data 610 described above. It is determined whether to output the PR channel decoder 301.

  FIG. 7 is a diagram illustrating a selection rule of the selector 605 that selects output data based on the detection results of the first spike noise detector 602 and the second spike noise detector 603. For example, the selector 605 selects the output data 611 of the divider when the output of the first spike noise detector 602 is 0 and the output of the second spike noise detector 603 is also 0. The selector 605 selects the data 612 output from the memory 401 when the output of the first spike noise detector 602 is 0 and the output of the second spike noise detector 603 is 1. The selector 605 selects the data 610 of the output of the memory 402 when the output of the first spike noise detector 602 is 1 and the output of the second spike noise detector 603 is 0. The selector 605 outputs a zero value as output data when the output of the first spike noise detector 602 is 1 and the output of the second spike noise detector 603 is also 1. In this case, outputting the zero value can be executed by clearing the contents of the memory 401 and the memory 402 to zero.

FIG. 8 is a diagram illustrating a first example of sample data. In FIG. 8, a signal 801 indicates a reproduction signal when the first data before retry is reproduced, and a signal 802 indicates a reproduction signal when the second data after retry is reproduced. The signal 801 corresponds to the data y _i described above, and the signal 802 corresponds to the data y _i _new. A sample within a range surrounded by reference numeral 803 indicates a sample where a spike noise has occurred. When ± 2 is set as the spike noise determination threshold value of the first spike noise detector 602 and the second spike noise detector 603 shown in FIG. 6, it is larger than this threshold value. Samples in the range of sample numbers 18 to 21 having an amplitude value are determined to be spike noise because the amplitude value greatly deviates from the threshold value during the first reproduction. In the second reproduction, the amplitude value of the reproduction signal corresponding to the sample determined as the spike noise is in the range of −1 to +1. Accordingly, the case where the output of the first spike noise detector 602 is 1 and the output of the second spike noise detector 603 is 0 shown in the selection rule of FIG. The selector 605 selects and outputs the data 610 output from the memory 402. That is, data determined as spike noise is not output to the PR channel decoder 301. Therefore, the sample data of sample numbers 1 to 17 and 22 or more are decoded using the data 611 indicating the average value, and the sample data of 18 to 21 are decoded using only the value of the data 610 (ie, the reference number). 804 range not used).

  FIG. 9 is a diagram illustrating a second example of sample data. In FIG. 9, a signal 901 indicates a reproduction signal when the first data before retry is reproduced, and a signal 902 indicates a reproduction signal when the second data after retry is reproduced. FIG. 9 shows an example in which both the first reproduction data before retry and the second reproduction data after retry are determined to be spike noise.

  As shown in FIG. 9, when both the first sample data before retry and the sample data at the time of retry are determined to be spike noise, the above-described selection rule shown in FIG. Corresponding to the case where the output of one spike noise detector 602 is 1 and the output of the second spike noise detector 603 is 1, the selector 605 outputs a zero value as output data.

  The iterative decoding unit 129 decodes the data having the zero value as described above output from the memory unit 128 and outputs a signal indicating that a spike noise error has occurred to the controller 130 in FIG. Then, the controller 130 notifies the user of this, and prompts the user to clean the optical disk 122 or the optical head, for example.

  Further, in the second embodiment described above, not only the range determined as the spike noise indicated by the reference numeral 804 in FIG. 8 but also sampling data having a predetermined constant length on both sides of the range. It is also possible to provide a range not used for data decoding. That is, as indicated by reference numeral 805 in FIG. 8, the area where it is determined that spike noise exists within a range of ± 1 on both sides of sample numbers 18 to 21 is expanded. That is, sample data with sample numbers 1 to 16 and 23 or more is decoded using average value data 611, and sample data with sample numbers 17 to 22 is decoded using only the value of data 610 ( That is, the range of reference number 805 is not used).

It is a figure which shows the structure of the recording / reproducing apparatus of the optical disk using iterative decoding. It is a figure which shows the structural example of the encoder used in order to encode a turbo code. It is a figure which shows the basic composition of an iterative decoder. It is a figure which shows the Example which shows the basic composition of this invention. It is a figure which shows the 1st Example of the calculation unit of this invention. It is a figure which shows the 2nd Example of the calculation unit of this invention. It is a figure which shows the selection rule of a selector. It is a figure which shows the 1st example of sample data. It is a figure which shows the 2nd example of sample data.

Claims (5)

In a data recording / reproducing system that records and reproduces a recording signal in which data is encoded by a convolutional code through a partial response channel and decodes the data from the reproduced signal using iterative decoding using likelihood information.
A memory unit for storing the reproduction signal;
Based on the first time reproduction signal by reproducing, and a retry determining section for determining whether it is necessary to play again,
When the retry determination unit determines that the reproduction is necessary again, the amplitude of the reproduction signal by the first reproduction and the amplitude of the reproduction signal reproduced by the reproduction are predetermined values. It is determined whether it has the above value, and based on the determination result, the data among the reproduction signal stored in the memory unit by the first reproduction and the reproduction signal reproduced by the reproduction again data recording and reproducing system, characterized in that the reproduced signal or use both the decoding, either to use one to determine not to use both. 2. The data recording / reproducing system according to claim 1, wherein whether to clear or clear sample data stored in the memory unit is determined according to a decoding result or a state in the middle of decoding. When it is determined that both of the reproduction signals are used for decoding the data, the data is decoded using an average value of the reproduction signal from the first reproduction and the reproduction signal reproduced by the reproduction again. The data recording / reproducing system according to claim 1, wherein: 2. A signal indicating that the reproduction signal is an error that cannot be used for decoding the data when it is determined that both of the reproduction signals are not used for decoding the data. The data recording / reproducing system described in 1. If the amplitude of the reproduction signal by the first reproduction or the reproduction signal reproduced by the reproduction again has a value greater than or equal to a predetermined value, the amplitude of the reproduction signal is greater than or equal to a predetermined value. 2. The data recording / reproducing system according to claim 1, wherein a predetermined range of the reproduction signal having the value of the first reproduction or the reproduction signal reproduced by the reproduction is not used for decoding the data.

Images