JP2005032312A - Information reproducing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information reproducing device provided with a shockproof function which is more reliable than that of the conventional device by preventing a shockproof memory from underflowing. <P>SOLUTION: The information reproducing device calculates a likelihood difference between a maximum likelihood state transition column at each time when decoding the maximum likelihood and the 2nd maximum likelihood state transition column, and performs retry control of data storage operation to the shockproof memory on the basis of the likelihood difference. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

【００３３】
ここでｙ〔ｋ〕：ＰＲ出力値、ｄ_ｋ：現在の時刻ｋにおける記録符号、ｄ_ｋ−１：１クロック前の時刻ｋ−１における記録符号、ｄ_ｋ−２：２クロック前の時刻ｋ−２における記録符号、ｄ_ｋ−３：３クロック前の時刻ｋ−３における記録符号である。
【００３４】
図６に（１、７）ＲＬＬとＰＲ（１、２、２、１）に基づく状態遷移図を示す。各状態はＰＲ値に応じて次の状態に遷移する。続いて、状態遷移に応じて復号ビットを出力する。また、図６の状態遷移図を時間軸方向に展開したトレリス図を図７に示す。
【００３５】
ビタビ復号では、図７に示すトレリス図に従って、各状態の時刻ｋでのパスメトリックｍ_０００［ｋ］〜ｍ_１１１［ｋ］は、時刻ｋ−１の所定状態のメトリックｍ_０００［ｋ−１］〜ｍ_１１１［ｋ−１］と時刻ｋでの実際のＰＲ出力値ｙ［ｋ］とを用いて以下の計算式Ｘで求められる。
【００３６】
【外２】

【００３７】
状態Ｓ（ｄ_ｋ−２、ｄ_ｋ−１、ｄ_ｋ）とは、現時刻の復号データがｄ_ｋ、１時刻前の復号データがｄ_ｋ−１、２時刻前の復号データがｄ_ｋ−２であることを示す。
【００３８】
状態Ｓ０００、状態Ｓ００１、状態Ｓ１１０及び状態Ｓ１１１では、各時刻毎に２つのパスが合流するため、合流する２つのパスのうち、パスメトリック値が小さい方を生き残りパスとして選択する。ここで、パスとはそれまで経由した状態の履歴であり、状態遷移列とも呼ぶ。
【００３９】
また、図７のトレリス線図に示すように時刻ｋ−１から時刻ｋの各状態に遷移する際には、各遷移により定まるＰＲ（１、２、２、１）の理想出力値と実際の再生信号をＰＲ（１、２、２、１）処理した出力値とのユークリッド距離をブランチ・メトリックとして加算する。なお、ＰＲ（１、２、２、１）の理想出力値は０、１、２、３、４、５、６の７値となる。
【００４０】
次にビタビ復号３０２ならびにＳＡＭ値算出３０５の動作を説明する。図８は、ビタビ復号３０２とＳＡＭ値算出３０５の機能ブロックを示したものである。図８において８０１はメトリック算出、８０２は加算／比較／選択、８０４はパスメモリ、８０５はメトリック差算出、８０６は標準偏差算出、８０７、８０８はレジスタである。
【００４１】
ＰＲ（１、２、２、１）から出力されたサンプル値はビタビ復号３０２に入力されメトリック算出８０１においてブランチ・メトリックが算出される。ブランチ・メトリックは加算／比較／選択８０２において、前記計算式ｘが実施される。
【００４２】
入力された時刻ｋにおけるブランチ・メトリックと時刻ｋ−１での各状態のパスメトリック値を加算し、その加算結果を比較して、小さい方の値を時刻ｋにおける各状態のパスメトリック値として選択し、この新しいパスメトリック値をレジスタ８０７に格納する。選択されたパスメトリック値の状態遷移に応じて、パスメモリ８０４は最も確からしいデータ系列を出力する。
【００４３】
つまり、パスメトリック値を更新してゆくことで、ｎ時刻前からのパス（状態遷移列）の確からしさが計算できる。その中で所定の区間における最小のパスメトリック値に対応する状態遷移列のデータ系列を出力することで、最も確からしいパスに対応するデータ系列を復号結果として出力する。
【００４４】
続いて、ＳＡＭ値算出３０５について説明する。ＳＡＭ（ＳｅｑｕｅｎｃｅｄＡｍｐｌｉｔｕｄｅ Ｍａｒｇｉｎ）はビタビ復号において最も確からしいパスと次に確からしいパスとの差であり、たとえば、Ｔｉｍ Ｐｅｒｋｉｎｓ ａｎｄ Ｚａｃｈａｒｙ Ａ．Ｋｅｉｒｎ、“Ａ Ｗｉｎｄｏｗ−Ｍａｒｇｉｎ−Ｌｉｋｅ Ｐｒｏｃｅｄｕｒｅ ｆｏｒ Ｅｖａｌｕａｔｉｎｇ ＰＲＭＬ Ｃｈａｎｎｅｌ Ｐｅｒｆｏｒｍａｎｃｅ”、ＩＥＥＥ Ｔｒａｎｓａｃｔｉｏｎｏｎ Ｍａｇｎｅｔｉｃｓ、Ｖｏｌ．２１ Ｎｏ．２、Ｍａｒｃｈ １９９５などに報告されている。
【００４５】
図８において、メトリック差８０５は、加算／比較／選択８０２出力結果から最も確からしいパスと次に確からしいパスのパスメトリック差をレジスタ８０８に格納する。
【００４６】
ビタビ復号においては、最も確からしいパス、つまりデータ系列が各時刻でもつサンプル値と波形等化理想値との累積誤差（パスメトリック値）が最小となるパスを選択するものである。したがって、最も確からしいパスと次に確からしいパスのメトリックの差が大きいほど前者の信頼性が高いので復号データの信頼性も高い。逆に、最も確からしいパスと次に確からしいパスのメトリック差が小さい場合、二者択一に誤りが生じる可能性が高く、復号データの信頼性が低くなる。メトリック差算出８０５の出力値をヒストグラム化したものを図４に示す。図４において、横軸はメトリック差、縦軸は頻度数であり、各サンプル点におけるメトリック差を累積した複数のヒストグラムのなかで、最もメトリック差が少ない領域に位置するヒストグラムを抽出例示している。図４において、再生信号の品質を変えたものを３種類プロット表示している。横軸のメトリック差が小さい領域は、最も確からしいパスと次に確からしいパスのメトリックが近接するため、ビタビ復号でパスの識別エラー、すなわち復号データの誤りを起こしやすい。したがって、図４の４ａ、４ｂ、４ｃにおいて、メトリック差が小さい領域の頻度に着目すると、４ｃ＞４ｂ＞４ａであることから、再生信号４ｃがビタビ復号において誤判定する可能性がもっとも高い。換言すれば信号品位が悪いということである。続いて再生信号４ｂ、そして最も誤判定の確率が低い、つまり信号品位が良好なのは再生信号４ａである。
【００４７】
図８の標準偏差算出８０６は、図４に示されるメトリック差のヒストグラムから標準偏差を求める機能を有する。図４の例では、標準偏差は４ｃ＞４ｂ＞４ａとなり、標準偏差の程度によって再生信号の品質を判定することができる。
【００４８】
図５は、ＳＡＭ値とエラーレート（ビット誤り率）の相関を示した図である。同図のとおり、ビタビ復号から得られるＳＡＭ値は、エラーレートと高い相関を示す。
【００４９】
引き続き図３の再生プロセッサに戻る。復調３０３は復号されたデータ列を（１、７）ＲＬＬ復調する復調装置であり、後段のＥＣＣ３０４によってエラー訂正された情報再生信号がデータバス２１２に出力される。
【００５０】
＜リトライ動作＞
図１は、本発明の特徴を最もよく表す図であり、情報再生装置２００において再生動作中にＳＡＭ値を監視してエラー処理を行う動作フローチャートを示す。
○ステップＳ１０１：データ再生
上位コマンドを受けて、ディスクから指示アドレスのデータを再生し、メモリに一時蓄積ならびに一定時間で読み出しを行いながら情報再生処理を行う。
○ステップＳ１０２：ＳＡＭ値算出
ディスク媒体から再生した信号から、ＳＡＭ値を算出しレジスタに格納する。
○ステップＳ１０３：ＳＡＭ値判定
Ｓ１０２で得られたＳＡＭ値を、規定値ｋと比較する。
【００５１】
規定値ｋとは、後段のＥＣＣ３０４においてエラー訂正処理を行ってもエラーを完全に訂正できない臨界値を設定している。
【００５２】
ＳＡＭ値が規定値ｋより小さい場合、再生信号はＥＣＣにおいて完全にエラー訂正可能であるので、リトライ不要と判断してフロー終了する。
【００５３】
規定値ｋより大きい場合、再生信号はＥＣＣでエラーを訂正しきれず情報出力に支障があると判断する。ステップＳ１０４に移行する。
○ステップＳ１０４：アドレス照会
ステップＳ１０３でＳＡＭ値が規定値ｋ以上になったディスク媒体上のアドレス位置を算出する。アドレス位置算出は、ＳＡＭ値判定の際に、当該ＳＡＭ値の計測データ列と再生アドレス情報を付随して監視することによってアドレスを照会することができる。
○ステップＳ１０５：リトライ
ステップＳ１０４で照会されたアドレス位置にシークして、再生リトライ動作を実施する。
【００５４】
再生リトライによって新たに再生されたデータは、再びリトライ移行判断を監視する本実施形のフローにしたがってＳＡＭ値が算出される。新たに再生されたＳＡＭ値に異常がなければ、ステップＳ１０４で照会されたＳＡＭ値異常部分のみメモリ上のデータが上書きされる。かかる後、上書きされたデータを含めて復調処理、ＥＣＣ処理が施される。
【００５５】
（その他の実施形態）
本実施形のショックプルーフ機能は、ＳＡＭ値がエラーレートと相関高い精度で再生信号品位を判定できることから、ＳＡＭ値に応じてＥＣＣの訂正能力を変更し、リトライ回避を図るものである。
【００５６】
＜リトライ動作＞
図１２は、本発明の特徴を最もよく表す図であり、情報再生装置２００において再生動作中にＳＡＭ値を監視してエラー処理を行う動作フローチャートを示す。
○ステップＳ１２０１：データ再生
上位コマンドを受けて、ディスクから指示アドレスのデータを再生し、メモリに一時蓄積ならびに一定時間で読み出しを行いながら情報再生処理を行う。
○ステップＳ１２０２：ＳＡＭ値算出
ディスク媒体から再生した信号から、ＳＡＭ値を算出しレジスタに格納する。
○ステップＳ１２０３：ＳＡＭ値判定
Ｓ１２０２で得られたＳＡＭ値を、規定値ｋ、ｍ（ｋ＜ｍ）と比較する。
【００５７】
規定値ｋとは、後段のＥＣＣ３０４において通常のエラー訂正処理を行うことによってエラーを完全訂正できる値を設定している。
【００５８】
規定値ｍとは、後段のＥＣＣ３０４においてＣ１訂正Ｃ２訂正の繰り返しループを通常動作より増やしてエラー訂正能力を最大限に高めることによって完全訂正できる臨界値を設定している。
【００５９】
１）ｋ＞ＳＡＭ値のとき
ＳＡＭ値が規定値ｋより小さい場合、再生信号はＥＣＣにおいて完全にエラー訂正可能であると判断し、リトライ不要と判断してフロー終了する。
２）ｍ≧ＳＡＭ値≧ｋのとき
ＳＡＭ値が規定値ｋ以上、かつ規定値ｍ以下であるとき、再生信号は通常ＥＣＣ設定でエラーを訂正しきれず情報出力に支障があると判断する。ステップＳ１２０４に移行。
３）ＳＡＭ値＞ｍ
ＳＡＭ値が規定値ｍより大きい場合、再生信号は最大訂正能力を設定したＥＣＣでもエラーを訂正しきれず情報出力に支障があると判断する。ステップＳ１２０５に移行。
○ステップＳ１２０４：ＥＣＣ訂正能力変更
ＥＣＣ３０４のＣ１訂正Ｃ２訂正の繰り返しループを増やし、最大限繰り返すモードに設定する。
○ステップＳ１２０５：アドレス照会
ステップＳ１２０３でＳＡＭ値がＳＡＭ値＞ｍとなったディスク媒体上のアドレス位置を照会する。アドレス位置照会は、ＳＡＭ値判定の際に、当該ＳＡＭ値の計測データ列と再生アドレス情報を付随して監視することによって照会することができる。
○ステップＳ１２０６：リトライ
ステップＳ１２０５で算出されたアドレス位置にシークして、再生リトライ動作を実施する。
【００６０】
再生リトライによって新たに再生されたデータは、再びリトライ移行判断を監視する本実施形のフローにしたがってＳＡＭ値が算出される。新たに再生されたＳＡＭ値に異常がなければ、ステップＳ１２０５で照会されたＳＡＭ値異常部分のみメモリ上のデータが上書きされる。かかる後、上書きされたデータを含めて復調処理、ＥＣＣ処理が施される。
【００６１】
＜その他の判定指標＞
これまで詳述した実施形では、ビタビ復号処理から算出されるＳＡＭ値を監視してリトライ移行するものであったが、本実施形では、他の評価指標を用いてショックプルーフ動作の信頼性を高める。
【００６２】
その他の評価指標として、たとえば特開２００３−１４１８２３公報に開示されているような評価指標を用いても本発明を適用することができる。
【００６３】
具体的には、再生信号において図１３に示すような特定パターンを検出する。そして再生信号のなかで特定パターンが含まれている個所のみからメトリック差の検出を行う。図１３において特定パターンとは、ＰＲ（１、２、２、１）のビタビ復号において、最小ユークリッド距離１０を与えるパターンの組み合わせである。
【００６４】
メトリック差算出部では、ビタビ復号の復号データが図１３に示すパターンのいずれかに一致する場合、パスＡのメトリック及びパスＢのメトリックからメトリック差を算出する。図１４にパスＡとパスＢのトレリスを例示する。図１４のパスＡとパスＢのユークリッド距離は１０であるので、メトリック差の分布は図４に示すように平均値１０の正規分布のようになる。以降の処理は、前実施形と同様である。
【００６５】
特定パターンとは、エラー発生頻度の高いデータ列のことである。このように特定パターンの組み合わせに着目してメトリック差を抽出し、そのヒストグラムの偏りを検出することによって、評価指標の検出感度を上げることができる。すなわち、エラー発生しやすいパターンに限定しているので、より短いデータ長の演算処理で、再生信号品位を的確に把握することができる。
【００６６】
なお、本実施形におけるＳＡＭ値は、各時刻におけるメトリック差を累積したヒストグラムの偏りから標準偏差を求めてＳＡＭ値として抽出している。しかしながら、本発明の主旨から明白なように、ＳＡＭ値算出は前記標準偏差に限定されるものではない。一例として、図１０に示すとおり、メトリック差の小さい領域を“しきい値＝Ｔ”で判別し、しきい値Ｔ以下の頻度を勘案することでもＳＡＭ値相当の指標として適用可能である。
【００６７】
また、本実施形はハードウエアを踏まえた構成について説明したが、これに制約されることなく、ソフトウエアによるプログラム処理によっても実現可能である。
【００６８】
【発明の効果】
上述した通り、本発明の情報再生装置によれば、ディスク媒体から得られる再生信号の品位をＥＣＣ処理より早く判定することができるので、再生リトライの移行判断を瞬時に行うことができる。したがって、高精細映像情報のような転送レートの高い情報再生であっても、ショックプルーフメモリのアンダーフローを未然防止し、信頼性の高いショックプルーフ機能を実現することができる。
【００６９】
また、最尤復号の尤度差に基づいてＥＣＣ訂正能力を局所的、かつ選択的に変更することができるので不要なリトライ動作を回避することができる。
【００７０】
更に、ディスク媒体上の局所的なエラー発生個所を瞬時に特定することができるので、再生リトライの実施時間を必要最小限のデータ再生に抑制することができる。
【図面の簡単な説明】
【図１】本発明の情報再生装置の動作フローチャートである。
【図２】本発明の情報再生装置の機能ブロック図である。
【図３】図２図示の情報再生装置の再生プロセッサの機能ブロック図である。
【図４】ＳＡＭを説明する図である。
【図５】ＳＡＭとエラーレートの相関を示す図である。
【図６】ＰＲ（１，２，２，１）に基づく状態遷移図である。
【図７】ＰＲ（１，２，２，１）に基づくトレリス図である。
【図８】図３図示のビタビ復号とＳＡＭ値算出の詳細な機能ブロック図である。
【図９】ショックプルーフメモリの動作説明図である。
【図１０】ＳＡＭ値算出の他の実施例を説明する図である。
【図１１】ＥＣＣ処理ブロックを説明する図である。
【図１２】本発明の情報再生装置の他の動作フローチャートである。
【図１３】本発明における他の評価指標を説明する図である。
【図１４】本発明における他の評価指標を説明するトレリス図である。
【符号の説明】
２０１ ディスク
２０２ 光ピックアップ
２０３ スピンドルモータ
２０４ ＲＦアンプ
２０５ ＡＧＣ・Ｆｉｌｔｅｒ
２０６ Ａ／Ｄ
２０７ 再生プロセッサ
２０８ サーボＤＳＰ
２０９ ドライブコントローラ
２１０ メモリ制御
２１１ メモリ
２１２ データバス
３０１ 波形等化
３０２ ビタビ復号
３０３ 復調
３０４ ＥＣＣ
８０１ メトリック算出
８０２ 加算・比較・選択
８０４ パスメモリ
８０５ メトリック差算出
８０６ 標準偏差算出
８０７、８０８ レジスタ[0001]
BACKGROUND OF THE INVENTION
The present invention is applied to a reproducing apparatus that reads information from a disk-shaped medium, and particularly stores a signal read from the medium temporarily in a memory, and then reproduces the data while maintaining the continuity of the reproduced data from the memory. The present invention relates to an information reproducing apparatus including a proof memory.
[0002]
[Prior art]
A drive device using a disk-like medium has been developed as a large-capacity data storage device because it is mainly used as a peripheral device such as a personal computer and has high portability of the medium and high reliability of repeated recording and reproduction. In particular, it has data accessibility advantages that cannot be realized with tape-like recording media, such as that continuous data can be recorded or reproduced intermittently, or arbitrary partial data can be accessed at high speed and randomly from continuous length data. ing. Nowadays, the diameter and density of disk-shaped media have been further increased and the transfer speed has been increased, and as a main axis of multimedia equipment, it is applied to video equipment that assumes mobile operation in an outdoor environment.
[0003]
Such an information reproducing apparatus has a so-called shock proof function in which data is intermittently reproduced from a disk medium at a high speed, temporarily stored in a memory, and then transmitted at a real time rate.
[0004]
FIG. 9 is a diagram for explaining the shock proof function. FIG. 9 (9-1) shows the amount of storage in the memory that controls the shock proof operation on the vertical axis and the time transition of the remaining memory on the horizontal axis. FIG. 9 (9-2) is a timing signal for writing data from the disk medium to the memory. Only in the H section (0-t1, t2-t3, t4-t5), information is reproduced from the disk medium at high speed and stored in the memory. accumulate.
[0005]
FIG. 9 (9-3) shows information transmission timing from the memory, and information is read from the memory during the H section (t1-).
[0006]
The operation of FIG. 9 will be described. It is assumed that the reproducing operation is started at time 0 in the information reproducing apparatus. As shown in FIG. 9 (9-1), data is read from the disk medium at high speed and written to the memory between time 0 and time t1. When the remaining amount of memory reaches the a level (time t1), reading from the disk is temporarily stopped. Subsequently, reading from the memory is performed according to the timing signal shown in (9-3). Usually, the reading speed is from a fraction to several tenths of the writing speed to the memory. When the memory storage amount becomes b level, the timing signal (9-2) turns to H and data reading from the disk is resumed.
[0007]
This shock proof function can minimize the data access time to the disk medium, and can respond to various errors caused by sudden temperature changes or disturbances such as shaking, vibration, and shock when used as a mobile device. Thus, continuous reproduction of highly reliable information can be realized.
[0008]
Conventionally, many reproducing apparatuses having such a shock proof function have been proposed. In Japanese Patent Laid-Open No. 2002-183968, it is determined whether or not information reproduction has been performed normally by looking at the result of ECC (error correction processing) performed on a reproduction signal obtained from a disk medium. A technique for performing playback retry by changing parameters related to playback processing such as power is disclosed. Japanese Patent Application Laid-Open No. 2001-133501 discloses a technique for identifying the error occurrence position by monitoring the frequency of error occurrence and monitoring the fluctuation of the reproduction signal amplitude value when the reproduction signal is ECCed. Yes. In such a conventional technique, the quality of a reproduction signal is determined by ECC processing, an error occurrence or error correction is detected, and an error occurrence location in the disk medium is specified and reproduction retry is executed. Thus, there was an aim to increase the reliability of continuous information reproduction.
[0009]
On the other hand, when an information playback device is used for mobile applications, unexpected shock or vibration may be applied to the housing. In such a case, tracking that follows a predetermined track on the disk medium is lost, causing a servo error. there is a possibility. In this case, by monitoring the displacement of the tracking error signal in the servo processing system, the servo error flag is notified to the upper management mechanism, the servo is returned, and then reproduction retry is performed.
[0010]
[Patent Document 1]
JP 2002-183968 A
[Patent Document 2]
JP 2001-1335012 A
[0011]
[Problems to be solved by the invention]
However, the following problems are apparent in an information reproducing apparatus having a conventional shockproof function. The technique disclosed in the above-mentioned Japanese Patent Application Laid-Open No. 2002-183968 or Japanese Patent Application Laid-Open No. 2001-1335012 performs an ECC process on a reproduction signal, and determines whether or not to retry reproduction based on the decoding result.
[0012]
FIG. 11 is a diagram showing an ECC block. As is well known, the ECC decoding process performs error correction processing on reproduction data in units of ECC blocks as shown in FIG. The ECC block is encoded with C1 parity in the horizontal direction and C2 parity in the vertical direction. The ECC block horizontal error correction processing is called C1 correction, and the vertical error correction processing is called C2 correction, and C1 and C2 error correction are repeated. Error correction is performed by carrying out. Here, the data unit of the ECC block is generally about 32 KB to 64 KB, but the ECC decoding result is obtained because the information reproduction data satisfies the ECC block, and error correction processing such as C1 correction and C2 correction is performed. Later.
[0013]
That is, the signal reproduced from the disc is temporarily stored in the memory, and then subjected to decoding processing and ECC calculation processing. In the ECC calculation processing, correction processing is performed a plurality of times on data in units of ECC blocks. Due to these processes, a delay of at least several seconds is caused until the error condition of the reproduction signal is determined. If the reproduction retry transition determination that the error cannot be corrected is made after such a large delay, the processing delay further spreads from the identification of the address of the error portion to the execution of the reproduction retry. Since the increase in the transfer speed of information reproduction is steadily progressing, it is necessary to determine whether to perform the reproduction retry at the earliest possible stage in order to guarantee continuous data reproduction.
[0014]
In addition, the conventional playback retry operation that monitors servo errors can quickly detect disturbances such as shaking and vibration to the chassis, but is there any local signal quality degradation due to dust or dirt adhering to the disk medium? It is impossible to detect the deterioration of the reproduction signal due to environmental temperature change in mobile operation.
[0015]
The present invention has been made in view of the above problems, and an object of the present invention is to provide an information reproducing apparatus that prevents underflow of a shock proof memory and has a more reliable shock proof function than conventional ones. .
[0016]
[Means for Solving the Problems]
The object is to binarize the reproduced signal by maximum likelihood decoding that selects the most probable state transition sequence from m state transition sequences that transition from the first state at time k-n to the second state at time k. A maximum likelihood decoder, an arithmetic unit for calculating a likelihood difference between the most probable state transition sequence and the second most probable state transition sequence, and a shock proof memory based on the likelihood difference obtained by the arithmetic unit This is achieved by an information reproducing apparatus having a control means for performing retry control of the data storage operation.
[0017]
(Function)
According to the configuration described in claim 1 of the present invention, since the quality of the reproduction data can be determined earlier than the ECC processing, the address of the local error occurrence location on the disk medium is specified and the instantaneous operation to the reproduction retry operation is performed. The transition can be achieved and the reliability of the shock proof function can be improved compared to the conventional one.
[0018]
Further, according to the configuration described in claim 2, unnecessary reproduction retry can be avoided by enhancing the correction capability of the ECC processing.
[0019]
In addition, according to the configuration described in claim 3, since it is possible to limit the retry of the reproduction data only in the poor quality, the time required for the retry can be shortened to the minimum necessary.
[0020]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
[0021]
The present invention is applied to, for example, an information reproducing apparatus 200 as shown in FIG.
[0022]
<Overall Configuration of Information Reproducing Device 200 and Series of Operations>
The information reproducing apparatus 200 includes a disk-shaped medium (hereinafter referred to as “disk”) 201, an optical pickup 202, a spindle motor 203, a servo DSP (Digital Signal Processor) 208, an RF preamplifier 204, an AGC (Auto Gain Control) / Filter 205, A / D206, playback processor 207, drive controller 209, memory 211, memory control 210, and data bus 212.
[0023]
In the information reproducing apparatus 200, the memory 211 is a memory space used by time sharing in each functional block via the data bus 212, and is controlled and managed by the memory control 210 via the drive controller 209.
[0024]
The drive controller 209 includes a central processing unit (CPU), and receives and executes a user-specified command via the data bus 212 or executes a predetermined program to control the entire system of the information reproducing apparatus 200.
[0025]
The optical pickup 202 includes an optical coupling group (not shown), a laser element, an actuator, a light receiving element, and the like, and performs laser light irradiation and information detection on the disk 201. For information detection, the amount of light reflected from the disk medium is detected and the reflected light is photoelectrically converted to obtain a reproduced electrical signal.
[0026]
The servo DSP 208 has a function of controlling the entire drive control to the disk 201, and controls the position of the optical pickup 202 on the disk 201 at a predetermined address by a traverse motor (not shown). Further, focus control and tracking control are performed by controlling an actuator in the optical pickup 202. Further, the laser 201 is controlled to be driven at a predetermined number of rotations by the laser emission light amount control in the optical pickup 202 and the spindle motor 203.
[0027]
Information reproduction processing from a disc will be described. The weak reproduction signal obtained from the optical pickup 202 is amplified by the RF preamplifier 204. Thereafter, the AGC / Filter 205 performs gain control and band limitation to a predetermined level. The A / D 206 converts the input analog signal into a digital signal. These are optimally controlled to predetermined characteristics from a drive controller 209 by a control line (not shown). Subsequently, reproduction signal processing is performed in the reproduction processor 207 and information data is transmitted to the data bus 212.
[0028]
<Detailed functions and series of operations of the playback processor 207>
Next, reproduction signal processing of the reproduction processor 207 will be described in detail with reference to FIG. In FIG. 3, 301 is waveform equalization, 302 is Viterbi decoding, 303 is demodulation, 304 is ECC (Error Correct Code), 305 is SAM (Sequential Amplitude Margin) value calculation.
[0029]
The signal input to the reproduction processor 207 is subjected to a PLL (not shown) and a clock component synchronized with the reproduction signal is extracted. This clock is used for the processing timing of the reproduction processor.
[0030]
The disc information shown in the present embodiment is recorded by performing NRZI conversion on a known (1, 7) RLL code, and the minimum inversion interval is 2.
[0031]
Waveform equalization 301 performs waveform equalization having the characteristics of partial response PR (1, 2, 2, 1). PR (1, 2, 2, 1) is represented by the following formula.
[0032]
[Outside 1]

[0033]
Where y [k]: PR output value, d _k : Recording code at current time k, d _k-1 : 1 recording code at time k-1 before clock, d _k-2 : Recording code at time k-2 two clocks before, d _k-3 : A recording code at time k-3 three clocks before.
[0034]
FIG. 6 shows a state transition diagram based on (1, 7) RLL and PR (1, 2, 2, 1). Each state transitions to the next state according to the PR value. Subsequently, the decoded bit is output according to the state transition. FIG. 7 shows a trellis diagram obtained by developing the state transition diagram of FIG. 6 in the time axis direction.
[0035]
In the Viterbi decoding, the path metric m at time k in each state according to the trellis diagram shown in FIG. ₀₀₀ [K] to m ₁₁₁ [K] is a metric m in a predetermined state at time k−1. ₀₀₀ [K-1] to m ₁₁₁ Using [k−1] and the actual PR output value y [k] at time k, the following calculation formula X is used.
[0036]
[Outside 2]

[0037]
State S (d _k-2 , D _k-1 , D _k ) Means that the decoded data at the current time is d _k The decoded data one hour before is d _k-1 The decoded data two hours ago is d _k-2 Indicates that
[0038]
In the state S000, the state S001, the state S110, and the state S111, since two paths merge at each time, the path with the smaller path metric value is selected as the surviving path from the two paths that merge. Here, the path is a history of the state that has been passed, and is also called a state transition sequence.
[0039]
Also, as shown in the trellis diagram of FIG. 7, when transitioning from time k-1 to each state of time k, the ideal output value of PR (1, 2, 2, 1) determined by each transition and the actual The Euclidean distance from the output value obtained by PR (1, 2, 2, 1) processing of the reproduction signal is added as a branch metric. The ideal output value of PR (1, 2, 2, 1) is 7 values of 0, 1, 2, 3, 4, 5, 6.
[0040]
Next, operations of the Viterbi decoding 302 and the SAM value calculation 305 will be described. FIG. 8 shows functional blocks of the Viterbi decoding 302 and the SAM value calculation 305. In FIG. 8, 801 is a metric calculation, 802 is an addition / comparison / selection, 804 is a path memory, 805 is a metric difference calculation, 806 is a standard deviation calculation, and 807 and 808 are registers.
[0041]
Sample values output from PR (1, 2, 2, 1) are input to Viterbi decoding 302, and a metric calculation 801 calculates a branch metric. The branch metric is added / compared / selected 802 with the formula x implemented.
[0042]
Add the branch metric at time k and the path metric value of each state at time k−1, compare the addition results, and select the smaller value as the path metric value of each state at time k The new path metric value is stored in the register 807. In response to the state transition of the selected path metric value, the path memory 804 outputs the most likely data series.
[0043]
In other words, by updating the path metric value, it is possible to calculate the probability of the path (state transition sequence) from time n before. Among them, by outputting a data sequence of a state transition sequence corresponding to the minimum path metric value in a predetermined section, a data sequence corresponding to the most probable path is output as a decoding result.
[0044]
Next, the SAM value calculation 305 will be described. SAM (Sequenced Amplitude Margin) is the difference between the most probable path and the next most probable path in Viterbi decoding. For example, Tim Perkins and Zachary A. Keirn, “A Window-Margin-Like Procedure for Evaluating PRML Channel Performance”, IEEE Transactionon Magnetics, Vol. 21 No. 2, March 1995.
[0045]
In FIG. 8, the metric difference 805 stores the path metric difference between the most probable path and the next most likely path from the addition / comparison / selection 802 output result in the register 808.
[0046]
In Viterbi decoding, the most probable path, that is, the path that minimizes the accumulated error (path metric value) between the sample value of the data series and the waveform equalization ideal value at each time is selected. Therefore, the larger the difference between the most probable path and the next most probable path metric, the higher the reliability of the former, and the higher the reliability of the decoded data. Conversely, if the metric difference between the most probable path and the next most probable path is small, there is a high possibility that an error will occur in the alternative, and the reliability of the decoded data will be low. FIG. 4 shows a histogram of the output values of the metric difference calculation 805. In FIG. 4, the horizontal axis is the metric difference, and the vertical axis is the frequency number. Among the plurality of histograms obtained by accumulating the metric difference at each sample point, the histogram located in the region where the metric difference is the smallest is illustrated. . In FIG. 4, three types of plots in which the quality of the reproduced signal is changed are displayed. In the region where the metric difference on the horizontal axis is small, the most probable path and the next most probable path metric are close to each other, so that a Viterbi decoding is likely to cause a path identification error, that is, an error in decoded data. Therefore, in 4a, 4b, and 4c of FIG. 4, when attention is paid to the frequency of the region where the metric difference is small, since 4c>4b> 4a, the possibility that the reproduced signal 4c is erroneously determined in Viterbi decoding is the highest. In other words, the signal quality is poor. Subsequently, the reproduction signal 4b, and the reproduction signal 4a has the lowest probability of erroneous determination, that is, the signal quality is good.
[0047]
The standard deviation calculation 806 in FIG. 8 has a function of obtaining the standard deviation from the metric difference histogram shown in FIG. In the example of FIG. 4, the standard deviation is 4c>4b> 4a, and the quality of the reproduction signal can be determined by the degree of the standard deviation.
[0048]
FIG. 5 is a diagram showing the correlation between the SAM value and the error rate (bit error rate). As shown in the figure, the SAM value obtained from Viterbi decoding shows a high correlation with the error rate.
[0049]
Subsequently, the process returns to the reproduction processor of FIG. The demodulator 303 is a demodulator that performs (1, 7) RLL demodulation on the decoded data string, and an information reproduction signal that has been error-corrected by the ECC 304 at the subsequent stage is output to the data bus 212.
[0050]
<Retry operation>
FIG. 1 is a diagram that best represents the features of the present invention, and shows an operation flowchart in which an error processing is performed by monitoring a SAM value during a reproducing operation in the information reproducing apparatus 200.
Step S101: Data reproduction
In response to the upper command, the data of the designated address is reproduced from the disk, and the information reproduction process is performed while temporarily storing it in the memory and reading it at a predetermined time.
Step S102: SAM value calculation
The SAM value is calculated from the signal reproduced from the disk medium and stored in the register.
Step S103: SAM value determination
The SAM value obtained in S102 is compared with the specified value k.
[0051]
The specified value k is set to a critical value that cannot completely correct an error even if error correction processing is performed in the ECC 304 at the subsequent stage.
[0052]
If the SAM value is smaller than the specified value k, the reproduction signal can be completely error-corrected in ECC, so that it is determined that no retry is necessary, and the flow ends.
[0053]
If the value is larger than the specified value k, it is determined that the reproduced signal cannot correct the error by ECC, and there is a problem in information output. The process proceeds to step S104.
○ Step S104: Address inquiry
In step S103, the address position on the disk medium where the SAM value is equal to or greater than the specified value k is calculated. In the address position calculation, when the SAM value is determined, the address can be inquired by monitoring the measurement data string of the SAM value and the reproduction address information.
○ Step S105: Retry
Seek to the address position inquired in step S104, and perform a reproduction retry operation.
[0054]
The SAM value of the data newly reproduced by the reproduction retry is calculated according to the flow of the present embodiment for monitoring the retry transition determination again. If there is no abnormality in the newly reproduced SAM value, the data in the memory is overwritten only in the SAM value abnormality part inquired in step S104. After that, demodulation processing and ECC processing are performed including overwritten data.
[0055]
(Other embodiments)
The shock proof function of the present embodiment can avoid the retry by changing the ECC correction capability according to the SAM value since the reproduction signal quality can be determined with high accuracy in correlation with the error rate of the SAM value.
[0056]
<Retry operation>
FIG. 12 is a diagram that best represents the characteristics of the present invention, and shows an operation flowchart in which the SAM value is monitored and error processing is performed during the reproducing operation in the information reproducing apparatus 200.
○ Step S1201: Data reproduction
In response to the upper command, the data at the designated address is reproduced from the disk, and the information reproduction process is performed while temporarily storing it in the memory and reading it at a predetermined time.
Step S1202: SAM value calculation
The SAM value is calculated from the signal reproduced from the disk medium and stored in the register.
Step S1203: SAM value determination
The SAM value obtained in S1202 is compared with specified values k and m (k <m).
[0057]
The specified value k is set to a value that allows the error to be completely corrected by performing normal error correction processing in the ECC 304 at the subsequent stage.
[0058]
The specified value m is set to a critical value that can be completely corrected by increasing the error correction capability to the maximum by increasing the number of iteration loops for C1 correction and C2 correction from the normal operation in the subsequent ECC 304.
[0059]
1) When k> SAM value
If the SAM value is smaller than the specified value k, it is determined that the reproduced signal can be completely error-corrected in the ECC, it is determined that no retry is necessary, and the flow ends.
2) When m ≧ SAM value ≧ k
When the SAM value is not less than the specified value k and not more than the specified value m, it is determined that the reproduction signal cannot normally correct the error by the ECC setting and has a problem in information output. The process proceeds to step S1204.
3) SAM value> m
If the SAM value is larger than the specified value m, it is determined that the reproduced signal cannot correct the error even with the ECC for which the maximum correction capability is set, and the information output is hindered. The process proceeds to step S1205.
Step S1204: Change ECC correction capability
The repeat loop of C1 correction C2 correction of ECC304 is increased, and the mode is set to repeat as much as possible.
○ Step S1205: Address inquiry
In step S1203, the address position on the disk medium in which the SAM value becomes SAM value> m is inquired. The address position inquiry can be made by observing the measurement data string of the SAM value and the reproduction address information when determining the SAM value.
○ Step S1206: Retry
The playback retry operation is performed by seeking to the address position calculated in step S1205.
[0060]
The SAM value is calculated for the data newly reproduced by the reproduction retry according to the flow of this embodiment for monitoring the retry transition determination again. If there is no abnormality in the newly reproduced SAM value, the data in the memory is overwritten only in the SAM value abnormality part inquired in step S1205. After that, demodulation processing and ECC processing are performed including overwritten data.
[0061]
<Other judgment indicators>
In the embodiment described in detail so far, the SAM value calculated from the Viterbi decoding process is monitored and the retry shift is performed, but in this embodiment, the reliability of the shock proof operation is improved by using another evaluation index. Increase.
[0062]
As another evaluation index, for example, an evaluation index as disclosed in JP-A-2003-141823 can be used to apply the present invention.
[0063]
Specifically, a specific pattern as shown in FIG. 13 is detected in the reproduction signal. Then, the metric difference is detected only from the portion where the specific pattern is included in the reproduction signal. In FIG. 13, the specific pattern is a combination of patterns that gives the minimum Euclidean distance 10 in the Viterbi decoding of PR (1, 2, 2, 1).
[0064]
The metric difference calculation unit calculates a metric difference from the path A metric and the path B metric when the decoded data of Viterbi decoding matches any of the patterns shown in FIG. FIG. 14 illustrates a trellis of path A and path B. Since the Euclidean distance between the path A and the path B in FIG. 14 is 10, the distribution of the metric difference is like a normal distribution with an average value of 10 as shown in FIG. The subsequent processing is the same as in the previous embodiment.
[0065]
The specific pattern is a data string having a high error occurrence frequency. In this way, by extracting the metric difference by paying attention to the combination of the specific patterns and detecting the bias of the histogram, the detection sensitivity of the evaluation index can be increased. In other words, since the pattern is limited to a pattern in which an error is likely to occur, the reproduction signal quality can be accurately grasped by a calculation process with a shorter data length.
[0066]
Note that the SAM value in this embodiment is extracted as the SAM value by obtaining the standard deviation from the bias of the histogram in which the metric differences at each time are accumulated. However, as is clear from the gist of the present invention, the SAM value calculation is not limited to the standard deviation. As an example, as shown in FIG. 10, an area having a small metric difference is discriminated by “threshold = T”, and the frequency equal to or lower than the threshold T can be taken into consideration as an index corresponding to the SAM value.
[0067]
Further, although the present embodiment has been described with respect to a configuration based on hardware, the present embodiment is not limited to this and can be realized by program processing by software.
[0068]
【The invention's effect】
As described above, according to the information reproducing apparatus of the present invention, the quality of the reproduction signal obtained from the disk medium can be determined earlier than the ECC processing, so that it is possible to instantaneously determine whether to retry the reproduction retry. Therefore, even when information is reproduced with a high transfer rate such as high-definition video information, underflow of the shock proof memory can be prevented and a highly reliable shock proof function can be realized.
[0069]
Further, since the ECC correction capability can be locally and selectively changed based on the likelihood difference of maximum likelihood decoding, unnecessary retry operation can be avoided.
[0070]
Furthermore, since a local error occurrence location on the disk medium can be identified instantaneously, the reproduction retry execution time can be suppressed to the minimum necessary data reproduction.
[Brief description of the drawings]
FIG. 1 is an operation flowchart of an information reproducing apparatus of the present invention.
FIG. 2 is a functional block diagram of the information reproducing apparatus of the present invention.
3 is a functional block diagram of a reproduction processor of the information reproducing apparatus shown in FIG. 2. FIG.
FIG. 4 is a diagram illustrating SAM.
FIG. 5 is a diagram illustrating a correlation between a SAM and an error rate.
FIG. 6 is a state transition diagram based on PR (1, 2, 2, 1).
FIG. 7 is a trellis diagram based on PR (1, 2, 2, 1).
FIG. 8 is a detailed functional block diagram of Viterbi decoding and SAM value calculation shown in FIG. 3;
FIG. 9 is an explanatory diagram of the operation of the shock proof memory.
FIG. 10 is a diagram for explaining another embodiment of SAM value calculation.
FIG. 11 is a diagram for explaining an ECC processing block;
FIG. 12 is another operation flowchart of the information reproducing apparatus of the present invention.
FIG. 13 is a diagram for explaining another evaluation index in the present invention.
FIG. 14 is a trellis diagram illustrating another evaluation index in the present invention.
[Explanation of symbols]
201 disks
202 Optical pickup
203 Spindle motor
204 RF amplifier
205 AGC ・ Filter
206 A / D
207 Playback processor
208 Servo DSP
209 Drive controller
210 Memory control
211 memory
212 Data bus
301 Waveform equalization
302 Viterbi decoding
303 Demodulation
304 ECC
801 Metric calculation
802 Addition, comparison, selection
804 path memory
805 Metric difference calculation
806 Standard deviation calculation
807 and 808 registers

Claims (3)

Maximum likelihood of binarizing a reproduction signal by maximum likelihood decoding that selects the most probable state transition sequence from m state transition sequences that transition from the first state at time k-n to the second state at time k. A decoder;
An arithmetic unit for calculating a likelihood difference between the most probable state transition sequence and the second most probable state transition sequence;
An information reproducing apparatus comprising control means for performing retry control of a data storage operation in a shock proof memory based on a likelihood difference obtained by the arithmetic unit. An ECC circuit that corrects an error included in the binarized maximum likelihood decoding output, wherein the control means determines the reliability of the reproduction signal from the likelihood difference obtained by the computing unit, 2. The information reproducing apparatus according to claim 1, wherein the error correction capability of the ECC circuit is changed based on the information. When the likelihood difference obtained by the arithmetic unit exceeds a predetermined threshold value, the control means inquires the address on the medium of the corresponding reproduction data, replays the data at the address, and shock-proof memory 2. The information reproducing apparatus according to claim 1, wherein the corresponding reproduced data is overwritten.

Images