JP3600352B2 - Optical disc apparatus and optical disc reproduction waveform detection method - Google Patents

Description

【００２７】
まず、セクタあたりの容量を１０２４バイトの案を考える。案１、２、３はトラックピッチを０．８５ミクロンとし、図２０に示すように、ユーザ使用領域の最内周の記録位置を従来の３０ｍｍから２７ｍｍに拡大し、ユーザ使用領域の外周の記録半径位置をそれぞれ従来の６０ｍｍから６１．５、６１．０、６２．０ｍｍとしたものである。図２０では外周の半径位置は６１．５の場合を示した。線密度は干渉量から見た前述の限界範囲に入っている。案４は内周を従来通りとした場合で、線密度の限界値に押さえるための外周の値を求めると、６２．０ｍｍとなる。次に、セクタあたりの容量を２０４８バイトにし、エラー訂正のＥＣＣ（エラー コレクション コード）の構造を１０２４バイト変えないものにすると、データの実効効率が約５％程度向上する。以下この場合について、述べる。案５、６は線密度をできる限り緩くするように最内周と最外周の組合せを求めたものである。案７は案６に対してトラックピッチを緩めて、線密度方向に詰めたものである。案８、９、１０はそれぞれ最内周を従来通りとし、外周を変えた場合である。これらの案を決めるときの条件として、最外周は円板のインジェクション工程において内周から押された樹脂が外周で押しとどめれられるために、歪みが発生し、複屈折が増加し光磁気ディスクでは使用できなくなる。その限界は６２ｍｍ程度であり、６１．５ｍｍ以下が好適である。線密度は前述の如く、下限として０．３８５ミクロンであり、望ましくは０．４ミクロン以上が好適である。以上の観点から、セクタあたりの容量が１０２４バイトでは案１が好適であり、セクタあたりの容量が２０４８バイトでは内周を従来通りとすると案８か９であり、内周と外周を同時に変更する場合にはトラックピッチ０．８５ミクロンの案６か、トラックピッチ０．８８ミクロンの案７が好適である。
【００２８】
（トラック構造）
図２（ａ）に示す様に、従来では溝１０（グルーブ）間の領域１１（ランド部）に光スポットを位置決めする連続溝方式を用いているが、この方式ではトラックピッチを狭めると、溝幅が一定なのでマークを記録する領域１１（ランド部）の幅が狭くなり、記録マーク１２の幅が狭くなると同時に溝１０（グルーブ部）が再生スポットにかかりはじめ、グルーブで発生するノイズの混入量が増加する。
【００２９】
そこで、ディスク溝構造を従来のランド部１１とクルーブ部１０を逆転させ、グルーブ部１０の幅を広げ、ランド部１１の幅を狭める。記録部分は広くなったグルーブ部１０に行う。これによりスポットがグルーブ部の端にかかる割合を低減でき、グルーブの壁に発生する変動（これらの原因はレジストの荒れ、パワー変動によって生じる微小な変動である）によるノイズを低減できる。また。記録領域が狭められることなく、記録マーク１２の幅の制限を取り除くことができる。この効果を図３に示す。再生波長は６８５ｎｍ、レンズの開口数は０．５５で、記録マークのビットピッチは０．３５ミクロンである。従来溝の構造のままトラックピッチを詰めていくと、ビットピッチから決まる最短マークとギャップの繰り返しパターンからの信号とノイズの相対比と示すＣ／Ｎ（キャリヤー トウ ノイズレシオ）が急激に低下するが、本発明の構造ではトラックピッチ０．８５ミクロンまでほとんど低下しない。
【００３０】
（記録波形制御）
従来の線密度は０．５１ミクロン／ビットであり、最短マーク長Ｔｍｉｎは０．６８ミクロンとなっている。１ー７変調コードではＴｍｉｎは検出窓幅Ｔｗの２倍である。従来の記録波形は図４に示す様に、本発明者が考案した方式で、最短マーク長２Ｔｗを３Ｔｗ／２長パルスで記録し、次にＴｗづつ増加するマークはＴｗ／２の休止後、Ｔｗ／２のパルスを追加していくものであったが、本発明では最短マークをＴｗ長のパルスで記録し、その後Ｔｗ／２休止、Ｔｗ／２長のパルスを追加する。この理由は以下の通りである。
【００３１】
このマークを０．５１ミクロン相当のパルス幅３／２Ｔｗを持つパルスで記録し、幅として０．６ミクロン程度の楕円型となっている。本発明の線密度は約０．４ミクロン／ビット前後の値であり、同じスポットを用いて最短マーク長として約０．５３ミクロンを記録しなくてはならない。同様に３／２Ｔｗのパルス幅を用いて記録すると、楕円型になって長さを０．５３ミクロンに制御すると幅はこれより必ず狭くなる。マーク幅は信号の大きさに比例することからできる限り大きくしておかなくてはならない。そこで最短マークの記録パルス幅を短くすることになるが、クロックに同期し、クロックから作成できる事を考慮するとこのパルス幅はＴｗ／２の整数倍が望ましい。このことから最短マークの記録パルス幅はＴｗかＴｗ／２となる。この中でＴｗ／２にすると記録パワーが大きくなることからＴｗが好適である。これで記録すると最短マークはほぼ円形状となり、長さを制御しながら、最大のマーク幅を実現できる。長いマークを記録するときは図５に示す様に、従来と同様に最短マークを記録するパルスのあとに各コードデータ長に対応してＴｗ／２のパルスを付け足した記録波形を用いる。
【００３２】
（対物レンズへの入射光学系）
対物レンズへの入射光束の分布を変化させることにより、実効的な開口数を変える事ができる。例えば、対物レンズの入射側の有効開口径ａに対して入射光束の広がりを狭くすると光スポット形成に寄与する有効径が実効的に狭くなり、開口数が減少したと同様の効果が光スポットに現れる。データを再生する円周方向には開口数を維持したままクロストークに効く半径方向には開口数を減少させるために、入射分布を円周方向に長軸を持ち、半径方向に短軸の楕円分布とする。
【００３３】
前記分布と通常の分布による信号読み出し特性を図６、７に示す。０．４ミクロン／ビットの線密度の最短マークとギャップの繰り返しパターンを記録したときの再生信号振幅とディスク傾きとの関係を図６に、隣接トラックに最長マークとギャップの繰り返しパターンを記録し、隣接トラックからの漏れ込みと傾きとの関係を測定した結果を図７に示す。いずれも、信号量のスケールは任意スケールである。図８には上記図６、７のデータからクロストークの影響を最短マークとギャップパターンの振幅に対する最長マークとギャップパターンの漏れ込みの比で定義し、クロストークと傾きの関係を実効開口数の組合せで表した楕円分布をパラメータとして示した。
【００３４】
クロストークの特性としては、以下の特徴がある。コマと非点、焦点ずれ収差による隣接データからの応答は高周波域が極端に低下する。従って、当該トラックを再生するのに障害になるのは隣接のデータのなかでも低周波成分を含むデータパターンのときである。具体的には６Ｔ−７Ｔの様な長いマークと長いギャップの組み合わせが最悪となる。漏れ込みの波形は応答が高域に延びていないため正弦波状になっている。
【００３５】
最短繰り返しパターンからの信号振幅は開口数を大きくすればするほど大きくなり、傾きに対して大きく影響をうける。楕円分布でもトラック方向（／の右側に示す）の開口数と信号振幅の依存性は上記と同様な傾向をもつ。漏れ込みは開口数を大きくすればするほど傾きに対して大きく影響をうける。楕円分布によりトラック方向の開口数が等方的な場合に比較して傾きに対する漏れ込みの量は大きくなるが、開口数が０．５５の場合はその増加はほとんど無視でき、楕円分布と等方分布の漏れ込み特性はほぼ等しい。以上の結果、図８に示すように実効開口数比０．６／０．５５の楕円分布は傾きが無いときの値は開口数０．６の等方分布よりクロストーク特性は良くないが、他の等方分布、楕円分布に比較すればよい。また、ディスク傾きが増加してくると、開口数０．６の等方分布は急激にクロストークが増加するが、実効開口数比０．６／０．５５の楕円分布はその増加が少なく、０．４度あたりでは逆に実効開口数比０．６／０．５５の楕円分布の方が約２ｄＢ程度少なくなる。従って、総合的なクロストーク特性から、本発明では実効開口数比０．６／０．５５の楕円分布を用いる。
【００３６】
（試し書きパターン）
装置毎に記録されるマークが変動することを防止し、記録マークの特性を合わせるために記録前にあらかじめ記録条件を最適に合わせる試し書きを行う。試し書きのパターンとしては従来最密パターン（最短マークとギャップの繰り返しパターン）と最疎パターン（最長マークとギャップの繰り返しパターン）の組み合わせであったが、今回は波形間の干渉が強いため、図９の様に、最密パターンの次に長いパターン（例えば、３Ｔｗマークとギャップ）と最長パターンより短く、かつ飽和レベルを持つパターン（５Ｔｗマークとギャップ）の組み合わせにより試し書きをおこなう。試し書きでは図のように記録パワーを順次変えて記録再生し、最疎、最密パターンの平均値のズレ△Ｖを検出し、これを零とするように最適記録パワーを決定する。試し書きにおいては再生時の等化動作をかけないか、かけても波形間干渉を取る程度とし、再生の条件が試し書きのパワー設定に影響を与えないようにする。本発明の線密度では３Ｔｗのギャップ長では、０．８ミクロンとなりスポットサイズで規格化したときの値は０．７とであり、図１から見るとほとんど干渉が無視できる。
【００３７】
（回路によりクロストークの影響を低減）
後述する等化係数の学習動作も同じ回路構成により実現できるので合わせて図１０に示す。光ディスク上に予め設けられたプリピット部にはデータを記録する区切れ目であるセクタの先頭を示すマーク、番地を読むための同期信号、番地情報が記録されている。光ヘッド検出系と信号処理の組合せにより、プリピット信号と光磁気信号が出力される。プリピット信号検出部からのプリピット信号はプリピットマーク検出部２００によりデータの記録はじめを示すタイミング信号を発生し、クロック発生回路２０７においてデータに同期したクロックを発生するＶＦＯ（バリアブル フレクエンシ オシレータ）を起動する。セクタの先頭を示す信号を発生し、データ部の信号処理を制御するタイミングを発生する回路２１６に入力する。
【００３８】
光磁気信号は後述する等化定数設定回路２０１により、適切な等化が行われ、レベル合わせ回路２０２に入力される。ここでは等化後の波形に対して後の処理のために、レベルを付加する。レベル付加後の信号はレベルスライス回路２０３、２０４、２０５にそれぞれ入力される。レベルスライス回路２０３では図１１に示す様に、再生信号の真ん中のスライスレベルＬ１と検出信号とのクロス点にパルス１００を発生させ、該パルスをクロック発生回路２０７に入力し再生クロックＣＫ１を発生する。レベルスライス回路２０４ではレベルスライス回路２０３のスライスレベルより上にずれたスライスレベルＬ２と検出信号とのクロス点にパルス１０２を発生させる。レベルスライス回路２０５ではレベルスライス回路２０４のスライスレベルとは、レベルスライス回路２０３のスライスレベルを挟んで対称の位置にあるスライスレベルＬ３と検出信号とのクロス点にパルス１０１を発生させる。なお、スライスレベル回路におけるスライスレベルは、既知の慣用技術を用いて、スライスレベルＬ１，Ｌ２，Ｌ３を適宜に設定し得るものである。
レベルスライス回路２０４と２０５によって発生したパルス１０２、１０１をそれぞれ、位相比較器２０８、２０９に入力し、再生クロックとの位相を比較する。位相比較して検出窓幅にパルスが存在するか否かを判別する。このために位相比較器ではデータ検出のクロックＣＫ１（データ検出範囲では“１”の値をとる）とは、反転したクロックＣＫ２（データ検出範囲では“０”、データ検出範囲を超えると“１”の値をとる）パルス１０２、１０１の立ち上がりでクロックＣＫ２のレベルを保持し、クロックＣＫ１の立ち上がりでリセットする。このようにすると、図１１のケース２のようにスライレベルＬ１が再生波形の中心と一致していると、レベルＬ２、レベルＬ３の立ち上がりが検出窓幅の中にあるため、位相比較器２０８、２０９の出力は現れない。しかし、ケース１のように再生波形の真ん中がスライスレベルＬ１より低くなると、位相比較器２０９の出力１０４にパルスが発生するが、位相比較器２０８の出力１０３にはパルスが生じない。また、ケース３の様に再生波形の真ん中がスライスレベルＬ１より高くなると、位相比較器２０９の出力１０４にパルスは発生しないが、位相比較器２０８の出力１０３にはパルスが生じる。出力１０３、１０４を切り替え回路２１０に入力し、後述する等化係数学習に使用するかクロストーク対策に使用するかにより後続する処理系への接続を変える。切り替え動作の指令は再生処理系を制御している上位コントローラによって発生される。
なお、図１１において、図示するスライスレベルＬ３とＬ２が信号振幅の０．３と０．７に設定されているが、これは模式的な図示例であって、これらのスライスレベルにおいて、図１２に示す最短マークの再生波形でもクロスできるレベルでなければならないことは当然である。このことは、スライスレベルＬ１が再生波形の中心に一致していると、スライスレベルＬ２、スライスレベルＬ３のパルスの立ち上がり位置が検出窓幅の中にあるとの上述の記載からも明らかである。また、図１１においては、再生波形の立ち上がり時におけるクロスの態様を図示しているが、再生波形の立ち下がり時についても同様の処理がなされ得るのは自明であり、これについては図１１では省略している。
【００３９】
クロストーク対策用が切り替え回路２１０で選択されると、パルス１０３、１０４は補正スライス量発生回路２１１に入力され、パルス１０４があればマイナスの一定補正量を出力し、パルス１０３があればプラスの一定補正量を出力する。この補正量をレベル合わせ回路２０２に入力し、再生波形の全体レベルを補正量の分だけシフトさせる。補正スライス制御回路２１３は再生データを回路２１７で弁別した後、回路２１５でデータを復調し、エラー訂正回路２１９でエラー状況を監視して、再生エラーが生じたら補正スライス量発生回路２１１の一定補正量を増減し、補正効果をさらに向上させる。また、切り替え動作指令とともにスライスレベル設定回路２０６に入力され、スライスレベルＬ１とＬ２，Ｌ３のレベルを変えて実効的に比較位相差を変えることにより、補正動作を制御する。
【００４０】
実際にクロストークが生じたときの動作を図１２（a)と（ｂ）を用いて述べる。最短マークとギャップの組合せパターンに続いて６Ｔｗマークとギャップの組合せパターンが存在しているとき、隣接トラックにデータが無いときには信号３０２が再生される。隣接トラックに６Ｔｗマークとギャップの組合せパターンを連続して記録すると、隣接トラックから信号３０１が漏れ込み、再生信号３０２に重畳され信号３０３となる。クロックＣＫ１は図の様に検出される。ここで、前述のスライス補正を行うと、レベル合わせ回路２０２の入力信号３０３に対してレベル合わせ回路２０２に入力され、レベル変動量の補正量は３００となり、信号３０１に比例した変動量を打ち消すようにレベル合わせを行う。
【００４１】
レベル合わせを行う方法の他に、データ分別のクロックの位相をずらして補正を行う方法を図１３に示す。ブロック図のほとんどは図１２で説明したものである。補正スライス量発生回路２１１の代わりに補正位相量発生回路２２１を用い、位相比較器２０８、２０９の出力から補正スライス量を検出したのと同様に、補正位相量を検出し、これを位相合わせ回路２２０に入力し、クロックＣＫ１の位相をずらす。ずらしたクロックを用いてデータ弁別を行うことにより、クロストークの影響を低減して安定にデータを検出できる。位相ずらし量は補正位相制御回路２２２により再生エラーの状態を見ながら調整する。
【００４２】
上記実施例では位相をフィードフォワード制御により補正する方法であったが、別の実施例ではクロストークによる位相ズレをフィードフォワードにより低減することができる。図１０のクロック発生回路２０７の帯域は通常データの周波数帯域よりも狭くとり、データによってクロック周波数が変動しないように構成されている。本実施例ではエラー発生が無いときには追従帯域は通常と同じようにクロストークの主成分の周波数よりも低く選んでおき、エラーが増加したときに、エラー情報に従って、クロック発生回路の帯域をクロストークの主成分の周波数をカバーするように設定し、追従帯域を増加させる。例えば、回転数を３６００ｒｐｍにすると８Ｔｗのマークとギャップの組合せパターンからの周波数は５ＭＨｚ程度になる。クロック発生回路の追従帯域を前記周波数より上げると欠陥等により検出パルスの間隔が大きく変動するとクロック発生回路が暴走する事がある。これを回避するために、本発明では図１８のようなクロック発生回路の構成となっている。レベルスライス回路２０３により再生波形とスライスとの交点でパルスを発生し、該パルスを位相比較器４００に入力し、ＶＣＯ４０３からのクロックパルスと位相比較し、その結果を位相補償回路４０１に入力する。該位相補償回路４０１はクロック発生回路の帯域と応答特性を決めるように補償係数を制御する。補償回路４０１の出力はホールド回路４０２に入力され論理和回路４０５の制御信号４１０に応じてＶＣＯ４０３を駆動する電流をホールドするか、位相補償回路４０１の出力をそのままＶＣＯ４０３に入れるかを制御する。ここで、欠陥等の影響がクロック発生回路２０７に及ばないように、２つの監視回路を設ける。一つは再生データパルスの間隔を監視する周期監視回路４０６であり、ここではデータパルスの間隔がデータ変調コードにあるものから外れることが無いか監視し、外れる事があると、外れた期間だけホールド回路４０２を起動し、同時に位相比較器にデータパルスが入力されないように切り替え回路４１３により切り替えすることにより、基準クロック４１２を代わりに位相比較器４００に入力する。これにより位相補償回路４０１が飽和することなく、データ間隔が正常に復帰したときにクロック発生回路２０７もすぐに動作開始ができる。もう一つの監視回路は再生信号の振幅を監視する振幅異常検出回路４０７であり、レベルスライス回路２０４、２０５の出力１０１、１０２を見ておき、該出力が特定期間来ないときには再生出力が低下したと見なし、ＶＣＯ４０３を駆動する電流をホールドする。以上の２つの監視回路により欠陥等によるクロック発生回路が暴走することを無くすることができ、クロック発生回路の追従帯域を向上させることができる。クロック発生制御回路４０８にはエラー訂正回路２１９からの信号によりエラーが増加したときに、エラー情報に従って、切り替え回路４０４が指令を出し、位相補償回路４０１の補償係数を変化させ、追従帯域を増加させる。エラー発生が無いときには追従帯域は通常と同じようにクロストークの主成分の周波数よりも低く選んでおく。
【００４３】
図１９を用いて実際の回路動作を説明する。等化定数の設定動作に関しては図１３と同じなので省略し、クロストークの対策について述べる。クロストーク対策用が切り替え回路４０９で選択されると、パルス１０１、１０２はクロック発生回路４０８に入力され、前述の回路処理を行う。
【００４４】
（等化係数の学習）
従来よりも波形間干渉が増加することから等化定数を学習し、設定する事が必要である。波形間の干渉により信号振幅は低下することから、該低下量を求め、干渉量を低減するように等化係数を変化させる。振幅低下は前述のスライスレベルを３つ設け、真ん中のスライスレベルと検出信号とのクロス点にパルスを発生させ、該パルスによってデータ再生のクロックを発生する。また真ん中のレベルを中心として上下に対称な２つのスライスレベルと再生信号とのクロス点にパルスを発生させ、これらのパルスと前記クロックのとの位相差を測定することにより求められる。図１３において、比較器２０８、２０９の出力パルス１０４、１０３をそれぞれ計測器２１２により等化データが記録された期間計測する。等化定数制御回路２１４では後述する制御フローに従って、補正等化量発生回路２１８を制御し、デフォルト値をセットしたり、等化係数を増減したりする。等化定数設定回路２０１は図１６に示すような構成になっている。再生信号はバッファアンプ４００を介して遅延線４０１とゲイン補正器４０４に供給される。遅延線４０１の出力は遅延線４０２と加算回路４０３に入力される。遅延線４０２の出力はゲイン補正器４０５を介して、加算回路４０３に入力される。ゲイン補正器４０４の出力も加算回路４０３に入力される。等化係数Ｋ１、Ｋ２はゲイン補正器４０４、４０５で設定される値である。補正等化量発生回路２１８の指令値に従ってゲイン補正器４０４、４０５に値が設定される。
【００４５】
学習方法を図１４と図１６をもとに説明する。等化係数Ｋ１，Ｋ２はまず、装置出荷時に設定された初期値をディフォルト値として用いる。データの記録単位であるセクタは図１５にしめす構成となっている。予め設けられたプリフォーマット部のあとに、データ記録領域がある。データ記録領域は同期パターンからなる同期部に続いてデータと再同期パターンが繰り返し連続する。図１０のプリピットマーク検出回路２００とタイミング発生回路２１６により、記録データの領域を示す信号４１０を発生し、これをもとに等化係数を変化させる。この制御は等化定数制御回路２１４によって行われる。
【００４６】
この制御は、上記位相ずれが特定の値を超えたらパルスが出るようにしておき、このパルスの数が特定の値を超えたら等化係数を増加または減少させるようにする。図１４のフローに従って、等化定数制御回路２１４が指令を補正等化量発生回路２１８に出しながら、ディスクが回転したときに計測セクタ（斜線で示した）において、位相ずれエラーの計測値を計測器２１２から取り込み、最小等化係数を探す。図１５では増減回数が３回の場合を示した。
【００４７】
この学習はまず、装置が起動されたときに行われる。前述の試し書きにより求めたパワーにより長マーク後に最密マークがあるパターンを記録する。このパターンとして試し書きのパターンを用いてもよい。等化係数を増加または減少させながら、位相ずれが特定値を越えたことを示すパルスを計測して、パルス数が最も少なくなる等化係数の組を捜す。この組をデフォルト値として装置に設定する。
【００４８】
次の機会は再生エラーが発生し、再試行を行う場合である。図１７に示すフローにより学習を行う。まずディフォルト値でエラーの発生したセクタを読み出し、エラー訂正回路２１９によりエラー訂正可能かどうかを判定する。エラー訂正可能かどうかの判定はエラー処理ブロック（コードワードという）ごとのエラーの数を基に判定する。エラー訂正が不能の場合に、まず、等化係数を増加または減少させ、上記エラー処理ブロック（コードワードという）ごとのエラーの数を計測し、エラー訂正が可能になるように等化係数を合わせる。この処理によりエラー訂正できないときは前述の位相ずれを示すパルスの数を測定し、エラーを生じたセクタに記録されたデータを用いて位相ずれを示すパルスの数が最小となる等化係数を捜す。
以上説明したように、本発明の実施の形態では、レベル合わせ回路での再生波形のレベル変動処理と等化定数設定回路での再生波形の等化処理を個別に説明したが、図１０の回路構成を参照すると、当該レベル変動処理と等化処理を切り替え回路の切り替えによって適宜のタイミングで使い分け実施し得ることは明らかである。
【００４９】
【発明の効果】
以上により、従来の記録再生方式とフォーマットを大幅に変更することなく、１３０ｍｍ径の記録媒体の片面で２．６ＧＢ以上の容量を有する光磁気ディスク装置及び媒体を実現できる。
【図面の簡単な説明】
【図１】マークの存在による干渉量とマーク間隔の関係を示す図である。
【図２】本発明による媒体の溝構造を示す図である。
【図３】本発明による媒体の溝構造の効果を示す図である。
【図４】本発明による記録マーク形状制御方法を示す図である。
【図５】１−７変調符号のデータパルス波形に対応した記録波形を示す図である。
【図６】実効開口数の変化による最密パターン信号の傾きの依存性を示す図である。
【図７】実効開口数の変化による最疎パターンの漏れ込みの傾き依存性を示す図である。
【図８】実効開口数の変化によるクロストークの傾き依存性を示す図である。
【図９】試し書き動作を説明する図である。
【図１０】本発明の実施例の回路ブロック図である。
【図１１】波形ずれを測定する動作説明図である。
【図１２】漏れ込みがある時のレベル合わせ回路の動作説明図である。
【図１３】本発明の実施例の回路ブロック図である。
【図１４】等化係数を合わせるフロー図である。
【図１５】等化係数を合わせる動作の説明図である。
【図１６】等化定数設定回路の構成を示す図である。
【図１７】等化係数を合わせるもう一つの実施例のフロー図である。
【図１８】本発明のクロック発生回路のブロック図である。
【図１９】本発明のもう一つの実施例の回路ブロック図である。
【図２０】本発明の記録領域と従来の記録領域とを説明する図である。
【符号の説明】
１０ グルーブ部 １１ ランド部 １２ マーク １３ 基板
１００ 検出信号とのクロス点を示すパルス
１０３、１０４ 検出窓幅を越えた事を検出するパルス
２００ プリピットマーク検出回路 ２０１ 等化定数設定回路
２０２ レベル合わせ回路
２０３、２０４、２０５ レベルスライス回路
２０６ スライスレベル設定回路 ２０７ クロック発生回路
２０８、２０９ 位相比較器 ２１０ 切り替え回路
２１１ 補正スライス量発生回路 ２１２ 計測器
２１３ 補正スライス制御回路 ２１４ 等化定数制御回路
２１５ データ復調回路 ２１６ タイミング発生回路
２１７ データ弁別回路 ２１８ 補正等化量発生回路
２１９ エラー訂正回路[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an optical disk device that optically records and reproduces information on a rotating disk, and particularly relates to a magneto-optical disk having a diameter of 130 mm without significantly changing the recording / reproducing method and format. The present invention relates to an optical disk device and a medium for increasing the capacity.
[0002]
[Prior art]
In the magneto-optical disk apparatus of the light intensity modulation system, the current storage capacity is 130 GB in diameter and 1.3 GB on one side, the recording area is from 30 mm in inner circumference to 60 mm in outer circumference, and the surface density is 1.1 Gbit / in2. Attempts to increase the density have the following problems.
[0003]
(1) Limit of s / n in light intensity modulation method
Since higher density requires writing of finer marks, the signal magnitude s proportional to the mark width decreases. However, in the magneto-optical disk modulation, which is another type of magneto-optical disk, the width can be increased, so that s is not easily deteriorated. In this respect, the light intensity modulation method has a limit.
[0004]
(2) The problem of noise increase in the continuous groove method
Conventionally, a continuous groove method for positioning a light spot in a region (land portion) between grooves (grooves) is used. However, in this method, when the track pitch is narrowed, the groove width is constant, so that a mark recording region (land portion) is used. ) Becomes narrower, the recording mark width becomes narrower, and at the same time, a groove (groove portion) starts to be applied to the spot, so that the amount of noise generated in the groove increases.
[0005]
(3) The problem of increasing crosstalk
When the track pitch is reduced, signal leakage from an adjacent track increases, and jitter at the time of signal detection increases.
[0006]
(4) When the linear density is reduced, the interference between waveforms becomes stronger, and when trying to read a mark recorded by another device, the amount of interference differs due to the different recording state, and the reproduction condition is kept constant. Otherwise, a detection error will occur.
[0007]
[Problems to be solved by the invention]
Without changing the recording / reproducing method of the optical disk and the medium structure, it is necessary to maintain the compatibility with the optical disk which has been realized so far, to increase the capacity more than twice and to set the maximum transfer rate to 4 MB / s or more. A possible optical disk device and medium are realized. What is the conventional method?,
(1) Light intensity modulation recording method,
(2) Continuous groove tracking method,
(3) The substrate thickness is 1.2 mm, so the cartridge thickness is not different from the conventional onething,
(4) Revolution speed is over 3000 RPMBeing,
It is.
[0008]
Other conditions are as follows: (1) The format is almost common. This is because the configuration of the logic lsi can be changed only minutely. (2) A common disk making technique. This is because the disk making process requires only minor changes. (3) The configuration of the recording / reproducing circuit is substantially equal, and can be dealt with by changing the parameter or by making small changes.
[0009]
[Means for Solving the Problems]
The following is a solution for increasing the recording capacity by a factor of two or more under the above-mentioned limiting conditions.
[0010]
1. For how to improve s / n,
(1) The shortest mark length is made as long as possible to increase the mark width.
For this purpose, the modulation method is selected and the recording waveform is controlled. Of these, the modulation method uses the same 1-7 RLL as the conventional one. In the conventional recording waveform, the shortest mark length 2Tw devised by the present inventor is recorded by a 3Tw / 2 long pulse, and the mark increasing by Tw is added by a Tw / 2 pulse after the Tw / 2 pause. However, in the present invention, the shortest mark is recorded by a pulse of Tw length, and then a pulse of Tw / 2 pause and Tw / 2 length is added.
[0011]
(2) Reduce interference between waveforms and increase effective S / N.
The shortest mark length Tmin is related to the spot diameter Ws so that Tmin / Ws exceeds 0.45 at worst. Below this, the waveform interference between the marks increases, and it is essential to learn the equalization coefficient for waveform equalization. The reason is that the response from the mark on the optical disc is determined by the mark shape and the shape of the light spot. In a compatible medium such as an optical disk, the above two factors fluctuate. We have to be as robust as possible with this change.
[0012]
(3) Enlarge the area of the recording unit.
The disc groove structure is reversed between the conventional land portion and the crew portion, so that the width of the groove portion is increased and the width of the land portion is reduced. The recording part is performed in the conventional groove part. As a result, it is possible to reduce the ratio of the spot to the end of the groove portion, and it is possible to reduce the noise due to the fluctuation occurring on the groove wall—roughness of resist and fluctuation due to power fluctuation. Also,The limitation on the width of the mark can be removed without reducing the recording area.
[0013]
(4) Increase the resolution.
The numerical aperture NA of the conventional objective lens is 0.55, but by increasing it to 0.6, the resolution can be improved. However, when the numerical aperture is increased, side lobes generated by the tilt of the disk in the track direction are applied to adjacent tracks, and crosstalk is likely to occur. Therefore, the following crosstalk reduction method is used.
[0014]
(A) A method in which a measure is taken by regarding the change as a slice level.
The signal response from the adjacent data due to the side lobe is extremely low in the high band. Therefore, the crosstalk component has a significant effect when the data pattern includes a low-frequency component among adjacent data. Specifically, the combination of a long mark such as 6T-7T and a long gap is the worst. The crosstalk waveform is sinusoidal. It is superimposed on the signal from the data of the read track. Therefore, the level fluctuation of the low-frequency component of the original waveform is detected by using the symmetry of the amplitude level of the reproduced waveform centered on the edge portion, and its influence is removed.
[0015]
As a method, three slice levels are provided, a pulse is generated at a cross point between the middle slice level and the detection signal, and a clock is generated by the pulse. Further, a pulse is generated at a cross point between another slice level and the detection signal, and the phase difference between these pulses and the clock is measured.
[0016]
When the phase shift exceeds a specific value, a pulse is generated, and when a pulse occurs, the slice level is shifted by a specific amount. Alternatively, the phase of a clock used for data discrimination is shifted.
[0017]
Further, when an error still occurs, a retry for reading the same sector is performed again, and at this time, the shift amount is increased.
[0018]
(B) Another solution to the problem associated with high resolution.
Since the tilt in the track radial direction is a problem, the size of the side lobe is reduced by lowering the effective NA in this direction. The method of realizing this is to make the incident distribution an elliptical distribution having a major axis in the circumferential direction and a minor axis in the radial direction. This reduces the amount of crosstalk that occurs when the disk tilt is large, but increases crosstalk when the tilt is small. This reduces the effect of crosstalk by the method described above.
[0019]
(5) Perform recording mark control for maintaining compatibility.
In order to prevent the marks recorded for each apparatus from fluctuating and to make the characteristics of the recorded marks uniform, test writing is performed before recording to optimize the recording conditions. Conventionally, the test writing pattern was a combination of the densest pattern and the sparsest pattern, but this time the interference between the waveforms was strong, so the next longest pattern after the densest pattern and shorter than the longest pattern, and had a saturation level Perform test writing by combining patterns. In trial writing, recording and reproduction are performed while sequentially changing the recording power, and the optimum recording power is determined. In the test writing, the equalization operation at the time of reproduction is not performed, or even if the operation is performed, interference between waveforms is taken so that the reproduction condition does not affect the power setting of the test writing.
[0020]
(6) An equalization coefficient as a reproduction condition is determined by trial reading and learning of automatic equalization.
The data on the optical disk is managed and information is recorded and reproduced in sector units. Therefore, information may be recorded only in one sector, and the recording condition may be significantly different from the condition of the adjacent sector or track. Since the response characteristics of the signal from the optical disc are determined by the read spot and the shape of the recording mark, even if the equalization coefficient is obtained in accordance with the combination of the mark and the read spot under different recording conditions, the response is different if the recording device and the recording medium are different. Are different. Therefore, it is necessary to find an equalization coefficient for each sector.
[0021]
The following properties are used as the detection principle. Signal amplitude decreases due to interference between waveforms. Since the amount of reduction increases in accordance with the amount of interference, the amount of reduction is determined, and the equalization coefficient is changed so as to reduce the amount of interference. As a method for realizing the detection principle, three slice levels are provided, a pulse is generated at a cross point between the middle slice level and the detection signal, and a clock for data reproduction is generated by the pulse. In addition, a pulse is generated at a cross point between the reproduced signal and two slice levels vertically symmetric with respect to the center level, and the phase difference between the pulse and the clock is measured.
[0022]
The learning is performed according to the following procedure. First, a value set at the time of shipping the device is used as a default value for the equalization coefficient. When the phase shift exceeds a specific value, a pulse is generated. When the number of pulses exceeds a specific value, the equalization coefficient is increased or decreased.
[0023]
(A) A pattern having a densest mark after a long mark is recorded by the power obtained by test writing. A test writing pattern may be used as this pattern. While increasing or decreasing the equalization coefficient, a pulse indicating that the phase shift has exceeded a specific value is measured, and a set of equalization coefficients with the smallest number of pulses is searched for. This set as default value
(I) When a replay error occurs and a retry is to be performed, first, the sector in which the error has occurred is read using the default value, and it is determined whether error correction is possible. The determination as to whether error correction is possible is made based on the number of errors for each error processing block (referred to as a code word). If error correction is not possible, first increase or decrease the equalization coefficient, measure the number of errors for each error processing block (called a codeword), and adjust the equalization coefficient so that error correction is possible. . If the error cannot be corrected, the number of pulses indicating the above-described phase shift is measured, and an equalization coefficient that minimizes the number of pulses indicating the phase shift is searched for using data recorded in the sector where the error has occurred.
[0024]
BEST MODE FOR CARRYING OUT THE INVENTION
(format)
The format configuration of the present invention will be described.
The shortest mark length and interval that determine the density in the linear direction are as shown in FIG. 1 when the inter-waveform interference amount specified by the spot size is calculated. FIG. 1A shows a reproduction waveform with respect to the spot center position normalized by the diameter Ws when a long mark is read out from the spot with the diameter Ws. Assuming that the end of the mark is the origin of the movement of the center of the spot, the negative area indicates the amount of interference due to the presence of the long mark as a function from the end of the mark. Further, when the interference amount and the differential coefficient of the interference amount are expanded from 0.35 to 0.6 in the normalized movement distance, the result is as shown in FIG. FIG. 1C further shows the waveform interference amount in an enlarged manner. When the saturation level of the mark is 1 and the interference amount is 10%, the normalized distance is 0.45. Therefore, the shortest mark length Tmin is set so that Tmin / Ws becomes 0.45 or more in relation to the spot diameter Ws. Below this, the waveform interference between the marks increases, and the waveform interference amount is greatly affected by the fluctuation of the mark spot due to the fluctuation of the light spot and the fluctuation of the recording power. In an optical disk, the response from a mark is determined by the mark shape and the shape of the light spot. In a compatible medium such as an optical disk, the above two factors fluctuate. We have to be as robust as possible with this change.
[0025]
If the wavelength of the laser to be used is 685 nm and the effective numerical aperture in the circumferential direction is 0.6 as described later, the spot diameter Ws is 1.14 microns. Then, as described above, the shortest mark length Tmin must be 0.51 μm or more. When a 1-7 RLL (Run Length Limited) code is used for the modulation method, the bit pitch becomes 0.385 microns / bit or more. When the characteristics related to the format are examined under the limited conditions, the cases shown in Table 1 are candidates. FIG. 20 shows the current recording radius position and the name of the recording area.
[0026]
[Table 1]

[0027]
First, consider a case where the capacity per sector is 1024 bytes. In the cases 1, 2, and 3, the track pitch is set to 0.85 microns, and as shown in FIG. 20, the recording position on the innermost circumference of the user use area is increased from 30 mm to 27 mm, and the recording on the outer circumference of the user use area is increased. The radial positions are respectively 61.5, 61.0, and 62.0 mm from the conventional 60 mm. FIG. 20 shows a case where the radial position of the outer circumference is 61.5. The linear density falls within the above-mentioned limit range viewed from the amount of interference. Case 4 is a case where the inner circumference is the same as the conventional case, and the outer circumference value for holding down to the limit value of the linear density is 62.0 mm. Next, if the capacity per sector is set to 2048 bytes and the ECC (error correction code) structure for error correction is not changed to 1024 bytes, the effective data efficiency is improved by about 5%. Hereinafter, this case will be described. In Cases 5 and 6, the combination of the innermost circumference and the outermost circumference is determined so as to make the linear density as loose as possible. In case 7, the track pitch is loosened and the line pitch is reduced in the line density direction. Cases 8, 9 and 10 are the cases where the innermost circumference is the same as the conventional one and the outer circumference is changed. As a condition for deciding these plans, the resin pressed from the inner circumference in the disk injection process is kept at the outer circumference in the outermost circumference, so distortion occurs, birefringence increases, and in the magneto-optical disk, Can no longer be used. The limit is about 62 mm, and preferably 61.5 mm or less. As described above, the lower limit of the linear density is 0.385 μm, preferably 0.4 μm or more. From the above viewpoint, the case 1 is preferable when the capacity per sector is 1024 bytes, and when the capacity per sector is 2048 bytes, the case 8 or 9 is assumed assuming that the inner circumference is the same as before, and the inner circumference and the outer circumference are changed simultaneously. In this case, the case 6 having a track pitch of 0.85 microns or the case 7 having a track pitch of 0.88 microns is preferable.
[0028]
(Truck structure)
As shown in FIG. 2A, in the related art, a continuous groove method for positioning a light spot in a region 11 (land portion) between grooves 10 (grooves) is used. Since the width is constant, the width of the mark recording region 11 (land portion) becomes narrower, and the width of the recording mark 12 becomes narrower. At the same time, the groove 10 (groove portion) starts to be applied to the reproduction spot, and the amount of noise mixed in the groove is mixed. Increase.
[0029]
Therefore, the conventional land portion 11 and the crew portion 10 are reversed in the disk groove structure, the width of the groove portion 10 is increased, and the width of the land portion 11 is reduced. The recording part is performed on the widened groove part 10. As a result, the ratio of the spot to the edge of the groove portion can be reduced, and noise due to fluctuation occurring on the groove wall (these causes are minute fluctuations caused by roughening of resist and power fluctuation) can be reduced. Also. The limitation on the width of the recording mark 12 can be removed without reducing the recording area. This effect is shown in FIG. The reproduction wavelength is 685 nm, the numerical aperture of the lens is 0.55, and the bit pitch of the recording mark is 0.35 microns. If the track pitch is reduced with the conventional groove structure, the C / N (carrier-to-noise ratio), which indicates the relative ratio of the signal to the noise from the shortest mark and gap repetition pattern determined by the bit pitch and the carrier-to-noise ratio, rapidly decreases. With the structure of the present invention, the track pitch hardly decreases to 0.85 microns.
[0030]
(Recording waveform control)
The conventional linear density is 0.51 micron / bit, and the shortest mark length Tmin is 0.68 micron. For the 1-7 modulation code, Tmin is twice the detection window width Tw. As shown in FIG. 4, the conventional recording waveform uses the method devised by the present inventors to record the shortest mark length 2 Tw with a 3 Tw / 2 long pulse. Although the pulse of Tw / 2 is added, in the present invention, the shortest mark is recorded by a pulse of Tw length, and then a pulse of Tw / 2 pause and Tw / 2 length is added. The reason is as follows.
[0031]
This mark is recorded with a pulse having a pulse width of 3/2 Tw corresponding to 0.51 micron, and has an elliptical shape with a width of about 0.6 micron. The linear density of the present invention is about 0.4 microns / bit, and the same spot must be used to record the shortest mark length of about 0.53 microns. Similarly, when recording is performed using a pulse width of 3/2 Tw, the width becomes narrower if the length is controlled to 0.53 μm as an elliptical shape. Since the mark width is proportional to the signal size, it must be as large as possible. Therefore, the recording pulse width of the shortest mark is reduced. However, considering that the pulse width can be generated from the clock in synchronization with the clock, the pulse width is desirably an integral multiple of Tw / 2. From this, the recording pulse width of the shortest mark is Tw or Tw / 2. Tw is preferable because Tw / 2 increases the recording power. When recording is performed in this manner, the shortest mark has a substantially circular shape, and the maximum mark width can be realized while controlling the length. When recording a long mark, as shown in FIG. 5, a recording waveform is used in which a pulse for recording the shortest mark and a pulse of Tw / 2 corresponding to each code data length are added as in the conventional case.
[0032]
(Optical system for incidence on objective lens)
By changing the distribution of the light beam incident on the objective lens, the effective numerical aperture can be changed. For example, when the spread of the incident light beam is narrowed with respect to the effective aperture diameter a on the incident side of the objective lens, the effective diameter contributing to the formation of the light spot is effectively narrowed, and the same effect as the decrease in the numerical aperture is obtained in the light spot. appear. Regenerate data Maintain the numerical aperture in the circumferential direction while maintaining the numerical aperture In order to reduce the numerical aperture in the radial direction that works for crosstalk, the incidence distribution has a major axis in the circumferential direction and an ellipse with a minor axis in the radial direction Distribution.
[0033]
FIGS. 6 and 7 show signal readout characteristics based on the distribution and the normal distribution. FIG. 6 shows the relationship between the reproduction signal amplitude and the disc tilt when a repetitive pattern of the shortest mark and gap having a linear density of 0.4 micron / bit was recorded. FIG. 6 shows the repetition pattern of the longest mark and gap recorded on an adjacent track. FIG. 7 shows the result of measuring the relationship between the leakage from the adjacent track and the inclination. In any case, the scale of the signal amount is an arbitrary scale. In FIG. 8, the influence of the crosstalk is defined by the ratio of the leakage of the longest mark and the gap pattern to the amplitude of the shortest mark and the gap pattern from the data of FIGS. 6 and 7, and the relationship between the crosstalk and the inclination is defined as the effective numerical aperture. Elliptic distributions represented by combinations are shown as parameters.
[0034]
The characteristics of the crosstalk include the following characteristics. The response from adjacent data due to coma, astigmatism, and defocus aberration has an extremely low frequency range. Therefore, an obstacle to reproducing the track is a data pattern including a low frequency component among adjacent data. Specifically, the combination of a long mark such as 6T-7T and a long gap is the worst. The leakage waveform has a sinusoidal shape because the response does not extend to high frequencies.
[0035]
The signal amplitude from the shortest repetition pattern increases as the numerical aperture increases, and is greatly affected by the inclination. Even in the elliptical distribution, the dependence of the numerical aperture and signal amplitude in the track direction (shown on the right side of /) has the same tendency as described above. Leakage has a greater effect on tilt as the numerical aperture increases. The elliptical distribution increases the amount of leakage relative to the tilt as compared with the case where the numerical aperture in the track direction is isotropic, but when the numerical aperture is 0.55, the increase is almost negligible, and the elliptical distribution is isotropic. The leakage characteristics of the distributions are approximately equal. As a result, as shown in FIG. 8, the value of the elliptic distribution having an effective numerical aperture ratio of 0.6 / 0.55 when there is no inclination has a lower crosstalk characteristic than the isotropic distribution having a numerical aperture of 0.6. What is necessary is just to compare with other isotropic distributions and elliptical distributions. Also, as the disk tilt increases, the isotropic distribution with a numerical aperture of 0.6 sharply increases the crosstalk, whereas the elliptical distribution with an effective numerical aperture ratio of 0.6 / 0.55 has a small increase. Conversely, at around 0.4 degrees, the elliptical distribution having an effective numerical aperture ratio of 0.6 / 0.55 is smaller by about 2 dB. Therefore, from the overall crosstalk characteristics, the present invention uses an elliptic distribution having an effective numerical aperture ratio of 0.6 / 0.55.
[0036]
(Trial writing pattern)
In order to prevent the marks recorded for each apparatus from fluctuating and to make the characteristics of the recorded marks uniform, test writing is performed before recording to optimize the recording conditions. Conventionally, the test writing pattern was a combination of the densest pattern (repeated pattern of the shortest mark and gap) and the sparse pattern (repeated pattern of the longest mark and gap). As shown in FIG. 9, test writing is performed using a combination of the next longest pattern (for example, 3 Tw mark and gap) and the pattern having a saturation level shorter than the longest pattern (5 Tw mark and gap). In the test writing, recording and reproduction are performed while sequentially changing the recording power as shown in the figure, the deviation ΔV of the average value of the sparsest and densest patterns is detected, and the optimum recording power is determined so as to make this zero. In the test writing, the equalization operation at the time of reproduction is not performed, or even if the operation is performed, interference between waveforms is taken so that the reproduction condition does not affect the power setting of the test writing. In the linear density of the present invention, the gap length of 3 Tw is 0.8 μm, which is 0.7 μm when normalized by the spot size. From FIG. 1, the interference can be almost ignored.
[0037]
(Effect of crosstalk is reduced by circuit)
FIG. 10 also shows an equalizing coefficient learning operation to be described later, which can be realized by the same circuit configuration. In a pre-pit portion provided in advance on the optical disc, a mark indicating the head of a sector which is a break for recording data, a synchronization signal for reading an address, and address information are recorded. A pre-pit signal and a magneto-optical signal are output by a combination of the optical head detection system and signal processing. The pre-pit signal from the pre-pit signal detection unit generates a timing signal indicating the start of data recording by the pre-pit mark detection unit 200, and activates a VFO (Variable Frequency Oscillator) which generates a clock synchronized with the data in the clock generation circuit 207. . A signal indicating the head of the sector is generated and input to a circuit 216 for generating timing for controlling signal processing of the data portion.
[0038]
Magneto-optical signals are equalized as described below.constantAppropriate equalization is performed by the setting circuit 201, and the result is input to the level matching circuit 202. Here, a level is added to the equalized waveform for later processing. The level-added signal is input to the level slice circuits 203, 204, and 205, respectively. In the level slice circuit 203, as shown in FIG. 11, a pulse 100 is generated at the cross point between the middle slice level L1 of the reproduction signal and the detection signal, and the pulse is input to the clock generation circuit 207 to generate the reproduction clock CK1. . The level slice circuit 204 generates the pulse 102 at the cross point between the slice level L2 shifted above the slice level of the level slice circuit 203 and the detection signal. The level slice circuit 205 generates a pulse 101 at a cross point between the detection signal and a slice level L3 which is symmetrical with respect to the slice level of the level slice circuit 203 with respect to the slice level of the level slice circuit 203.As the slice level in the slice level circuit, the slice levels L1, L2, and L3 can be appropriately set using a known conventional technique.
The pulses 102 and 101 generated by the level slicing circuits 204 and 205 are input to phase comparators 208 and 209, respectively, and compared with the phase of the reproduced clock. The phases are compared to determine whether a pulse exists in the detection window width. Therefore, in the phase comparator, the data detection clock CK1 (having a value of “1” in the data detection range) is the inverted clock CK2 (“0” in the data detection range, and “1” when exceeding the data detection range). The level of the clock CK2 is held at the rise of the pulses 102 and 101, and reset at the rise of the clock CK1. In this case, when the sly level L1 coincides with the center of the reproduced waveform as in Case 2 in FIG. 11, the rising edges of the levels L2 and L3 are within the detection window width. The output of 209 does not appear. However, when the middle of the reproduced waveform becomes lower than the slice level L1 as in Case 1, a pulse is generated at the output 104 of the phase comparator 209, but no pulse is generated at the output 103 of the phase comparator 208. When the middle of the reproduced waveform is higher than the slice level L1 as in Case 3, no pulse is generated at the output 104 of the phase comparator 209, but a pulse is generated at the output 103 of the phase comparator 208. The outputs 103 and 104 are input to a switching circuit 210, and the connection to the subsequent processing system is changed depending on whether the outputs 103 and 104 are used for equalization coefficient learning or crosstalk measures described later. The command for the switching operation is issued by a higher-level controller that controls the reproduction processing system.
In FIG. 11, the slice levels L3 and L2 shown are set to the signal amplitudes of 0.3 and 0.7, however, this is a schematic example, and in these slice levels, FIG. It is natural that the reproduction waveform of the shortest mark shown in FIG. This is clear from the above description that when the slice level L1 coincides with the center of the reproduced waveform, the rising positions of the pulses of the slice level L2 and the slice level L3 are within the detection window width. Although FIG. 11 shows the crossing mode at the time of the rise of the reproduction waveform, it is obvious that the same processing can be performed at the time of the fall of the reproduction waveform, and this is omitted in FIG. are doing.
[0039]
When the countermeasure for crosstalk is selected by the switching circuit 210, the pulses 103 and 104 are input to the correction slice amount generation circuit 211. If the pulse 104 is present, a negative constant correction amount is output. Outputs a constant correction amount. This correction amount is input to the level matching circuit 202, and the entire level of the reproduced waveform is shifted by the correction amount. The correction slice control circuit 213 discriminates the reproduction data by the circuit 217, demodulates the data by the circuit 215, monitors the error status by the error correction circuit 219, and corrects the reproduction slice when a reproduction error occurs.amountThe fixed correction amount of the generation circuit 211 is increased or decreased to further improve the correction effect. Also, the correction operation is input to the slice level setting circuit 206 together with the switching operation command, and by changing the slice levels L1, L2, and L3 to effectively change the comparison phase difference.
[0040]
The operation when crosstalk actually occurs will be described with reference to FIGS. When there is a combination pattern of a 6 Tw mark and a gap following a combination pattern of a shortest mark and a gap, and when there is no data in an adjacent track, the signal 302 is reproduced. When a combination pattern of a 6 Tw mark and a gap is continuously recorded on an adjacent track, a signal 301 leaks from the adjacent track and is superimposed on a reproduction signal 302 to become a signal 303. Clock CK1 is detected as shown. Here, when the above-described slice correction is performed, the input signal 303 of the level matching circuit 202 is input to the level matching circuit 202, and the correction amount of the level fluctuation amount becomes 300, so that the fluctuation amount proportional to the signal 301 is canceled. Adjust the level to.
[0041]
FIG. 13 shows a method of performing correction by shifting the phase of a clock for data separation in addition to the method of performing level adjustment. Most of the block diagram has been described with reference to FIG. The corrected slice amount is detected from the outputs of the phase comparators 208 and 209 by using the corrected phase amount generation circuit 221 instead of the corrected slice amount generation circuit 211.NotoLikewise,correctionphaseThe amount is detected and input to the phase matching circuit 220 to shift the phase of the clock CK1. By performing data discrimination using the shifted clock, the influence of crosstalk can be reduced and data can be detected stably. The phase shift amount is adjusted by the correction phase control circuit 222 while observing the state of the reproduction error.
[0042]
In the above embodiment, the phase is corrected by the feedforward control. However, in another embodiment, the phase shift due to the crosstalk can be reduced by the feedforward control. The band of the clock generation circuit 207 in FIG. 10 is narrower than the frequency band of the normal data, and the clock frequency is not changed by the data. In this embodiment, when no error occurs, the tracking band is selected to be lower than the frequency of the main component of the crosstalk as usual, and when the error increases, the band of the clock generation circuit is changed according to the error information. Is set so as to cover the frequency of the main component, and the tracking band is increased. For example, when the rotation speed is set to 3600 rpm, the frequency from the combination pattern of 8 Tw marks and gaps becomes about 5 MHz. If the follow-up band of the clock generation circuit is raised above the above-mentioned frequency, the clock generation circuit may run out of control if the interval between the detection pulses greatly changes due to a defect or the like. In order to avoid this, the present invention has a configuration of a clock generation circuit as shown in FIG. The level slice circuit 203 generates a pulse at the intersection of the reproduced waveform and the slice, inputs the pulse to the phase comparator 400, compares the phase with the clock pulse from the VCO 403, and inputs the result to the phase compensation circuit 401. The phase compensation circuit 401 controls a compensation coefficient so as to determine a band and a response characteristic of the clock generation circuit. The output of the compensation circuit 401 is input to the hold circuit 402 and controls whether to hold the current for driving the VCO 403 in accordance with the control signal 410 of the OR circuit 405 or to input the output of the phase compensation circuit 401 to the VCO 403 as it is. Here, two monitoring circuits are provided so that a defect or the like does not affect the clock generation circuit 207. One is a cycle monitoring circuit 406 for monitoring the interval of the reproduced data pulse. Here, it is monitored whether the interval of the data pulse deviates from that in the data modulation code. The reference circuit 412 is input to the phase comparator 400 instead by activating the hold circuit 402 and simultaneously switching by the switching circuit 413 so that the data pulse is not input to the phase comparator. This allows the clock generation circuit 207 to immediately start operating when the data interval has returned to normal without the phase compensation circuit 401 being saturated. Another monitoring circuit is an amplitude abnormality detection circuit 407 for monitoring the amplitude of the reproduction signal. The output 101 and 102 of the level slice circuits 204 and 205 are observed, and when the output does not come for a specific period, the reproduction output decreases. , And holds the current for driving the VCO 403. The above two monitoring circuits can prevent the clock generation circuit from running away due to a defect or the like, and can improve the tracking band of the clock generation circuit. When an error is increased by the signal from the error correction circuit 219 to the clock generation control circuit 408, the switching circuit 404 issues a command according to the error information, changes the compensation coefficient of the phase compensation circuit 401, and increases the tracking band. . When no error occurs, the tracking band is selected to be lower than the frequency of the main component of the crosstalk as usual.
[0043]
The actual circuit operation will be described with reference to FIG. The operation for setting the equalization constant is the same as that in FIG. When the countermeasure for crosstalk is selected by the switching circuit 409, the pulses 101 and 102 are input to the clock generation circuit 408, and perform the above-described circuit processing.
[0044]
(Learning of equalization coefficient)
It is necessary to learn and set the equalization constant because the inter-waveform interference increases as compared with the related art. Since the signal amplitude decreases due to the interference between waveforms, the amount of the decrease is determined, and the equalization coefficient is changed so as to reduce the amount of interference. To reduce the amplitude, three slice levels are provided, a pulse is generated at a cross point between the middle slice level and the detection signal, and a clock for data reproduction is generated by the pulse. Also, a pulse is generated at the cross point between the reproduced signal and two slice levels vertically symmetrical with respect to the center level, and the phase difference between the pulse and the clock is measured. FigureThirteenAt,The output pulses 104 and 103 of the comparators 208 and 209 are measured by the measuring device 212 for a period during which the equalized data is recorded. The equalization constant control circuit 214 controls the correction equalization amount generation circuit 218 according to a control flow described later to set a default value or increase or decrease the equalization coefficient. The equalization constant setting circuit 201 has a configuration as shown in FIG. The reproduced signal is supplied to the delay line 401 and the gain corrector 404 via the buffer amplifier 400. The output of the delay line 401 is input to the delay line 402 and the addition circuit 403. The output of the delay line 402 is input to the addition circuit 403 via the gain corrector 405. The output of the gain corrector 404 is also input to the addition circuit 403. The equalization coefficients K1 and K2 are values set by the gain correctors 404 and 405. The values are set in the gain correctors 404 and 405 according to the command value of the correction equalization amount generation circuit 218.
[0045]
The learning method will be described with reference to FIGS. As the equalization coefficients K1 and K2, first, initial values set at the time of device shipment are used as default values. The sector as a data recording unit has a configuration shown in FIG. After the pre-format part provided in advance, there is a data recording area. In the data recording area, the data and the re-synchronization pattern are repeated successively after the synchronizing section including the synchronization pattern. A signal indicating a recording data area is generated by the pre-pit mark detection circuit 200 and the timing generation circuit 216 in FIG.410Is generated, and based on this, the equalization coefficient is changed. This control is performed by the equalization constant control circuit 214.
[0046]
In this control, when the phase shift exceeds a specific value, a pulse is emitted, and when the number of pulses exceeds a specific value, the equalization coefficient is increased or decreased. According to the flow of FIG. 14, the equalization constant control circuit 214 outputs a command to the correction equalization amount generation circuit 218, and measures the phase shift error measurement value in a measurement sector (shown by oblique lines) when the disk is rotated. From the input unit 212 and search for the minimum equalization coefficient. FIG. 15 shows a case where the number of changes is three.
[0047]
This learning is first performed when the device is activated. A pattern having a densest mark after a long mark is recorded by the power obtained by the test writing. A test writing pattern may be used as this pattern. While increasing or decreasing the equalization coefficient, a pulse indicating that the phase shift has exceeded a specific value is measured, and a set of equalization coefficients with the smallest number of pulses is searched for. This set is set in the device as a default value.
[0048]
The next opportunity is when a playback error occurs and a retry is performed. Learning is performed according to the flow shown in FIG. First, the sector in which an error has occurred is read using the default value, and it is determined by the error correction circuit 219 whether the error can be corrected. The determination as to whether error correction is possible is made based on the number of errors for each error processing block (referred to as a code word). If error correction is not possible, first increase or decrease the equalization coefficient, measure the number of errors for each error processing block (called a codeword), and adjust the equalization coefficient so that error correction is possible. . If the error cannot be corrected by this process, the number of pulses indicating the above-mentioned phase shift is measured, and the data recorded in the sector where the error has occurred is used to search for an equalization coefficient that minimizes the number of pulses indicating the phase shift. .
As described above, in the embodiment of the present invention, the level variation processing of the reproduced waveform in the level matching circuit and the equalization processing of the reproduced waveform in the equalization constant setting circuit have been individually described. Referring to the configuration, it is clear that the level variation processing and the equalization processing can be selectively used at appropriate timing by switching the switching circuit.
[0049]
【The invention's effect】
As described above, without significantly changing the conventional recording / reproducing method and format, a single side of a 130 mm diameter recording medium can be used..A magneto-optical disk device and a medium having a capacity of 6 GB or more can be realized.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a relationship between an interference amount due to the presence of a mark and a mark interval.
FIG. 2 is a diagram showing a groove structure of a medium according to the present invention.
FIG. 3 is a diagram showing an effect of a groove structure of a medium according to the present invention.
FIG. 4 is a diagram showing a recording mark shape control method according to the present invention.
FIG. 5 is a diagram showing a recording waveform corresponding to a data pulse waveform of a 1-7 modulation code.
FIG. 6 is a diagram illustrating the dependence of the gradient of a close-packed pattern signal on the change in the effective numerical aperture.
FIG. 7 is a diagram showing the slope dependence of the leak of the sparsest pattern due to a change in the effective numerical aperture.
FIG. 8 is a diagram illustrating the dependence of crosstalk on inclination due to a change in effective numerical aperture.
FIG. 9 is a diagram illustrating a test writing operation.
FIG. 10 is a circuit block diagram of an embodiment of the present invention.
FIG. 11 is an operation explanatory diagram for measuring a waveform shift.
FIG. 12 is an explanatory diagram of the operation of the level matching circuit when there is leakage.
FIG. 13 is a circuit block diagram of an embodiment of the present invention.
FIG. 14 is a flowchart for adjusting an equalization coefficient.
FIG. 15 is an explanatory diagram of an operation of adjusting an equalization coefficient.
FIG. 16 is a diagram showing a configuration of an equalization constant setting circuit.
FIG. 17 is a flowchart of another embodiment for adjusting an equalization coefficient.
FIG. 18 is a block diagram of a clock generation circuit according to the present invention.
FIG. 19 is a circuit block diagram of another embodiment of the present invention.
FIG. 20 is a diagram illustrating a recording area according to the present invention and a conventional recording area.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 Groove part 11 Land part 12 Mark 13 Substrate
100 Pulse indicating the cross point with the detection signal
103, 104 Pulse for detecting that the detection window width is exceeded
200 Prepit mark detection circuit 201 Equalization constant setting circuit
202 Level matching circuit
203, 204, 205 level slice circuit
206 Slice level setting circuit 207 Clock generation circuit
208, 209 Phase comparator 210 Switching circuit
211 Correction slice amount generation circuit 212 Measuring instrument
213 Correction slice control circuit 214 Equalization constant control circuit
215 Data demodulation circuit 216 Timing generation circuit
217 Data discrimination circuit 218 Correction equalization amount generation circuit
219 Error correction circuit

Claims (6)

An equalization constant setting circuit for performing an equalization process on a reproduction waveform reproduced from the optical disc;
A level matching circuit that varies the level of the reproduced waveform after the equalization processing,
A pulse is generated at a cross point between the middle slice level of the amplitude of the reproduced waveform from the level matching circuit and the reproduced waveform, and a reproduced clock having a detection window width is generated by a clock generating circuit that receives the pulse as an input. One level slice circuit for setting one slice level;
A pulse is generated symmetrically with respect to the one slice level with respect to the signal amplitude, and a pulse is generated at an intersection with the reproduction waveform. The other pulse is generated so that the position of the generated pulse is within the detection window width. Another level slice circuit for setting a slice level,
A phase comparator that compares the phase of the pulse generated from the other level slice circuit and the recovered clock that forms the detection window width,
The phase comparator determines whether the pulse generated from the another level slice circuit exists within the range of the detection window width,
The optical disc apparatus according to claim 1, wherein the level matching circuit changes the level of the reproduced waveform based on a correction amount corresponding to a result of the determination by the phase comparator. An equalization constant setting circuit for performing an equalization process on a reproduction waveform reproduced from the optical disc;
A level matching circuit that varies the level of the reproduced waveform after the equalization processing,
One slice level for generating a pulse at a cross point between the middle slice level of the amplitude of the reproduced waveform from the level matching circuit and the reproduced waveform and generating a reproduced clock for forming a detection window width via a clock generating circuit. One level slice circuit for setting
Another level slice circuit that is set symmetrically to the one slice level, generates a pulse at an intersection with the reproduction waveform, and sets another slice level;
A phase comparator that compares the phase of the pulse generated from the other level slice circuit and the recovered clock that forms the detection window width,
The phase comparator determines whether the pulse generated from the another level slice circuit exists within a range of a detection window width,
Providing a measuring device for measuring the number of the pulses existing outside the range of the detection window width determined by the phase comparator over a predetermined period,
An optical disc device, wherein the equalization constant setting circuit controls an equalization coefficient for the reproduced waveform so as to minimize the measured number by an output from the measuring device. The optical disc device according to claim 2,
The equalization constant setting circuit controls the equalization coefficient of the equalization constant setting circuit when the number of pulses existing outside the range of the detection window width measured by the measuring device exceeds a threshold. An optical disc device characterized by the above-mentioned. The optical disc device according to claim 1,
Switching the control process for changing the level of the reproduced waveform by the level matching circuit,
The equalization constant setting circuit, by the output from a measuring instrument that measures the number of the pulses existing outside the range of the detection window width determined by the phase comparator over a predetermined period, the measured number is An optical disc device, which performs a control process for controlling an equalization coefficient for the reproduced waveform so as to minimize the reproduction waveform. Performs equalization processing and level matching on the playback waveform played back from the disc,
One slice level for generating a pulse at a cross point between the middle slice level of the amplitude of the reproduced waveform and the reproduced waveform, which has been subjected to the equalization processing, and generating a reproduced clock having a detection window width, and Setting another slice level that is set symmetrically with respect to the one slice level in terms of signal amplitude and generates a pulse within the detection window width at an intersection with the reproduction waveform,
Comparing the phase of the pulse generated at the set other slice level and the phase of the reproduced clock forming the detection window width,
Determine whether the pulse is within the range of the detection window width by comparing the phase,
A reproduction waveform detection method for an optical disk, wherein the level adjustment is performed by varying a level of the reproduction waveform based on a correction amount corresponding to a result of the determination. Performs equalization processing and level matching on the playback waveform played back from the disc,
A slice level for generating a reproduction clock for forming a detection window width by generating a pulse at a cross point between the middle slice level of the amplitude of the reproduction waveform and the reproduction waveform, which has been subjected to the equalization processing, and setting the slice level. Setting another slice level that is set symmetrically with respect to the one slice level with respect to the signal amplitude and generates a pulse at an intersection with the reproduction waveform,
Comparing the phase of the pulse generated at the set other slice level and the phase of the reproduced clock forming the detection window width,
By comparing the phases, the number of deviations of the compared phases exceeding a predetermined value is measured,
A method for detecting a reproduced waveform of an optical disk, wherein an equalization coefficient for the reproduced waveform is controlled so that the measured number is minimized.

Images