200818152 • 九、發明說明： . 【發明所屬之技術領域】 本發明係有關於光學儲存裝置 ^ “ <寫入朿略微調（write strategy rmg)，_關於利料對時脈邊緣偏差（data-to-el〇ckedge _wn)來_ (tune)寫人策略參數之方法及系統。 【先前技術】 、由於夕媒體朗肋發展，儲存大量數位資料之需求亦快速地 成長□此，儲存谷㈣且體積小巧之光學儲存媒體，例如：光 碟片（Compact Disc，CD )或數位多樣化光碟（邮⑹200818152 • Nine, invention description: . [Technical field of the invention] The present invention relates to an optical storage device ^ <write strategy rmg, _ about the margin of the clock edge deviation (data- To-el〇ckedge _wn) _ (tune) method and system for writing human strategy parameters. [Prior Art] Due to the development of the eve media, the demand for storing large amounts of digital data has also grown rapidly. Compact optical storage media, such as compact discs (CDs) or digitally diverse discs (mail)
Versatile Disc， DVD) ’ ,沈非系流行，且光學儲存裝置，例如··光碟機（⑶此… 或數位多樣化光碟機（DVDdrive)，已成為個人電腦之標準配備， 用來進行上述之多媒體應用。 以上述之光碟機為例，當光碟機被控制以將資料寫入可記錄式 光碟片（CD-Recordable disc，CD-R disc)時，光碟機中之雷射二 極體之寫入功率（writingpower)通常係被設為特定值，且對應於 資料之寫入脈衝（write pUise)被用於在可記錄式光碟片之溝槽 (groove)上將資料記錄成複數個凹坑（pit)與平面（land)。寫 入功率之特定值可通過最佳功率校正（optimal power calibration， OPC)程序來取得。另一方面，通過寫入策略微調（write strategy 5 200818152 层）裎序亦稱為燒錄策略微調C recording strategy tuning )程 序’、改變用來控制寫入脈衝的寬度之寫入策略參數，可增加可記 錄式光碟#上卿奴凹坑解面的長紅射度。請參考可記 錄式光碟#規格之侧文件，例如：橘皮書第—部b⑻^ Parti)，以取得更多資訊。 依據相關技術，特絲置（勤^波器）可姻於寫入策略微 調程和通常，在預先之試誤性寫人程序之後，依據示波器上所 顯不之複數個重建波形(亭〇duced wavef_ )的眼圖（啊油咖） 來控制寫人脈麟寬度之新的—組寫人策略參數通常係由工程師 或研究員基於鎌來決定。然而的程序必須針對各種 可應用的賴以及不_記錄速度而錢地進行，重複進行的程 序至少包括寫人賴資料、檢視示波^所顯狄重建波形的= 圖、以及藉由經驗依據眼圖來決定新的一組寫入策略參數，因此 使用此方法會㈣球師或研究貞許乡日销。喊藉 來決定這些寫人策略參數並非自動的運作，因此上述之寫入2 微調程序相當耗時。另外，由於㈣經驗依據該關來決定新的 -組寫入策略參數並不是定量的，因此上述之寫人策略微調程序 是不明確的（indefinite)。在某錄況下’含财清_圖會干擾 寫入策略微調程序甚至會使寫入策略微調程序失效。 足 特定儀器，例如：時間間距分析_imeim_ia啤激Τα) 或抖動計量器（一酬er)，可能有助於取得用來決定新的一組 6 200818152 寫入策略參數的資訊。然而，仍然需要進行類似的例行工作，且 若時間間距分析儀或抖動計衫健簡單地被_簡量測之用 而不設置額外的控制系統’則手動微調程序所造成之相同缺點仍 然存在。另外，自特定儀器所取得之資訊的意義通常是隱含的 因此，需要耗費經驗豐富的工程師或研究員許多時間。 【發明内容】 為解決以上技術問題，本發賴供了—_來微調光學儲存裝 置之寫入策略參數之方法及其系統。 、本發明提供了-_來難光學儲存裝置之寫人策略參數之 方法，包含：伽彳複數個長度，每—長度對應於光學儲存裝置所 存取之光學儲存媒體上之—凹坑（pit)或—平面（iand);進行對 應於稷數個㈣集類型（data set咖）之計算，以及產生分別對 (data-to-clock edge tati°n>其中資料集類型包含至少一凹坑平面凹坑⑽and-pit) 資料木類51或至）_平面凹坑平面資料集類型。 以及使用雜對時脈邊緣偏差以微調寫人策略參數，其中寫入策 略參數分別對應於資料集類型。 /本發明另提供了一種用來微調光學儲存裂置之寫入策略參數 之糸統、’，含偵測器、計算模組、與控制器，其中計算模組係柄 接至债H控制別她接至計算模組。伽丨器量測複數個長度， 200818152 每-長度對應於藉由光學储存裝置所存取之光學儲存媒體 凹坑或-平面。計算模組進行對應各種資料翻型之計瞀， 產生分別·於這歸觸類狀資料對時脈槪偏差， 料集類型包含至少1坑平面凹㈣贿_或至少—平面凹坑 平面資料集_。控繼彻這㈣料對時脈邊緣偏差以微調分 別對應於这些資料集類型之寫入策略參數。 本發明另提供了—_來微調光學儲存裝置之寫人策略參數 ，方法’包含.侧複數個長度’每—長度對應於軸光學儲存 裝置:存取之光學儲存舰上之—凹坑或_平面；進行對應於複 數個貝料木翻之打以及產生分賴應於資料無型之複數個 資料_脈邊緣偏差。進行對應於資贿_之計算以及產生分 別對應於龍集_之:#料對時脈邊緣偏差之步.包含：取得 (derive) >料對日樣邊緣偏差大於第—門檻值（threshed)之資 料术類5L ’於資料對時脈邊緣偏差大於第一門檻值之資料集類型 中，取得資_發生次數多於第二門檻值所指示之次數之資料集 齡，以及a丨异對應於資料集發生次數多於第二門檻值所指示之 資料木類型之寫入桌略參數的調整量。以及利用調整量以 微调寫入策略參數。 本發明提供之用來微調光學儲存裝置之寫人策略參數之方法 及/、系統，藉由依據資料對時脈邊緣偏差來微調寫入策略參數， 而不而要特定外部裝置的協助，亦不需要耗費球師或研究員過 200818152 多時間，即可微調光學儲存裝置之寫入策略參數。 【實施方式】 本發明提供用來微調（tune)光學儲存裝置之寫入策略參數 (write strategy parameter )之系統。這些系統係用來微調寫入策略 參數之電路，該電路係置於該光學儲存裝置中。這些系統的一些 實施例也可以在實質上（substantially)為該光學儲存裝置本身。 為了簡明起見，在以下的說明中係採用將該系統實施於一電路 中。然而，其他實施方式亦可應用於這些詳細的實施例。 請參考第1圖，第1圖為依據本發明第一實施例之用來微調光 车儲存裝置100之寫入策略參數之系統1⑻c的方塊圖，其中系統 100C係為置於光學儲存裝置100中之電路。光學儲存裝置〗⑻可 對光學儲存媒體102進行資料存取。為了簡明起見，本實施例採 用可§己錄式光碟片（CD_Recordable disc，CD-R disc)作為光學儲 存媒體102，並採用光碟機（CDdrive)作為光學儲存裝置1〇〇來 進行說明。熟悉此項技藝者應可理解，其它種類之光學儲存媒體， 例如·· DVD-R 規格、DVD_rw 規格、DVD+R 規格、dvd rw 規格、或DVD-RAM規格之數位多樣化光碟，以及其對應之光學 儲存裝置，例如：數位多用途光碟機（DVD drive)，皆可應用本 發明達到類似功效。 如第1圖所示，於光學儲存裝置100之讀取模式中，光學儲存 200818152 裝置100之光學讀取頭（optical pickup) 110自光學儲存媒體102 。賣取資料’以產生原始射頻訊號（raW ra(ji〇丘叩邱加乂 Signa〗，瓜界Kp signal) 111。光學儲存裝置1⑻之波形等化器（wavef〇rmeq皿ηζ6γ) 112專化原始射頻訊號hi以產生重建訊號（网⑽啦皿1)， 於本實施例中重建訊號係為射頻訊號113。另外，光學儲存裝置 100之分切器（siicer) 114分切（slice)射頻訊號113以產生分切 訊號115。因為光學讀取頭110、波形等化器112、以及分切器114 之運作原理均為熟悉此項技藝者所知悉，故不在此贅述其細節。 於第1圖所示之光學儲存裝置100中，調變器16〇、寫入脈衝 (write pulse )產生器 162、與發射源驅動器（ra(jiati〇n source driver ) 164會依據寫入策略參數共同驅動光學讀取頭11〇 ;依據分切訊號 115，系統l〇〇c通過控制訊號151來微調寫入策略參數。調變器 160係耦接於光學儲存裝置1〇〇之編碼器（未顯示），用來調變編 碼器所輸出之編碼資料以產生調變訊號161，調變訊號161載有 (carry )八轉十四调變（eight-|；0_f〇ur^een m〇duiati〇n，efm )資訊。 寫入脈衝產生器162依據上述寫入策略參數，產生對應於調變訊 唬161所載的八轉十四調變資訊之寫入脈衝，並輸出由寫入脈衝 汛唬163所載之寫入脈衝。另外，發射源驅動器164依據寫入脈 衝减163來產生驅動訊號165以驅動光學讀取頭11〇。調變器 160、寫入脈衝產生器162、與發射源驅動器164之運作原理均為 熟悉此項技藝者所知悉，故不在此贅述其細節。 200818152 • 依據本貝施例’糸統1⑽C包含·鎖相迴路（phase-locked loop， • PLL) 120、偵測器（例如：第1圖所示之八轉十四調變長度偵測 器130)、计鼻板組140、以及控制器（例如··第1圖所示之寫入 脈衝控制器150)。計算模組140包含型樣依附分類器（pattem dependency classifier) 142與資料對時脈邊緣偏差計算器 (data-to_clock edge deviation calculator) 144。鎖相迴路 120 依據 分切訊號115藉由鎖定分切訊號115之通道位元率（1/T)(channel ’ blt rate)來產生一個八轉十四調變資料時脈（EFM加故d〇ck) CLK，其中八轉十四調變資料時脈CLK之週期係被視為丨丁。八 轉十四調變長度偵測器13 〇依據八轉十四調變資料時脈CLK來擷 取分切訊號115所載之八轉十四調變資訊，並偵測複數個長度， 其中每一長度對應於記錄在光學儲存媒體1〇2上之一凹坑（扣丈） 或一平面（land)。典型的分切訊號115係為方波，其上升邊緣(rising edge)與下降邊緣⑽ingedge)之_間距〇nt魏以及下降 邊緣與上升邊緣之間的間距均可有各種不同長度。於本實施例 中\轉十四口周變長度偵測器13〇量測分切訊號出之上升邊緣 與下降邊緣之間的間距及分切訊號m之下降邊緣與上升邊緣之 關間距，來作為上述之長度，其中每—_係對應於一凹坑或 .7平面。因此，長度包含對應於複數個凹坑之複數個凹坑長度p， 以及對應於複數個平面之複數個平面長度L。每個凹坑長度P代 表d著光學儲存媒體102上之溝槽（_ve)所記錄之一凹坑，每 個平面長度L代表沿著溝槽所記錄之一平面。需要注意的是，本 發明之另一實施例的分切訊號ιΐ5可載有加強型八轉十四調變 11 200818152 訊 (EFMPlus，EFM+)資訊（例如：應用DVD_R規格之實施例）或 其它相容於人轉切調變/加強型人轉十四機的變化規格之資 七7 〇 β依據第-實施例，於可記錄式光碟狀理想狀況下，由分切訊 號II5中細取的凹j：几長度與平面長度均為時脈週期τ之倍數，且 這些凹坑長度與平面長度的分佈範圍係從3T至11τ。也就是說， 凹坑=長度Ρ或平面之長度L可為3Τ、4Τ、...、或ητ。所以， 用來量測這些職長额平面長度之參考減（例如：上述之八 轉十四調㈣料時脈CLK)的合理週期應小於或等於τ。依據本 實施例，輸人Μ轉十_變長度偵·⑽之參考訊號係為八 轉十四爾倾日械CLK，所以參考訊號的職係為τ。於可吃 錄式光碟狀實況下，由八射四調縣度侧器ηο之輸 出訊號131所載的長度L與ρ通常並非為Τ的整數倍。計算模也 刚可分別對複數個資料集類型（datasettype)作計算，並產生分 別對應於這些資__之資贿時脈邊緣偏差㈤㈣韻 edgedevia㈤’其中這些資料對時脈邊緣偏差係由資料對時脈邊 緣偏差計Μ 144之輸出訊號145所載，f料集_至少包含一 凹坑平面凹坑（pit_land_pit) f_麵或至少包含—平面凹坑平 面㈤d-_and)資料集類型。每—f料集類型係對應於至少一 特定目標凹坑長度（例如：3T、4T、...、11τ)與—特定目標平面 長度（例如：31^、...、叫與另—特定目標凹坑長度（例如： 3Τ、4Τ、...、11Τ)之組合，或至少—蚊目標平面長度盘至少一 12 200818152 特定目標凹坑長度與另—特定目標平面長度之組合。 型樣依附分類器142將複數個資料集分類為各資料集類型。於 本實施例中’每一資料集，即(P1，L，P2)或(u，p，L2)，包含有兩個 長又中為料集(P1，L，P2)係指對應於一平面之長度與兩個分別 與其相_ (adjaeent)凹坑之μ，啼料集(L1，p，L2)係指對應 於一凹坑之長度與兩個分別與其相鄰的平面之長度。在此，如(Ρητ, LmT，P1T)或(LnT，pmT，LlT)之標示法係用來表示上述資料集類型，其中 nT、mT或it係以時脈週期τ為單位來表示長度；於本實施例 中 ’ n = 3、4、...、或 11，m = 3、4、...、或 U，且1=3、4、、 或11。資料集類型(LnT，PmT Lit)中之每一者，例如：n =⑽、以=秦 且1 1〇之資料集類型(Ln〇*T，PmO*T，L1()*T)，係被用於對應於目標平 面長度為nO*T之平面、與緊_平面之後且目標凹坑長度為 mO*T之凹坑、以及緊_凹坑之後刻標平面長度為1㈣之平 面之複數個資料集(L1，P，L2)進行分類。相似地，資料集麵& LmT，P1T)中之每一者，例如：n = n〇、m = m〇、且！ = 1〇之資料集類 型(Ρη〇*τ，Lm〇*T，Pl〇*T)，係被用於分類對應於目標凹坑長度為η〇*τ 之凹坑、與緊隨該凹坑之後且目標平面長度為祕了之平面、以 及緊k A平面之後且目;^凹坑長度為1Q* 了的複數個資料集(PI，L P2)。需要注意的是，複數織料集類型(LnT，PmT，L1T)中之每一者，’ 例如.貪料集類型(1^，1>111(^，]：1()叮）’係對應於特定目標平面長度 n〇*T、特定目標凹坑長度mG*T、與特定目標平面長度1㈣之组 合(n〇*T，m〇*T，10*T)。複數個資料集類型I LmT，⑸中之每一 13 200818152 者，例如：資料集類型(Ρη〇*τ，WtIVt)，係對應於特定目伊凹坑 長度η〇*Τ、特定目標平面長度m〇*T、與特定目標凹坑長度 之組合(n〇*T，m〇*T，1〇*τ)。由於㈣與！各有九個可能的又值〇 到11)，.所以對資料集類型(LnT，PmT，L1T)而言有(9 * 9 * 9)個組合， 而對貝料集類型(PnT，LmT，PlT)而言也有(9 * 9 * 9)個組合，因此資 集類型之總數係為(9 * 9 * 9 * 2) = 1458。 、’ 於實作上可賤擇某她合；寫人策略麵職據這些所選擇 的組合來調整，而非依據所有可能的組合來調整。 另外’若倾_，Ρ，增長度;U、p、與㈣足下列條件， 則型樣依附分鋪142可將這缩帽L1，P，L2)分類進資料集類 型0^11〇*1^，卩111〇*7'，11()*丁）· (n0 - 0.5) * T <L1 <(n〇 + 0.5) * T ； (m0 - 〇·5) * Τ 立 gmO + 0.5) * Τ ;且 (10 — 0·5) * Τ 红2 410 + 0·5) * Τ。 相似地，若資料集(P1，L，P2)中的長度⑴匕與㈣足下列條件， 則型樣依附分類II⑷可將這些資料雜分類進資料集類 型(PnO*T，Lm〇*T，Ρΐ〇*τ) · (ηΟ 0.5) * Τ <Ρ1 <(η〇 + 0.5) * Τ ； 14 200818152 (m〇 — Ο·5) * T 红 ^(m0 + 0·5) * τ ;且 (10 - 0·5) * Τ 分2 或10 + 0·5) * Τ。 資料對時脈邊緣偏差計算器Μ4可如 應於資料集類型一 差。資料對日械邊緣偏差計算n 144計算魏個熟對時脈邊= 長度（data-to_clockedgelength)，其中每一個資料對時脈邊緣長户 係為上述之參考訊號（於本實施射即八轉十，㈣料時^ CLK)之上升邊緣或下降邊緣以及分切訊號ΐΐ5之轉變邊緣 ^ansitiG'edge)之間的間距。另夕卜’資料對時脈邊緣偏差計算 器144計算複數個差值’以產生分別對應於各個資料集類型 PmT’L1T)與(PnT，LmT，PlT)之資料對時脈邊緣偏差。上述差值中的每一 個係為-資料對時脈邊緣長度與—目標㈣對時脈邊緣長度之間 的差值，其中目標資料對時脈邊緣長度係為對應於特定資料集類 型(Ln〇*T，PmG*T，L1G*T)或(PnG*T，LmG*T,p1()*T)之預定值。 以第2圖所示之情況為例來進一步說明。第2圖為依據第一實 施例之利用資料對時脈邊緣偏差來進行長度補償的示意圖，其中 凹坑A對應於目標長度(即凹坑之目標長度為4丁）。如第2圖 所示，Ttopr與Tlast分別表示用來控制凹坑的開始位置與結束位 置之寫入策略參數。於第2圖所示之情況中，凹坑a被視為凹坑 PT，其結束位置於此情況中並不完美；換言之，於凹坑ρΤ之結束 位置上可測量到邊緣偏差。位於凹坑ρ丁之前的先前平面（previous 15 200818152 land) PL有目標長度W (即平面之目標長度為4Τ)，位於先前平 面PL之前的先前凹坑（previous pit) ΡΡ有目標長度Ρ3Τ。另外， 位於凹坑ΡΤ之後的後續平面（f〇ll〇wing iand)FL有目標長度l5T， 位於後績平面FL之後的後續凹坑（f〇n〇wing pit) FP有目標長度 Ρπ。在此，型樣（PP，PL，PT)與（PT，FL，FP)對應於上述之至少 一凹坑平面凹坑資料集類型，以及型樣(PL，ΡΤ，FL)對應於上述之 至少一平面凹坑平面資料集類型。 依據計算模組140所進行、且分別對應於型樣(pp，pL，pT)、(pL， PT，FL)、與(PT，FL，FP)的資料集類型之計算（尤其為統計方面的 計算）’統計結果可被擷取以指出型樣(pp，PL，PT)、（pL，pT，FL)、 與(PT，FL，FP)中之一者主導著凹坑ρτ的不完美結束 之現象，例如，型樣(PT，FL，FP)主導著凹坑ρτ的不完美結束之 現象，則相對應之寫入策略參數Tlast(n，m，丨）（其代表用來控制對 應於負料集類型(PnT，LmT，P1T)之凹坑的結束位置之寫入策略參婁文） 應被微調，以調整凹坑打之不完美的結束，其中依據第2圖所示 之情況，（n，m，1) = (4, 5, 4)。 而要/主思的疋，:貝料對時脈邊緣偏差計算器144可計算資料對 時脈邊緣長度dl。於本實施财，龍_脈姐長度dl係為分 切=號115之下降邊緣的時間點D (即對應於凹坑A的結束位置 之間點）、从八轉十四機體時脈CLK巾隨狀上升邊緣 (例士日寸間點d)之間的間距。請注意，時間點D實質上係為 16 200818152 ‘射頻訊號113之值跨越（贿S)某-預定值（例如 • 11114的分切位準之值）時的同-時間點。資_ 計鼻器144藉由_分切訊號Π5的位準由高至低^ f 時間點D。同樣的方法可以被應用於計算對應資料隼類‘ ^ 〜爾料對時脈邊緣長度。此外，魏對時脈邊緣偏 144 爲差。於L貫施财’ f料對時脈邊緣偏差計算器⑷藉 计分析多靖應於特定:#料㈣型之差值，來產^魅 型之資料對時脈邊緣偏差。統計分析可為對這些差值進== 員 异，或找出這些差值之眾數（m〇stfrequentvalue)。上述差值中之 t個可⑽-資料對時脈邊緣長度與目標資料對時脈邊緣長度 (例如.弟2圖所示之情況中為〇5T)之間的差值。因為對座於 凹坑Α的結綠置之時_在理綠況下應該是時_〜 於集_(P4T，L5T，Ρ4τ)的目標⑽對雜邊緣長度係為 Ιλ5 丁0 …以—第圖所枚情況為例來進一步說明。第3圖為依據本發明 弟-貫域m騎時脈雜偏絲進行長度補償的示意 圖，其中凹坑b亦對應於目標長度心。於第3圖所示之情況中， 凹坑B麵如坑PT’啊ρτ_触胁赠況巾並不完美； 換吕之，於凹坑ΡΤ之開錄置上可測量到邊緣偏差。 相似地，依據計算· 140所進行、且分別對應於型樣(pp，PL， 17 200818152 ΡΤ)、（PL，PT，FL)、與(PT，FL，FP)的資料集類型之計算，統計結 果可被擷取以指出型樣(PP，PL，PT)、（PL，PT, FL)、與(PT，FL，Fp) 中之一者主導著凹坑PT的不完美開始之現象，例如，型樣(ρρ，·ρ^ ΡΤ)主導著凹坑ΡΤ的不完美開始之現象，則相對應之寫入策略泉 數丁topr(n，m，1)(其代表用來控制對應於資料集類型(PnT， 之凹坑的開始位置之寫入策略參數）應被微調，以調整凹坑ρτ< 不元美開始，其中依據苐3圖所示之情況，（n，m，1) = (3, 4, 4)。 需要注意的是，資料對時脈邊緣偏差計算器144計算資料對時 脈邊緣長度d2。於本實施例中，資料對時脈邊緣長度汜係為分切 訊號II5之上升邊緣的時間點E (即對應於凹坑B的開始位置之 時間點）以及於八轉十四調變資料時脈CLK+隨後之上升邊礙例 如：時間點e所指之處)之間的間距。請注意，時間點e實質上 係為射頻訊號113之值跨越預定值（例如：對應於分切器ιΐ4的 物立準之值）時的同-時間點。麵對時脈邊緣偏差計算哭144 =侧分切訊號Π5的位準由低至高之轉變來侦測時間㈣。同 樣的方法可以被應計刺應於㈣細 =_值嫩嫩輪败__植邊緣偏差4 迷差值中的每-個係為―資料對時脈邊緣 對時脈邊緣長度之間的差值。於第3圖 二目心貝抖 於凹坑B的開始位置之時間點在理想狀=，因為對應 以對應於__(~，L4T，P4糊桿資，所 ^貝枓對時脈邊緣長度係 18 200818152Versatile Disc, DVD) ', Shen Fei is popular, and optical storage devices, such as CD-ROM ((3) this... or digitally diverse CD player (DVDdrive), have become standard equipment for personal computers, used to carry out the above multimedia For example, when the optical disk drive is controlled to write data into a CD-Recordable Disc (CD-R disc), the writing of the laser diode in the optical disk drive is taken as an example. The writing power is usually set to a specific value, and a write pUise corresponding to the data is used to record the data into a plurality of pits on the groove of the recordable optical disc (pit). And the plane. The specific value of the write power can be obtained by the optimal power calibration (OPC) program. On the other hand, the write strategy fine-tuning (write strategy 5 200818152 layer) is also called For the programming strategy to fine-tune the C recording strategy tuning program, change the write strategy parameters used to control the width of the write pulse, and increase the long redness of the recordable CD-ROM. Please refer to the side file of the recordable disc # specification, for example: Orange Book Part B(8)^Parti) for more information. According to the related technology, the special wire (industry wave device) can be used to write the strategy fine-tuning process and usually, after the pre-trial trial and error writing procedure, according to the complex reconstruction waveform displayed on the oscilloscope The eye diagram of the wavef_) (the ah oil coffee) to control the new width of the write temperament - the group writer strategy parameters are usually determined by the engineer or researcher based on 镰. However, the program must be carried out for various applicable and non-recording speeds. The repeated procedures include at least writing the data, viewing the oscilloscope, reconstructing the waveform = graph, and using the empirical basis. The graph determines the new set of write strategy parameters, so using this method will (4) the golfer or the researcher Xu Xuan daily sales. Calling to determine these writer policy parameters is not automatic, so the above-mentioned write 2 fine-tuning process is quite time consuming. In addition, since (4) experience determines that the new-group write strategy parameters are not quantitative based on the relationship, the above-mentioned writer strategy fine-tuning procedure is indefinite. In a certain recording situation, the inclusion of the _ _ graph will interfere with the write strategy fine-tuning program and even invalidate the write strategy fine-tuning program. Specific instruments, such as time-interval analysis _imeim_ia beer Τα or jitter metering, may help to obtain information used to determine a new set of 6 200818152 write strategy parameters. However, similar routine work is still required, and if the time-spacing analyzer or jitter meter is simply used for simplification without setting up an additional control system, the same shortcomings caused by manual fine-tuning still exist. . In addition, the meaning of the information obtained from a particular instrument is often implicit. Therefore, it takes a lot of time for an experienced engineer or researcher. SUMMARY OF THE INVENTION In order to solve the above technical problems, the present invention provides a method and system for fine-tuning a write strategy parameter of an optical storage device. The present invention provides a method for writing a human strategy parameter of a difficult optical storage device, comprising: a plurality of lengths of gamma, each length corresponding to a pit on an optical storage medium accessed by the optical storage device (pit) Or - plane (iand); performs calculations corresponding to the number of sets of data sets, and produces separate pairs (data-to-clock edge tati°n> where the data set type includes at least one pit plane Pit (10) and-pit) Data type wood 51 or to) _ plane pit plane data set type. And using the mismatched clock edge offset to fine tune the write strategy parameters, where the write strategy parameters correspond to the data set type, respectively. The invention further provides a system for fine-tuning the write strategy parameters of the optical storage split, including a detector, a calculation module, and a controller, wherein the calculation module handle is connected to the debt H control She is connected to the computing module. The gamma meter measures a plurality of lengths, and the length of each of the 200818152 corresponds to an optical storage medium pit or plane accessed by the optical storage device. The calculation module performs the calculation of the corresponding data type, and generates the deviation of the clock data from the data of the type of the contact type, and the type of the material set includes at least one pit plane concave (four) bribe _ or at least - plane pit plane data set _. This control (4) feeds the clock edge deviation to fine-tune the write strategy parameters corresponding to these dataset types. The present invention further provides -_ to fine-tune the writer strategy parameters of the optical storage device, the method 'includes. The side plurality of lengths' each length corresponds to the axis optical storage device: access to the optical storage ship - pit or _ Plane; performs a plurality of data corresponding to a plurality of shells and produces a plurality of data _ pulse edge deviations that are dependent on the data. Performing the calculation corresponding to the bribe _ and generating the steps corresponding to the dragon set _: #料对脉脉边偏差. Including: derived > material to the daily edge deviation is greater than the first threshold (threshed) The data class 5L 'in the data set type of the data with the clock edge deviation greater than the first threshold value, the number of times the number of acquisitions is more than the number of times indicated by the second threshold value, and a difference corresponds to The number of occurrences of the data set is greater than the adjustment amount of the written parameter of the data type indicated by the second threshold value. And use the amount of adjustment to fine tune the write strategy parameters. The invention provides a method and/or system for fine-tuning a write strategy parameter of an optical storage device, by fine-tuning a write strategy parameter according to a data edge offset, instead of assisting a specific external device, or It takes a lot of time for the player or researcher to spend more than 200818152 to fine tune the write strategy parameters of the optical storage device. [Embodiment] The present invention provides a system for fine tune the write strategy parameter of an optical storage device. These systems are used to fine tune the circuitry that writes the strategy parameters that are placed in the optical storage device. Some embodiments of these systems may also be substantially the optical storage device itself. For the sake of brevity, the system is implemented in a circuit in the following description. However, other embodiments are also applicable to these detailed embodiments. Please refer to FIG. 1. FIG. 1 is a block diagram of a system 1(8)c for fine-tuning a write strategy parameter of a terrarium storage device 100 according to a first embodiment of the present invention, wherein the system 100C is placed in the optical storage device 100. The circuit. The optical storage device (8) can access the optical storage medium 102 for data access. For the sake of brevity, the present embodiment uses a CD_Recordable Disc (CD-R disc) as the optical storage medium 102, and uses a CD drive as an optical storage device. Those skilled in the art should be able to understand other types of optical storage media, such as DVD-R specifications, DVD_rw specifications, DVD+R specifications, dvd rw specifications, or DVD-RAM specifications, and their corresponding digital discs. Optical storage devices, such as digital drive (DVD drive), can be applied to achieve similar effects. As shown in FIG. 1, in the read mode of optical storage device 100, optical storage 200818152 optical pickup 110 of device 100 is from optical storage medium 102. Selling data 'to generate the original RF signal (raW ra (ji〇丘叩邱加乂Signa〗, Kap signal) 111. Optical storage device 1 (8) waveform equalizer (wavef〇rmeq dish ηζ6γ) 112 specialization original The RF signal hi is used to generate a reconstruction signal (network (10) dish 1). In this embodiment, the reconstructed signal is the RF signal 113. In addition, the Siicer 114 of the optical storage device 100 slices the RF signal 113. The splitting signal 115 is generated. Since the optical pickup 110, the waveform equalizer 112, and the divider 114 operate on a principle known to those skilled in the art, the details are not described herein. In the optical storage device 100, the modulator 16〇, the write pulse generator 162, and the source driver (ra) 163 drive the optical reading according to the write strategy parameters. The head 11〇; according to the cut signal 115, the system l〇〇c fine-tunes the write strategy parameter by the control signal 151. The modulator 160 is coupled to the optical storage device 1〇〇 encoder (not shown) for The code output by the modulation encoder In order to generate the modulation signal 161, the modulation signal 161 carries (carry) eight-to-fourteen modulation (eight-|; 0_f〇ur^een m〇duiati〇n, efm) information. The write pulse generator 162 is based on The write strategy parameter generates a write pulse corresponding to the eight-turn fourteen modulation information carried by the modulation signal 161, and outputs the write pulse carried by the write pulse 汛唬 163. In addition, the source driver 164 generates drive signal 165 based on write pulse minus 163 to drive optical pickup 11. The operation of modulator 160, write pulse generator 162, and transmit source driver 164 are well known to those skilled in the art. Therefore, the details are not described here. 200818152 • According to this example, the system 1 includes the phase-locked loop (PLL) 120 and the detector (for example, the eight-turn ten shown in Figure 1 a four-modulation length detector 130), a nasal plate set 140, and a controller (for example, the write pulse controller 150 shown in Fig. 1). The calculation module 140 includes a pattern dependent classifier (pattem dependency) Classifier) 142 and data pair clock edge offset calculator (data-to_ Clock edge deviation calculator) 144. The phase-locked loop 120 generates an eight-turn fourteen-modulation data according to the channel bit rate (1/T) of the switching signal 115 according to the switching signal 115. Pulse (EFM plus d〇ck) CLK, in which the period of eight to fourteen modulated data clock CLK is regarded as a Kenting. The eight-to-fourteen-modulation length detector 13 撷 is based on the eight-to-fourteen-modulation data clock CLK to capture the eight-to-fourteen modulation information contained in the switching signal 115, and detects a plurality of lengths, each of which One length corresponds to a pit (depression) or a land recorded on the optical storage medium 1〇2. A typical splitting signal 115 is a square wave, and the spacing between the rising edge and the falling edge (10) ingedge and the spacing between the falling edge and the rising edge can be of various lengths. In the embodiment, the four-port peripheral variable length detector 13 measures the distance between the rising edge and the falling edge of the switching signal and the closing distance between the falling edge and the rising edge of the switching signal m. As the above length, each of them corresponds to a pit or a .7 plane. Thus, the length includes a plurality of pit lengths p corresponding to a plurality of pits, and a plurality of plane lengths L corresponding to the plurality of planes. Each pit length P represents a pit recorded by a groove (_ve) on the optical storage medium 102, and each plane length L represents a plane recorded along the groove. It should be noted that the switching signal ιΐ5 of another embodiment of the present invention may carry enhanced eight-to-fourteen modulation 11 200818152 (EFMPlus, EFM+) information (for example, an embodiment of the application DVD_R specification) or other phase. According to the first embodiment, in the case of the recordable disc-shaped ideal condition, the concave cut by the slitting signal II5 is adopted in the change specification of the human-transformed/reinforced human-turning fourteen machine. j: Several lengths and plane lengths are multiples of the clock period τ, and the distribution of these pit lengths and plane lengths ranges from 3T to 11τ. That is, the pit length Ρ or the length L of the plane may be 3 Τ, 4 Τ, ..., or ητ. Therefore, the reasonable period for measuring the reference subtraction of these lengths of the length of the job (for example, the above eight to fourteen (four) material clock CLK) should be less than or equal to τ. According to the embodiment, the reference signal of the input _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ In the case of a recordable compact disc, the lengths L and ρ carried by the output signal 131 of the eight-tone four-degree county side ηο are usually not integer multiples of Τ. The computational model can also calculate the datasettypes of the datasettypes separately, and generate the marginal deviations of the bribes corresponding to these resources_(5)(4) rhyme edgedevia(5), where the data is related to the clock edge deviation by the data pair The output of the clock edge deviation meter 144 is contained in the output signal 145, and the f-set includes at least one pit plane pit (pit_land_pit) f_ plane or at least - plane pit plane (f) d-_and) data set type. Each type of f set corresponds to at least one specific target pit length (for example: 3T, 4T, ..., 11τ) and - a specific target plane length (for example: 31^, ..., called and another - specific A combination of target pit lengths (eg: 3Τ, 4Τ, ..., 11Τ), or at least—a mosquito target plane length disc of at least one 12 200818152 combination of a specific target pit length and another specific target plane length. The classifier 142 classifies the plurality of data sets into the data set types. In this embodiment, 'each data set, ie (P1, L, P2) or (u, p, L2), contains two long and medium The set of materials (P1, L, P2) refers to the length corresponding to a plane and two μ of the adjeedent pits respectively, and the set of materials (L1, p, L2) corresponds to a pit. The length is the length of two planes adjacent to each other. Here, the labeling method such as (Ρητ, LmT, P1T) or (LnT, pmT, LlT) is used to indicate the above data set type, where nT, mT or It is expressed in units of the clock period τ; in the present embodiment, 'n = 3, 4, ..., or 11, m = 3, 4, ..., or U, and 1 = 3, 4, , or 11. Each of the data set types (LnT, PmT Lit), for example: n = (10), with = Qin and 1 1〇 data set type (Ln〇*T, PmO*T, L1()*T) is used for the plane corresponding to the target plane length nO*T, the pit after the compact_plane and the target pit length is mO*T, and the engraved plane after the compact_pit A plurality of data sets (L1, P, L2) of a plane of length 1 (four) are classified. Similarly, each of the data set faces & LmT, P1T), for example: n = n〇, m = m〇, And! = 1〇The data set type (Ρη〇*τ, Lm〇*T, Pl〇*T) is used to classify the pit corresponding to the target pit length η〇*τ, and follow the pit After that, the target plane length is the secret plane, and the data set (PI, L P2) with the pit length of 1Q* after the compact k A plane. It should be noted that each of the plural fabric set types (LnT, PmT, L1T), for example, the greedy set type (1^,1>111(^,]:1()叮)' corresponds to The combination of a specific target plane length n〇*T, a specific target pit length mG*T, and a specific target plane length 1 (four) (n〇*T, m〇*T, 10*T). Multiple data set types I LmT Each of (5) 13 200818152, for example: data set type (Ρη〇*τ, WtIVt), corresponds to a specific mesh pit length η〇*Τ, a specific target plane length m〇*T, and a specific target The combination of pit lengths (n〇*T, m〇*T, 1〇*τ). Since (four) and ! each have nine possible values to 11), so for the dataset type (LnT, PmT, In the case of L1T), there are (9 * 9 * 9) combinations, and for the type of bait collection (PnT, LmT, PlT), there are also (9 * 9 * 9) combinations, so the total number of collection types is ( 9 * 9 * 9 * 2) = 1458. , 'On the implementation can choose a certain combination; the written strategy is adjusted according to these selected combinations, rather than adjusting according to all possible combinations. In addition, if 'pour _, Ρ, growth degree; U, p, and (4) are the following conditions, then the model attachment 142 can classify the shrink cap L1, P, L2) into the data set type 0^11〇*1 ^,卩111〇*7',11()*丁)·(n0 - 0.5) * T <L1 <(n〇+ 0.5) * T ; (m0 - 〇·5) * Τ 立 gmO + 0.5 ) * Τ ; and (10 — 0·5) * Τ Red 2 410 + 0·5) * Τ. Similarly, if the lengths (1) and (4) of the data set (P1, L, P2) are as follows, then the type II (4) can be classified into the data set type (PnO*T, Lm〇*T, Ρΐ〇*τ) · (ηΟ 0.5) * Τ <Ρ1 <(η〇+ 0.5) * Τ ; 14 200818152 (m〇— Ο·5) * T red^(m0 + 0·5) * τ ; And (10 - 0·5) * Τ points 2 or 10 + 0·5) * Τ. The data vs. clock edge deviation calculator Μ4 can be as good as the data set type. The data is calculated for the Japanese machine edge deviation n 144 to calculate the length of the clock edge = length (data-to_clockedgelength), where each data for the clock edge long family is the above reference signal (in this implementation, the eight-turn ten (4) The spacing between the rising edge or falling edge of the ^CLK) and the transition edge ^ansitiG'edge of the switching signal ΐΐ5. Further, the data-to-clock edge deviation calculator 144 calculates a plurality of difference values' to generate data-to-clock edge deviations corresponding to the respective data set types PmT'L1T) and (PnT, LmT, PlT), respectively. Each of the above differences is the difference between the data-to-clock edge length and the -target (four)-to-clock edge length, where the target data versus clock edge length corresponds to a particular data set type (Ln〇). *T, PmG*T, L1G*T) or a predetermined value of (PnG*T, LmG*T, p1()*T). The case shown in Fig. 2 will be further described as an example. Fig. 2 is a view showing the length compensation of the clock edge deviation using the data according to the first embodiment, wherein the pit A corresponds to the target length (i.e., the target length of the pit is 4). As shown in Figure 2, Ttopr and Tlast represent the write strategy parameters used to control the start and end positions of the pit, respectively. In the case shown in Fig. 2, the pit a is regarded as the pit PT, and its end position is not perfect in this case; in other words, the edge deviation can be measured at the end position of the pit ρ. The previous plane (previous 15 200818152 land) PL before the pit ρ has a target length W (i.e., the target length of the plane is 4 Τ), and the previous pit before the previous plane PL has a target length Ρ3 Τ. In addition, the subsequent plane (f〇ll〇wing iand) FL located after the pit 有 has a target length l5T, and the subsequent pit (f〇n〇wing pit) FP located after the posterior plane FL has a target length Ρπ. Here, the patterns (PP, PL, PT) and (PT, FL, FP) correspond to at least one pit plane pit data set type described above, and the patterns (PL, ΡΤ, FL) correspond to at least the above A planar pit planar dataset type. Calculated according to the type of data set performed by the calculation module 140 and corresponding to the patterns (pp, pL, pT), (pL, PT, FL), and (PT, FL, FP) (especially for statistical purposes) Calculation) 'Statistical results can be extracted to indicate that one of the patterns (pp, PL, PT), (pL, pT, FL), and (PT, FL, FP) dominates the imperfect end of the pit ρτ The phenomenon, for example, the pattern (PT, FL, FP) dominates the imperfect end of the pit ρτ, then the corresponding strategy parameter Tlast(n, m, 丨) is written (the representative is used to control the corresponding The write strategy of the end position of the pit of the negative material set type (PnT, LmT, P1T) should be fine-tuned to adjust the imperfect end of the pit, according to the situation shown in Figure 2, (n,m,1) = (4, 5, 4). And the thinking/mining 疋: the bait-to-clock edge deviation calculator 144 can calculate the data pair clock edge length dl. In this implementation, the length dl of the dragon_pulse sister is the time point D of the falling edge of the cutting = number 115 (ie, the point corresponding to the ending position of the pit A), from the eight to fourteen body clock CLK towel The spacing between the rising edges (points d). Please note that the time point D is essentially the same-time point when the value of the RF signal 113 crosses (bribe) a certain value (for example, the value of the slitting level of 11114). The _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The same method can be applied to calculate the corresponding data class ‘ ^ ^ 尔 对 时 时 。 。 。 。 。. In addition, Wei is poor on the edge of the clock edge 144. In the L-Financing, the material-to-clock edge deviation calculator (4) borrows the analysis of the difference between the specific: #料(四) type, and produces the data of the enchant type to the edge deviation of the clock. Statistical analysis can be done by == different for these differences, or by finding the mode of these differences (m〇stfrequentvalue). The t of the above differences may be (10) - the difference between the data edge length and the target data versus the clock edge length (for example, 〇 5T in the case shown in Figure 2). Because the green of the pits is placed in the green hole, it should be time _~ in the set _ (P4T, L5T, Ρ 4τ) target (10), the length of the miscellaneous edge is Ιλ5 □ 0 ... The case of the figure is taken as an example for further explanation. Fig. 3 is a schematic view showing the length compensation of the clock-in-phase hybrid yarn according to the present invention, wherein the pit b also corresponds to the target length center. In the case shown in Fig. 3, the pit B surface such as the pit PT' ah ρτ_ flank gift is not perfect; for the Lu, the edge deviation can be measured on the open pit. Similarly, according to the calculation of 140, and corresponding to the type (pp, PL, 17 200818152 ΡΤ), (PL, PT, FL), and (PT, FL, FP) data set type calculation, statistics The result can be captured to indicate that one of the patterns (PP, PL, PT), (PL, PT, FL), and (PT, FL, Fp) dominates the imperfect start of the pit PT, for example The pattern (ρρ,·ρ^ ΡΤ) dominates the phenomenon of the imperfect start of the pit, and the corresponding strategy is written to the number of the tops (n, m, 1) (the representative is used to control the corresponding data The set type (PnT, the write strategy parameter of the start position of the pit) should be fine-tuned to adjust the pit ρτ< not to start with the US, where (n, m, 1) = (3, 4, 4) It should be noted that the data is calculated by the clock edge deviation calculator 144 for the clock edge length d2. In this embodiment, the data is the slit length signal II5. The time point E of the rising edge (ie, the time point corresponding to the start position of the pit B) and the rising edge of the eight-turn fourteen-modulation data clock CLK+ : The distance between the point in time e). Note that the time point e is essentially the same-time point when the value of the RF signal 113 crosses a predetermined value (for example, the value corresponding to the object of the divider ι 4). In the face of the clock edge deviation calculation cry 144 = the side of the side cut signal Π 5 level from the low to high transition to detect the time (four). The same method can be accrued to (4) fine = _ value tenderness defeat __ plant edge deviation 4 each of the difference is "the difference between the data edge of the clock edge and the edge of the clock edge" value. In Fig. 3, the time point at which the heart beats at the start position of the pit B is in the ideal shape = because the corresponding corresponds to __(~, L4T, P4 paste, and the edge length of the clock edge Department 18 200818152
被決定為0.5T 而要〜的疋因為從貧料對時脈邊緣 寫入脈衝控制器150之訊於是齡㈣^ 144傅輸至 八魅哭m A 叙數位的，所咕有需要，型樣依附 ^生之分類資訊可通過料對時脈邊緣偏差計算器 被傳达至寫入脈衝控制器15〇。相似地為 器142傳輸至資料對時脈邊 巧处樣依附刀類 因此若有1之訊號是數位的， 口此右有而要’八轉十四調變長度偵測器咖 可通過型樣依附_ 142被傳 2生之彻]、、、°果 ^ ^ 中寫入脈衝控制器150可通過直 連接而耦接至型樣依附分類器142’且資 器144也可通過直接連接而輪 谓遺緣偏差打 接至八轉十四調變長度偵測器130。 —於第-實施例之-變化例中，被用於計算差值以產生對應於特It is decided to be 0.5T and it is ~ because 从 写入 对 对 对 对 对 对 对 对 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 写入 写入 写入 写入 写入 写入 写入 写入 写入 写入 写入 144 144 144 144 144 144 144 144 144 144 144 The classification information of the attachment can be communicated to the write pulse controller 15 by the feed-to-clock edge deviation calculator. Similarly, the device 142 transmits the data to the clock to the edge of the knife. Therefore, if the signal of 1 is digital, the mouth has the right and the 'eight to fourteen modulation length detector can pass the pattern. Attached _ 142 is transmitted 2, 、, ° ° ^ ^ The write pulse controller 150 can be coupled to the model dependent classifier 142' by a straight connection and the 144 can also be connected by direct connection The margin deviation is connected to the eight-to-fourteen modulation length detector 130. - in the variant of the first embodiment - is used to calculate the difference to produce a corresponding
Pm〇" 科對《邊緣長度，可以是對應於特定資料細 W)的複數個資料對時脈邊緣 "η°τ’ m0*T， 算差值以產生騎於特行。她地，被用於計 時脈邊緣偏紅目鮮簡s_Ur 、则料對 料隼類碰L P 長度，Μ是軸於特定資 值。、μ 的複數個資料對時脈邊緣長度之平均 時脈邊緣偏差之目標資料對時脈邊j :T，LmQ*T，p_)的資料對 啊脈姐長度，可以是對應於特定資 19 200818152 料集類型(LnG*T，Pm㈣，Lig*t)的彳政個⑽對時崎緣長度以及對 應於特料細麵(P，，LmG*T，Pig*t)_跋個資料對時脈邊緣 長度之平均值。 、 需要注意的是，上述寫入策略參數，例如：Tt〇pr(n，m，麟 Tlast(n，m，l)，皆可以被自動地微調，這是由於本發明不再需要一 些特定裝置（例如：上述之示波||)。另外，在沒有外部裝置的協 助下，寫入脈#ί控制器15〇可依據計算模組14〇所產生之資料對 日守脈邊緣偏差來微調寫入策略參數，所以依據本發明，寫入策略 參數可於綠或晶>}上被自動地微調。通雜據轉對時脈邊緣 偏差來微調寫入策略參數，促使對應於被寫到光學儲存媒體1〇2 上的最新資料（其係利用最近更新的寫入策略參數而寫上的）之 凹坑長度或平面長度可接近或達到Τ的目標倍數。 第4圖為依據本發明一實施例用來微調光學儲存裝置的寫入 策略參數之方法910的流程圖。方法910於步驟91〇s開始並可通 過第1圖所不之糸統100C來實施。 於步驟912中，在光學儲存裝置100中之微處理單元 (micm-pr〇CeSSingimit，MI>U)所執行之韌體碼的控制下，光學儲 存裝置100利用對應於光學儲存裝置1〇〇之特定轉速之寫入策略 參數的初始值，將資料寫到光學儲存媒體1〇2上。 20 200818152 於步=914中，光學儲存裝置1〇〇讀取被寫到光學儲存媒 102上之資料以產生分切訊號115。 ’、一 ;Y驟916巾’系統聰之八轉十四調變長度偵測器Μ 由1測分切訊號115來偵測凹坑之長度p與平面之長度L。曰 :驟918巾，計魏组14_對應靖彻貞型⑹，b =、（PnT U Μ之資料對時脈邊緣偏差，其資料集類型如前面 H 少—凹坑平面凹坑資料集_或至少—平面凹坑平 面資料集_，其中於本_中，η=、… 凡千 或11 於步驟920中，執行拿刀體碼之微處 ==碼^處理單元決定需要微調寫入策略參數，㈣微 入步驟92“㈣2 職處理單元可決定直接進 所轳、十、 仃竭920之檢查。雖然如步驟920盥922中 麟林發批限定。如果 數，則步驟_ 922可被描述 ^^參 類似的選擇-個或複數個寫 略參數型式。 明將不再贅述。 3㈣參數型式的貫施選擇之重複說 21 200818152 若執行勒體碼之微處理單元決定進入步驟92 如前面所述，利用資料對時脈邊緣偏差來微調寫人策略=露 於步驟924中，於執行韌體碼之微處理 存裝置微調後的寫人策略錢值（㈣t 寫入策略參數值)，將資料寫到先學儲=::驟922後之 4注意’依據本發明應用於 _睛規格之數”樣 類型⑹，〜丄)與叫，L·,，Ρ,τ)之總數量可通過:取得貝科集 2000 ; 10 *10 木 10 木 十，η 了Γ對_ R規袼或DVD+R規格之數位多樣化光碟而 二4 4、…、n、或 1“ = 3、4、…、m14，且“ d 4、…、1卜或14。 。周正’而不疋依據所有可能的組合來調整。 分別:二:::她务明不同的實施例藉由利用長度偏差統計來 、、、朿略參數的示意圖，其中第5圖所示之寫入策略參 22 200818152 數可應用於寫入DVD-R規格之光碟片，且於第5圖中，用於多脈 衝之寫入策略的寫人策略參數與用於單—脈衝之—寫人策略的寫 入策略參數係分別以理想串列數位訊號（㈣seHai咖如 )來絲。寫人策略參數細卜TtQp2、τι^、τμ2、 Ttopr、Todf、Todr與Tlast分別對應於某些邊緣延遲（或邊緣位 移）’且寫入策略參數Τηφ則對應於某一脈衝寬度。另外，寫入策 略參數，例如第5圖所示之過驅動功率（overdry，〇d power)、寫入功率(write _r)、以及偏壓功率卿^係 分別對應於某些功率位準。 第6圖為依據本發明不同的實施例藉由利用長度偏差統計來 分別微調寫入策略參數的示意圖，其中第6圖所示之寫入策略來 數可應寫人議.規格光則，朋於第—寫人策略（即第6 圖中所不之寫入策略1 )的寫入策略參數與用於第二寫入策 略(即第6圖中所示之”寫入策略2”）的寫入策略參數係分別以 理想串列數細絲示於第6圖巾。寫人策略參數卿、喊、 Tlast卜T!ast2、與Tc(X)i分別對應於某些邊緣延遲（或邊緣位 移），且寫入策略參數Tmp則對應於某一脈衝寬度。另外，寫入 策略參數’例如第6 _示之以择抹除轉（erasep〇wer)、 與偏壓功率分別對應於某些功率位準。 第7圖為依據本發明之第二實施例之帛來微調光學儲存裝置 200之寫人策略參數之系統觀的示意圖。本實施例與第-實施 23 200818152 例相似’其差舰明如下。於第二實關巾，輸人至八轉十四調 變長度偵測器130之參考訊號係為振盪器220所產生之參考時: CLK2 °參考時脈CLK2的頻率並不需要與八轉十四調變資料時脈 CLK的頻率相等。 、 第8圖為依據本發明之第三實關之用來微調光學儲存裳置 300之寫入策略參數之系統3qqc的示意圖。本實施例與第一實施 例相似，其差異說明如下。系統3GGC包含取樣電路（sampling circuit)，輕接於波形等化器112以接收重建訊號（例如：射頻 訊號113)。取樣電路係被驗轉重建訊號以產生數位訊號，·於 本貫施例中，數位訊號係為數位射頻訊號315。如第8圖所示，取 樣電路包含舰數_換器（analGg—tQ—咖加⑽鮮細， ADC)314與鎖相迴路32〇。類比數位轉換器314依據參考時脈 對賴《 113進行類比數位轉換，以產生數位射頻訊號315，且 鎖相迴路320依據數位射頻訊號315來產生參考時脈ακ3。 系統300C更包含八轉十四調變長度偵測器33〇、計算模組 340、與寫入脈衝控制器35〇，其中計算模組謂包含型樣依附分 類器342射料對時脈邊緣偏差計算器344。在此，被使用於偵測 長度的訊號係為數位射頻訊號315，而非分切訊號115。入轉十四 週雙長度制器33G藉由觀測（observe)數位射頻訊號315之值 來偵測時間點之間的間距，並產生間距的長度，其巾每—間距係 對應於-個凹坑或-個平面。這些間距的邊界可藉由預定值來決 24 200818152 定；預定值可以是數位射頻訊號315所載之最大值與最小值之間 2間值’例如·最大值與最小值之平均值。中雖相當於前面 各實施例所述之分切訊號。 =圖為於重建訊號（例如：射頻訊號113)之複數個取樣點 乂㊉」的记號來標示）的示意圖，其中特定取樣點的值 與預定值（例如：上述巾p卩枯、卩q 」 上述中間值）之間的差值d3可作為用來指出資 料對時脈邊緣偏差d4的指標（indiGatiGn)。依據第9圖所示之 射頻訊號的波形’大部分跨顧纽之取樣財都會完美Pm〇" Section of the "edge length, which can correspond to a specific data fine W" to the clock edge "η°τ' m0*T, the difference is calculated to produce a special line. She is used to calculate the s_Ur of the edge of the clock, which is the length of the material, and the axis is the specific value. , μ of a plurality of data on the average edge of the clock edge length of the target data on the clock edge j: T, LmQ * T, p_) data pair y pulse length, may correspond to a specific capital 19 200818152 The type of material set (LnG*T, Pm (4), Lig*t) is the length of the time and the corresponding fine surface (P,, LmG*T, Pig*t) The average of the length of the edge. It should be noted that the above write strategy parameters, for example: Tt〇pr(n, m, Lin Tlast(n, m, l), can be automatically fine-tuned, because the present invention no longer needs some specific devices. (For example: the above oscilloscope||). In addition, without the assistance of an external device, the write pulse #ί controller 15〇 can fine-tune the write to the edge of the pulse according to the data generated by the calculation module 14〇 Into the policy parameters, so according to the present invention, the write strategy parameters can be automatically fine-tuned on the green or crystal>}. The data is fine-tuned to the clock edge deviation to fine-tune the write strategy parameters, causing the corresponding to be written to the optical The pit length or plane length of the latest material on the storage medium 1 2 (which is written using the recently updated write strategy parameters) may be close to or reach the target multiple of Τ. FIG. 4 is an embodiment of the present invention. A flow chart of a method 910 for fine-tuning a write strategy parameter of an optical storage device. Method 910 begins at step 91 〇s and can be implemented by the system 100C of Figure 1. In step 912, in optical storage Microprocessing unit in device 100 (m Under the control of the firmware code executed by icm-pr〇CeSSingimit, MI>U), the optical storage device 100 writes the data to the optical using the initial value of the write strategy parameter corresponding to the specific rotational speed of the optical storage device 1〇〇. The storage medium is 1 〇 2. 20 200818152 In step = 914, the optical storage device 1 reads the data written on the optical storage medium 102 to generate a slitting signal 115. ', one; Y 916 towel' system Congzhi's eight-to-fourteen-modulation length detector Μ The length of the pit p and the length of the plane L are detected by the 1-point cut signal 115. 曰: 918 巾 towel, counting Wei group 14_ corresponding to Jing Che-type (6), b =, (PnT U Μ data on the clock edge deviation, its data set type such as the front H less - pit plane pit data set _ or at least - plane pit plane data set _, which in this _ , η =, ... Where the thousand or 11 in step 920, the implementation of the body code of the knife body == code ^ processing unit determines the need to fine-tune the write strategy parameters, (d) micro-in step 92 "(4) 2 job processing unit can decide to directly enter Execution of the 920, 十, exhausted 920. Although as in step 920 盥 922 Lin Lin issued a batch of restrictions. For the number, the step _ 922 can be described as ^^ reference to a similar selection - or a plurality of write parameters. The description will not be repeated. 3 (4) Repeat the description of the choice of the parameter type 21 200818152 If the implementation of the in vivo code The microprocessor unit determines to proceed to step 92, as described above, using the data to fine-tune the write-off strategy for the clock edge deviation = as shown in step 924, after performing the fine-tuning of the firmware of the firmware code, the write strategy value ( (4) t write strategy parameter value), write the data to the first learn storage =:: after the 922, note 4 'apply to the number of _ eye specifications according to the invention' (6), ~丄) and call, L·,, The total number of Ρ, τ) can be obtained by: obtaining Beco set 2000; 10 *10 wood 10 wood ten, η Γ Γ R R R or + DVD+R specifications of digital diversified discs and two 4 4,...,n , or 1 " = 3, 4, ..., m14, and "d 4, ..., 1 Bu or 14. . Zhou Zheng’s adjustments are based on all possible combinations. Respectively: 2::: She clarifies that the different embodiments use the length deviation statistics to summarize the parameters of the parameters. The write strategy shown in Figure 5 can be applied to write DVD- R-format optical disc, and in Figure 5, the write strategy parameter for the multi-pulse write strategy and the write strategy parameter for the single-pulse-write strategy are respectively ideal serial digital signals ((4) seHai coffee, such as) come to silk. The write strategy parameters TtQp2, τι^, τμ2, Ttopr, Todf, Todr, and Tlast correspond to some edge delays (or edge shifts) respectively, and the write strategy parameter Τηφ corresponds to a certain pulse width. In addition, write strategy parameters, such as overdry (〇d power), write power (write _r), and bias power, as shown in Figure 5, correspond to certain power levels, respectively. Figure 6 is a schematic diagram of fine-tuning the write strategy parameters by using the length deviation statistics according to different embodiments of the present invention, wherein the number of write strategies shown in Figure 6 can be written. The specification light is The write strategy parameter of the first-write strategy (ie, write strategy 1 in Figure 6) and the write strategy for the second write strategy (ie, write strategy 2 shown in Figure 6) The input strategy parameters are shown in the sixth figure with the ideal number of filaments. The write strategy parameters Qing, Shout, Tlast, T!ast2, and Tc(X)i correspond to some edge delays (or edge shifts), respectively, and the write strategy parameter Tmp corresponds to a certain pulse width. In addition, the write strategy parameter 'e.g., the sixth is shown as an erasep〇, and the bias power corresponds to some power level, respectively. Figure 7 is a schematic illustration of a system view of a human strategy parameter for fine-tuning an optical storage device 200 in accordance with a second embodiment of the present invention. This embodiment is similar to the first embodiment 23 200818152. In the second real cleaning towel, the reference signal input to the eight-to-fourteen-modulation length detector 130 is the reference generated by the oscillator 220: CLK2 ° reference clock CLK2 frequency does not need to be eight to ten The frequency of the four modulation data clocks CLK is equal. Figure 8 is a schematic diagram of a system 3qqc for fine-tuning the write strategy parameters of the optical storage shelf 300 in accordance with the third embodiment of the present invention. This embodiment is similar to the first embodiment, and the differences are explained below. The system 3GGC includes a sampling circuit that is coupled to the waveform equalizer 112 to receive the reconstructed signal (e.g., RF signal 113). The sampling circuit is inspected and reconstructed to generate a digital signal. In the present embodiment, the digital signal is a digital RF signal 315. As shown in Fig. 8, the sampling circuit includes a ship number_changer (analGg-tQ-Caga (10) fresh, ADC) 314 and a phase locked loop 32A. The analog-to-digital converter 314 performs analog-to-digital conversion on the reference clock according to the reference clock to generate the digital RF signal 315, and the phase-locked loop 320 generates the reference clock ακ3 according to the digital RF signal 315. The system 300C further includes an eight-to-fourteen-modulation length detector 33〇, a calculation module 340, and a write pulse controller 35〇, wherein the calculation module includes a sample-dependent classifier 342 to shoot the clock edge deviation Calculator 344. Here, the signal used for detecting the length is the digital RF signal 315 instead of the divisor signal 115. The four-week dual-length controller 33G detects the spacing between the time points by observing the value of the digital RF signal 315, and generates the length of the spacing, and the spacing of the towels corresponds to a pit. Or - a plane. The boundaries of these spacings may be determined by a predetermined value; the predetermined value may be the value between the maximum and minimum values carried by the digital RF signal 315, such as the average of the maximum and minimum values. Although it is equivalent to the slitting signal described in the previous embodiments. = The figure is a schematic diagram of the number of sampling points of the reconstruction signal (for example: RF signal 113), wherein the value of the specific sampling point is a predetermined value (for example, the above-mentioned towel is 卩, 卩q The difference d3 between the above intermediate values can be used as an index (indiGatiGn) for indicating the data to the clock edge deviation d4. According to the waveform of the RF signal shown in Figure 9, most of the sample money will be perfect.
資料時脈的下降邊緣’所以大部分資料對時脈邊緣 、，、、、令上述特定取樣點的值是指於特定取樣時被 讀值，而該數值係由數位射頻訊號315所载。取樣 J 装r/Uf (例如.上述差值d3)可表示資料對時脈邊緣偏 料對時脈邊緣偏差。計算模組34G可藉由計算預定^^與: 述之中間值）以及當數位射頻訊號315之值跨越預定值間 在此，型樣依附分類請2與型樣依附分類器142 功能’而八針，細測謂射輸㈣輪出訊: 所載之長度L與P ’其中輸出訊號331係與輸出訊號1 : 實施例之貧料對時脈邊緣偏差計算器344藉由使用上述之接义近^ 25 200818152 直線的關係，來計算資料對時脈邊緣·。另外 益350具有與寫入脈衝控制器15〇相同的功能，而計算模組_ 則可輸出由輸出訊號345所載之資料對_邊緣偏差，其中輪出 訊號345係與輸出訊號145相似。 第10圖為第8圖所示之實施例之一變化例的示意圖，其中使 用了内插器（interpolator) 416 於類比數位轉換器犯與 鎖相迴路320之間。鎖相迴路320依據内插器416所產生之内插 訊號（interpolated signal) 417來產生參考訊號⑶以，而内插 器416則依據數位射頻訊號315與參考訊號CLK4來進行内插運算 (interpolation operation)。於本變化例中，八轉十四調變長 度偵測器330之輸入可被内插訊號417所替換。内插器416之運 作原理係為熟悉此技藝者所知悉，故不在此贅述其細節。 第11圖為依據本發明一實施例之統計計算之詳細實施的方法 930的流程圖，其中第11圖所示之詳細實施的方法開始於步驟 930S，並可被應用於第4圖所示之實施例，尤其是步驟918、920、 與 922。 於步驟932中，進行對應於型樣(PP，PL，PT)、（PL，PT，F1)、 及/或(PT，FL，FP)的複數個資料集類型之統計計算，並取得資 料對時脈邊緣偏差大於門檻值Th—A(例如：Th-A = 0· 3T)之資料集 類型。 26 200818152 、於步驟934中，於資料對時脈邊緣偏差大於門檻值Th_A之資 料^型#中’取得資料集發生次數多於Th_B (即門健Th』The falling edge of the data clock 'so the majority of the data for the edge of the clock, , , , , and the value of the particular sampling point described above refers to the value read at the time of the particular sampling, and the value is contained by the digital RF signal 315. Sampling J with r/Uf (for example, the difference d3 above) can indicate the deviation of the data from the edge of the clock edge to the edge of the clock. The calculation module 34G can calculate the predetermined ^^ and: the intermediate value) and when the value of the digital RF signal 315 crosses the predetermined value, the model is attached to the classification 2 and the model is attached to the classifier 142 function 'eight Needle, fine measurement of the transmission (four) round of the message: the length L and P ' contained therein, the output signal 331 and the output signal 1: the poor balance of the embodiment of the clock edge deviation calculator 344 by using the above-mentioned Near ^ 25 200818152 The relationship between the lines, to calculate the data on the edge of the clock. In addition, the benefit 350 has the same function as the write pulse controller 15A, and the calculation module_ can output the data pair_edge deviation carried by the output signal 345, wherein the round signal 345 is similar to the output signal 145. Figure 10 is a schematic illustration of a variation of the embodiment shown in Figure 8, wherein an interpolator 416 is used between the analog-to-digital converter and the phase-locked loop 320. The phase-locked loop 320 generates the reference signal (3) according to the interpolated signal 417 generated by the interpolator 416, and the interpolator 416 performs the interpolation operation according to the digital RF signal 315 and the reference signal CLK4. ). In the present variation, the input of the eight-to-fourteen-modulation length detector 330 can be replaced by the interpolated signal 417. The operation of the interpolator 416 is known to those skilled in the art and the details thereof are not described herein. 11 is a flow diagram of a method 930 of a detailed implementation of statistical calculations in accordance with an embodiment of the present invention, wherein the method of the detailed implementation illustrated in FIG. 11 begins at step 930S and can be applied to FIG. Embodiments, particularly steps 918, 920, and 922. In step 932, statistical calculations are performed on a plurality of data set types corresponding to patterns (PP, PL, PT), (PL, PT, F1), and/or (PT, FL, FP), and data pairs are obtained. The clock edge deviation is greater than the data set type of the threshold value Th—A (for example, Th-A = 0·3T). 26 200818152 In step 934, the number of occurrences of the data set is greater than Th_B (ie, the gate is Th) in the data of the data edge deviation greater than the threshold value Th_A.
所心疋之錄）之資料翻型，其巾服值Th—B係為正整數 如：Th β 二哪。 J 於步驟936中，計算對應於資料集發生次數多於Th—B之資料 集類型之寫人策略參數的調整量。並於步驟93GE中結束。 〜而要主思的疋’依據本發明之某些實施例，可進行統計計算以 、疋對應長度L及/或p的分佈曲線。分佈曲線之某些特徵的資 訊，例如：分佈曲線的形狀、半高寬（_ -鲥⑽让）、及 被門植值Th_A肖Th—B所蝴之區域，可祕決定是賴調寫入 策略參數以及決定寫入策略參數之調整量。 第12圖與第13圖為依據本發明之一實施例之分別於微調寫入 策略參數之前與之後的取樣數（sample _t)相對於資料對時 脈邊緣長度之曲線示意圖。如第12圖所示，關於對應於(ρτ，孔） <4T，5T)之組合「PT(4T) + FL(5T)」，其曲線以某一值為中心（例 如：零）。另外，對應於(PL，PT，FL) = (3T，4T，5T)之組合「PL(3T) + PT(4T) + F1(5T)」之曲線係以該某一值減掉S2為中心，而對 應於(PL，PT，FL) = (4T，4T，5T)之組合「pl(4T) + ρτ(4Τ) + FL(5T)」之曲線係以該某一值加上si為中心，且對應於(ρ[ ρτ 27 200818152 FL) = (5Τ，4Τ, 5T)之組合「PL(5T) + ρτ(4Τ) + FL(5T)」之曲 線亦以某一值為中心。因此，依據本實施例，微調寫入策略參數， 當如第12圖所示之三個較低曲線被集中於一處以使它們以同一值 或彼此相近的值為中心’則可窄化（narr〇w)如第12圖所示之較 咼曲線。因此，如第13圖所示差距S1與從被縮小了。 需要注意的;％，本發明可藉由使用具有多個元件組合而成的硬 體、或藉由使用執行軟體或軔體程式之電腦來實施。另外，於申 請專利範圍或上述說明中所揭露的系統元件當中，有些元件可藉 由使用同一硬體或軟體裝置來實現。 以上所述僅為本發明之較佳實施例，凡依本發明申請專利範 圍所做之鱗變化與修飾，f應屬本發明之涵蓋範圍。 【圖式簡單說明】 調光學儲存裝置之寫入策 料對時脈邊緣偏差來進行 第1圖為依據本發明一實施例之用來微 略參數之系統的方塊圖。 第2圖為依據本發明一實施例之利用資 長度補償的示意圖。 資料對時脈邊緣偏差來進行 苐3圖為依據本發明一實施例之利用 長度補償的示意圖。 第4圖為依據糊-實微_____ 28 200818152 略參數之方法的流程圖。 第5圖為依據本發明不同的實施例之藉由利用長度偏差統計來分 別微調寫入策略參數的示意圖。 第6圖為依據本發明不同的實施例之藉由利用長度偏差統計來分 別微調寫入策略參數的示意圖。 第7圖為依據本發明一實施例之用來微調光學儲存裝置之寫入策 略參數之系統的示意圖。 第8圖為依據本發明一實施例之用來微調光學儲存裝置之寫入策 略參數之系統的示意圖。 第9圖為於重建訊號上之複數個取樣點的示意圖。 第10圖為第8圖所示之實施例之一變化例的示意圖。 第11圖為依據本發明一實施例之統計計算之詳細實施的流程圖。 第12圖為依據本發明一實施例之於微調寫入策略參數之前的取樣 數相對於負料對時脈邊緣長度之曲線的示意圖。 第13圖為依據本發明一實施例之於微調寫入策略參數之後的取樣 婁丈相對於資料對時脈邊緣長度之曲線的示意圖。 【主要元件符號說明】 W^〇〇73〇MOO Τ〇2^ 光學儲存裝置 系統 ^¥儲存媒體 光學讀取頭 29 200818152 112 波形等化器 114 分切器 120, 320 鎖相迴路 130, 330 八轉十四調變長度债測器 140, 340 計算模組 142,342 型樣依附分類器 144, 344 資料對時脈邊緣偏差計算器 150,350 寫入脈衝控制器 160 調變器 162 寫入脈衝產生器 164 發射源驅動器 220 振盪器 314 類比數位轉換器 416 内插器 111， 113, 115, 131， 143, 145， 151，161，163, 165, 315, 331， 343, 345, 351 訊號 CLK，CLK2, CLK3, CLK4 時脈 d, e, D, E, Do, Eo 時間點 dl，d2 資料對時脈邊緣長度 d3 差值 d4 資料對時脈邊緣偏差 30 200818152The data of the heart of the record), the towel value Th-B is a positive integer such as: Th β two. J In step 936, the amount of adjustment of the writer policy parameter corresponding to the data set type of the data set occurrence number more than Th-B is calculated. And ends in step 93GE. 〜 主 主 疋 疋 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 依据 。 。 。 。 。 。 。 。 。 The information of some characteristics of the distribution curve, such as the shape of the distribution curve, the full width at half maximum (_ - 鲥 (10) let), and the area of the threshold value of the threshold value Th_A Xiao Th-B, the secret decision is the write Policy parameters and the amount of adjustment that determines the write policy parameters. Fig. 12 and Fig. 13 are graphs showing the number of samples (sample _t) before and after the fine tuning of the write strategy parameters, respectively, with respect to the data versus the edge length of the clock, in accordance with an embodiment of the present invention. As shown in Fig. 12, regarding the combination "PT(4T) + FL(5T)" corresponding to (ρτ, hole) <4T, 5T), the curve is centered on a certain value (for example, zero). In addition, the curve corresponding to the combination of (PL, PT, FL) = (3T, 4T, 5T) "PL(3T) + PT(4T) + F1(5T)" is centered on the value minus S2. And the curve corresponding to the combination of (PL, PT, FL) = (4T, 4T, 5T) "pl(4T) + ρτ(4Τ) + FL(5T)" is centered on the value plus si And the curve corresponding to the combination of (ρ[ ρτ 27 200818152 FL) = (5Τ, 4Τ, 5T) "PL(5T) + ρτ(4Τ) + FL(5T)" is also centered on a certain value. Therefore, according to the present embodiment, the write strategy parameters are fine-tuned, and when the three lower curves as shown in Fig. 12 are concentrated in one place so that they are centered at the same value or close to each other, they can be narrowed (narr 〇w) A more detailed curve as shown in Figure 12. Therefore, as shown in Fig. 13, the gap S1 and the slave are reduced. It is to be noted that the present invention can be implemented by using a hardware having a combination of a plurality of elements or by using a computer executing a software or a program. In addition, some of the system components disclosed in the scope of the claims or the above description may be implemented by using the same hardware or software device. The above description is only the preferred embodiment of the present invention, and the scale changes and modifications made by the patent application scope of the present invention should be within the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS The writing of the optical storage device is performed on the clock edge deviation. Fig. 1 is a block diagram of a system for abbreviating parameters according to an embodiment of the present invention. Figure 2 is a schematic diagram of the use of length compensation in accordance with an embodiment of the present invention. The data is performed on the clock edge deviation. The Fig. 3 is a schematic diagram of the use of length compensation in accordance with an embodiment of the present invention. Figure 4 is a flow chart of the method based on the paste-real micro_____ 28 200818152 slightly parameter. Figure 5 is a schematic illustration of the fine tuning of write strategy parameters by utilizing length deviation statistics in accordance with various embodiments of the present invention. Figure 6 is a schematic diagram of the fine tuning of write strategy parameters by utilizing length deviation statistics in accordance with various embodiments of the present invention. Figure 7 is a schematic illustration of a system for fine tuning the write strategy parameters of an optical storage device in accordance with one embodiment of the present invention. Figure 8 is a schematic illustration of a system for fine tuning the write strategy parameters of an optical storage device in accordance with one embodiment of the present invention. Figure 9 is a schematic diagram of a plurality of sampling points on the reconstructed signal. Fig. 10 is a schematic view showing a modification of the embodiment shown in Fig. 8. Figure 11 is a flow chart showing a detailed implementation of statistical calculations in accordance with an embodiment of the present invention. Figure 12 is a graph showing the number of samples before the fine-tuning of the write strategy parameters versus the negative-to-clock edge length in accordance with an embodiment of the present invention. Figure 13 is a diagram showing the plot of the sampling versus the data versus the edge length of the clock after fine-tuning the write strategy parameters in accordance with one embodiment of the present invention. [Main component symbol description] W^〇〇73〇MOO Τ〇2^ Optical storage device system ^¥Storage media optical pickup 29 200818152 112 Waveform equalizer 114 Divider 120, 320 Phase-locked loop 130, 330 VIII Turn the fourteen variable length debt detector 140, 340 calculation module 142, 342 type dependent classifier 144, 344 data to the clock edge deviation calculator 150, 350 write pulse controller 160 modulator 162 write pulse generator 164 launch Source driver 220 oscillator 314 analog to digital converter 416 interpolator 111, 113, 115, 131, 143, 145, 151, 161, 163, 165, 315, 331, 343, 345, 351 signal CLK, CLK2, CLK3, CLK4 clock d, e, D, E, Do, Eo time point dl, d2 data vs. clock edge length d3 difference d4 data vs. clock edge deviation 30 200818152
Ttopr，Tlast，Todr，Todf，Ttopl， Ttop2, Tlastl，Tlast2, Tcool， 丁 mp 寫入策略參數 S1，S2 分佈曲線的中心值之間的差 距 910,930 方法 910S，910E，912〜924,930S， 930E，932 〜936 步驟 31Ttopr, Tlast, Todr, Todf, Ttopl, Ttop2, Tlastl, Tlast2, Tcool, Ding mp Write strategy parameter S1, S2 The difference between the center values of the distribution curve 910, 930 Method 910S, 910E, 912~924, 930S, 930E, 932 ~936 Step 31