• d 了 • d 了 經濟部智慧財產局員工消費合作社印製 A7 --- ---B7 五、發明説明（i ) 發明背景 ' 本發明係有闞如申請專利範圍第1項之前文所引用的方法。 Berlekamp et al之美國專利案號4 559 625及扪£111111 et al乙美國專利案號5 299,2〇8係揭露間條及錯誤防護資 Λ子的解碼處理，其中在第一字所發現的錯誤圖案會提供 線索，以定位在相同字群組中的另一字錯誤。該等參考係 使用標準化格式及在各種不同字上具有多符號錯誤猝發的 錯灰杈组。在特殊字中的錯誤發生機率較強烈，而錯誤係 發生在下一字或某些字上所指向的相對符號位置s此程序 時常會提高所更正的錯誤數目。本發明家已確認此原理的 問嘁.當線索芊已完全更正的時候，線索才可實現。 發明概要 結果，本發明的目的是要提供編碼格式，其中線索字會 以較目標字爲更確定的程度來解碼。因此，根據它其中的 一觀點，本發明的特徵如申請專利範園第1項之所描述特徵 部份，所發現的線索會造成或指向到抹除符號3隨著如此 的指向，錯誤更正便會以更強而有力處理。事實上，當沒 有錯誤位置知道的時候，許多碼便會更正最個錯誤。所 提供的抹除位I，通常’ e > t的許多抹除會予以更正。揉 除符號特徵以外的其它類型可實行。猝發和任意錯誤組合 的防護將也會改良，或者，抹除定位的提供只需使用較少 數目的錯誤符號，如此便可將計算簡化。原則上，本發明 可用在儲存環境及傳輸環境3 本發明也關於用以解碼如此編碼資訊之方法、編碼及/或 * 4' 本紙張尺度適用中國國家標準（CNS ) A4規格（210x^97公釐) --- —.--.-----种衣------ir------il (請先閱讀背面之注意事項再填寫本頁) 4257 7 3 Α7 Β7 經濟部智慧財產局員工涓費合作社印製 五、發明説明（ 解碼以上述方法所使用之裝置、及用以與如此编碼及/或解 碼所形成界面之資訊載體。本發明的進一步有利的觀點是 在申请專利範圍内引述。 圖式之簡單說明 本發明的這些及進一步觀點和優點隨後會在揭示較佳具 體實抱例及參考附圖中更詳細地討論，其中： 圖1是具有編碼器、載體、和解碼器的系統； 圖2碼格式原理； 圖3是乘積碼格式： 圖4是具有猝發偵測的長距離碼； 圖5是警戒碼和猝發指示器子碼： 圖6猝發指示器子碼格式； 圖7警戒碼及其乘積子碼； 圖8是各種不同的進—步觀點： 圖9是另一格式： 圖1 0是間條處理的細節3 較佳具體實施例之詳細説明 圖丨係根據發明而顯示廣泛的系統，其提供編碼器 '載韹 、和解碼器，眾具體實施例係用於編碼、儲存、及最後解 碼得自聲頻或影像信號或來自資料之取樣或多位元符號的 序列3終4機2 0係接收具有八位元大小的連續符號。分離 器22會反覆及週期性地將線索字所意欲的第—符號轉移至 编，器24。此外，分離器22會將所有其它的符號轉移至爲 m 6。在編碼器2 4中’該等線索字係籍著將所結合的資 本紙婊尺度適用中國國家標準（CNS ) Μ規格（2!〇>< 297公釐 . -^IT4a—--- (請先閲讀背面之注意事項再填寫本頁) A7 B7 425773 五、發明説明（ 料編碼成第一多符幹奴4 φ t α Β 姽巧误更正碼的碼字而形成。此碼可以 疋 Reed-Soloinon 石民 - . ， 馬、乘積碼'X錯碼、或上述的組人3 在編碼器2 6中，該答θβ — 一. ° ^目標子係精著編碼成第二多符號錯誤 更正碼的碼字而开{士. .. 而形成。在具體實施例中，所有的碼字具有 相同的長度，但這並韭3 " & 卫非疋咸格的需束。理想上，兩種碼會 士具=二碼之第—予碼的Reed_s〇i_…參考圖; 會更清楚，該等線索字大體上具有較大的錯誤防護程度， 而且相對包括較少的非多餘符號。 在方塊~ 8，如此所形成的該等碼字會轉移至其已指示亂 數的一或多個輸出，所以稍後會討論的媒體分配會變成相 同的。方塊3 0係表示接收編碼過資料的媒體本身3此事實 上是以適當的寫原理與媒體组合而寫入3或者，該媒體可 作爲來自諸如註記的原始編碼媒體的副本而實現s理想上 ’錯存是疋全連續的，但其它結構也可使用3在方塊3 2， 各種不同的字會再一次從媒體讀取。然後，第—碼的該等 線索字會傳送到解碼器3 4，並基於固有的重複性而解碼3 此外’於jk便圖2的討論會更明白，如此的解碼會在錯誤的 位I上出現線索，而不是在這些線索字上3方塊3 5會接收 這些線索，及包括使用一或多個不同策略的程式，用以锊 如此線素轉換成抹除定位，該等目標字是在解碼器3 6中解 碼2在抹除位置的控制下，該等目標字的錯誤防護會提高 到可接受的位準a最後，所有的解碼字會藉由符合於輸出 4 〇的最初格式之元件3 8而解多工。爲了簡化’用以形成技 此各種子.τ纟先界面的機械結構已劣略3 -6 本紙&尺度適用中國國家標準（CNS ) A4规格（2丨0X297公釐） . . 沖衣 I1T------^-- {諳先¾讀背面之注意事項再填寫本買) 經濟部智慧財產局員Η消f合作社印製 五 發明説明（ Α7 Β7 經濟部皙慧財雇忌員工浈f合作社印製 圖2係描述相對簡化的m政斗. . 匕的碼格式。如圖所示，該編碼資訊在 覜念上是配置在1 6列X # 。 J -糊付唬的方塊内，即是5 1 2個符 號°在媒體上的儲存异说細 疋〜欄連續開始於左上糊3所安排的 區域係包括撿查符號，而字〇、4、8、和12的其中每一個 具有8個檢查符號’並構成線索字。其它字的其中每一個係 包括4個檢查符號’並構成目標字。整個方塊係包括43 2個 資訊符號和8〇個檢查符號。後者能夠以更流通的方法而在 一們的相打字上疋乜3該等資訊符號部份可以是虛擬符號 3 itReed-Solomnnjs. ^ ^ ^ >- n碼允許在母個線索字上更正多達4個 符號錯誤5 f際的符號錯誤是以X指示=結果，所有的線索 卞L正$地解碼，因當它們從未超過4個錯誤。明顯地，然 而，字2和3不能夠只基於它們本身的多餘符號而解碼^現 在在圖2中，除了 6 2、6 6、6 8以外的所有錯誤是表示錯 ^ ‘名而’只有至少3個連續線索字的線條5 2和5 8係認 爲錯誤择發’所以至少所有中間符號位置會得到抹除旗號 。而且’在件發的第—線索字錯誤之前的該等目標字和猝 發的最後.¼索子錯誤之後的該等目標字會在那位置上獲得 抹除琪號’其係決定在所遵循的策略。線條54並未認爲是 猝發’因爲它太短。 知不’子4的兩個錯誤會在兩所結合的襴中產生抹除旗號 :此會提供可更正的字2和3，而其中每一個具有單一錯誤 符號及兩饲抹除符號。然而，不包括線條54的任意錯誤 6 2 6 8係概成字5、6、7的線索，因爲它們之中的每一個 只包括早一線索字3在某些情況’抹除會造成零錯誤圖案 本紙楱尺料财關家料（CNS ) Μ規格（训χ 297公釐） --.-----焚.------ΐτ-----!,银 (請先閲讀背面之注意事項再填寫本頁) 425773 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明説明（5 ) ，因爲在S位元符號中的任意錯誤具有]/256機率重新產生 正確的符號=> 同樣地，長猝發的x特殊線索字會產生正確的 符號3籍著在相同猝發的先前和隨後線索符號之間的橋接 策略，此正確符號然後會結合在猝發’而以錯誤線索符號 的相同方法轉換成適當目標符號的抹除値。上面的決定可 根據解碼策略修改，可進一步由其它的參數控制3 實際格式的討論 此後，實際的格式會討論。圖3係表示乘積碼格式。字是 水平和垂直的，並配置核檢位元組。圖4係表示所謂在上面 某些字具有特殊猝發偵測的長距離碼，而這些上面的某些 字具有更多的核檢位元組3本發明係提出所謂的警戒碼， 其構成可以是如同圖3和4之原理的组合。窝係始終沿著圖 3 ' 4所顯示的箭號而循續進行3 本發明的實際觀點是引用數位光學儲存的較新方法3特 殊的特徵在於讀取上面的傳送層的基體是〗〇 〇微米厚度5 孩通道位元具有相同0 . 1 4微米的大小，所以在2 / 3通道失 上的資料位元組只有1 . 7微米的長度3在頂部表面上的光束 直徑具有大约丨2 5微米的直柽=磁碟盒或其封套能減少發 生大猝發的機率。然而，小於5 〇微米的非一致性粒子會造 成短路錯誤，發明家具有特別所使用的錯誤模組，A /、T理 由錯誤傳播的如此錯誤會争致2 0 0微米的捽發，其并八、 大約1 2 0 位元組。特別是，發明家已使用具有丨2 〇 B、* IzJ 定大小捽發錯誤模組’其任意地開始於每一位元也 2.6 *丨〇'5、或平均每3 2 k B區塊一猝發的機率3本發明已 -8 - (請先閱讀背面之注意事項再楨寫本頁) 裝 •1Γ 線 本紙張尺度適用中國g家標隼（CNS ) Λ4規格（210X297公釐） 經瀆部智慧財產局員工消費合作社印製 4^5775 A7 _ B7_ 五、發明説明（6 ) 朝向光碟儲存的發展，但是諸如多軌磁帶的其它結構及諸 如磁性和光學磁性的其它技術也可從此所描述的改良方法 而狻益5 圖5係顯示警戒碼和猝發指示器子碼。警戒碼是由兩子碼 A和B所组成3猝發指示器子碼（B I S )係包括線索字。藉由 格式化，它是非常深地X錯長距離碼，該長距離碼允許定 位多重猝發錯誤的位置，如此所發現的錯誤圖案會處理， 以獲得該等目標字的抹除資訊，而該等目標字係配置在乘 積子碼（P S )的具體實施例=該乘積子碼會經由使用來自猝 發指示器子碼所獲得的抹除旗號來更正多猝發錯誤的组合。 下列是所提議的格式： • [ 3 2 k B ’的方塊係包括I 6個D V D相容磁區 •每個如此磁區係包括2 0 6 4 = 2 0 4 8 + 1 6個位元组資 料 •在E C C編碼之後’每個磁區包括2 3 6 8個位元组 •因此，該編碼率是0.S72 •在方塊中’ 2 5 6個同步區塊會是如下所示而格式化 • 每個磁區係包括1 6個同步區塊 • 每個同步區塊是由4馆群體3 7 B所组成 •每組3 7 B係包括丨B的高度交錯猝發指示子碼及進一 步3 6 B的乘積子碼= 如圖5所示，列是從磁ί菜循續地讀取，開始於前述的同步 圖案，每列係包括在安排方法中所顯示的4 Β的β I S，並依 續編，且由3 6個其它的位元组所隔開3 1磁區有！ 6列，而 ___ _ _ 9- 本紙張又度適财賴緖準（CNS Μ4規格（) " ----.-----壯衣------，玎------^ (請先閲讀背面之注意事項再填寫本頁} 經濟部智慈財產局8工消費合作社印製 4 25 A7 ____ B7 - 1 __ __ 五、發明説明（7 ) —同步區塊有2 5 6列= 圖6係顯示圖5之每磁區相同是64個编號位元组之專有的 猝發指示器子碼格式，而構造如下所示： •有16列，而每列是t = μ的[64，32，33]以碼： •如箭號所示得自磁碟的循序欄，所以四個欄群組係得 自快速定址的單一磁區； •每個BIS可指示至少5 92Β (〜丨公釐）的16猝發； • B I S係包括每一磁區有3 2 B資料' 4欄的B I S、及特別 的丨6B DVD資料頭，在資料頭上有53的核檢，以允許快 速的定址讀出、及丨1 B的使用者資料3 圖7顯示警戒碼及其乘積子碼，而乘積子碼是建立在該等 目標字3當乘積子碼位元组從磁碟讀取而忽略B丨S位元組 的時候，乘積子碼位元纟且便會循續地编鼠3 圖8係顯示乘積子碼的具體實施例之各種不同進一步觀點 。特別是’該乘積子碼是Reed，Solomon碼的[256，228 ，2 9〗* [ 1 4 4 ’ I 4 3，2 ]乘積碼，資料位元組數目是 228*143 = 32604’即是16倍於（2048 + II)使用者位元 組+ I 2個多餘的位元组3 圖9係顯示圖8的另一格式，其係省略整個水平的1166(^ S ο 1 ο ηι ο η碼：該水平區塊大小是3 6個位元组（圖7的四分之 一），並使用[256 ’ 224，33] Reed-Solomon 碼。每匍 磁區具有2368位元組’而且不需要需虛擬位元組3 第一欄中的碼是以兩步驟形成3從每個磁區，1 6資枓頭 位元组是首先以[2 0，1 6，5 ]編碼，以允許快速的位址取 -10- 本紙乐尺度適用中國國家標準（CNS M4規格（2丨0X297公釐） I. —11 1 I 裝〆 I . » ^ n 1. 1. (铐先閎讀背面之注意事項再填寫本頁) 425773 Α7 ___ Β7 五、發明説明（g ) 回3結果所產生的2 0個位元組+每磁區的進一步3 2個使 用者位元組係形成資料位元組，而整個會進一步編碼=— 個2 K磁區的資料符號可位在如下所示的只有一個實際磁區 =每欄[2 5 6，2 2 4 ’ 3 3 ]碼係包括每2 K磁區有8個核檢符 號）此外’每個[256 ’ 208，49]碼具有每2K磁區丨2個核 檢符號及4個核檢符號[2 0，i 6，5 ]碼，以便獲得具有4 8 個多餘位元组的[與2 5 6，2 0 8，4 9 ]碼。 圖1 0係顯示此間條的細節。在此，，*，係表示資料頭位 元组，口 疋[2 〇，1 6 ]碼的核檢，’·’是3 2個“進一步” 資料位元組、及[2 5 6，2 0 δ ]碼的1 2個核檢位元组。 f— II -I I I ---11^^—f·-- 〈請先閱讀背面之注意事項再填寫^育) 訂 線 經濟部智慈財產局8工涓費合作社印製 本纸張又度適用中國國家標準（〇^)/\4現格（210><297公釐）• d, • d, printed A7 by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs --- --- B7 V. Description of the invention (i) Background of the invention 'The present invention is as cited in the previous paragraph of the scope of patent application Methods. Berlekamp et al. U.S. Patent No. 4 559 625 and £ 111111 et al. B. U.S. Patent No. 5 299,208 disclose the decoding process of stripe and error protection data. Among them, the one found in the first word Error patterns provide clues to locate another word error in the same word group. These reference systems use a standardized format and a set of error gray branches with multiple symbol error bursts on various words. Errors in special words are more likely to occur, and errors occur at the relative symbol positions pointed to by the next word or words. This procedure often increases the number of errors corrected. The inventors have confirmed the question of this principle. The clues can only be realized when the clues have been completely corrected. SUMMARY OF THE INVENTION As a result, an object of the present invention is to provide an encoding format in which a clue word is decoded to a more certain degree than a target word. Therefore, according to one of its viewpoints, the features of the present invention are as described in the feature part of the first patent application park. The clues found will cause or point to the erasure symbol 3. With such a direction, errors are corrected. Will be dealt with more powerfully. In fact, many codes will correct the most errors when no error location is known. The provided erase bit I, usually many erases of ' e > t, will be corrected. Other types of kneading are possible. The protection of bursts and arbitrary error combinations will also be improved, or the provision of erasure positioning will only require a smaller number of error symbols, thus simplifying calculations. In principle, the present invention can be used in storage and transmission environments. 3 The present invention also relates to methods, codes and / or * 4 'used to decode such coded information. This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210x ^ 97). ) --- --.--.----- seed coat ------ ir ------ il (Please read the precautions on the back before filling out this page) 4257 7 3 Α7 Β7 Economy Printed by the staff of the Intellectual Property Bureau of the Ministry of Foreign Affairs of the People's Republic of China. 5. Description of the invention (decoding the device used in the above method, and the information carrier used to interface with the encoding and / or decoding. Quoted within the scope of the patent application. These and further perspectives and advantages of the present invention will be discussed in more detail later in the disclosure of preferred specific examples and with reference to the accompanying drawings, where: FIG. 1 is an encoder, Carrier and decoder system; Figure 2 code format principle; Figure 3 is a product code format: Figure 4 is a long distance code with burst detection; Figure 5 is a guard code and a burst indicator subcode: Figure 6 Burst indicator Subcode format; Figure 7 Guard code and its product subcode Fig. 8 is a variety of different advanced views: Fig. 9 is another format: Fig. 10 is the detail of the stripe processing 3 Detailed description of the preferred embodiment 丨 is a system showing a wide range according to the invention, which provides coding And a decoder, embodiments of which are used to encode, store, and finally decode a sequence of samples or multi-bit symbols derived from audio or video signals or from data. 8-bit continuous symbol. The separator 22 will repeatedly and periodically transfer the first symbol of the clue word to the editor, 24. In addition, the separator 22 will transfer all other symbols to m 6. In the encoder 24, 'these clues are based on the Chinese paper standard that the combined capital paper size applies to the Chinese National Standard (CNS) M specification (2! 〇 > < 297 mm.-^ IT4a —--- (Please read the precautions on the back before filling this page) A7 B7 425773 V. Description of the invention (Material coded as the first multi-symbol slave 4 φ t α Β 姽 It is formed by mistakenly corrected code code. This code can be 疋Reed-Soloinon Shi Min-., Horse, product code 'X wrong code, or on The group of people 3 in the encoder 26, the answer θβ — a. ° ^ target sub-system is well coded into the second multi-symbol error correction code codeword and is formed by {士. ..... In a specific embodiment All the codewords have the same length, but this is not a requirement of Wei Fei's Hammer. Ideally, the two types of codewords = the second code—the code of Reed_s〇i_ … Refer to the figure; it will be clearer that these clue words generally have a greater degree of error protection and relatively include less non-redundant symbols. At blocks ~ 8, the codewords thus formed are transferred to one or more outputs where they have indicated random numbers, so the media allocations discussed later will become the same. Box 3 0 indicates that the media itself that received the encoded material 3 was actually written 3 in combination with the appropriate writing principles and media 3 Alternatively, the media can be ideally implemented as a copy from the original encoded media such as annotations' The staggered memory is completely continuous, but other structures can also use 3 in block 3 2 and various words will be read from the media again. Then, the clue words of the first code will be transmitted to the decoder 3 4 and decoded based on the inherent repetitiveness. In addition, the discussion in Figure 2 will be more clear in JK, and such decoding will be on the wrong bit I There are clues instead of 3 clues on these clue words. 3 5 5 will receive these clues, and include programs that use one or more different strategies to convert the line into erasure positioning. The target words are being decoded. Under the control of the erasing position in the decoder 3 6, the error protection of these target words will be increased to an acceptable level a. Finally, all the decoded words will pass the element 3 that conforms to the original format of the output 4 〇 8 and solve multiple tasks. In order to simplify the mechanical structure used to form the various interfaces, the τ 纟 interface has been inferior to 3 -6 paper & dimensions applicable to the Chinese National Standard (CNS) A4 specification (2 丨 0X297 mm)... I1T- ----- ^-{谙 Read the notes on the back before filling out this purchase) Member of the Intellectual Property Bureau of the Ministry of Economic Affairs printed the five invention notes printed by the cooperative (Α7 Β7 The Ministry of Economic Affairs ’s fair-minded staff hired employees 浈 f cooperatives Figure 2 is printed to describe the relatively simplified code format of m .. The code format of the dagger. As shown in the figure, the encoding information is arranged in 16 columns of X #. J-paste the box, that is, It is 5 1 2 symbols ° Stored on the media. The column starts consecutively from the top left. The area arranged by 3 includes the check symbol, and each of the words 0, 4, 8, and 12 has 8 Check symbols 'and form clue words. Each of the other words includes 4 check symbols' and form target words. The entire block system includes 43 2 information symbols and 80 check symbols. The latter can be used in a more circulating way. On one of them, the information symbol part 3 can be a virtual symbol 3 itReed-Solomnnjs. ^ ^ ^ >-The n code allows correction of up to 4 symbol errors on the parent clue word. 5 Symbol errors on the f are indicated by X = result. All clues are decoded positively, because When they never exceed 4 errors. Obviously, however, words 2 and 3 cannot be decoded based only on their own redundant symbols ^ Now in Figure 2, all errors except 6 2, 6 6, 6 8 are Represented wrongly ^ 'Name and' only have at least 3 consecutive clues. Lines 5 2 and 5 8 are considered to be incorrectly selected 'so at least all intermediate symbol positions will get erased flags. And' in the first line of clues — The target words before the error and the end of the burst. The target words after the error will be erased at that position, which is the strategy determined to follow. Line 54 is not considered a burst. 'Because it's too short. Do you know?' The two mistakes of Sub 4 will cause erasure flags in the two combined crickets: this will provide correctable words 2 and 3, each of which has a single error symbol and two feeds. Erase the symbol. However, any errors that do not include line 54 6 2 6 8 series Form clues of 5, 6, and 7 because each of them only includes the clue 3 of the earlier clue. In some cases, 'erasing will cause zero errors. Training χ 297 mm) --.----- burn .------ ΐτ ----- !, silver (please read the precautions on the back before filling this page) 425773 A7 B7 Ministry of Economy Wisdom Printed by the Consumer Cooperative of the Property Bureau V. Invention Description (5), because any error in the S-bit symbol has a] / 256 probability to regenerate the correct symbol = > Similarly, a long burst of x special clue words will produce The correct symbol 3 is a bridging strategy between the previous and subsequent clue symbols of the same burst. This correct symbol is then combined with the burst 'and converted into the erasure of the appropriate target symbol in the same way as the wrong clue symbol. The above decision can be modified according to the decoding strategy, and can be further controlled by other parameters. 3 Discussion of the actual format After this, the actual format will be discussed. Fig. 3 shows a product code format. Words are horizontal and vertical, and check bytes are configured. FIG. 4 shows the so-called long-distance codes with special burst detection on some of the words, and some of these words have more verification bytes. The present invention proposes a so-called alert code, which can be composed of This is a combination of the principles of Figures 3 and 4. The nesting system continues along the arrows shown in Fig. 3'4. 3 The practical point of the present invention is to refer to the newer method of digital optical storage. 3 A special feature is that the substrate for reading the transport layer above is 0. The thickness of the 5 micron channel bit has the same size of 0.14 micron, so the data byte lost on the 2/3 channel is only 1. 7 microns in length. 3 The beam diameter on the top surface has about 2 5 Micron mullions = cartridge or its case can reduce the chance of large bursts. However, non-uniform particles smaller than 50 microns will cause short-circuit errors. The inventor has a special error module used. Such errors propagated by A / T errors will cause a burst of 200 microns. Eight, about 120 bytes. In particular, the inventor has used a module with a size of 丨 2 〇B, * IzJ to issue errors, which arbitrarily starts with each bit also 2.6 * 丨 0'5, or an average of one every 32 KB blocks. Probability of burst 3 The present invention has been -8-(Please read the precautions on the back before writing this page) Packing • 1Γ The size of the paper is applicable to the Chinese g family standard (CNS) Λ4 specification (210X297 mm) Printed by the Intellectual Property Bureau's Consumer Cooperatives 4 ^ 5775 A7 _ B7_ V. Description of the Invention (6) Development towards optical disc storage, but other structures such as multi-track magnetic tape and other technologies such as magnetism and optical magnetism can also be described from here Improved method and benefits 5 Figure 5 shows the alert code and burst indicator subcode. The alert code is composed of two sub-codes A and B. The 3 burst indicator sub-code (B I S) includes a clue word. By formatting, it is a very deep X-error long-distance code. The long-distance code allows the location of multiple burst errors. The error patterns found in this way will be processed to obtain the erasure information of the target words. A specific embodiment in which the target word is arranged in a product subcode (PS) = the product subcode corrects the combination of multiple burst errors by using the erasure flag obtained from the burst indicator subcode. The following are the proposed formats: • [3 2 k B 'block system includes 16 DVD compatible sectors • each such sector system includes 2 0 6 4 = 2 0 4 8 + 16 bytes Information • After ECC encoding, 'Each sector includes 2 3 6 8 bytes. Therefore, the encoding rate is 0.S72. • In a block, 2 5 6 sync blocks will be formatted as shown below. • Each magnetic zone system includes 16 sync blocks. • Each sync block is composed of 4 library groups 3 7 B. • Each group of 3 7 B series includes a highly interlaced burst indicator subcode and further 3 6 The product subcode of B = As shown in Figure 5, the columns are read sequentially from the magnetic dishes, starting from the aforementioned synchronization pattern. Each column includes the β IS of 4 Β shown in the arrangement method, and A sequel, and separated by 3 6 other bytes. 3 1 sector exists! 6 columns, and ___ _ _ 9- This paper is also financially suitable for Lai Xuzhun (CNS Μ4 specifications () " ----.----- Zhuang Yi ------, 玎 --- --- ^ (Please read the notes on the back before filling out this page} Printed by the 8th Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 4 25 A7 ____ B7-1 __ __ V. Description of the invention (7) — Synchronized blocks have 2 5 6 columns = Figure 6 shows that each sector in Figure 5 is the same burst indicator subcode format of 64 numbered bytes, and the structure is as follows: • There are 16 columns, and each column is [64, 32, 33] with t = μ in code: • The sequential columns from the disk are shown as arrows, so the four column groups are derived from a single sector of fast addressing; • Each BIS can indicate 16 bursts of at least 5 92B (~ 丨 mm); • BIS system includes BIS with 3 2 B data in each magnetic field, 4 BIS columns, and special 6B DVD data header. There is a 53 check on the data header. To allow fast address reading, and 1 B user data 3 Figure 7 shows the guard code and its product subcode, and the product subcode is built on the target word 3 when the product subcode bytes are removed from the disk When reading and ignoring B 丨 S bytes The product subcode is bitwise and will be programmed sequentially. Figure 8 shows various further views of the specific embodiment of the product subcode. In particular, 'the product subcode is Reed, Solomon's [256, 228, 2 9〗 * [1 4 4 'I 4 3, 2] Product code, the number of data bytes is 228 * 143 = 32604', which is 16 times (2048 + II) user bytes + I 2 redundant Figure 3 shows another format of Figure 8, which omits the entire horizontal 1166 (^ S ο 1 ο ηι ο η code: the horizontal block size is 36 bytes (Figure 7 One-quarter), and use [256 '224, 33] Reed-Solomon codes. Each magnetic field has 2368 bytes' and does not require virtual bytes. The code in the first column is formed in two steps. 3 From each magnetic sector, the 16-byte header byte is first encoded with [2 0, 1 6, 5] to allow fast address retrieval. -10- This paper scale is applicable to Chinese national standards (CNS M4 specifications) (2 丨 0X297mm) I. —11 1 I Decoration I. »^ N 1. 1. (Read the precautions on the back before filling out this page) 425773 Α7 ___ Β7 5. Description of the invention (g) Back 3 results The generated 20 bytes + further 32 user bytes per sector form a data byte, and the whole will be further encoded =-the data symbols of 2 K sectors can be located as shown below Only one actual magnetic field = each column [2 5 6, 2, 2 4 '3 3] code system includes 8 check symbols per 2 K magnetic field) In addition' each [256 '208, 49] code has each 2K magnetic field 丨 2 check symbols and 4 check symbols [2 0, i 6,5] codes in order to obtain [with 2 5 6, 2 0 8, 4 9] with 4 8 extra bytes code. Figure 10 shows the details of this bar. Here, * indicates the data header byte, the verification of the [2 0, 16] code, and "·" are the 3 "further" data bytes, and [2 5 6, 2 0 δ] code of 12 check bytes. f— II -III --- 11 ^^ — f ·-<Please read the precautions on the back before filling in ^ Education) Threaded paper printed by the Ministry of Economic Affairs, Intellectual Property Bureau, 8th Cooperative Fee Cooperative, and applicable again Chinese national standard (〇 ^) / \ 4 is present (210 > < 297 mm)