1307850 九、發明說明： 【發明所屬之技術領域】 本發明關於一種利用雜湊値來檢查資料量之裝置和方 法，以及算出可例如爲電腦檔之資料量的比較雜湊値之裝 置和方法。 【先前技術】 ' 對於大半不同之應用情形，例如，在數位權利管理當 * 中，於使用前期望或必須檢查檔案或位元流之認證性或完 φ 整性。檢查資料量之完整性之大槪範圍可例如爲要檢查之 電腦的RAM或ROM，或大量儲存裝置上之檔案。檢查時 可發現對於資料之處理，或者在另外處理資料前可辨認記 憶體或寫入之錯誤。 尤其是在音訊應用或視訊應用中，甚至如果資料是以 壓縮形式存在時，資料量則非常大。例如，一首典型的MP3 編碼歌曲產生播放時間爲每分鐘1 MB之資料量。 爲了例如以雜湊演算法檢查檔案之完整性，經常以雜 • 湊演算法計算電腦檔案，並與一參考値比較其結果，即雜 - 湊値。 . 以下將說明利用加密雜湊法之資料集的這種完整性檢 查。此處，加密雜湊値亦稱爲總和雜湊値。雜湊法以界定 及唯一方式，從未定長度之輸入値計算出所謂雜湊之某一 輸出値。該雜湊値例如爲長度20個位元組之字串。雜湊功 能之特性在於爲各任意輸出値算出唯一相關之輸出値，無 法從該輸出値回算輸入値。 1307850 首先以雜湊演算法進行處理完整之資料量，致構成資 料量之雜湊，其中’針對該完整資料量計算雜湊値。對於 稍後之完整性檢查，以雜湊演算法再次完整地處理要檢查 之資料量。如果要檢査之資料量提供與參考經過中相同之 雜湊，則可認爲對資料集無任何改變。 雜湊功能之主要工作爲檢查並確保數位資料之完整性 。應用情形之範圍從核對和計算到簽章方法。這意謂著直 接使用雜湊作爲核對和或使雜湊額外簽章，取代原先之資 ® 料量。對於完整性檢查，直接使用雜湊値作爲核對和。 實質上可將雜湊功能之需求列爲以下三點。另一方面 ’各雜湊値必須時常對等地存在。這意謂著雜湊値之槪率 在各種輸入値中必須相同。而且，輸入値中之小變更應會 造成雜湊値之變更。而且，要產生碰撞並不輕鬆。這意謂 著對於一指定之輸入値，非常難以發現具有相同雜湊値之 第二輸入値。符合所提及之所有三種需求之雜湊功能稱爲 加密雜湊功能。 最重要之加密雜湊功能包含SHA-l、MD4、MD5以及 RIPE-MD-160。目前最重要之雜湊功能SHA-1(SHA =安全性 雜湊演算法，SECURE HASH ALGORITHM)處理長度爲512 位元之資料區，產生長度爲160位元之雜湊値。對於SHA-l ’所謂鏈變數，以及所謂壓縮功能之五個3 2位元變數扮演 重要之角色。 在雜湊功能SHA-1中，首先將輸入値分成長度爲512 1307850 位元之資料區。然後，壓縮功能收納五個鏈變數以及五個 5 1 2位元資料區並將該等對映至次五個3 2位元碼。現在以 四回合之各20個相同運算執行該作用，當中依據預界定之 計算運作使個別位元位移。最後，使五個鏈變數之內容輸 出爲雜湊値。 例如，在規格書"開放行動聯盟，OMA DRM規格y2.0 草本版2.0-2004年4月10日”中說明使用雜湊法作爲完整 ' 性檢查。例如，在"網際網路串流媒體聯盟，加密及驗證規 • 格，1.0版，2004年2月”規格書中說明另一個檢查法。 經常上，資料集之完整性檢查未終止於其本身。而是 ，僅將完整性檢查安置在資料集實際使用之上游。因完整 性檢查與造成額外成本和延誤資料集使用之花費有關，因 此完整性檢查之努力不利於資料集之使用。 尤其是，完整性檢查和所需之資源使用需大量時間， 特別是執行核對和計算之計算能力，這大爲不利。在依據 是否產生資料集完整性完成任何敘述前，因首先必須不斷 ® 檢查整個資料集，故所需之努力特別有關係。這在具有音 ' 訊和視訊檔案之非常大資料集中特別有影響，且例如可分 - 別在音訊資料和視訊資料之再生中啓動延遲。 另一實質缺點是由於首先永遠必須檢查整個資料集之 必要性所產生之高能量消耗。如可能只要使用部分之資料 ’這亦有可能。例如，當在可携式播放裝置馬上播放或快 轉或回轉DRM保護之音訊片段或視訊片段時，首先因此亦 需分別檢查完整之音訊片段或視訊片段。尤其是可携式裝1307850 IX. Description of the Invention: [Technical Field] The present invention relates to an apparatus and method for checking the amount of data using a hash, and a device and method for calculating a data mixture which can be, for example, a computer file. [Prior Art] 'For most of the different application scenarios, for example, in digital rights management when *, it is desirable or necessary to check the authenticity or integrity of the file or bit stream before use. The scope of checking the integrity of the data volume can be, for example, the RAM or ROM of the computer to be inspected, or the file on a large number of storage devices. During the inspection, it may be found that the processing of the data or the error of the memory or the writing may be recognized before the additional processing of the data. Especially in audio applications or video applications, even if the data is in compressed form, the amount of data is very large. For example, a typical MP3 encoded song produces a data volume of 1 MB per minute. In order to check the integrity of the file, for example, by a hash algorithm, the computer file is often calculated by a hash algorithm, and the result is compared with a reference, that is, mixed. This integrity check of the data set using the cryptographic hash method will be explained below. Here, the cryptographic hash is also known as the summation hash. The hash method calculates a certain output of the so-called hash in a defined and unique way, from an input of an undetermined length. The hash is, for example, a string of 20 bytes in length. The characteristic of the hash function is that the only relevant output 値 is calculated for each arbitrary output 値, and the input 値 cannot be calculated from the output 値. 1307850 First, the complete amount of data is processed by the hash algorithm, which constitutes a hash of the amount of information, where 'the hash is calculated for the complete data amount. For later integrity checks, the amount of data to be checked is again completely processed in a hash algorithm. If the amount of data to be inspected provides the same hash as in the reference, there is no change to the data set. The main task of the hash function is to check and ensure the integrity of the digital data. Application scenarios range from check and calculation to signature methods. This means that the hash is used directly as a check and or the hash is additionally signed to replace the original amount of material. For integrity checks, use hashes directly as checksums. In essence, the requirements of the hash function can be listed as the following three points. On the other hand, 'the various hashes must exist reciprocally. This means that the rate of hashing must be the same in all input ports. Moreover, small changes in the input 应 should result in a change in the hash. Moreover, it is not easy to create a collision. This means that for a given input, it is very difficult to find a second input 具有 with the same hash. The hash function that meets all three of the requirements mentioned is called the cryptographic hash function. The most important cryptographic hash functions include SHA-l, MD4, MD5, and RIPE-MD-160. At present, the most important hash function SHA-1 (SHA = security hash algorithm, SECURE HASH ALGORITHM) handles a data area of 512 bits in length, resulting in a hash of 160 bits. The so-called chain variable of SHA-l', and the five 32-bit variables of the so-called compression function play an important role. In the hash function SHA-1, the input port is first divided into data areas of length 512 1307850 bits. The compression function then stores five chain variables and five 5 1 2 bit data fields and maps them to the next five 3 2 bit codes. This effect is now performed in 20 identical operations in four rounds, with individual bits being shifted according to a predefined calculation operation. Finally, the contents of the five chain variables are output as hashes. For example, in the specification "Open Action Alliance, OMA DRM Specification y2.0 Herb Edition 2.0-April 10, 2004, the use of hashing is described as a complete 'sex check. For example, in "Internet Streaming Media Alliance, Encryption and Verification Regulations, Version 1.0, February 2004, another specification is described in the specification. Often, the integrity check of a data set does not end on itself. Instead, only the integrity check is placed upstream of the actual use of the data set. Because the integrity check is related to the cost of additional costs and delays in the use of the data set, the integrity check effort is not conducive to the use of the data set. In particular, the integrity check and the required resource usage take a lot of time, especially the ability to perform checks and calculations, which is highly disadvantageous. The effort required is particularly relevant since the entire data set must be continually checked before the completion of any statements based on the integrity of the data set. This is particularly influential in very large datasets with audio and video archives, and for example, can be divided into delays in the regeneration of audio and video data. Another substantial drawback is the high energy consumption due to the necessity of first checking the entire data set. If you can use some of the information, it is possible. For example, when a portable playback device plays or quickly rotates or revolves a DRM-protected audio clip or video clip, it is first necessary to separately check the complete audio clip or video clip. Especially portable
1307850 置，這造成縮短電池壽命。 另一缺點爲只要部分資料存在，有關資料集的敘述就 可能不存在。這缺點是因實際之處理在負面之檢查結果下 可多半必須被省略掉。 而且，不利的是部分使用資料需要相同的檢查努力作 爲資料之完整利用。當要馬上只播放或快轉或回轉音訊/視 訊片段達一片段時亦會產生起動延遲。 【發明內容】 本發明之目的在提供資料量檢查用之裝置和方法以及 用於算出雜湊値之裝置和方法，其能以低耗力來加速檢查 程序。 藉由用於檢查申請專利範圍第1項之資料量的裝置、 用於檢查申請專利範圍第6項之資料量的方法、以及申請 專利範圍第7項之電腦可讀取媒體達成這目的。 本發明提供一種用於檢查資料量之裝置，其中該資料 量具有多個子集及每個子集具有部分雜湊値，該裝置包含 用於算出雜湊値之手段，其構成係從子集所提供之部 分雜湊値算出共同雜湊値；以及 比較手段，其構成係用以比較該共同雜湊値與提供給 該比較手段之比較雜湊値。 本發明還提供從具有多個子集之資料量算出比較雜湊 1307850 値之裝置，該裝置包含： 用於算出每個子集之部分雜湊値的手段；以及 從該子集之部分雜湊値算出該比較雜湊値之手段。 本發明依據之結果爲資料集之分部態樣或部分完整性 檢查造成減低完整性檢查或建立雜湊値所需之努力。 依據本發明方法，未在整個資料集上構成雜湊値，卻 在資料集之子集上構成雜湊値。這些子集之雜湊値是儲存 * 在資料集中或個別分開。而且，額外之雜湊値係構成在部 • 分雜湊値上。其爲有利的，因爲爲了檢查該資料集之完整 性，可針對額外之雜湊値首先檢查部分之雜湊値。然後借 助於與個別子區域相關之核對和可檢查個別之子區域。 透過依據發明方法之分段態樣來檢査完整性之可能性 ，形成一連串的優點。尤其是，當只檢查資料集之第一分 段而更未檢查整個資料集時可能已使用資料。例如’可因 此開始虛擬地播放一音訊/視訊片段而無時間延遲。而且， 亦可選擇性地只部分使用資料集’不必檢查整個資料集。 ® 當快轉和回轉或跳至該音訊/視訊片段時，分段態樣之檢查 ' 是有利的。透過分段態樣之檢査略過不需要之分段檢查。 ' 因此，節省時間和資源。 依據另一實施例，與完整性檢查倂行可分段態樣地實 際利用該資料。因此’ 一旦爲了只接著檢查該分段之完整 性且然後立即處理它，僅必須從磁碟機載入，例如’資料 段。這可節省時間、資源和能量。 1307850 依據另一實施例，亦可向下游方向檢查。當可接受已 能使用檔案之個別片段而不需檢查的時候，這是有可能的 。例如’使用音樂檔之第一片段或另一片段且與使用倂行 地加以檢查。這具有在執行期間檢查要檢查之資料的優點 。而且’只必須從硬碟取出要使用之資料一次。如果檢查 成功，則釋放下一段加以處理。否則，在剛執行之片段後 * 即停止處理。 而且’藉由預先檢查之資料資料集的子分段之雜湊値 φ ，本發明方法能快速預先檢查資料集。因此，例如，可快 速槪觀大量之資料集。只在雜湊値之檢查後接著發生利用 該雜湊値之資料集本身的實際檢查。 依據另一實施例，從個別子集之部分雜湊値可構成另 一中介雜湊値。從該中介雜湊値，其數値小於部分雜湊値 數値’依序構成一共同雜湊値。其優點爲在大量子集之下 ，藉由與從該中介雜湊値所算出之雜湊値比較的共同雜湊 値可產生快速之預檢。只接著爲針對伴隨之中介雜湊値所 • 檢查的部分雜湊値。有如必須從部分雜湊値立即計算出共 . 同雜湊値般’從中介雜湊値算出共同雜湊値係顯著地與較 省力有關。 有利的是，可使本發明方法有利地使用有關DCF格式 (DCF = DRM內容格式）。在”開放行動聯盟；DRM內容格式 V2.0;草稿版2.0-2004年4月20日”中說明DCF格式且以 MPEG-4資料格式之加密音訊/視訊資料使用該DCF格式。 -10- 1307850 【實施方式】 在本發明較佳實施例之隨後說明中，在各種圖式中所 說明之類似代理元件將使用相同或類似之參註編號，其中 ，略過這些元件之反覆說明。 第1圖表示用於檢查資料量之裝置的示意說明，該裝 置包含用於算出雜湊値之手段102，比較手段103，及使用 ' 資料量之手段1〇4。用於檢查資料量之裝置係構成爲藉由 * 部分雜湊値和比較雜湊値，而在使用前先檢查資料量。該 φ 部分雜湊値和該比較雜湊値係與該資料量一起被供給至用 以檢查資料量的裝置。 用於算出雜湊値之手段1 02係構成用以接收資料量 112。資料量112分成多個子集（圖中未顯示）。對於資料量 1 1 2之各個子集’已經計算出部分雜湊値1 1 4。亦提供部分 雜湊値114給用以算出雜湊値的手段1〇2。而且，已從該 部分雜湊値II4形成亦被提供給用於檢查資料量之該裝置 的比較雜湊値118。在用以算出雜湊値的該裝置中，已從 • 第2圖中所示之資料量來算出，例如部分雜湊値丨！ 4、以 . 及比較雜湊値1 1 8，並和該資料量一起被閂鎖。 _ 用於算出雜湊値之手段102係構成爲從部分雜湊値 1 1 4來計算共同雜湊値i丨6並將它提供給比較手段丨03。比 較手段1 03更構成爲用以接收比較雜湊値i i 8。從該部分 雜湊値1 1 4亦已形成該比較雜湊値丨！ 8。對於比較雜湊値 1 1 8以及共同雜湊値丨丨6之構成，使用相同之計算作用。 該比較手段103係構成爲用以比較該共同雜湊値116和該 -11- 1307850 比較雜湊値1 1 8。如本實施例，依據該共同雜湊値〗丨6和 該比較雜湊値1 1 8間之比較結果，比較手段1 〇 3係構成爲 用以提供比較信號1 2 0給用以算出雜湊値的手段1 〇 2。該 比較信號120表示該比較雜湊値118是否符合該共同雜湊 値116。該共同雜湊値116和該比較雜湊値118符合時， 可認爲自從構成該比較雜湊値1 1 8後更未改變該部分雜湊 値1 1 4，其中，從該部分雜湊値1 1 4形成該共同雜湊値i i 6 〇 根據這實施例，在比較信號1 2 0表示比較雜湊値〗i 8 和該共同雜湊値116符合時，用於算出雜湊値之手段1〇2 係構成爲從資料量112之第一子集來計算第一個另一部分 雜湊値122，並將它提供給該比較手段103。該比較手段 1 〇3亦構成爲用以接收該部分雜湊値1 1 4。回應接收到另一 個部分雜湊値1 22，該比較手段1 03係構成爲用以比較該 另一個部分雜湊値1 22和相對應的部分雜湊値1 1 4，其中 ’該部分雜湊値114係從與另一個部分雜湊値122相同之 子集所形成。依據該部分雜湊値114和該另一個部分雜湊 値1 22之比較，比較手段丨〇3係構成爲用以提供檢查結論 1 24給使用該資料量的手段。檢查結論〗24表示該部分雜 湊値1 1 4是否符合所伴隨之另一個部分雜湊値1 22。依據 與該部分雜湊値114相同之演算法，已從所伴隨之子集構 成另一個部分雜湊値122。因此，在該部分雜湊値114符 合該另一個部分雜湊値122時，可認爲自從計算所伴隨之 該部分雜湊値i丨4後，更未改變子資料量，其中，對該子 -12- 1307850 資料量計算該另一個部分雜湊値122。 在檢查結論124表示該部分雜湊値114和該另 分雜湊値122符合時，使用資料量之手段1〇4係構 用資料量1 1 2之子集，其中對於該資料量1 1 2，構 個部分雜湊値1 22並在比較手段1 03中加以比較。 另外，用於算出雜湊値之手段102可構成爲算 . 量112之子集的該另一個部分雜湊値122，且平行 — 們提供給比較手段1 〇3作爲該比較信號1 20之結果 B ，用以算出雜湊値之手段102可構成爲在已經比較 雜湊値116和該比較雜湊値118之前，就已先建立 部分雜湊値122。 部分雜湊値1 1 4可和該資料量1 1 2分別被提供 雜湊値的手段102。另外，可將部分雜湊値114整 料量1 1 2中。比較雜湊値1 1 8亦可和該資料量1 1 2 提供至用以檢查該資料量的裝置。 如果提供部分雜湊値1 1 4給用以檢査資料量的 B ，只在已經提供資料量112之子集用以算出雜湊値 . 102後，用於算出雜湊値之手段1〇2可構成爲事先 收到之子集的另一個部分雜湊値122。在這情況下 當已經從用於算出雜湊値之手段102收到部分雜湊 時來算出共同雜湊値116。 在已收到檢查結論124前’用於使用資料量之弓 可構成爲已經利用資料量112或資料量112之子集 情況下，如果檢查結論1 2 4表示部分雜湊値1 1 4和 一個部 成爲使 成另一 出資料 地將他 。而且 該共同 另一個 給算出 合在資 一起被 裝置時 之手段 形成已 ，只有 値114 三段10 4 。在這 另一個 -13- .1307850 部分雜湊値1 2 2不符合或共同雜湊値1 1 6和比較雜湊値1 1 8 不符合時，用於使用該資料量之手段104可構成爲中斷使 用該資料量。 第2圖表示依據本發明之實施例，用以算出雜湊値之 裝置的示意說明。 尤其是，用於算出雜湊値之裝置係構成爲提供部分雜 湊値1 1 4和比較雜湊値1 1 8，爲了檢查資料量之完整性， ' 該部分雜湊値1 1 4和該比較雜湊値1 1 8可被用於檢查第1 φ 圖中所示之用於檢查資料量之裝置。 用於算出雜湊値之裝置包含用於算出部分雜湊値之手 段206、用於算出比較雜湊値之手段207、及整合手段208 。用於算出雜湊値之裝置係構成爲接收資料量112。該資 料量1 1 2包含多個子集。用於算出部分雜湊値之手段206 係構成從已接收資料量1 1 2之子集形成部分雜湊値1 1 4， 並將之提供給用以算出比較雜湊値之手段2 0 7。因此，對 於資料量1 1 2之各個子集，提供部分雜湊値1 1 4給手段207 φ 來算出比較雜湊値。藉由用於算出部分雜湊値之手段206 . ，依據預定之制定演算法，從該子集算出該部分雜湊値1 1 4 〇 如果該資料量1 1 2未被分成子集，則用於算出部分雜 湊値之手段206可包含用於將該資料量（圖中未顯示）分成 多個子集之手段。 用於算出比較雜湊値之手段207係構成爲依據預定之制 定規則，從部分雜湊値1 1 4來算出並提供比較雜湊値1 1 8。 -14- 1307850 分別儲存並提供部分雜湊値114和比較雜湊値1 1 8供 進一步之處理。依據本實施例’提供部分雜湊値114給整 合手段208。整合手段208係構成爲接收資料量112，將部 分雜湊値114包含在資料量112中並將其提供爲具有部分 雜湊値的資料量212。 依據實施例，藉由整合手段208 ’將該部分雜湊値1 1 4 • 插入至該資料量112，故將其安排成均等地分佈在資料量 - 212中。除此之外，可將該部分雜湊値114 —起安排在預 φ 定位置之該資料量212中。 如果該部分雜湊値1 1 4是散佈在該資料量2 1 2中時， 藉由檢查該部分雜湊値，依照在儲存裝置中或該資料量212 之另外傳輸是否已發生錯誤，可事先以高槪率算出該部分 雜湊値。該部分雜湊値114在該資料量212固定位置中之 儲存具有優點爲，在該資料量212之隨後檢查中，可首先 讀出該部分雜湊値並可針對該比較雜湊値1 1 8加以比較。 除此之外，將該資料量1 1 2和該部分雜湊値1 1 4分別 φ 儲存。該比較雜湊値118亦可被整合在該資料量112中或 . 亦可被分別儲存。 依據一實施例，利用加密雜湊功能來算出該雜湊値。 依據該實施例，利用雜湊演算法、雜湊値或”匆忙地”檢查 ，則可進行例如爲Μ P 3格式歌曲之大檔案完整性的分段態 樣檢査。該資料被分成數段，例如爲對應於子集之MPEG-4 編碼音訊資料之資料段或存取單元。 在用於算出雜湊値之裝置中，分別雜湊個別之子分段 -15- 1307850 而產生雜湊。然後將從分段所算出之雜湊値儲存在例如與 該資料儲存在一起之表中。針對這雜湊値之表構成的新雜 湊並將之使用爲所謂主雜湊之實際雜湊。該雜湊値對應於 該部分雜湊値，而該雜湊對應於該比較雜湊値。 將子分段分成適當或預期大小之子分段。在這當中， 於粒度與額外雜湊値額外所需之儲存空間之間取得妥協。 * 在雜湊該子分段後，將子分段之雜湊値儲存在一個或更多 - 個表中。隨後針對該表或具有部分雜湊値之表來形成雜湊 φ 値。然後可將該主雜湊儲存在外並成爲稍後針對它加以檢 查之參考。 對於在用於檢查資料量之裝置中的雜湊檢查，首先以 該雜湊預先檢查雜湊表。亦即，計算針對該表或具有部分 雜湊値之表的雜湊並與該主雜湊比較。從該表中使用雜湊 値，然後檢查個別的子分段。直到最後，來自該表之該部 分雜湊値被用來個別檢查檔案之子分段。在該主雜湊和針 對該部分雜湊値所計算之該雜湊之間爲均等之後，在這實 • 施例中檢查該子分段。依此方式，可已經使用已檢查過之 . 子分段，而其它者則更未加以檢查。 依據另一個實施例，構成那麼多的子分段使得具有部 分雜湊値之表變成太大。在這情況下，可以階層式使用數 個雜湊表。這意謂著該第一表包含雜湊値，該雜湊値可爲 下級表依序作爲主雜湊使用。作爲其一種特殊情況，可產 生一連續清單。這意謂著表中之該最後雜湊値爲下一個表 之該主雜湊等等。 -16- 1307850 依據另一實施例，該雜湊表和該主雜湊是爲依據該 DCF(DRM內容格式）來加密之物件而說明，其中之DCF大 多包含以MPEG-4資料格式所加密之音訊/視訊資料。DCF 物件被分成預期之子分段數量，所謂的資料段’其中’各 組合一個或更多個存取單元。將具有這資訊之資料段表插 入至該DCF物件中。針對該資料段以雜湊演算法SHA-1計 算各個雜湊。從所計算之所有雜湊，構成一個表’並將該 表插入DCF物件中作爲MPEG-4基元。針對該雜湊値之表 ，構成該主雜湊，將該雜湊値儲存在該DCF物件外作爲參 考値。 藉由從DCF物件，以雜湊値表讀出MPEG-4基元來檢 查該完整性。然後針對該表計算雜湊値並與該主雜湊比較 。均等時可進行使用DCF物件，否則會剔除DCF物件。如 果連續使用DCF物件，則從該DCF物件接著搜尋預期之資 料段，並處理該資料段，例如，具有同時檢查該資料段中 雜湊之再生。同時檢查隨附資料段之雜湊係被稱爲"匆忙地” 檢查。如果資料段之雜湊値符合來自該雜湊表中之對應値 ，則可處理另一個資料段。否則，會因爲D C F物件之修正 而拒絕進一步之處理。 在先前說明中’即使已經說明利用雜湊演算法之雜湊 値的計算，但明顯地本發明方法並不限於雜湊作用，而是 可構成任何種類之雜湊値或核對和。例如，可實施同位之 計算。而且’對於需要檢查資料完整性之所有應用可使用 本發明方法。這種應用例如可爲電腦系統或數位信息傳輸 -17- 1307850 系統。例如，在電腦系統中’當儲存資料時可產生雜湊値 並共同地加以儲存。在後續讀出及利用該資料當中，亦讀 出該雜湊値並用來檢查該資料。在傳統系統中’在資料傳 輸前可直接計算該雜湊値且然後與資料一起傳輸該雜湊値 並在接收器中加以估算。因此’可確定已正確地傳輸資料 〇 ' 對於數位資訊之分段態樣之完整性檢査，將資料分成 ' 子集。該子集可彼此獨立或可重疊。尤其是，其可以彼此 • 獨立地將子集解碼或依構造地分析子集。用於檢查資料量 之裝置可爲部分之編碼器而用於算出雜湊値之裝置爲部分 之解碼器。 依據該條件，可以硬體或軟體執行檢查資料量之發明 方法，以及算出雜湊値之發明方法。該執行可實施在數位 儲存媒體上，尤其是具有能與可程式化電腦系統互動之電 子可讀取控制信號的軟碟或CD上，以致能執行對應之方 法。通常，當在電腦上執行電腦程式產品時，本發明因此 # 亦在於將程式碼儲存在用於實施本發明方法之機器可讀取 - 之載體上的電腦程式產品。另言之，當在電腦上執行電腦 程式時，本發明可因此以具有用於實施該方法之程式碼的 電腦程式加以實現。 【圖式簡單說明】 以下參考隨圖將更加詳細說明本發明之較佳實施例， 其中： 第1圖爲依據本發明實施例，用於雜湊値資料量之裝 -18- 1307850 置的方塊電路圖；以及 第2圖爲依據本發明另一個實施例，用於算出雜湊値 之裝置的方塊電路圖。 【主要元件符號說明】1307850 set, which results in shortened battery life. Another disadvantage is that as long as some of the information exists, the description of the relevant data set may not exist. The disadvantage is that the actual processing must be omitted from the negative inspection results. Moreover, it is disadvantageous that partial use of the data requires the same inspection effort as a complete utilization of the data. A start delay is also generated when a piece of audio/video clip is played or fast-forwarded. SUMMARY OF THE INVENTION An object of the present invention is to provide an apparatus and method for data amount inspection and an apparatus and method for calculating a hash, which can speed up an inspection procedure with low power consumption. This is achieved by a device for inspecting the amount of information in the first application of the patent scope, a method for inspecting the amount of information in the scope of the patent application, and a computer readable medium applying the seventh item of the patent scope. The present invention provides an apparatus for inspecting a data amount, wherein the data amount has a plurality of subsets and each subset has a partial hash, and the apparatus includes means for calculating a hash, the composition of which is provided from the subset Part of the hash is to calculate the common hash; and the comparison means is used to compare the common hash and the comparison we provide to the comparison means. The present invention also provides an apparatus for calculating a comparison hash 1307850 from a data amount having a plurality of subsets, the apparatus comprising: means for calculating a partial hash of each subset; and calculating the partial hash from the subset More cumbersome means. The results of the present invention are based on the fact that the partial or partial integrity check of the data set results in reduced integrity checks or the establishment of hashes. According to the method of the present invention, no hashes are formed on the entire data set, but a hash is formed on a subset of the data sets. The hashes of these subsets are stored * in the data set or individually. Moreover, the extra hashes are made up of parts and pieces. This is advantageous because in order to check the integrity of the data set, some of the hashes can be checked first for additional hashes. The individual sub-areas can then be checked by checking against individual sub-areas. A series of advantages are formed by examining the possibility of completeness in accordance with the segmentation aspect of the inventive method. In particular, data may have been used when only the first segment of the data set was examined and the entire data set was not examined. For example, 'there may be a virtual play of an audio/video clip without time delay. Moreover, it is also possible to selectively use only the data set' without having to check the entire data set. ® When checking for fast and swivel or jumping to the audio/video clip, the check of the segmentation pattern is advantageous. The segmentation check is skipped by the inspection of the segmentation pattern. 'So save time and resources. According to another embodiment, the data is actually utilized in a piecemeal manner with the integrity check. Therefore, once it is only to check the integrity of the segment and then process it immediately, it must only be loaded from the disk drive, for example, the data segment. This saves time, resources and energy. 1307850 According to another embodiment, it is also possible to check in the downstream direction. This is possible when it is acceptable to use individual segments of the file without checking. For example, 'use the first segment or another segment of the music file and check it with the use of it. This has the advantage of checking the data to be checked during execution. And 'only need to take the information to be used once from the hard drive. If the check is successful, the next segment is released for processing. Otherwise, the processing is stopped after the segment just executed. Moreover, the method of the present invention can quickly pre-check the data set by the hash φ of the sub-segments of the pre-examined data set. So, for example, you can quickly see a large collection of data. The actual inspection of the data set itself using the hash is then carried out only after the inspection of the hash. According to another embodiment, a portion of the hashes from the individual subsets may constitute another intervening hash. From the intermediary, the number 値 is smaller than the partial hash number 値' to form a common hash. The advantage is that under a large number of subsets, a fast preflight can be generated by a common hash compared to the hashes calculated from the intervening hashes. It is only a part of the hash that is checked for the accompanying media. For example, it is necessary to calculate the total from the partial weeds. The same as the weeds, the calculation of the common hashes from the intermediaries is significantly more dependent on the effort. Advantageously, the method of the invention can advantageously be used with respect to the DCF format (DCF = DRM content format). The DCF format is described in "Open Action Alliance; DRM Content Format V2.0; Draft Version 2.0-April 20, 2004" in the DCF format and encrypted audio/video data in the MPEG-4 data format. -10- 1307850 [Embodiment] In the following description of the preferred embodiments of the present invention, similar proxy elements illustrated in the various figures will use the same or similar reference numbers, wherein the repeated description of these elements is omitted. . Fig. 1 is a schematic illustration of an apparatus for checking the amount of data, the means including means 102 for calculating hashes, means for comparison 103, and means for using 'data amount 1'. The device for checking the amount of data is configured to check the amount of data before use by using *partial hashing and comparing the hashes. The φ partial hash 値 and the comparative hash 与 are supplied to the device for checking the amount of data together with the data amount. The means for calculating the hash 1 is constructed to receive the amount of data 112. The data volume 112 is divided into a plurality of subsets (not shown). For the subsets of data volumes 1 1 2, a partial hash 値 1 1 4 has been calculated. A partial hash 114 is also provided for the means 1 to 2 for calculating the hash. Moreover, a comparison has been made from the portion of the hash 値 II4 which is also supplied to the apparatus for checking the amount of data. In the device for calculating the hash, it has been calculated from the amount of data shown in Fig. 2, for example, a partial hash! 4, with . and compare the hash 値 1 18 8, and with the amount of data is latched. The means 102 for calculating the hash is configured to calculate the common hash 从i丨6 from the partial hash 値 1 1 4 and supply it to the comparison means 丨03. The comparison means 103 is configured to receive the comparison 値i i 8 . From this part, the 値1 1 4 has also formed this comparison! 8. For the comparison of the composition of the hash 値 1 1 8 and the common mash 値丨丨 6, the same calculation is used. The comparison means 103 is configured to compare the common hash 116 and the -11-1307850 to compare the hash 1 1 8 . According to the present embodiment, according to the comparison result between the common hash 丨 丨 6 and the comparison 杂 値 181, the comparison means 1 〇 3 is configured to provide a comparison signal 1 2 0 to the means for calculating the hash 値. 1 〇 2. The comparison signal 120 indicates whether the comparison hash 118 corresponds to the common hash 116. When the common hash 値 116 coincides with the comparison 値 118, it can be considered that the partial 値 1 1 4 is not changed since the constituting the comparative 値 1 1 8 , wherein the partial 値 1 1 4 is formed from the partial 値 1 1 4 According to this embodiment, when the comparison signal 1 2 0 indicates that the comparison hash 値i 8 and the common hash 値 116 coincide, the means for calculating the hash 〇 1 〇 2 is configured as the data amount 112. The first subset is used to calculate the first other partial hash 値 122 and provide it to the comparison means 103. The comparison means 1 〇3 is also configured to receive the partial hash 値1 1 4 . In response to receiving another partial hash 値 1 22, the comparison means 103 is configured to compare the other partial hash 値 1 22 and the corresponding partial hash 値 1 1 4, wherein 'the partial hash 値 114 is from Formed with the same subset of another partial hash 122. Based on the comparison of the portion of the hash 値 114 and the other portion of the hash 値 1 22, the comparison means 系 3 is configured to provide an inspection result 1 24 to the means for using the amount of data. The check conclusion 24 indicates whether the portion of the hash 値 1 1 4 conforms to the other partial hash 値 1 22 that is accompanied. According to the same algorithm as the partial hash 114, another partial hash 122 has been constructed from the accompanying subset. Therefore, when the portion of the hash 値 114 conforms to the other portion of the hash 値 122, it can be considered that since the calculation of the portion of the mash 値 i 丨 4, the amount of sub-data is not changed, wherein the sub--12- 1307850 The amount of data is calculated by the other part of the hash 値 122. When the inspection conclusion 124 indicates that the partial hash 値 114 and the other multiplexer 値 122 are in conformity, the data amount means 1 〇 4 is used to construct a subset of the data amount 1 1 2 , wherein for the data amount 1 1 2, a Partially hashed 1 22 and compared in comparison means 103. In addition, the means 102 for calculating the hash 可 can be constructed as the other partial hash 値 122 of the subset of the quantities 112, and the parallels are provided to the comparison means 1 〇 3 as the result B of the comparison signal 1 20 . The means 102 for calculating the hash may be configured to establish the partial hash 122 first before the hash 116 and the comparison hash 118 have been compared. Partially entangled 値1 1 4 can be provided with the means 102 of the hash 分别, respectively, with the amount of data 1 1 2 . In addition, a portion of the hash 値 114 can be made in a total amount of 1 1 2 . The comparison 杂1 1 8 can also be provided with the amount of data 1 1 2 to the device for checking the amount of data. If a partial hash 値1 1 4 is provided for B to check the amount of data, only after a subset of the data amount 112 has been provided for calculating the hash 102 102, the means for calculating the hash 〇 1 〇 2 may be Another part of the subset is hashed 122. In this case, the common hash 116 is calculated when a partial hash has been received from the means 102 for calculating the hash. Before the inspection conclusion 124 has been received, the bow for the use of the data amount may be configured to have used the subset of the data amount 112 or the data amount 112, if the inspection conclusion 1 2 4 indicates that the partial hash 値 1 1 4 and one portion become Make him another information. And the other is formed by the means of calculating the combined capital when it is installed together, only 値114 three paragraphs 10 4 . In the case that the other 13-1.373050 part of the hash 値1 2 2 does not match or the common hash 値1 1 6 and the comparison 値1 1 8 does not match, the means 104 for using the data amount may be configured to interrupt the use of the The amount of data. Fig. 2 is a schematic illustration of a device for calculating a hash according to an embodiment of the present invention. In particular, the means for calculating the hash 构成 is configured to provide a partial hash 値 1 1 4 and a comparison 値 1 1 8 , in order to check the integrity of the data amount, 'the part of the 値 1 1 4 and the comparison 値 1 1 8 can be used to check the device for checking the amount of data shown in the 1 φ figure. The means for calculating the hash is to include a means 206 for calculating a partial hash, a means 207 for calculating a comparative hash, and an integration means 208. The device for calculating the hash is configured to receive the data amount 112. The amount of information 1 1 2 contains multiple subsets. The means 206 for calculating the partial hash 构成 constitutes a partial hash 値 1 1 4 from the subset of received data quantities 1 1 2 and provides it to the means 2 7 7 for calculating the comparative hash 。. Therefore, for each subset of the data amount 1 1 2, a partial hash 値 1 1 4 is supplied to the means 207 φ to calculate a comparison hash. By means of a means for calculating a partial hash 206 206 , the partial hash 値 1 1 4 is calculated from the subset according to a predetermined algorithm, and if the data amount 1 1 2 is not divided into subsets, it is used to calculate The partially hashed means 206 can include means for dividing the amount of data (not shown) into a plurality of subsets. The means 207 for calculating the comparison hash is configured to calculate and provide a comparison hash 1 1 8 from the partial hash 値 1 1 4 in accordance with a predetermined rule. -14- 1307850 Store and provide a partial hash 値 114 and a comparison 杂 1 1 8 for further processing. A partial hash 114 is provided to the integrating means 208 in accordance with the present embodiment. The integration means 208 is configured to receive the data amount 112, and the partial hash 114 is included in the data amount 112 and provided as a data amount 212 having a partial hash. According to the embodiment, the partial hash 値 1 1 4 • is inserted into the data amount 112 by the integration means 208 ', so that it is evenly distributed in the data amount - 212. In addition to this, the portion of the hash 値 114 can be arranged in the data amount 212 at the pre-φ position. If the portion of the hash 値1 1 4 is interspersed in the data amount 2 1 2, by checking the partial hash 値, according to whether another error has occurred in the storage device or the data amount 212, the high value may be high in advance. The rate is calculated for this part of the hash. The storage of the portion of the hash 114 in the fixed position of the data amount 212 has the advantage that, in subsequent inspections of the data amount 212, the portion of the hash can be read first and can be compared for the comparison hash 1 1 8 . In addition to this, the data amount 1 1 2 and the partial hash 値 1 1 4 are respectively stored as φ. The comparison hash 118 can also be integrated into the data volume 112 or can be stored separately. According to an embodiment, the hash is calculated using an encryption hash function. In accordance with this embodiment, a segmentation pattern check, such as a large file integrity of a ΜP3 format song, can be performed using a hash algorithm, hash or "hurried" check. The data is divided into segments, for example, data segments or access units corresponding to the subset of MPEG-4 encoded audio material. In the device for calculating the hash, the individual sub-segments -15 - 1307850 are respectively mixed to produce a hash. The hashes calculated from the segments are then stored, for example, in a table stored with the data. A new mixture of the hashed tables is used and used as the actual hash of the so-called main hash. The hash 値 corresponds to the partial hash 値, and the hash corresponds to the comparison hash. Sub-segments are divided into sub-segments of appropriate or expected size. In this, compromises are made between the granularity and the extra storage required for additional hashes. * After hashing the sub-segment, store the hash of the sub-segment in one or more - one table. The hash φ 値 is then formed for the table or for a table with partial hashes. The main hash can then be stored and referenced for later review. For the hash check in the device for checking the amount of data, the hash table is first checked in advance by the hash. That is, the hash for the table or the table with partial hashes is calculated and compared to the main hash. Use hashes from the table and check individual subsections. Until the end, the portion of the hash from the table was used to individually examine sub-segments of the file. After the main hash and the hash are equalized between the hashes calculated for the portion of the hash, the sub-segment is checked in this embodiment. In this way, the sub-segments that have been checked can already be used, while others are not checked. According to another embodiment, so many sub-segments are formed such that the table with partial hashes becomes too large. In this case, several hash tables can be used hierarchically. This means that the first table contains hashes, which can be used as the main hash for the subordinate table. As a special case, a continuous list can be produced. This means that the last hash in the table is the main hash of the next table, and so on. -16- 1307850 According to another embodiment, the hash table and the main hash are described for objects encrypted according to the DCF (DRM content format), wherein the DCF mostly includes audio encrypted in the MPEG-4 data format/ Video material. The DCF object is divided into the expected number of sub-segments, and the so-called data segments 'where' each combines one or more access units. Insert the data segment table with this information into the DCF object. The hashes are calculated for the data segment by the hash algorithm SHA-1. From all the hashes calculated, a table is formed' and inserted into the DCF object as an MPEG-4 primitive. For the table of the hash, the main hash is constructed, and the hash is stored outside the DCF object as a reference. This integrity is checked by reading the MPEG-4 primitive from the DCF object in a hashed table. A hash is then calculated for the table and compared to the main hash. DCF objects can be used when they are equal, otherwise DCF objects will be removed. If the DCF object is used continuously, then the DCF object is then searched for the expected data segment and the data segment is processed, for example, with the simultaneous inspection of the hash regeneration in the data segment. Also check the hash of the accompanying data section is called "hurry" check. If the hash of the data segment matches the corresponding flaw from the hash table, then another data segment can be processed. Otherwise, it will be due to DCF objects. The modification refuses further processing. In the previous description 'even though the calculation of the hash using the hash algorithm has been explained, it is apparent that the method of the present invention is not limited to the hashing effect, but may constitute any kind of hash or checksum. For example, co-location calculations can be implemented, and 'the method of the invention can be used for all applications that need to check the integrity of the data. Such applications can be, for example, computer systems or digital information transmission -17- 1307850 systems. For example, in computer systems' When the data is stored, the hash can be generated and stored together. In the subsequent reading and use of the data, the hash is also read and used to check the data. In the conventional system, the hash can be directly calculated before the data is transmitted. And then transmit the hash with the data and estimate it in the receiver. So 'can be sure it is correct Transmitting data 〇 ' For the integrity check of the segmentation aspect of digital information, the data is divided into 'subsets. The subsets can be independent or overlapable with each other. In particular, they can mutually decode the subsets independently or according to each other. Structurally analyzing the subset. The means for checking the amount of data may be part of the encoder for calculating the hashed device as part of the decoder. According to the condition, the invention method for checking the amount of data may be performed by hardware or software. And an invented method of computing the hash. The execution can be implemented on a digital storage medium, particularly a floppy disk or CD having an electronically readable control signal that can interact with the programmable computer system to perform the corresponding method. Generally, when a computer program product is executed on a computer, the present invention therefore also resides in a computer program product for storing the code on a machine readable medium for carrying out the method of the present invention. When the computer program is executed, the present invention can be implemented by a computer program having a code for implementing the method. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will be described in more detail with reference to the accompanying drawings in which: FIG. 1 is a block circuit diagram of a -18- 1307 850 for a smashed data volume according to an embodiment of the present invention; According to another embodiment of the present invention, a block circuit diagram of a device for calculating a hash is performed. [Description of main component symbols]
102 手 段 103 比 較 手 段 1 04 手 段 112 資 料 量 114 部 分 雜 湊 値 116 共 同 雜 湊 値 118 比 較 雜 湊 値 120 比 較 信 號 122 部 分 雜 湊 値 124 檢 查 結 論 206 手 段 207 手 段 208 手 段 2 12 資 料 量 -19-102 Section 103 Comparison Section 1 04 Hand Section 112 Material Quantity 114 Partial Miscellaneous 値 116 Common Miscellaneous 値 118 Comparison Miscellaneous 値 120 Comparison Signal 122 Part Miscellaneous 値 124 Inspection Conclusion 206 Hand 207 Hand Segment 208 Hand Segment 2 12 Information Quantity -19-