CN102196008A - Peer-to-peer download method, video device and content delivery method - Google Patents

Abstract

The invention provides a downloading method, and relates to the related technical fields of Internet technology, P2P transmission technology, content division and the like. When a user downloads a data file or multimedia program content from the internet, a content provider or a content sharer defines different segments in the same file as different attributes according to different attributes of specific content of the program, classifies the segments, and provides classified information for sharing to other users. When the user downloads the content from the internet, the downloading user can preferentially select the segment list according with the attribute requirement of the user according to the known classification information, preferentially download and use the segments, expand the downloading of the segment classification information according with the requirement of the user, and provide the segment-level right control and management strategy.

Description

Peer-to-peer download method, video device and content delivery method

technical field

The invention relates to the fields of Internet application, digital broadcasting, network transmission, P2P downloading technology, audio and video technology and the like. It mainly uses P2P network transmission technology, multimedia compression and transmission technology, etc.

Background technique

With the popularity of the Internet, Internet users can share rich data and multimedia resources on the Internet. A commonly used method of data sharing on the Internet is that the content provider places the content on the server, the user connects to the server where the content resides from the client, and downloads the shared file located on the server from the server. Due to the sharing of more and more multimedia content on the Internet, the transmission of large amounts of data has caused a large amount of network transmission resources to be occupied, which not only affects the smoothness of Internet transmission for other users, but also affects the downloader's local resources and server resources. There is a lot of consumption. For example: a large amount of audio and video resources occupy a large amount of server and client storage space, and it takes a long time to download the entire content file. The content of network sharing varies from good to bad. In many cases, due to the lack of understanding of the file content in advance, the downloaded file may be wrong or fail to meet the requirements. Sometimes the download cannot be completed due to the interruption of the server during the download process, which affects the user's use. Therefore, a peer-to-peer (P2P: Peer to Peer) download technology provides a one-to-many download method to solve the load on the server and optimize the rational use of transmission resources for the file sharing of large amounts of data.

The principle of P2P downloading is: the content provider divides the content into fragments according to certain rules, divides the content into multiple fragments of a certain size, and then generates an index file based on the division rules of the fragments and the size information of the fragments. file, the download client can download the fragments in units, and reassemble these fragments according to the index file information to generate a file consistent with the original file. Taking BT (BitTorrent) as an example, its index file is the seed file of BT. The size of the seed file is very small compared to the target file to be downloaded, and users can easily download it from the Internet through traditional methods. In addition to segment division rule information, the BT seed file also includes tracking server (Tracker server) address information and the like. Content providers upload torrent files to the Internet for other users to download. After an ordinary user downloads a torrent file from the Internet, he accesses the Tracker server according to the content of the torrent file, and then the Tracker server provides a list of clients having the target file corresponding to the torrent file. The download client visits each client that owns the content according to the list, and requests them to download a segment in the target file. Finally, when all the segments are downloaded, the download client assembles these segments into The file that is consistent with the original download target file completes the download process of the entire target file.

In P2P downloading, the download client can connect with multiple content-owning clients at the same time to realize one-to-many download, and according to the network transmission optimization algorithm, the client with the fastest transmission speed between the download requester and the download requester will be given priority Data requests are made there for maximum efficiency in network utilization and shortest download times. Due to the one-to-many mode of P2P downloading, the download speed is fast and the reliability is high, and now it has become one of the main download methods on the Internet.

However, in P2P downloading, it is required that the download requester must download the entire file before he can use the file normally. However, in actual situations, it often happens that some fragments in the entire file cannot be successfully downloaded and the entire file cannot be used because the number of owners of some fragments is very small or they are offline. At the same time, since a complete file needs to be downloaded in P2P downloading, the amount of download is fixed, and the network traffic that each user pays for this is also fixed. As a result, users have to pay for the network traffic of the entire file even if they only need part of the content. Fees and download waiting times.

At the same time, with the deepening of copyright awareness, the definition of the rights of files shared on the Internet is becoming more and more detailed. Content providers hope to define the scope of users' rights in more detail through detailed definitions of copyright protection. Specifically, for a movie file, although the entire movie cannot be released to all users for free due to commercial reasons, some parts of it, such as clips, content previews, and advertisement clips, etc., the content provider hopes to provide free of charge. To all users, in order to attract more people to know the content of the film and make reasonable commercial downloads. The current solution is to extract this part of the content independently as an additional file, and release it along with the movie or independently of the movie file. Free users can download this part of the flower or trailer, but cannot download or play the complete movie. . This virtually increases the number of downloaded files and the amount of data downloaded and transferred. Similarly, different content in the movie may be suitable for different users to watch. For example, some content involving horror and violence is not suitable for minors to watch. Distribution will inevitably lead to complicated management and greater download traffic, which is inconvenient for program production and user download.

In short, in the current P2P downloading, the download speed is prioritized to optimize the network transmission performance, and it does not fully consider the download requester's demand for content and the download requester's network usage status. All users need to complete the download of the target file. . At the same time, it cannot provide any personalized advice on how to use the downloaded target file. The current P2P download method is to give priority to speed, but does not fully consider the degree of demand of each segment on the user.

In order to make up for some limitations of the current P2P download, many new implementation methods have expanded it to support special applications. For example, in the case of streaming media watched in real time, it is necessary to schedule the priority of the clips according to the order in which they appear in the streaming media, so that the clips with more urgent needs have the priority to download even if the download speed is slow and the urgency is low. Fragments that have a higher download speed are also set to a lower download priority. This type of solution is a known example: P2P protocol-based media file playback control method and device [CN101227590A].

This known example discloses a P2P download control method and device for on-demand media files. , preferentially downloading the control information segment where the file control information is located and the index information segment where the index information is located, so as to implement downloading and playing of the media file under the P2P protocol. The key point is that according to the current playback position of the streaming media, the segment closest to the current playback position is preferentially downloaded, that is, the priority download order of the segments is determined according to the time distance of the current playback position.

In addition, as another known example [Japanese Patent Application Laid-Open No. 2001-344242], a video device is disclosed that receives a part of content from a server and reproduces it, and then receives the entire content as necessary. For example, in this patent document, a portable terminal device is described. After receiving a trailer of a movie from a server, if the user indicates that he wants to watch the full movie, he starts to receive the full movie independent of the trailer.

Invention content:

The known example [CN101227590A] provides a media file application for audio and video on demand, according to the characteristics of its application, the priority order of segment downloads in P2P downloads is dynamically adjusted and downloaded, which solves the problem that ordinary P2P cannot meet the actual demand for on-demand services Some deficiencies have effectively ensured the real-time performance and fluency of playing media files in the on-demand service. However, this known example cannot provide personalized downloading and viewing methods for different users.

In some streaming media applications, such as the download speed of the client is limited, even the optimized download sequence or continuous download cannot guarantee the normal playback of the streaming media. Pause of the video. At the same time, for non-streaming media files, it is also necessary to download all the file fragments, and the downloading user can reassemble all the fragments according to the order of the seed file to form a complete file before starting to use the file. And before all the fragments are downloaded, the file still cannot be used, or there are certain problems in use. In view of the characteristics of audio and video media files, the content that users pay more attention to is often part of the content. For example, in many cases, for a movie, the user may prefer to watch the highlights, or hope to download the highlights for viewing first, shorten the waiting time for downloading all content, and then choose according to the needs in free time. Watch other parts. Or you want to download the preview part of the content to get a general idea of the storyline of the film, or to confirm the content and picture quality of the film, etc. These requirements of being interested in downloading part of the target file actually exist, but the current P2P technology cannot provide a good solution. Therefore, the present invention proposes a P2P downloading method, which supports the download and use of some clips, and also supports the user to customize the essence of the movie after watching it, so as to watch it repeatedly or recommend others to watch it. It also provides different downloading and playing levels for different users, which is conducive to parental control or copyright protection.

Therefore, an object of the present invention is to provide a P2P downloading method, so that downloading users can selectively download and play required content according to their specific needs.

In addition, the video equipment disclosed in the known example [Japanese Patent Application Laid-Open No. 2001-344242], before receiving a complete movie, first understands the story outline of the movie, and then judges whether to receive the complete movie, so it is not necessary to receive the complete movie first. However, this video device has the following problem: After receiving and reproducing partial content such as a trailer, if the complete movie is received again, since the part of the movie content belongs to both the part of the trailer and the part of the complete movie, it is difficult to download it again. This partially overlapping content will waste communication resources and communication time. Therefore, another object of the present invention is to provide a video device and a content transmission method capable of saving communication resources and communication time when receiving content such as movies.

A kind of P2P downloading method provided by the present invention is characterized in that: based on the personalized content definition that carries out personalized division to content, carry out personalized download to the target file that is divided into a plurality of fragments, described method comprises the following steps: A ) The download client sends a download request to the content provider, and downloads the fragments belonging to the specified personalization level in the target file; B) The download client combines the downloaded fragments to form a personalization that meets the requirements of the personalization level content

A kind of video equipment provided by the present invention is characterized in that: based on the personalized content definition for personalized division of content, the target file divided into multiple segments is downloaded and played personalizedly, including: a downloading device, The provider sends a download request to download segments belonging to the specified personalization level in the target file; the combination device combines the downloaded segments to form personalized content that meets the requirements of the personalization level; The personalized content combined by the combination device is played.

Another video device provided by the present invention, after receiving a part of the content of the target file via the network for reproduction, receives another part for reproduction, and is characterized in that: a part of the content of the target file is contained in the content of the target file A plurality of content fragments in the content fragments of the target file, the other part is a part or all of the content fragments of the target file except a part of the content of the target file.

Another video device provided by the present invention, after receiving a part of the content of the target file via the network for reproduction, receives another part for reproduction, and is characterized in that: the content includes a plurality of content fragments, and an There is segment header data, the segment header data includes attribute information representing the attribute of the content segment, and when the segment header data received by the video device contains the specified attribute information, the received additional After reproducing these received content fragments, receive the content fragments whose fragment header data does not contain the specified attribute information, and for the content fragments containing the specified attribute information and those that do not contain The content segment of the specified attribute information is reproduced.

Another video device provided by the present invention receives and reproduces a part of the content of the target file via the network, and then receives another part for reproduction, and is characterized in that the video device can be connected to a plurality of unspecified terminals to perform the reproduction of the content The content includes a plurality of content fragments, and these content fragments are stored in a plurality of unspecified terminals, and the video device stores a plurality of attribute information representing the attribute of the content and the content contained in each attribute information The segment number network server receives the content segment number contained in the predetermined attribute information, receives the address of the terminal having the content segment number from the management server that manages the content segments owned by the unspecified plurality of terminals, and receives the content segment number from the management server. After the terminal at the address receives and reproduces the content fragments, it receives a part or all of the content fragments except the received content fragment serial numbers for reproduction.

A method for transmitting content provided by the present invention, the video equipment that reproduces a part of the content of the received target file and then reproduces the other part transmits the content via the network, characterized in that: the content contained in the target file The header of the content fragment is appended with fragment header data, and the fragment header data includes attribute information indicating the attribute of the content fragment.

The invention described in this article is based on the basic principle of P2P, by defining the segment attributes of specific files according to the content contained in the segment itself, different segment combinations are generated, and these combinations are passed through the seed file, Tracker server, segment attribute definition or other methods, recommend the definition distribution of segment combination to download users. Downloading Users can refer to personalized fragment combinations according to their specific needs, and selectively download the desired fragments. This method provides a humanized and personalized download method, optimizes the download priority of download segments, reduces the amount of network transmission data, reduces the space required for local storage, and speeds up the download speed. Methods of downloading, playing and managing.

Moreover, using the video equipment of the present invention can save communication resources and communication time when receiving content such as movies, and improve communication speed.

Description of drawings:

Figure 1: Basic principle of the invention

Figure 2: The main components of a standard BT seed file

Figure 3: The P2P download method shown in this method

Fig. 4: the main composition of BT seed file shown in the present invention

Figure 5: Introduction to MPEG frame structure

Figure 6: The data size of an MPEG frame and the segment division in P2P downloading

Figure 7: Sequence of downloaded clips and content during playback

Figure 8: Download client workflow 1

Figure 9: Definition of the torrent file

Figure 10: Fragment with Fragment Control Header

Figure 11: Fragment structure definition

Figure 12: Download client workflow 2

Figure 13: The download client makes a content customization request in the form of attributes

Figure 14: Information that can be included in a fragment descriptor

Fig. 15: A schematic diagram of a video device applying the present invention.

Detailed ways

The present invention is based on the existing basic principle of P2P, that is, divides the target file into segments, and reduces the download unit from the file as the unit to the segment as the unit for downloading. This method can realize downloading different fragments from multiple content owners for a file at the same time, and realizes a one-to-many downloading mode, thus improving the downloading speed.

BT download is one of the foundations of the present invention, first of all we briefly introduce its principle. In BT download, as the key index information of file fragmentation, the role of seed (seed) file is very important, which stipulates the file information, file name, directory name, length, etc. of the target file to be downloaded, and then specifies the target The fragmentation status of the file (such as the segment size piece_length) and the check code (hash value) of each segment are used to verify the integrity and correctness of the downloaded segment data. Simultaneously, the Tracker server (announce tracker) that downloads the client and obtains the client list that has the content of the target file is also defined in the seed file. Through the server, the download client can obtain a list of other users who own the content, as well as the owner of each segment in the target file to be downloaded. Through these information, the download client can directly request other clients to obtain the specified target file fragments, so as to realize the parallel download of different fragments.

In BitTorrent download, information exchange is carried out between the data requester and the data provider through the BitTorrent protocol. The basic principles of BT can refer to the official agreement definition of BT. Here we will list some of the important content of the interactive information protocol.

Have information, used to indicate the currently downloaded and verified data segment ID

Bitfield information, which is used to indicate the location bitmap of all fragment sets currently owned locally, and is used to indicate the state of fragments owned locally.

Request information, which is used to download the requested download information from the requesting end to the content provider, including the requested segment ID, the starting position in the requested data segment (the offset relative to the segment header) and the length of the requested data to be downloaded.

The Piece information is used by the content provider to provide the data it needs to the download requester.

Through the above main information, the download requester and the content provider can know each other's fragments and can transfer the actual data of the fragments.

Based on the above BT download principle and protocol brief introduction, the present invention will provide a P2P download mode, support the file fragment download mechanism, and support the combined download and playback of different fragments at the same time, by defining different fragments according to their specific content attributes, the huge It facilitates various applications based on file fragments.

In the present invention, by dividing and playing the fragments according to different content levels in the seed file, it is realized in multiple ways, and the fragment list corresponding to each personalized playback level can be directly defined in the seed file, or from the file The offset of the header, or define the segment attributes used for personalized playback. By cooperating with the segment attributes in the segment, the segment sequence that matches its attributes can be effectively selected for download and playback. Therefore, in the present invention, there are many specific implementation methods, which can be defined by defining a personalized seed file, which can be an additional personalized content of the seed file, or a user-defined sequence of fragments, or from other users or content providers. Or, the recommended information on the network includes the content of segment classification information in the target file.

After obtaining the fragment list, the download client obtains the fragment list in the target file that needs to be downloaded according to its selection of the personalization level (Profile), and then only downloads the fragments defined in the personalized profile that belong to the user's selection, while Not all fragments belonging to the object file are downloaded at all. During playback, these downloaded segments (or all the segments in the target file) are rearranged according to the definition of segment sequences in the profile to form part of the content in the target file that conforms to the definition in the profile, Then it is presented to the user in the form of a file (and in common BT download, it is presented to the user in the form of a target file only after all the segments have been downloaded). The user can play the content in the target file defined by the profile.

For example, in Figure 1, profile-1 defines "trailer". During the download process, the download client requests the content provider to download the fragment list belonging to the "trailer profile" according to the segment list defined in the "trailer profile". , which are then assembled into the trailer content of the target file. The content provider still exists in the form of a complete target file, and does not need to provide an independent trailer file to occupy additional space. The download client can rely on the definition of the profile to generate various forms of content level files from the target file. , and at the same time it does not occupy additional disk space, but the target file exists in the form of a fragment library, and some fragments are taken out for playback according to the definition of the profile during the download and playback process.

If the user selects a complete download (also a division level of the personalized profile: complete download), all segments belonging to the target file will be downloaded and stored, which is similar to the existing common P2P download method. Therefore, the invention is fully compatible with existing download methods, while providing more diverse personalized download and playback controls.

It should be noted that although the present invention is based on the BT download principle, its application scenario is not limited to the BT download process, and the method described herein can be fully utilized in the entire P2P download mode (including BT download, emule Download and point-to-point download methods such as Thunder, Express, etc.), and similar network transmissions, can also be used in audio and video broadcasting and downloading, and other downloading methods in units of packages or segments, and can also be used in In network transmission or broadcast signal transmission for data exchange in units of fragments or packets, such as IP network or digital broadcast signal transmission. The method manually or automatically selects part of the packages or fragments for downloading according to the user's selection or depending on the current transmission situation, and comprehensively considers various factors to provide the user with multiple personalized selection methods. It should be pointed out that the selection of content packages or fragments can be specified by the user, recommended by the server, recommended by the content provider or content owner, or automatically recommended by the system according to the current situation, etc. etc., can be pre-defined or can be changed in real time.

Figure 1 shows the basic principle of the invention. In Fig. 1, the content provider publishes the original file that wishes to share, and for the download requester, it is the target file that wishes to download, including all the content in the target file, according to the P2P download method, firstly the target file Carry out fragment division processing, divide the original file provided by the content provider into several fragments of equal size (for different network applications, the division rules of packets or fragments are different, and the divided fragments may also have different sizes). In the standard BT download method, these fragments will be connected through the seed file, and the download client will download the content of different fragments from different content providers at the same time, and then perform verification and combination according to the definition in the seed file, and finally form the same The file provided by the content provider is consistent with the target file, enabling the complete download process.

However, in the method of the present invention, the content provider or the Tracker server provides multiple downloading or personalized selection modes including playing to the downloading client, such as profile-1, profile-2, etc. as shown in FIG. 1 . Different profiles define different combinations of content fragments, and these fragments are also part of the original file provided by the content provider. For example, the attribute of profile-1 is the trailer of the complete target file, which includes the first and second fragments, etc., while other fragments are not included in the definition of profile-1. Also as shown in profile-4, most of the clips belong to the definition in profile-4, and some clips are not included in profile-4 because the content may not be suitable for minors to watch. The content corresponding to these profiles does not contain all the fragments in the target file, so if the download request client chooses these download methods, it only needs to download the fragment content defined in the selected profile to enjoy all Selected profile content, and play to enjoy. According to this method, users only need to download part of the content to enjoy the playback mode of their selected level, which greatly reduces the network bandwidth for downloading, the waiting time for downloading, and the downloading fees paid for it. . Depending on how the selected grade is downloaded/played, users can subjectively judge whether the target file is worth a full download based on these partial contents. If the user selects part of the download according to the initial grade, and wants to fully enjoy these movies after playing, he only needs to re-select the Profile with richer content to download, and at this time, because some clips have been defined for the previously selected To download, at this time, it is only necessary to perform a simple differential comparison between the profile required now and the previous profile, determine the additional content segments that need to be downloaded, and download these additional segments. After the download is complete, it will be reassembled according to the newly selected profile to provide users with brand-new appreciation content.

In the definition such as profile-3, it contains all the content fragments belonging to the original file, and at the same time some fragments are re-appended at the end of the file, so profile-3 needs to download all the fragments in the original file, and at the same time When playing, it provides additional exciting memories, and these clips belong to the original file, so there is no need to add the size of the downloaded content, and it provides a more free way of enjoying during playback.

Using the downloading method described in the present invention does not increase the size of additional downloading content but provides a variety of freely selectable downloading and appreciating levels, reduces network occupation, download time and local disk space occupation, and provides users with more downloading and appreciating levels. way more free. In the common P2P downloading method, different levels of appreciation content need to download different files respectively, and most of the content in these files may be repeated. Using the method described in this invention, repeated downloading of data should be avoided as much as possible. And provide a variety of personalized download / playback options.

In the present invention, a P2P downloading mode for individualized selection of downloading/playing segments is provided. Although the original P2P download method divides the target file into segments, there is no segment-based division method when downloading and playing, but still needs to download and play the entire file, that is, the download/play method with the file as the unit. However, in the present invention, the way of downloading and playing is to use the segment downloaded by P2P as the unit to download and play the segment personalized selection, and these personalized customization information directly define the attribute information with the segment as the unit, that is, Profile Definition. And in the present invention, do not limit the definition mode of these personalized fragment attribute information, can be directly defined in the BT seed file, can be inserted before or after or among the fragment content as extra information of each fragment, also It can be embedded directly within the content of each segment, or exist as a separate file. The information defined for each segment attribute can be a part of the seed file distributed to each download user along with the seed file, or it can exist independently of the seed file, through the content provider, Tracker server or provider. Downloads are provided by other clients, or are defined by the downloader, or can be implemented by the user based on other information (such as comments on the Internet or oral recommendations from other users, etc.).

Fig. 2 shows the standard form of the BT seed file taking BT as an example, which defines the Tracker server address, the downloaded directory and file name list, the segment length divided by each segment, and the check code (Hash value) of each segment )etc.

Through this BT seed file, the user can get the address of the Tracker server, and access the Tracker server through this address (there can be multiple Tracker servers). Then the Tracker server provides the download client with the address list of other clients that own the target file and its useful segment status belonging to the target file. The download client can access other clients through this list to connect, and then perform the actual segment data exchange.

Fig. 3 shows the connection status of the general BT download mode, which is also one of the connection methods shown in the present invention. The download client obtains the seed file of the target file (or folder) to be downloaded from the Internet through other network methods (HTTP, FTP, etc.), and the structure of the seed file is defined in Figure 2. In Fig. 3, the download client (Downloaderclient) first connects to the Tracker server shown in the torrent file through the BT torrent file that has been downloaded to the local as shown in Fig. 2, and obtains other content providing customers who have the content through this server end list (Content Provider-1, ..., n), and then the download client connects to other content provider clients respectively, and then requests the content provider client to download segment data according to the BT protocol. The length of each segment and the hash value used for verification are stored in the BT seed file. Every time a complete fragment content is downloaded from other clients, it is verified by the verification hash value corresponding to this fragment to determine the correctness of the fragment. If it is correct, the fragment file downloaded locally is marked as a successful download. If the verification is wrong, the segment needs to be downloaded again.

Through the BT protocol mentioned above, the information exchange between the download client and other content provider clients can have a variety of information content, including the status of the fragment list owned by the download client/content provider client, and the request The segment ID, and the specific data in the requested segment (that is, the Piece protocol in the BT protocol, including the starting position in the segment, the length of the requested data, etc.). Similarly, the content provider client sends the specific content of the requested segment information to the download client. Then the download client combines the contents that belong to the same segment, and then performs hash verification. After the verification is successful, the entire segment content is stored in the local storage and the actual segment ownership status is updated. The download client can apply to multiple content provider clients for content belonging to different segments at the same time, and can also apply for content belonging to different locations of the same segment at the same time, so as to realize simultaneous downloading of multiple segments of content.

It should be pointed out that the download client and other content provider clients may become content download clients or content provider clients at any time as needed, and update their status information to the Tracker server at any time/regularly as needed, so that the Tracker server It is possible to grasp the current list of clients that own the content and their status in the entire network.

FIG. 4 shows an implementation method of the method of the present invention, that is, an implementation method in which the content classification information of the target file is embedded in the BT seed file. As shown in FIG. 4 , the seed file shown in the present invention includes all the content in the standard BT seed file ( FIG. 2 ), and the profile information is defined in the seed file at the same time. Two Profiles (Profile-1 and Profile-2) are defined in FIG. 4 , which respectively represent two different personalized options of "trailer" and "highlight chapter". Afterwards, some specific attributes of the entire personalized segment are defined, such as the time length of the personalized segment (or the length in bytes, etc.), and the suitable audience (Audience). The most important thing is the segment list corresponding to the personalized content (ie, the profile), and the segments described here are consistent with the segments shown in FIG. 1 and the standard BT. However, the difference from the standard BT seed file is that the standard BT seed file contains all segments of the entire target file, that is, a total of n segments from 1 to n (the last segment). However, in the method shown in the present invention, its fragments are not all n fragments, but a combination of multiple fragments, many of which may be discontinuous, such as not including the third in profile-1. fragment. Similarly, other profile definitions may only contain some of the n fragments.

Simultaneously, in the BT seed file shown in Fig. 4 the present invention, may also include a definition (piece_playback_offset) for the state when this profile is played, as Fig. 4 defines for each segment to be used for playing under this profile state Time on-chip data. For example, for the first segment, the offset in the segment is 100, that is, the content of 0-100 bytes in the segment is not used for playback, while for the fourth segment, the offset is -100, that is, the content at the end of the segment 100 bytes are not used for playback. Similarly, it is also possible to skip the front and back parts of some segments and only play some data in the middle of the segment. For example, in the xth fragment, only the data between 200 bytes and -300 bytes will be played.

The reason for doing this is: First, in order to maintain consistency with the P2P download protocol, the segment is the smallest unit of P2P download, which is downloaded and used for Hash value verification to determine whether the entire segment is completely downloaded, and it cannot be realized in the segment. Random part of the content download checksum. Secondly, considering that BT's division of file fragments is not based on content, but only divided into equal sizes according to the length of the file, and the fragment containing the required data may be located inside a certain fragment, and some of the content before this fragment is not There is no actual playback purpose, and BT download requires downloading the entire segment, so it is necessary to define the download segment and the offset of the content in the segment due to the playback content in the download segment, so as to facilitate access during playback. At the same time, in many audio and video files, there is a certain correlation between many video frames according to the codec requirements. For example, in MPEG video coding, the playback of P frames depends on the previous I frames, so the I frames must be included In the specific playback, it may not start to play from the I frame, but directly from a certain P frame or B frame, but the data of these I frames is still needed, but directly across the frame during playback The I frame is played only. Therefore, there is a certain relationship between personalized download and personalized playback, but not necessarily completely consistent.

Give an example to illustrate the function of piece_playback_offset. Fig. 5 shows a brief introduction to the frame structure of the MPEG encoding method commonly used in current video. In the MPEG system, video frame data basically includes three types, namely, I frame, B frame and P frame, wherein I frame is an independently coded frame, B frame and P frame are backward prediction frame and forward prediction frame respectively, and B frame and P frames are non-independent coding frames, that is, the complete data of B frames and P frames need to rely on the data of I frames for restoration. As shown in Figure 5, during video playback, I frame, B frame and P frame are mutually intersected to form video frames with equal time intervals and output to the display screen. However, there is a big difference in the amount of data in each frame. As shown in FIG. 6 , since the I frame is an independently coded frame and can be completely displayed without relying on the data of other frames, the data volume of the I frame is larger than that of the B frame or the P frame. However, if B frames and P frames need to be restored to complete frame data for display, they depend on their related complete I frame data. Therefore, if there is no corresponding I frame data, B frames and P frames cannot be decoded. cannot be fully displayed. Usually in this case, depending on the strategy of different devices, there will be a large mosaic on the screen or no display at all.

In network transmission, especially in P2P transmission based on the equal-size fragmentation mechanism, network transmission data does not care about the internal frame structure in the encoding process. In this case, if there is an I-frame data that is not completely downloaded, then It will cause other frames associated with it to be unable to be decoded and displayed, that is, unable to be decoded and displayed for a period of time. If according to the existing P2P download method, if some of the clips are not downloaded, it will cause many clips associated with it to be unplayable, and in many cases the entire video cannot be played. One of the reasons why the entire video cannot be played due to downloading. As shown in Figure 6, the boundary between Piece-1 and Piece-2 is divided inside the 7th frame. If Piece-1 is not successfully downloaded and Piece-2 is successfully downloaded, the I frame labeled 1 will not be downloaded. Downloaded to the local, affected by it, 2, 3, 4, 5, 6, 7, 8, 9, 10 that rely on the I frame data labeled 1 to be decoded cannot be decoded and played. Due to the successful download of piece-2, the I-frame labeled 13 can be decoded, and the frames 11, 12, and 14 that depend on the existence of the I-frame labeled 13 can be decoded and displayed normally.

Of course, there are many reasons why the playback cannot be played, such as related data errors due to failure to download the verification data, or partial or entire video files cannot be played due to incomplete data in other encoding methods, and so on.

From the relationship between segment division and actual video frame structure shown in Figure 6, it can be seen that in piece-2, although the previous frame data such as 7, 8, 9, and 10 are downloaded along with piece-2, they are not It can be decoded and displayed, so if you specify the playback offset playback_offset in the piece, you can let the player automatically skip the data that cannot be decoded and displayed normally, and decode and display from the 11th frame without causing any harm to the player error impact. Therefore, by defining the in-slice playback offset, the data used for playback in the segment defined in the downloaded personalized content definition can be reconnected to form a playable personalized content file. That is to say, these clips are not simply superimposed, but the data that can be decoded and displayed normally are organically connected to form a personalized content that can be played.

As shown in Figure 7, personalized content customization requires X, Z, M, and N fragments to be downloaded, and at the same time, the fragment playback offset makes only part of the content of these fragments available for playback, as shown in Figure 7. The content section to play. However, the "downloaded data not used for playback" in these segments will also be saved locally for data verification in units of segments and future file expansion.

Figure 8 shows the basic flow of personalized downloading and playing of the download client. First the user has obtained the individualized BT seed file from the Internet or in other ways, then downloads the client software to open and read the individualized BT seed file (S01); and identify (S02) according to some modes defined in the seed file; Can identify some personalized information defined in Figure 4, and then provide these personalized options to the user for selection (S03), such as the user can choose Profile-1 (trailer) or Profile-2 (excellent chapter), etc. Etc. Afterwards, the download client downloads according to the segment list defined in the Profile selected by the user (S04); after each segment has been downloaded, it is verified according to the Hash value of the corresponding segment contained in the BT seed file. If it is correct, it means that the segment download is successful. After all the segments corresponding to the profile are successfully downloaded, these segments are combined according to the order and mode specified by the Profile (S05); then according to the profile defined in the personalized seed file The playing mode is played (S06), realizing personalized downloading and personalized playing. Since the personalized definition of the segment corresponds to a very small amount of data corresponding to the data of the segment itself, this personalized definition method has very little impact on the BT seed file, and is fully compatible with the existing BT download method, and does not affect the Downloading the object file itself has no effect.

As the second implementation of the present invention, as shown in Figure 9, the specific fragment content classification belonging to each Profile in the seed file is defined in the form of fragment attributes, such as the fragment attribute of Profile-1 is A, and The segment attribute for Profile-2 is B. In the actual target file division process, by embedding piece attribute control information inside the piece content, as shown in Figure 10, adding a piece descriptor (Piece_descriptor) before the specific data of each piece, thereby increasing the more specificity of the piece. Multiple descriptions allow each segment to be described individually according to its content. For example, segment 1 belongs to attribute A, segment 2 belongs to attribute B, and segment n may belong to both attribute A and attribute B. In this way, the structure definition of each fragment is similar to that shown in FIG. 11 . That is, the fragment descriptor is added before the specific data of each fragment.

In the specific downloading process, as shown in FIG. 12 . The download client first reads the personalized BT seed file (S11); and identifies the personalized BT seed file (S12); by identifying the grading of the content in the BT seed file, a selection strategy (S13) to the content is provided to the user; After the user selects, determine the specific attributes of the segment to be downloaded. Then the download client needs to download the segment descriptor (S14) of each target file segment, obtain the specific attributes of each segment, to determine which segments (S15) the user wishes to download in the actual target file; then according to (S13) The user's selection and the actual attribute definition (S15) of each segment are matched to determine which segments the user wishes to download and which do not wish to download (S16); then the segments wishing to download are downloaded and combined (S17 ), and for content that does not belong to the Profile level selected by the user, the download of the remaining specific segment data is abandoned (S19). Finally, play according to the method defined by the Profile (S18), thereby realizing personalized download and personalized playback.

It should be pointed out that the data volume of the piece descriptor (Piece_descriptor) is very small relative to the specific content (Data) of the piece. For example, the piece_descriptor may only have a few bits or bytes in length, while the specific piece content data may be several megabytes in length, and the piece descriptor is almost negligible compared to the specific data, so this method is suitable for The division of fragments has almost no impact, and the increase in the amount of downloaded data is negligible. Alternatively, such segment attribute control information can be directly embedded into the original content (data) of the segment, that is, occupying several bit positions in Data to describe the properties of the segment, so that the division method and content of the original target file will not be increased at all. Size, and embedded control information for user's personalized download and personalized playback, although this method of embedding in data has a little modification to the original file. In audio and video files, because of its specific reality and playback effect, there will be no obvious changes in picture and sound quality due to the change of several bits, so this method is completely allowed in the download and playback of audio and video files. in use.

In the method of the present invention, the transmission protocol used in the actual transmission process can also adopt the transmission protocol used in BT transmission. For example, do the same:

1) State exchange protocol: the current local has successfully downloaded and verified the correct block list;

2) Data request protocol: the download request end will wish to download the ID of the block in the target file, the internal data start position of the requested block and the length of the requested data;

3) Data transmission protocol: the specific data in the target file transmitted by the content provider to the download request end, including the ID of the block in the target file, the starting position of the internal data of the transmitted block and the length of the requested data, and then the specific block internal data content.

The three protocols mentioned above are standard protocols in the BT transfer protocol, and are also applicable to the use of this method.

In the first implementation method of the present invention, due to the expansion of the BT seed file, more information can be provided in the seed file. The seed file mentioned here can be the expansion of the seed file as shown in Figure 4, that is, the definition of personalized Profile information is added in the traditional seed file, or it can be a traditional seed file, through some Profile information defined externally (It can be in the form of a file, or some settings) instead of inside the seed file, or a part defined inside the seed file, while other parts are defined outside the seed file and so on. Some personalized Profile content is defined through these additional information, and these personalized content can be identified by the download tool used by the download client, and converted into a combination of various blocks, so as to provide different content to content owners according to different needs. Data requests to meet the needs of different customers or different levels. This requires the download client to support selective block data requests based on personalized information or personalized user configurations, reassemble the received data according to personalized needs, and support the exchange between different personalized configurations. switch.

As another form of realization of the second method. For Figure 10, in addition to the content download request client first downloading the fragment descriptor of each fragment for its own judgment, the request attribute can also be transmitted to the content providing client through the request protocol, and the content providing client receives the When requesting the content of attributes, first match the requested content attributes with the attributes recorded in the fragment descriptors in all fragments, then sort out the sequence of fragments that meet the requirements of the content requester, and then send the content of these fragments to the content requester . In this way, the judgment of the content of the fragment is completed on the side of the content owner.

As shown in FIG. 13 , since different content attribute codes have been marked in the personalized content definition to correspond to different content categories, for example, "A" stands for "content introduction". If the user wants to download the content belonging to the "content introduction", he only needs to notify one or more content providers of the attribute "A" through the network (S21), and the content provider can know the content after receiving the request information from the download client. The content provider needs to download the demand of the requester, and the content provider already has the specific content of some fragments, so the content provider can check and match the demand of the download requester with the data it owns, and obtain the content it owns. A segment list (S22) that can be provided to the download requester and meets the requirements of the download requester (S22), and then this list is delivered to the download requester (S23). In this way, the download requester can request the fragments in these lists from the content provider or other content providers through the sequence number list of fragments described in the list (S24). Then one or more content providers transmit the data requested by the download requester to the download requester (S25). The download requester reassembles the fragments belonging to the attribute "A" to generate the required content belonging to the attribute "A" (S26).

For the piece description method described in FIG. 10 , the piece description information (piece_descriptor) is added on the basis of the piece data (Data part). As for the added position, it may be added before the original segment information as shown in FIG. 14 , or it may be other positions, such as the end of the segment. Since the segment description information occupies very little space compared to the segment data, it can also be embedded in the original segment data Data. For the length of the segment, it can be the total length of the new segment after adding the segment descriptor, or the original length. This strategy can be defined according to the data part definition corresponding to the check value in the seed file. It is convenient to verify the downloaded data. For example, what is described in the seed file is only the verification of the original data Data, although the complete segment includes the segment descriptor and the original segment data, and the verification is only for the original data. Similarly, if the verification corresponds to the combination of the fragment description information and the original data, the unified verification is performed after combining the two.

In the method described above, the original P2P fragmentation mechanism is extended, that is, content control is carried out in units of fragments. This control information can be embedded in the original distribution file by the content provider, or it can be carried out as an additional file. Download, that is, the content provider provides 1) the original file that only contains all the segment data Data, which can be downloaded through P2P, and also includes 2) the auxiliary file of the segment description information set of each segment, or the auxiliary file of this segment description information set Due to its small file size, it can be downloaded through P2P or other methods, or delivered to users through torrent files, that is, the torrent file not only describes the check value of each segment, but also contains the The fragment description information for .

Figure 14 shows the specific usage method of the piece description information (Piec_descriptor). In the figure, the segment description information is located at the front end of the segment specific data Data (as mentioned above, its position can be other positions or embedded in Data). Its content may include a variety of information, such as 1) content classification information, 2) copyright control information, 3) other information, etc. as shown in the figure. The content classification information is the content division method shown in the present invention, for example, attribute A represents trailers, attribute B represents exciting chapters, and so on. The attribute identification is described and recorded by one or more words, letters or numbers. The copyright control information can implement copyright control in units of fragments, such as the number of times of copying, the number of times of playing, etc., and can also include parental control information, etc. Other parts can describe many other relevant information according to the needs of the content provider. In the present invention, the segment description information may include one or more types of description information described above. In the method for downloading according to content classification in the present invention, it is necessary to use the content classification information therein to determine whether the segment content contains the content required by the user. For a clip, it can belong to multiple content classification levels at the same time, for example, it can belong to the content of attribute A "trailer" or the content of attribute B "excellent chapter". These information can be recorded in the content classification information , the specific recording method can be implemented in the way of bit recording, or in other ways.

According to the bit-based recording method, in the content classification information in the content description information, different content classification levels are recorded in different positions. For example: if (0000, 0001) means that the segment belongs to attribute A "trailer" content, and (0000, 0010) means that the segment belongs to attribute B "excellent chapter" content, then (0000, 0011) Indicates that the fragment belongs to both attribute A and attribute B content. Similarly, multiple content divisions can be defined for a segment.

The method described in this invention not only supports the user to select content in the downloaded file according to his own needs (based on a profile), but also supports switching between multiple different content levels, that is, from profile-1 to profile-2 or between other profiles switch between them at will. For example, for Profile-1, the third block is not included in profile-1, but for peofile-2, the third block is included in the definition of profile-2. When the user starts to download and play all the clips of profile-1, he also hopes to download and play in profile-2. In this way, the download client compares profile-1 and profile-2 and finds the third piece of data content Since there is no download at the profile-1 level, but it is required in profile-2, the download client only needs to download the third piece of content at this time. The basic method is to make a difference between profile-1 and profile-2, and then additionally download the required content without downloading all the content in profile-2. Save download space and time. After the fragments that are desired but not available locally are successfully downloaded to the local, all fragments previously owned and downloaded this time are reassembled according to the playback rules required by the new profile to obtain the content that belongs to the complete download.

In the same way, the switch from profile-2 to profile-3 level is the same. If you have already downloaded the segment, you will not download it again. If you want to own it, you will re-download the segment that has not been downloaded before, and then follow the new content. A regroup is required. If the content contained in the previous content level has completely contained all the fragments of the new desired content level, then the fragments of the desired content level have been downloaded to the local, and there is no need to download new fragments, only need to Existing fragments can be reassembled according to the new content division level.

For the profile-3 as shown in Figure 1, the content belonging to some wonderful chapters is appended to the end of the entire file. In this way, although the content length of the entire profile-3 is longer than the original file, in profile-3, the content of the additional part belongs to the original file, but it is repeated and recombined, so the required download amount is still the same as The original file is consistent. In this way, even if the length of time for users to watch is increased, no additional downloads will be added.

In addition to the two implementation methods described above, there may be other ways to obtain personalized profile definitions, and at the same time, users may define them themselves.

Examples of video equipment to which the present invention is applied are explained below.

Fig. 15 shows a block diagram of a video device 1 for personalized download and playback. The video equipment 1 includes: a downloading device 11 , a combining device 12 , a display device 13 and a control device (CPU) for controlling these devices, and may also include a selection device 14 and a storage device 15 as required. The execution of various actions such as receiving, downloading, combining, and selecting involved in the downloading and playing method of the present invention is not limited to functional modules such as these receiving devices, downloading devices, combining devices, and selecting devices, and can also be controlled by a CPU. .

Hereinafter, the video device 1 implementing the present invention using these functional blocks will be briefly described. Taking BT download as an example, the selection device 14 opens and reads the obtained personalized BT seed file, and identifies it according to some modes defined in the seed file, and can recognize some personalized information defined in Fig. 4, and then These personalized options are provided to the user to choose, as the user can select Profile-1 (trailer) or Profile-2 (exciting chapters) and so on; then the downloading device 11 downloads according to the segment list defined in the Profile selected by the user , after downloading each segment similarly, check according to the Hash value of the corresponding segment contained in the BT seed file, if the verification is correct, it means that the segment download is successful, after the segment corresponding to all profiles is successfully downloaded, the combination device 12 These fragments are combined according to the order and method specified by the Profile, and the playback device 13 plays according to the playback method defined in the personalized seed file corresponding to the Profile. In addition, as shown in FIG. 12 , the downloading device 11 may also obtain the attribute of the segment according to the segment descriptor of the segment, and match the acquired attribute with the attribute of the segment belonging to the specified personalization level to determine the segment to be downloaded.

If the user selects part of the download according to the initial grade, and wants to fully enjoy these movies after playing, he only needs to re-select the Profile with richer content to download, and at this time, because some clips have been defined for the previously selected downloading, at this time, the downloading device 11 only needs to determine the content segments that need to be additionally downloaded by performing a simple difference comparison between the profile required now and the previous profile, and download these additional segments.

In addition, when the content segment to be downloaded is shown in FIG. 10 , and its head is appended with segment header data containing segment attributes, the download device 11 receives the content segment appended with segment header data containing specified attribute information. After the display device 13 reproduces the received content fragments, when the user still wants to watch a complete movie or exciting chapter, the download device 11 receives the content fragments whose segment header data does not contain the specified attribute information, and combines them In combination with the device 12, the display device 13 reproduces the twice received content segments.

In addition, the download device 11 may receive a content segment number included in predetermined attribute information from a network server that stores a plurality of attribute information indicating attributes and a content segment number included in each attribute information, and may receive a content file with the content segment number from a management server. The terminal address of the serial number receives the content fragments from each content provider of the terminal address. After the display device 13 reproduces these content fragments, the downloading device 11 receives part or all of the content fragments except the reproduced serial number as required. The content segment of is reproduced by the display device 13 .

The method and video equipment described above are merely examples, and various changes can be made without departing from the gist of the present invention.

The invention provides a method for providing, downloading and playing individualized content in units of fragments, and improves the control mode of the content provider and the user on the content.

This method can be used in the distribution stage of the content. For example, the publisher hopes to increase the user's interest in the content by providing the user with some free clips (profile-film) or content introduction (profile-profile) for free, and attract users to use it completely. The (profile-full) download method is used for downloading and viewing, and at the same time, a certain fee is charged to the user, and different profiles can also be used to provide users with different viewing resolutions, etc., to achieve hierarchical distribution of content. For the client, it is hoped that in this way, it can personalize its own download method, and according to its actual needs and actual conditions, such as parental control, network download speed, and network download waiting time, etc., it is convenient for its own graded download And enjoy the content, to achieve the desired download method.

In this way, the user can also define the playback mode during the playback phase. For example, the user has downloaded all fragment contents, that is, the complete target file. Some of the clips may not be suitable for children to watch. At this time, parents only need to define a list of clips suitable for children to watch. When children watch, they are only allowed to choose profiles suitable for children to watch, and they are not allowed to choose The playback method of the complete file, so that although the user has downloaded all the fragment content, different playback content can be realized for different users, and personalized methods such as parental control have been realized, which greatly facilitates the family's management of content.

Claims (20)

1. method for down loading is characterized in that:

Based on content being carried out personalized individualized content definition of dividing, the file destination that is divided into a plurality of fragments is carried out personalization download, said method comprising the steps of:

A) download client is sent download request to the content provider, belongs to personalized other fragment of level of appointment in the downloading object file;

B) download client makes up the fragment of being downloaded, and forms the individualized content that meets other requirement of described personalized level.

2. method for down loading as claimed in claim 1 is characterized in that:

Described individualized content definition comprises the fragment list definition, according to the fragment sequence number definition fragment list corresponding of each described fragment with personalized rank,

Download client is only downloaded the fragment with the fragment sequence number that belongs to corresponding fragment list according to specified personalized rank.

3. method for down loading as claimed in claim 1 is characterized in that:

Described individualized content definition comprises the fragment attributes definition, to each described fragment definition fragment attributes corresponding with personalized rank,

Download client is only downloaded the fragment with corresponding fragment attributes according to personalized rank.

4. as each described method for down loading of claim 1～3, it is characterized in that:

Download client provides the option corresponding respectively with a plurality of personalized ranks to the user, determines the fragment that needs are downloaded according to the option of user's appointment, sends described download request to the content provider.

5. method for down loading as claimed in claim 3 is characterized in that:

The content provider determines the fragment that needs are downloaded according to the download request that has the fragment attributes definition that download client is sent, and downloads for download client.

6. as claim 3 or 5 described method for down loading, it is characterized in that:

Obtain the attribute of described each fragment according to the segment descriptors of each fragment of file destination, with the attribute that obtained with mate corresponding to specified personalized other fragment attributes of level, need to determine the fragment of downloading.

7. method for down loading as claimed in claim 1 is characterized in that:

The definition of described individualized content comprises the broadcast side-play amount to described fragment definition, when described individualized content is play, skips a part of data in the described fragment according to described broadcast side-play amount.

8. method for down loading as claimed in claim 1 is characterized in that:

Download client makes up the partial content of being downloaded according to Combination Control information, and described Combination Control information comprises the position of the content that is used to make up in the sign of fragment and the fragment.

9. method for down loading as claimed in claim 7 is characterized in that:

Described Combination Control information comprises initial address and the length in the fragment.

10. method for down loading as claimed in claim 1 is characterized in that:

Described individualized content definition is stored in and is used for the alternative document that seed file is interior or seed file is outer that P2P downloads, and perhaps, described individualized content definition is specified by the user temporarily, perhaps is embedded in the inside of downloading object file.

11. a video equipment is characterized in that:

Based on content being carried out personalized individualized content definition of dividing, the file destination that is divided into a plurality of fragments is carried out personalization download and play, comprising:

Download apparatus sends download request to the content provider, belongs to personalized other fragment of level of appointment in the downloading object file;

Composite set makes up the fragment of being downloaded, and forms the individualized content that meets other requirement of described personalized level; With

Playing device is play the individualized content that described composite set made up.

12. video equipment as claimed in claim 11 is characterized in that: also comprise

Choice device provides the option corresponding respectively with a plurality of personalized ranks to the user, selects for the user,

Described download apparatus sends described download request according to the option of user's appointment to the content provider.

13. video equipment as claimed in claim 11 is characterized in that:

Described download apparatus according to specified personalized rank, is only downloaded and is had the fragment sequence number that belongs to corresponding fragment list or the fragment of corresponding fragment attributes.

14. video equipment as claimed in claim 13 is characterized in that:

Described download apparatus obtains the attribute of described each fragment according to the segment descriptors of each fragment of file destination, with the attribute that obtained with belong to specified personalized other fragment attributes of level and mate, need to determine the fragment of downloading.

15. a video equipment after reproducing via the part of the content of network receiving target file, receives another part and reproduces, and it is characterized in that:

The part of the content of described file destination is a plurality of contents fragments in the contents fragment that content comprised of described file destination, and described another part is part or all a contents fragment of the described file destination except that the part of the content of described file destination.

16. a video equipment after reproducing via the part of the content of network receiving target file, receives another part and reproduces, and it is characterized in that:

Described content comprises a plurality of contents fragments, the slice header data is arranged in that the head of this contents fragment is additional, and described slice header data comprise the attribute information of the attribute of representing described contents fragment,

Comprise in the described slice header data that described video equipment will receive under the situation of described attribute information of regulation, receive the additional contents fragment that described slice header data are arranged, after the contents fragment reproduction that these received, receive the contents fragment that described slice header data do not comprise the attribute information of described regulation, the contents fragment of the attribute information that comprises described regulation and the contents fragment that do not comprise the attribute information of described regulation are reproduced.

17. a video equipment after reproducing via the part of the content of network receiving target file, receives another part and reproduces, and it is characterized in that:

This video equipment can be connected the transmission of carrying out described content with not specific a plurality of terminals and receive, and described content comprises a plurality of contents fragments, and these contents fragments are stored in not specific a plurality of terminals,

The webserver of the contents fragment sequence number that described video equipment is comprised from a plurality of attribute informations of attribute of storing the described content of expression and each attribute information receives the contents fragment sequence number that attribute information comprised of regulation, receive terminal address from the management server that the described not specific contents fragment that a plurality of terminal had is managed with described contents fragment sequence number, after reproducing from the terminal received content fragment of this address, part or all the contents fragment that receives contents fragment except that the described contents fragment sequence number that receives reproduces.

18. each the described video equipment as claim 15～17 is characterized in that:

The amount of information of the part of the content of described file destination is littler than the amount of information of another part.

19. a content delivery method reproduces the back to the video equipment that another part reproduces to the part of the content that receives file destination, sends described content via network, it is characterized in that:

At the head of the contents fragment that content comprised of described file destination, accessory plate paragraph header data, described slice header data comprise the attribute information of the attribute of representing described contents fragment.

20. content delivery method as claimed in claim 19 is characterized in that:

The amount of information of the part of the content of described file destination is littler than the amount of information of another part.

Images