Connection mode controlling apparatus, connection mode controlling method, and connection mode controlling program

Connection mode control device, connection mode control method and connection mode control program

Technical field

[0001] The present invention relates to the technical field of a network control device, a connection mode control device, a network control method, a connection mode control method, a network control program, and a connection mode control program. More specifically, the present invention relates to a network control device for controlling a distribution mode of distribution information distributed from a distribution source in a network system while gradually relaying the distribution information in relay units connected to form a multi-level hierarchy , The technical field of connection mode control equipment, network control method, connection mode control method, network control program and connection mode control program.

Background technique

[0002] In recent years, with the increase in the speed of Internet lines used in homes, network systems have been widely used. In a network system, a network is constructed by connecting home personal computers and the like into a tree structure with a single distribution device as a distribution source at the apex, and so-called contents such as music and movies are distributed via the network from the distribution device as distribution information. From the perspective of the connection mode, the network is called a "topology". Each distribution device and personal computer constituting the network is generally called a "node".

[0003] For example, Patent Document 1 discloses the prior art of a network system. Patent Document 1: Japanese Patent Pending Publication No. 2003-169089.

[0004] In each node included in the existing network system represented by the network system of Japanese Patent Laid-Open No. 2003-169089, the content transmitted from the upper node is temporarily stored in a buffer memory and used for playback processing Wait. This structure attempts to reduce the influence of transmission speed fluctuations in the Internet circuits constituting the distribution path in the network system. For example, a FIFO (First In First Out) memory such as a ring buffer memory is used as the buffer memory.

[0005] On the other hand, in the network system, the node that constructs the system is a home personal computer or the like as described above. Therefore, there may be cases where the power of any node on the distribution path is cut off regardless of whether the content is being distributed. In this case, the function of relaying content to the lower node connected to the node whose power switch is turned off will be stopped. [0006] During the content distribution in the network system, when the relay function of any node on the distribution path is stopped, the topology is rebuilt to include nodes other than the node whose relay function is stopped (that is, , Rebuild the publishing path from the publishing device to the lower node directly connected to the node where the relay function is stopped and restart publishing).

SUMMARY OF THE INVENTION Problems to be solved by the present invention

[0007] However, in the topology reconstruction of the existing network system, processing such as searching for the shortest path from the publishing system and using the search results for connection switching is necessary. As a result, the distribution of content to lower-level nodes is temporarily interrupted. In the case where the distribution is interrupted, in the lower node, the processing of the playback content stored in the buffer memory is continued (in other words, the data stored in the buffer memory as content is consumed), but the new content is not supplied to buffer. As a result, problems will arise, causing the amount of storage in the buffer to gradually decrease, and in some cases, the contents of the lower-level nodes are heavy.

CN 1998199 Β

Play processing was interrupted.

[0008] In a case where the relay function of any node included in the network system is stopped, interruption of the playback process may cause deterioration of the reliability of the network system itself.

[0009] The implementation of the present invention takes these problems into consideration, and the purpose of the present invention is to propose a network control device, a network control method, and a network control program for controlling the publishing mode in the network system so that even if the network system includes When the relay function of any node stops, the content can be released reliably, and the reliability of the network system itself can be improved without affecting the processing in the lower nodes. The solution to the problem

[0010] In order to achieve this object, the present invention according to claim 1 relates to a network control device for controlling any one of a plurality of repeaters included in the following network system, the network system including as the release information The publisher of the publishing source and the plurality of repeaters connected to the publisher and forming a plurality of levels, and wherein the publishing information is published from the publisher to each repeater, the device includes: such as The detection device of the CPU is used to detect that in the issuance of the release information to the target repeater as the repeater to be controlled, the release is relayed in any one of the repeaters located upstream Whether the function of the information is stopped; and a consumption control device such as a CPU for controlling the unit time consumption as the consumption per unit time of the posted information stored in the target repeater so that it is smaller than the total The amount of consumption per unit time before the relay function is stopped, and the release information is consumed due to use in a process performed in the target relay when the relay function is detected to be stopped.

[0011] Therefore, when the relay function in the repeater located on the upstream side in the network system is stopped, the consumption per unit time of the posting information stored in the repeater located on the downstream side is controlled to be less than the stop The previous consumption. As a result, without increasing the burden on the network system itself, it is possible to prevent the processing interruption in the downstream repeater caused by the interruption of the distribution of the posted information and the posted information is still being consumed at a rate similar to that before the interruption. Or serious processing delays occur.

[0012] In order to achieve this object, the present invention as claimed in claim 2 relates to the network control device as claimed in claim 1, wherein the release information is image information, and the consumption control device uses The image information used for processing reduces the storage capacity of the image information in the cache device at a slower rate than the decrease before the relay function stops, so as to reduce the consumption per unit time, wherein the cache device Used to temporarily store the image information in the target repeater.

[0013] Therefore, in addition to the effect of the present invention according to claim 1, the posting information is image information, and by making the reduction speed of the storage amount of the image information in the cache device slower than the reduction speed before the relay function is stopped, To reduce the consumption per unit time. Therefore, without performing complicated speed control processing, a simple structure can be adopted to reduce the consumption per unit time.

[0014] In order to achieve this object, the present invention according to claim 3 relates to a network control device according to claim 2, wherein the image information is dynamic image information composed of a plurality of still images, and the consumption control device The reduction speed of the storage amount is reduced by repeatedly outputting the same still image from the buffer device multiple times.

[0015] Therefore, in addition to the effect of the present invention according to claim 2, the speed of reducing the storage amount is reduced by repeatedly outputting the same still image as the composition of the moving image information multiple times. Therefore, the consumption per unit time can be reliably reduced by simple processing.

[0016] In order to achieve this object, the present invention according to claim 4 relates to the network control device according to claim 1, wherein the release information is encoded image information, and the consumption control device The decoding speed of the image information in the processing in the target repeater is slower than the decoding speed before the relay function stops, so as to reduce

CN 1998199 Β

Reduce the consumption per unit time.

[0017] Therefore, in addition to the effect of the present invention according to claim 1, the unit time consumption is reduced by making the speed of decoding image information in the processing in the target repeater slower than the decoding speed before the relay function is stopped. Therefore, the consumption per unit time can be reliably reduced by simple processing.

[0018] In order to achieve this object, the present invention according to claim 5 relates to a network control device according to claim 1, wherein the distribution information is dynamic image information composed of a plurality of still images, and the consumption control device The unit time consumption is reduced by making the display time of the still image longer than the display time before the relay function is stopped.

[0019] Therefore, in addition to the effect of the present invention according to claim 1, the unit time consumption is reduced by making the display time of the still image that is the composition of the moving image information longer than the display time before the relay function is stopped. Therefore, the consumption per unit time can be reliably reduced through simple processing.

[0020] In order to achieve this object, the present invention according to claim 6 relates to a network control device according to any one of claims 1 to 5, wherein the release information is composed of a plurality of consecutive unit release information, so If the unit publishing information in the predetermined position in the publishing information is empty, the information volume of the publishing information used for processing is zero, and the device further includes: a reset device such as a CPU, After the issuance of the target repeater restarts, for resetting the unit time consumption in the target repeater to the same unit time consumption as before the relay function is stopped; and The release control device of the CPU, after resetting the unit time consumption in the target repeater, is used to issue empty unit release information to all repeaters except the following repeaters, so The repeater is the repeater located downstream in an announcement before the relay function of the repeater whose relay function is stopped is stopped.

[0021] Therefore, in addition to the effects of the present invention according to any one of claims 1 to 5, only empty cell release information is released to the ones except those before the relay function of the relay whose relay function is stopped. All repeaters other than the repeater located on the downstream side in the release. Therefore, the playback time base in the repeater becomes gradually closer to the playback time base in other repeaters. As a result, by reducing the unit time consumption in the repeater on the downstream side of the repeater whose repeater function is stopped, it is possible to reduce one of the playback positions of the repeater and other repeaters on the playback time base. Between the deviations.

[0022] In order to achieve this object, the present invention as set forth in claim 7 relates to a control method for controlling any one of a plurality of repeaters included in the following network system including the information The publisher of the publishing source and the multiple repeaters connected to the publisher and forming multiple levels, and wherein the publishing information is published from the publisher to each repeater, the method includes: detecting Step for detecting the function of relaying the published information in any one of the repeaters located upstream in the issuance of the published information to the target repeater that is the repeater to be controlled Whether to stop; and a consumption control step for controlling the unit time consumption as the consumption per unit time of the posted information stored in the target repeater to be smaller than before the relay function is stopped The amount of consumption per unit time, the release information is consumed due to use in the processing executed in the target repeater when the relay function is detected to stop.

[0023] Therefore, when the relay function in the repeater located on the upstream side in the network system is stopped, the consumption per unit time of the posting information stored in the repeater located on the downstream side is controlled to be less than The consumption before the stop. As a result, without increasing the burden on the network system itself, it is possible to prevent processing from occurring in downstream repeaters due to the situation where the distribution of the published information is interrupted and the published information is still consumed at a rate similar to that before the interruption. Interruption or severe processing delay.

CN 1998199 Β

[0024] In order to achieve this object, the present invention as recited in claim 8 causes a computer included in a network control device to function as follows for controlling a plurality of relays included in the following network system The network system includes a publisher as a publishing source for publishing information, and a plurality of repeaters connected to the publisher and forming multiple levels, and wherein the publishing information is transmitted from the publishing source. The device is released to each repeater, and the role is: a detection device for detecting that any upstream device is located in the target repeater as the target repeater to be controlled. Whether the function of relaying the posted information in one of the repeaters is stopped; and a consumption control device for controlling the amount of consumption per unit time of the posted information stored in the target repeater The unit time consumption amount is smaller than the unit time consumption amount before the relay function stops, and the release information is due to the processing performed in the target repeater when the relay function stop is detected Use and be consumed.

[0025] Therefore, when the relay function in the repeater on the upstream side in the network system stops, the computer functions so that the consumption per unit time of the posted information stored in the repeater on the downstream side changes. It must be less than the unit time consumption before the relay function stops. Therefore, without increasing the burden on the network system itself, it is possible to prevent processing in the downstream repeater caused by the interruption of the release of the released information and the release of the information still being consumed at a rate similar to that before the interruption. Interruption or severe processing delay.

[0026] In order to achieve this object, the present invention as recited in claim 9 relates to a connection mode control device for controlling a publisher as a publishing source of publishing information in a network system and connecting to the publishing in a tree structure And form a connection mode between multiple repeaters at multiple levels, wherein the published information is published in the network system, and the equipment includes: a retrieval device such as a CPU, when any one of the repeaters When the relay function is stopped, it is used to retrieve any one of the relays except the relay whose relay function is stopped and can relay the posted information; a connection device such as a CPU is used for Connecting the repeater receiving the posting information to the retrieved another repeater; and a posting continuation means such as a CPU for passing the posting information via the other repeater The publishing speed of is faster than the publishing speed before the relay function is stopped to continue the publishing of the published information via the connected relay.

[0027] Therefore, when the relay function in any one of the relays is stopped, another relay that can relay the posted information is searched. When the publication of the published information is continued via another retrieved repeater, the publication speed of the published information is made faster than the publication speed before the relay function is stopped. Therefore, it is possible to continue the processing of posting information in the repeater belonging to the hierarchy below the repeater whose relay function has been stopped.

[0028] In order to achieve this object, the present invention according to claim 10 relates to a connection mode control device according to claim 9, wherein the release continuation means continues to issue while gradually passing through another repeater of the connection The release speed of the release information is increased to the maximum value of the release speed specified in the network connecting the publisher and each repeater, as an upper limit.

[0029] Therefore, the distribution is continued while gradually increasing the distribution speed of the distribution information via another connected repeater to the maximum value of the distribution speed as the upper limit, so that the repeaters belonging to a lower level can be faster To get the necessary release information.

[0030] In order to achieve this object, the present invention according to claim 11 relates to a connection mode control device according to claim 9 or 10, wherein the release continuation means works by making the release faster than before the relay function is stopped. The distribution is continued at the distribution speed until the storage amount of the distribution information to be distributed in the repeater as the distribution destination becomes a predetermined amount.

CN 1998199 Β

[0031] Since the posting is continued by making the posting speed faster than the posting speed before the relay function is stopped, until the storage amount of the posted posting information in the repeater that is the posting destination of the posted posting information becomes predetermined Therefore, it is possible to reliably continue the processing of publishing information in the relays belonging to the hierarchy below the relay whose relay function has been stopped.

[0032] In order to achieve this object, the present invention as recited in claim 12 relates to a connection mode control device for controlling a publisher as a publishing source of publishing information in a network system and connecting to the publishing in a tree structure The connection mode between multiple repeaters and forms multiple levels, wherein the release information is released in the network system, and the device includes: a connection device such as a CPU for connecting the multiple Some of the repeaters are connected to one of the repeaters, thereby forming a plurality of paths for distributing the release information to the one repeater; and a release control device such as a CPU for using the via as one of the paths The main path of the relay is released to the release information of the one repeater to another repeater belonging to the level below the one repeater, and the release is provided for the external output processing in the one repeater Information, and publish the published information that is published to the one repeater via the main path as one of the paths or the sub-path of the other path to another one of the levels that belong to the one repeater Repeater.

[0033] Therefore, a main path and a sub-path are formed to connect a plurality of repeaters to one repeater, and the posting information issued via the main path is used for external output processing in the one repeater, and for posting to Another repeater belonging to the hierarchy below the one repeater. The release information distributed via the main path or the sub path as one of the paths is also used to distribute to another repeater belonging to a lower level. By connecting multiple lines to each node, redundancy can be increased when preparing to stop the relay function of any one node, and it is possible to prevent the external output processing in the repeater belonging to a lower level from being stopped.

[0034] In order to achieve this object, the present invention according to claim 13 relates to a connection mode control device according to claim 12, further comprising a switching device such as a CPU, when belonging to the one repeater on the main path When the relay function of the repeater of the hierarchical level stops, it is used to switch the posted information that is posted to the one repeater via the sub-path, so as to provide it to the one repeater External output processing.

[0035] Therefore, when the relay function of the repeater belonging to the level above the one repeater on the main path is stopped, it is used to switch the distribution information distributed via the sub-path to provide it to the outside. Output processing. Therefore, the external output processing in the one repeater will not be interrupted.

[0036] In order to achieve this object, the present invention according to claim 14 relates to a connection mode control device according to claim 13, further comprising retrieval means, which is issued to the one via the sub-path when switched by the switching means When the publishing information of the repeater is provided to the external output processing in the one repeater, it is used to retrieve the publisher or a new medium belonging to the hierarchy above the one repeater. A repeater, wherein the connecting device connects the issuer or the new repeater retrieved by the retrieval device to the one repeater, thereby forming a new path.

[0037] Therefore, when the posted information posted via the sub-path is switched to provide it to the external output processing, the issuer or a new repeater belonging to a hierarchy above the one repeater is retrieved, and the issuer or retrieved The new repeater is connected to the one repeater, thereby forming a new path. Therefore, redundancy can be ensured and maintained.

[0038] In order to achieve this object, the present invention according to claim 15 relates to a connection mode control device according to claim 12, further comprising retrieval means, when belonging to the one above the repeater on the sub-path When the relay function of the repeater of the hierarchy is stopped, it is used to retrieve the publisher or the new repeater belonging to the hierarchy above the one repeater, wherein the connection device will The publisher or the retrieval device retrieved the

CN 1998199 Β

The new repeater is connected to the one repeater, thereby forming a new sub-path.

[0039] Therefore, when the relay function of the repeater belonging to the level above the one repeater on the sub-path is stopped, search for the publisher or a new relay belonging to the level above the one repeater And connect the publisher or the retrieved new repeater to the one repeater, thereby forming a new sub-path. Therefore, even when the relay function in the repeater on the sub-path is stopped, a new sub-path can be formed, and redundancy can be ensured and maintained.

[0040] In order to achieve this object, the present invention according to claim 16 relates to a connection mode control device according to claim 14 or 15, wherein said retrieval means retrieves said publisher or said new repeater so that the one Another repeater on the hierarchy to which the repeater belongs is included in the new main path.

[0041] Therefore, a publisher or a new repeater is retrieved so that another repeater located on the level to which the one repeater belongs is included in the new main path, so that the repeater on the same level can be passed through To receive post information.

[0042] In order to achieve this object, the present invention according to claim 17 relates to a connection mode control device according to any one of claims 9 to 16, wherein the connection device connects each repeater on a different level to This one repeater thus forms multiple paths.

[0043] Therefore, since repeaters located on different levels are connected to the one repeater, thereby forming multiple paths to the one repeater, even when there is a possibility of a failure such as a stop of the relay function When different on different levels, sufficient redundancy can be reliably ensured.

[0044] In order to achieve this object, the present invention as recited in claim 18 relates to a connection mode control method for controlling a publisher as a publishing source of publishing information in a network system and connecting to the publishing in a tree structure And form a connection mode between multiple relays at multiple levels, wherein the release information is released in the network system, the method includes: a retrieval step, when any one of the relays in the middle When the function is stopped, search for any one of the repeaters except the repeater whose relay function is stopped and be able to relay the posted information; the connecting step will be used to receive all the posted information. The repeater is connected to the retrieved another repeater; and a post continuation step, as a post connection step for continuing to post the post information via the connected another repeater, is used to pass through the The release speed of the release information of the other repeater is faster than the release speed before the relay function is stopped to continue the release.

[0045] Therefore, when the relay function in any one of the relays is stopped, another relay that can relay the posted information is retrieved. When continuing the publication of the publication information via the retrieved another repeater, the publication speed of the publication information is made faster than the publication speed before the relay function is stopped. Therefore, it is possible to continue the processing of posting information in the relays belonging to the hierarchy below the relay whose relay function has been stopped.

[0046] In order to achieve this object, the present invention as recited in claim 19 relates to a connection mode control method for controlling a publisher as a publishing source of publishing information in a network system and connecting to the publishing in a tree structure And form a connection mode between multiple relays at multiple levels, wherein the release information is released in the network system, and the method includes: a connecting step for connecting the multiple relays Some of them are connected to one of the repeaters, thereby forming a plurality of paths for distributing the release information to the one repeater; and a release control step for distributing to the main path via the main path as one of the paths The posted information of the one repeater is posted to another repeater belonging to the level below the one repeater, the posted information is provided for the external output processing in the one repeater, and will be used as The main path of one of the paths or the sub-path of the other path is released to the one repeater and the release information is released to another repeater belonging to a hierarchy below the one repeater.

[0047] Therefore, a main path and a sub-path are formed to connect multiple repeaters to one repeater and publish via the main path

The published information is used for external output processing in the one repeater, and used for publishing to another repeater belonging to a hierarchy below the one repeater, and via the main path or sub-path as one of the paths The released information is also used to release to another repeater belonging to a lower level. Therefore, by using multiple lines for repeaters, redundancy can be increased when preparing to stop the repeater function of any one repeater, and it is possible to prevent external output processing in repeaters belonging to a lower level from being Stop.

[0048] In order to achieve this object, the present invention as recited in claim 20 causes a computer included in a control mode control device to function as follows, and the control mode control device is used to control the distribution as distribution information in a network system A connection pattern between a publisher such as a node of a source and a plurality of repeaters such as a plurality of nodes that are connected to the publisher in a tree structure and form multiple levels, wherein the published information is in the network Announced in the system, the function is: a search device, when the relay function of any one of the repeaters is stopped, used to search for any one of the media except the repeater whose relay function is stopped Relay and capable of relaying the posting information; connecting means for connecting the repeater that receives the posting information to the other retrieved repeater; and posting continuation means as a means for passing through the The distribution continuation means for continuing the distribution of the distribution information by another repeater connected is used to make the distribution speed of the distribution information via the other repeater faster than the distribution before the relay function is stopped Speed to continue publishing.

[0049] Therefore, when the relay function of any one of the repeaters is stopped, another repeater that can relay the posted information is searched. When continuing the publication of the published information via the retrieved another repeater, the function of the computer is to continue the publication by making the publication speed of the published information faster than the publication speed before the relay function was stopped. Therefore, it is possible to continue the processing of posting information in the relays belonging to the hierarchy below the relay whose relay function has been stopped.

[0050] In order to achieve this object, the present invention as set forth in claim 21 causes a computer included in a connection mode control device to function as follows, and the connection mode control device is used to control the distribution as distribution information in a network system The connection mode between the publisher of the source and the multiple repeaters that are connected to the publisher in a tree structure and form multiple levels, wherein the published information is published in the network system, and the role is : A connecting device for connecting some of the plurality of repeaters to one of the repeaters, thereby forming a plurality of paths for publishing the publishing information to the one repeater; and a publishing control device , For publishing the published information that is published to the one repeater via the main path as one of the paths to another repeater belonging to a hierarchy below the one repeater, for the one repeater The external output processing in provides the publication information, and publishes the publication information that is published to the one repeater via the main path as one of the paths or the sub-path as the other path to the one that belongs to Another repeater at the level below the repeater.

[0051] Therefore, a main path and a sub-path are formed to connect a plurality of repeaters to one repeater, and the posting information issued via the main path is used for external output processing in the one repeater, and for posting to The posting information that belongs to another repeater of a level below the one repeater and is distributed via the main path or the sub path as one of the paths is also used to distribute to another repeater belonging to a lower level. Therefore, by using multiple lines for repeaters, redundancy can be increased when preparing to stop the repeater function of any one repeater, and it is possible to prevent external output processing in repeaters belonging to a lower level from being Stop.

[0052] According to the invention of claim 1, when the relay function in the repeater located on the upstream side in the network system is stopped, each unit of the posted information stored in the repeater located on the downstream side is controlled Time consumption so that it is less than the consumption before stopping. As a result, without increasing the burden on the network system itself, it is possible to prevent the situation where the release of the posted information is interrupted while the posted information is still consumed at a rate similar to that before the interruption.

CN 1998199 Β

The processing in the downstream repeater is interrupted or severe processing delay occurs.

[0053] Therefore, even in the case where the relay function in any one of the repeaters included in the network system is stopped, it is possible to resume being stopped without exerting a large influence on the processing in the downstream repeater The relay function. Therefore, the released information can be released reliably, and the reliability of the network system itself is improved at the same time.

[0054] In the present invention according to claim 2, in addition to the effects of the present invention of claim 1, the posted information is image information, and the reduction speed of the storage amount of the image information in the cache device is slower than that of the present invention. The reduction rate before the relay function stops can reduce the consumption per unit time. Therefore, a simple structure can be adopted to reduce the consumption per unit time without performing complicated speed control processing.

[0055] In the present invention according to claim 3, in addition to the effect of the present invention of claim 2, the reduction speed of the storage amount is slowed by repeatedly outputting the same still image as a component of the moving image information multiple times. Therefore, simple processing can reliably reduce the consumption per unit time.

[0056] In the present invention according to claim 4, in addition to the effect of the present invention of claim 1, the decoding speed of the image information in the processing in the target repeater is slower than before the relay function is stopped. Decoding speed to reduce unit time consumption. Therefore, simple processing can reliably reduce the consumption per unit time.

[0057] In the invention according to claim 5, in addition to the effect of the invention of claim 1, by making the display time of the still image as a component of the moving image information longer than the display time before the relay function is stopped . Therefore, simple processing can reliably reduce the consumption per unit time.

[0058] According to the present invention of claim 6, in addition to the effects of the present invention of any one of claims 1 to 6, after the relay function is restored, the empty unit release information is released only to other than its relay The relay function of the repeater whose function is stopped stops all the repeaters other than the downstream repeater in the previous release. Therefore, the playback time base in the repeater becomes gradually closer to the playback time base in other repeaters. As a result, by reducing the unit time consumption in the repeater on the downstream side of the repeater whose repeater function is stopped, the playback position and other repeaters on the playback time base of the repeater can be eliminated The deviation between the playback positions.

[0059] According to the invention described in claim 7, when the relay function in the repeater located on the upstream side in the network system is stopped, the control of the distribution information stored in the repeater located on the downstream side The consumption per unit time is smaller than the consumption before the stop. Therefore, without increasing the burden on the network system itself, it is possible to prevent processing in the downstream repeater caused by the interruption of the release of the released information and the release of the information still being consumed at a rate similar to that before the interruption. Interruption or severe processing delay.

[0060] Therefore, even if the relay function of any one of the repeaters included in the network system is stopped, the stopped relay can be resumed without exerting a large influence on the processing of the downstream repeater. Features. Therefore, the released information can be released reliably, and the reliability of the network system itself is improved at the same time.

[0061] According to the invention described in claim 8, when the relay function in the repeater located on the upstream side is stopped, the computer functions so that the distribution information stored in the repeater located on the downstream side The consumption per unit time is less than the consumption before the relay function stops. Therefore, without increasing the burden on the network system itself, it is possible to prevent the downstream repeater from being processed by the downstream repeater due to the interruption of the release of the released information and the release of the information still being consumed at a rate similar to that before the stop. Interruption or severe processing delay.

[0062] Therefore, even when the relay function of any one of the repeaters included in the network system is stopped, the stopped relay can be resumed without exerting a large influence on the processing of the downstream repeater Features. Therefore, the released information can be released reliably, and the reliability of the network system itself is improved at the same time.

CN 1998199 Β

[0063] According to the invention described in claim 9, when the relay function in any one of the repeaters is stopped, another repeater capable of relaying and publishing information is retrieved. When the release of the released information is continued via another retrieved repeater, the release speed of the released information is made faster than the release speed before the relay function is stopped. As a result, it is possible to continue the processing of posting information in the relays belonging to the hierarchy below the relay whose relay function has been stopped.

[0064] Therefore, even another repeater connected under the repeater whose repeater function has stopped can use the posted information to continue processing such as playback without being affected by the function stop.

[0065] According to the invention described in claim 10, in addition to the effects of the invention described in claim 9, the distribution is continued while gradually increasing the distribution speed of the distribution information via another connected repeater to The maximum value of the publishing speed as the upper limit, so that the repeaters belonging to the lower level can get the necessary publishing information more quickly.

[0066] According to the present invention described in claim 11, in addition to the effects of the present invention described in claim 9 or 10, since the posting is continued by making the posting speed faster than the posting speed before the relay function is stopped, until The amount of storage in the repeater that is the distribution destination of the distributed information becomes a predetermined amount, so it is possible to reliably continue the distribution of information in the repeater belonging to the level below the repeater whose relay function has been stopped.Processing. Treatment. According to the invention of claim 4, the main path and the sub-path are formed to connect a plurality of repeaters to one repeater, and the release information issued via the main path is used for external output processing in the one repeater, And it is used to publish to another repeater belonging to the hierarchy below the one repeater. The release information distributed via the main path or the sub path as one of the paths is also used to distribute to another repeater belonging to a lower level. By connecting multiple lines to each node, redundancy can be increased when preparing to stop the relay function of any one node, and it is possible to prevent the external output processing in the repeater belonging to a lower level from being stopped.

[0067] According to the invention described in claim 12, a main path and a sub path are formed to connect a plurality of repeaters to one repeater, and the release information issued via the main path is used for the outside of the one repeater The output is processed and used for publishing to another repeater belonging to the hierarchy below the one repeater. The release information distributed via the main path or the sub path as one of the paths is also used to distribute to another repeater belonging to a lower level. By connecting multiple lines to each node, redundancy can be increased when preparing to stop the relay function of any one node, and it is possible to prevent the external output processing in the repeater belonging to a lower level from being stopped.

[0068] According to the present invention described in claim 13, in addition to the effects of the present invention described in claim 12, a relay belonging to a repeater at a level above the one repeater on the main path When the function is stopped, the information posted via the sub-path is switched to provide external output processing. Therefore, the external output processing in this repeater will not be interrupted.

[0069] According to the invention described in claim 14, in addition to the effects of the invention described in claim 13, when the posting information posted via the sub-path is switched to provide to the external output processing, it is retrieved that belongs to that one The publisher or new repeater of the hierarchy above the repeater, and the retrieved publisher or new repeater is connected to the one repeater, thereby forming a new path. Therefore, redundancy can be ensured and maintained.

[0070] According to the present invention described in claim 15, in addition to the effects of the present invention described in claim 12, a relay belonging to a repeater at a level above the one repeater on the sub-path When the function is stopped, a publisher or a new repeater belonging to a hierarchy above the one repeater is retrieved, and the retrieved publisher or new repeater is connected to the one repeater to form a new sub path. Therefore, even if the relay function in the repeater located on the sub-path stops, a new sub-path can be formed, and redundancy can be ensured and maintained.

[0071] The present invention according to claim 16, in addition to the effects of the present invention described in claim 14 or 15,

In addition, the publisher or repeater is retrieved so that another repeater on the level to which the one repeater belongs is included on the new main path, so that the posting information can be received via the repeater on the same level .

[0072] According to the invention described in claim 17, in addition to the effects of the invention described in any one of claims 9 to 16, repeaters located on different levels are connected to the one repeater , Thereby forming multiple paths to the one repeater. Therefore, even when the possibility of occurrence of a failure such as the stop of the relay function is different in different levels, sufficient redundancy can be reliably ensured.

[0073] According to the invention described in claim 18, when the relay function in any one of the repeaters is stopped, another repeater capable of relaying and publishing information is retrieved. When the release of the released information is continued via another retrieved repeater, the release speed of the released information is made faster than the release speed before the relay function is stopped. Therefore, it is possible to continue the processing of posting information in the relays belonging to the hierarchy below the relay whose relay function has been stopped.

[0074] Therefore, even another repeater connected under the repeater whose relay function has stopped can use the posted information to continue processing such as playback without being affected by the function stop.

[0075] According to the invention described in claim 19, a main path and a sub path are formed to connect a plurality of repeaters to one repeater, and the release information issued via the main path is used for the outside of the one repeater Output processing, and used for publishing to another repeater belonging to the level below the one repeater, and publishing information published via the main path or sub-path as one of the paths is also used for publishing to the lower level Another repeater in the hierarchy. By using multiple lines for repeaters, redundancy can be increased when preparing to stop the relay function of any one repeater, and it is possible to prevent external output processing in repeaters belonging to a lower level from being stopped.

[0076] According to the invention described in claim 20, when the relay function in any one of the repeaters is stopped, another repeater capable of relaying and publishing information is retrieved. When the distribution of the information is continued via the retrieved another repeater, the computer continues the distribution by making the publication speed of the publication information faster than the publication speed before the relay function is stopped. Therefore, it is possible to continue the processing of posting information in the relays belonging to the hierarchy below the relay whose relay function has been stopped.

[0077] Therefore, even another repeater connected under the repeater whose relay function has stopped can use the posted information to continue processing such as playback without being affected by the function stop.

[0078] According to the invention described in claim 21, the computer functions as follows: a main path and sub-paths are formed to connect a plurality of repeaters to one repeater, and the published information issued via the main path is used in the one The external output processing in the repeater and is used for publishing to another repeater belonging to the level below the one repeater, and the publishing information issued via the main path or sub-path as one of the paths is also used for Publish to another repeater belonging to a lower level. By using multiple lines for multiple repeaters, redundancy can be increased when preparing to stop the repeater function of any one repeater, and it is possible to prevent external output processing in repeaters belonging to a lower level from being Stop.

Description of the drawings

[0079] FIGS. 1A and 1B are block diagrams showing a schematic structure of a network system according to the first embodiment. Figure 1A is a block diagram showing the physical connection mode of the network system. Fig. 1B is a block diagram showing a connection mode of a network system as a topology. FIG. 2A shows the structure of nodes included in the network system according to the first embodiment, and FIG. 2B shows details of information stored in the storage unit 3. FIG. 3 is a flowchart showing the entire processing in the node according to the first embodiment. 4A to 4E are flowcharts (I), showing the details of processing in the node according to the first embodiment, and respectively show upper node connection processing, data receiving processing, lower node connection processing, exit message response processing, and Node exit processing. 5A and 5B are flowcharts (II), showing the details of processing in the node in the first embodiment, and respectively show

CN 1998199 Β

Π/34 page data transmission processing and data replay processing. Fig. 6 is a flowchart showing data display processing in a node according to the first embodiment. FIGS. 7A, 7B, and 7C are diagram (I) showing the packet release state in the network system according to the first embodiment and diagrams (i), (ii), and (iii) respectively showing state details. FIGS. 8A, 8B, and 8C are diagram (II) showing the state of packet distribution in the network system according to the first embodiment and diagrams (iv), (v), and (vi) respectively showing details of the state. 9A, 9B, and 9C are a diagram (III) showing the state of packet release in the network system according to the first embodiment and diagrams (vii), (viii), and (ix) respectively showing details of the state. 10A and 10B are diagrams (i) and (ii), respectively, showing details of the package release state in the case where node exit occurs in the network system according to the first embodiment. Figures and 11C are Figures (i), (ii), and (iii), respectively, showing details of the package publishing state in the case of reconnection after the node exit is completed in the network system according to the first embodiment. 12A, 12B, and 12C are diagrams (i), (ii), and (iii), respectively, showing the case of restoring the playback delay after node withdrawal and reconnection in the network system according to the first embodiment Details of the package release status. Figures 13A.13B and 13C show the decommissioning of nodes in the network system according to the first embodiment. The sending/receiving of data in the node under the condition that the replay delay is restored after exit and reconnection, and the state before the node exit, the state after the node exit, and the state of the replay delay recovery are respectively shown. 14A and 14B are flowcharts showing data display processing in nodes in the first modification and the second modification of the first embodiment, respectively. Fig. 15 is a block diagram showing a schematic structure of a network system according to the second embodiment. FIG. 16 is a block diagram showing the general structure of nodes included in the network system according to the second embodiment. 17A and 17B show the detailed structure of the node according to the second embodiment and show the details of the topology table and the operation of the buffer memory, respectively. FIG. 18 is a flowchart showing the conventional publishing operation in the network system according to the second embodiment. 19 is a block diagram showing the schematic structure of the network system according to the second embodiment after the relay function of some nodes is stopped. 20A, 20B and 20C are flowcharts showing the operation of the lower node when the relay function of some nodes is stopped, and showing the operation of the buffer memory. FIG. 21 shows the operation of the lower node when the relay function of some nodes is stopped. 22A and 22B show operations in a modified network system according to the second embodiment. FIG. 22A is a flowchart showing the operation, and FIG. 22B shows the operation of the buffer memory corresponding to the operation. Fig. 23 is a block diagram showing a schematic structure of a network system according to the third embodiment. Figure 24 is the box The figure shows the schematic structure of the network system according to the third embodiment after the relay function of some nodes is stopped. FIG. 25 is a flowchart showing the normal publishing operation of the network system according to the third embodiment. Fig. 26 is a flowchart showing the operation of lower-layer nodes when the relay function of some nodes is stopped.

Label description

[0080] 1,100CPU2 connection authentication unit 3 storage unit 4 input unit 5 output unit 6 message sending/receiving unit 7 data sending/receiving unit 8, 109 bus 10 first upper node information storage area 11 second upper node information storage area 12 transmission Data number information storage area 14 Received data number information storage area 15 Playback speed information storage area 16 Specific value storage area 17 Ring buffer area 102 Decoder 103 Table memory 104 Buffer memory 105 Broadband interface 106 Timer 107 Speaker 108 CRTNS network system S Server N, called, %, N₃, N₄, N₅, N₆, N₇, N₈, N₉, N₁₀, N_n, N₁₂, N₁₃, N₁₄,0-1,1-1,1-2,1-3, 2-1,2-2, 3-1,

3-2, 3-3, 3-4, 4-1, 4-2, 4-3 Node NT network L, L'line LM, LM^,Specific implementation of main line LS, LS' sub-line NT, NT2 network system

[0081] (1) First Embodiment The first embodiment of the present invention will be described below with reference to FIGS. 1A and 1B to 13A~13C.

CN 1998199 Β

The first embodiment to be described below relates to a case where the present invention is applied to a network control process in which the network control process is used to control the distribution mode of content in a network system in which content is distributed and all The network system includes: a server as a node, the node is a publishing device that is a publishing source of content that publishes information; and a plurality of nodes as user terminals, the user terminals are connected to include multiple levels below the server Tree structure.

[0082] FIGS. 1A and 1B show a schematic structure of a network system according to the first embodiment. 2A and 2B are block diagrams, showing detailed structures of nodes included in the network system. Figures 3 to 6 are flowcharts, each showing the publishing process performed in the node according to the present invention. FIGS. 7A to 7C to FIGS. 13A to 13C specifically show the issuance process.

[0083] In the case of viewing the network system of the first embodiment as a physical connection mode, as shown in FIG. 1A, a network system NS is constructed so that a server S and a plurality of nodes N serving as user terminals pass through as a wired circuit or The lines L of the wireless circuit are connected to each other to be able to transmit/receive information to/from each other via a network NT such as an Internet line. Considering the network system NS shown in FIG. 1A as a topology whose vertex is the server S, as shown in FIG. 1B, two nodes N are connected to the server S via a line L, and two nodes N are connected to Each of node N. Through the topology in which the nodes N are connected in a tree shape using the server S as the apex, the required content is published from the server S to the node N that needs the content.

[0084] The specific structure of the node N included in the network system S is described below with reference to FIGS. 2A and 2B.

[0085] As shown in FIG. 2A, the node N according to the first embodiment includes: a CPU 1 as a detection device, a consumption control device, a reset device, and a release control device; a connection authentication unit 2; a hard disk, a semiconductor memory, etc. Built-in storage unit 3; input unit 4 consisting of a mouse, keyboard, etc.; output unit 5 consisting of a monitor for displaying images, a speaker for outputting sound, and a decoder for decoding sound, etc.; message sending /Receiving unit 6 and data sending/receiving unit 7. The CPU 1, the connection authentication unit 2, the storage unit 3, the input unit 4, the output unit 5, the message sending/receiving unit 6, and the data sending/receiving unit 7 are connected to each other via a bus 8 to be able to send/receive information to each other.

[0086] The outline operation is described below. First, the message sending/receiving unit 6 is functionally divided into: a part connected to a node N (or server S) located at a higher level via a line L and sending/receiving messages from a node located at a higher level; and via The line L is connected to the node N located on the lower level and the part that sends/receives messages from the node located on the lower level. The message here indicates various control information necessary to distribute the required content in the network system NS. By receiving/sending messages, an environment for publishing actual content is formed.

[0087] On the other hand, like the message transmitting/receiving unit 6, the data transmitting/receiving unit 7 is functionally divided into: being connected to a node N located at a higher level via a line L, etc., and from being located at a higher level A part that transmits/receives data such as a node on the upper level; a part that is connected to a node N located on a lower level via a line L and transmits/receives data from a node located on a lower level. The data indicates data in the form of packets including sounds or images constituting the content itself distributed in the network system NS. In an environment formed by message transmission/reception, the slave server S publishes data as content to each node.

[0088] For data in the form of packets, it is assumed that consecutive packet numbers from the head of the content are assigned to data packets constituting a content block (for example, the content of a movie or the content of a piece of music). The data sent as packets is divided into small packets whose unit is smaller than the packet. In addition, for small packets, consecutive small packet numbers from the head of the content are assigned.

[0089] For example, in a case where the network system NS is a network system for distributing chargeable content, the connection authentication unit 2 passes in the case where the node N enters the network system NS or in which another node is newly connected In the case of N, the authentication information is sent/received from the server S to perform the so-called authentication process, which is used to determine whether the node N is in

CN 1998199 Β

The node to which content is allowed to be published under the control of the CPU 1.

[0090] In addition, the storage unit 3 temporarily or in a nonvolatile manner rewritably stores information to be described later, and if necessary, outputs the information to the CPU 10.

[0091] On the other hand, when an operation such as specifying content to be played back is performed by using the node N, the input unit 4 generates an operation signal corresponding to the performed operation, and outputs the operation signal to the CPU 110 CPU 1. The operation signal is received and the other components of the node N are controlled, thereby performing processing of publishing and replaying desired content.

[0092] The playback process is performed by using the output unit 5 under the control of the CPU 1.

[0093] The specific structure of the storage unit 3 will be described in detail with reference to FIG. 2B. FIG. 2B shows the storage unit 3 partitioned according to the type of information stored.

[0094] As shown in FIG. 2B, the storage unit 3 is composed of the following: a first upper-level node information storage area 10 for storing information that is located above the node N that includes the storage unit 3 in the topology shown in FIG. 1B The upper node information of the node N of the upper layer; the second upper node information storage area 11 is used to store the upper node information of the upper node N that is topologically located in the upper layer of the upper node N; the lower node information storage area 12 is used The storage means the lower-layer node information of one or more nodes N connected topologically to the node N including the storage unit 3, and the information number is equal to the lower-node number; the sending data number information storage area 13 is used to store the display The transmission data number information including the packet number of the data sent to the lower node; the received data number information storage area 14 is used to store the reception of the packet number representing the data received by the node N itself including the storage unit 3 Data number information; playback speed information storage area 15, used to store playback speed information representing the playback speed in the output unit 5 of images and sounds corresponding to the received data; specific value storage area 16, used It is stored as a specific value such as the minimum amount of stored data preset in the ring buffer area which will be described later; and the ring buffer area 17 is used to temporarily store data that is actually distributed as content.

[0095] The node information stored in the first upper node information storage area 10, the second upper node information storage area 11, and the lower node information storage area 12 specifically includes an IP (Internet Protocol) global address representing the node N.

[0096] The ring buffer area 17 is a ring buffer type storage area for temporarily storing data corresponding to distributed content in a FIFO manner. The ring buffer area 17 sequentially stores packets (and small packets) as content data and outputs the packets according to the storage order as a rule to provide them to the playback processing in the output unit 5.

[0097] Next, referring to the flowcharts shown in FIGS. 3 to 6, the processing of allowing the new node N to newly enter the network system NS, receiving the distribution data corresponding to the content, and replaying the content according to the first embodiment will be described, and The processing performed in the case where the function of relaying data to other nodes N connected to the new node N is stopped due to the withdrawal of the new node from the network system NS during replay is the processing performed in each node N .

[0098] In the case that the new node N is allowed to be located at the position of the next layer of any node N in the existing network system NS, enter the node N (the following will newly enter the network system NS and execute according to the flowchart of FIG. 3 The node N for data reception processing and replay processing is called "entry node N" to distinguish it from other nodes that have been connected to the network system NS) to perform the connection of the incoming node N to the upper node N (or the network system NS) The processing (steps S1 and S2) of the server S) located on the uppermost layer in the middle. When the connection request is rejected by the upper node N (No in step S2), the entry node N cannot join the existing network system NS at this time, and the processing of the entry node N is completed.

[0099] On the other hand, after the upper node N (or server S) is allowed to enter in the determination of step S1 and the necessary connection processing is performed through the upper node N (or server S) (Yes in step S2), it is determined that the entry node In N, whether the input unit 4 performs an operation of exiting from the network system NS in which the node N exists (step S3).

[0100] When the entry node N exits from the network system NS (Yes in step S3), exit is executed in the entry node N

CN 1998199 Β

Exit operation (step S4), and complete the processing of entering node N.

[0101] On the other hand, when the process of exiting from the network system NS is not performed but the data receiving process or the like is continuously performed (No in step S3), the entry node N executes the process of receiving necessary data from the upper node N (step S5), and reproduce the received data and output the reproduced data by using the output unit 5 (step S9). When another connected node N needs it, the entering node N executes an exit message response process for disconnecting the connection between another node N connected to the entering node N and the entering node N (step S8), returning to step S3 and The data reception and replay processing are continuously executed.

[0102] In parallel with the data replay process (step S9), the entry node N executes the process of connecting another node N to the next layer of the entry node N (step S6) and sends (relays) data to the connected node Processing of lower node N (step S7) ο

[0103] The processing in step S1 and steps S4 to S9 in FIG. 3 will be described in detail below in order.

[0104] First, the processing of step S1 will be described in detail by using the flowchart of FIG. 4A.

[0105] As shown in FIG. 4A, in the processing of step S1, first the entry node N sends a request message to the upper node N via the line L (step S10), and waits for a reply to the request from the upper node N (step S11) , And confirm the reply (step S12), wherein the request message is used to make the entering node N enter as the next layer node N of another node N that is the upper layer node.

[0106] When the entry node N is allowed to enter the network system NS in response to the reply (Yes in step S12), the entry node N stores the upper node information including the IP address corresponding to the upper node N in the entry node In the first upper node information storage area 10 in the storage unit 3 of N, the upper node information of the upper node including the IP address corresponding to another node above the upper node N is stored in the storage unit of the entry node N 3 in the second upper node information storage area 11 (step S13), and proceed to step S2 as shown in FIG. 3.

[0107] On the other hand, when the entry node N is not allowed to enter the network system NS in the determination of step S12 (No in step S12), the entry node N cannot display its function. Also in this case, after switching to step S2 shown in FIG. 2 (No in step S2), the process of entering the node N is completed.

[0108] Next, the data receiving process in step S5 shown in FIG. 3 will be described in detail by using the flowchart shown in FIG. 4B.

[0109] As shown in FIG. 4B, in the processing of step S5, first, the entering node N sends a message for requesting desired data to the upper node connected at this time via the line L (step S15), and confirms whether There is a transmission permission reply from the upper node N for this message (step S16).

[0110] When there is a transmission permission reply (Yes in step S16), the ingress node N receives a packet including the data desired by the ingress node N according to the reply (step S17), and proceeds to step S6 as shown in FIG. 3 Or S9. On the other hand, when the transmission permission response is not obtained from the upper node N in the determination of step S16 (NO in step S16), there is a possibility that a failure occurs on the line between the upper node N and the entry node N. As a result, the entering node N performs the above-mentioned processing in step S1 (refer to FIG. 4A) to eliminate the failure and can receive data, returns to the processing of step S15 again, and receives data from the upper node N that is connected again (step S15 or S17). ) Ο

[0111] Next, the lower node connection processing in step S6 shown in FIG. 3 will be described by using the flowchart shown in FIG. 4C.

[0112] As shown in FIG. 4C, in the processing of step S6, it is first determined whether the connection request message from the node N connected to the next layer of the entry node N has been sent to the entry node N (step S20). When the connection request message is not sent (No in step S20), the entry node N switches to step S7 in FIG. 3. When the connection request message has been sent

CN 1998199 Β

When the message is sent (Yes in step S20), for example, the connection authentication unit 2 included in each of the lower node N and the entry node N is used between the lower node N and the entry node N that have sent the connection request message. (Or between the lower node N and the server S via the entry node N) an authentication process is performed for determining whether the lower node N is a node N with the qualification to enter the network system NS (steps S21 and S22).

[0113] In the case where the lower node N cannot be allowed to enter the network system NS through the authentication process (No in step S22), the fact is sent to the lower node N as a connection prohibition message (step S25), and the procedure proceeds to Step S7 shown in FIG. 3.

[0114] On the other hand, when the lower node N is allowed to enter the network system NS in the authentication process in the determination of step S22 (YES in step S22), the fact is sent to the lower node N as a connection permission message (step S23), storing the lower-level node information including the IP address corresponding to the next-level node N, etc. in the lower-level node information storage area 12 in the storage unit 3 of the entry node N (step S24), and the program progresses to Step S7 shown in FIG. 3.

[0115] Next, the data transmission processing in step S7 shown in FIG. 3 will be described by using the flowchart shown in FIG. 5A.

[0116] As shown in FIG. 5A, in the processing of step S7, it is first determined whether a request for transmitting data to be reproduced in the lower node N has been received as a request from the next layer connected to the entry node N. The message of node N is sent (step S35). When the message is not received through the entry node N (No in step S35), the procedure proceeds to step S8 shown in FIG. 3.

[0117] On the other hand, when it is determined in step S35 that the message has been sent (Yes in step S35), next, in response to the message, it is determined whether the packet to be sent from the ingress node N is as described in detail below. An empty package (a package that only includes header information and does not include entity image information constituting content, etc.) (step S36). When the packet is not an empty packet (No in step S36), the packet that is not an empty packet is sent as content data to the lower node N via the line L (step S38), and the procedure proceeds to step S8 shown in FIG. 3.

[0118] On the other hand, when it is determined in step S36 that the packet to be transmitted is an empty packet (Yes in step S36), the data received by the ingress node N before the start of the processing in step S7 shown in FIG. 5A The packet number of is compared with the packet number of the physical data to be sent to the lower node N (that is, the physical data such as image data and sound data that are not included in the empty packet). In the case of discontinuous numbers (that is, empty packets are received), by concatenating the small packet numbers immediately before the empty packets and the small packet numbers immediately after the empty packets, the small packet numbers immediately following are arranged in the immediate vicinity. Following the previous small packet number (step S37), the data constituting the content in the small packet with the arranged small packet number is sent to the lower node N (step S38), and the procedure proceeds to step S8 shown in FIG. 3.

[0119] Next, the exit message response processing in step S8 shown in FIG. 3 will be described in detail by using the flowchart of FIG. 4D.

[0120] As shown in FIG. 4D, in the processing of step S8, it is first determined whether any one of the other nodes N connected to the entry node N (specifically, the one connected to the upper layer of the entry node N The upper node N or the lower node N connected to the next layer of the entry node N) receives the message that the node N will withdraw from the network system NS (step S26). In the case where the message has not been received (No in step S26), the procedure returns to step S3 in FIG. 3. On the other hand, when a message requesting exit is received (Yes in step S26), it is determined which of the node N of the upper layer and the node N of the lower layer has sent the message (step S27).

[0121] When the exit request message has been sent from the upper node N (the "upper node" in step S27), according to the representation stored in the second upper node information storage area 11 in the storage unit 3 of the entry node N Upper node N

CN 1998199 Β

The node information of the upper-layer node of the upper-layer node N is connected to the upper-layer node N (that is, the node N that sent the exit request message) to perform the upper-layer node connection processing in step S1 above, thereby restoring the topology (Step S1), and the procedure returns to step S3 in FIG. 3.

[0122] On the other hand, when the exit request message has been sent from the next node N ("lower node" in step S27), the entry in the storage unit 3 of the entry node N is deleted from the lower node information storage area 12. The node information of the lower node N is stored in the lower node information storage area 12 (step S28), and the procedure returns to step S3 in FIG. 3.

[0123] Next, referring to the flowchart of FIG. 4E, the node exit processing in step S4 shown in FIG. 3 (the processing of the entry node N exiting from the network system NS) will be described in detail.

[0124] As shown in FIG. 4E, in the processing of step S4, the above exit request message (refer to FIG. 4D) is first sent to the upper node N connected to the upper layer of the entry node N (step S30). Next, similarly, the above exit request message is sent to the lower node N connected to the next layer of the entry node N (step S31). After sending the message, the entry node N executes the process of exiting from the network system NS.

[0125] Finally, the data playback processing in step S9 shown in FIG. 3 will be described in detail by using the flowchart shown in FIG. 5B.

[0126] As shown in FIG. 5B, in the processing of step S9, the replay flag (that is, the replay flag indicating whether the replay processing using the output unit 5 is being executed) is first checked into the node N (step S40). When the playback flag is "off" (that is, playback processing is not currently performed in the output unit 5) (OFF in step S40), it is next determined whether the amount equal to or not has been stored in the ring buffer area 17 Or data larger than a specific value "t" which is a specific value of a preset storage amount (step S45).

[0127] When no data with a data amount equal to or greater than the specific value "t" is stored in the ring buffer area 17 (No in step S45), if the data playback processing is performed in this state (in other words, consumption If the data in the ring buffer area 17 is lost), the ring buffer area 17 enters a so-called underrun state, and there is a possibility that the playback of images and music will be interrupted. As a result, in order to further store data, the program returns to step S3 via step S8 shown in FIG. 3.

[0128] On the other hand, when it is determined in step S45 that the amount of data equal to or greater than the specific value "t" has been stored (YES in step S45), the playback flag that has been "off" is set to "On" (step S46), the process of replaying the stored data (for example, in the case of image information, the process of displaying image information) is executed in a manner similar to the related art (step S47), and the program progresses to FIG. Step S8 shown in 3.

[0129] On the other hand, when the playback flag is "on" in the determination of step S40 (that is, when playback processing is currently being performed in the output unit 5) (NO in step S40), In a manner similar to the above determination in step S45, it is determined whether the amount of data equal to or greater than the specific value "t" is currently stored in the ring buffer area 17 (step S41). When the amount of data equal to or greater than the specific value "t" has been stored (Yes in step S41), in a manner similar to step S47, as in the prior art, the playback mode such as the display speed of the image data is changed Set to the normal mode (step S42). The playback process is executed in accordance with this setting (step S44). After that, the procedure proceeds to step S8 shown in FIG. 3.

[0130] Conversely, when it is determined in step S41 that the amount of data equal to or greater than the specific value "t" has not been stored (No in step S41), the data playback mode is changed according to the storage amount (step S43), The playback process is executed in accordance with this setting (step S44), and the procedure proceeds to the process of step S8 shown in FIG. 3.

[0131] When the setting of the playback mode is changed in step S43, when data of an amount equal to or greater than the specific value "t" has not been stored, the playback mode is set to reduce the playback process in step S44. To reduce the data consumption in the ring

CN 1998199 Β

The reduction speed of the storage amount of the buffer area 17 (that is, the consumption amount per unit time of the data stored in the ring buffer area 17). Specifically, for example, the amount of data read from the ring buffer area 17 per unit time is set to reduce it to approximately 4/5 of the maximum value of the normal amount. More specifically, there is a method of controlling the reading amount so that the same packet is read twice in a cycle such as once every 5 packets. In this case, when the relay function in the upper node N is not stopped, it is sufficient to set the period to zero, and the period is shortened as the storage amount in the ring buffer area 17 decreases.

[0132] The degree of decrease in the amount of data read from the ring buffer area 17 per unit time may be such a degree that the user cannot visually perceive the decrease. The value of 4/5 can be changed to this extent.

[0133] Next, the data display processing in step S44 shown in FIG. 5B will be specifically described by using the flowchart of FIG. 6. The data display processing is based on a preset condition of changing the setting of the speed of reading data from the ring buffer area 17 in the processing of step S43.

[0134] As shown in FIG. 6, in the processing of step S44, first, the image is read from the ring buffer area 17 according to the reading amount per unit time (reading speed) set in the processing of step S43. Information data (step S50). The read data is decoded in the output unit 5 (step S51) and the decoded data is displayed on a display or the like not shown (step S52). Also in the case of sound information, data is read from the ring buffer area 17 at a reading speed set similarly in the case of image information (step S50). The read data is decoded and decoded data is generated from a speaker or the like not shown (steps S51 and S52).

[0135] Next, a series of network control processing will be described in more detail with reference to FIGS. 7A to 7C to FIGS. 13A to 13C including the flow of a packet composed of data as content.

[0136] In FIGS. 7A to 7C to 12A to 12C, it is assumed that in a topology whose vertex is server S, two nodes are called and N₂It is connected to the first level of the next level of server S. Two nodes Ν₃And Ν₄It is connected to the next layer of node Νι. Two nodes Ν₅And Ν₆Is connected to node N₂The next level. Two nodes Ν₇And %, two nodes Ν₉And Ν two nodes Ν_πAnd Ν₁₂And two nodes Ν₁₃And Ν₁₄Are respectively connected to the nodes %, Shan, Ν₅And Ν₆The next level. The following describes the distribution of image data stored in advance as content in the server S in the form of packets to the lower nodes N to N in the network system NS having the connection mode shown in the topology having this structure.₁₄And nodes Ν to Ν₁₄Node in Ν₄The processing executed in the case of exiting from the network system NS.

[0137] In addition, it is assumed that server S and nodes N to N₆Each of them has a ring buffer area 17 and also performs a process of relaying the packet P to other lower-level nodes (each having the ring buffer area 17) while storing the packet P as data (refer to steps S7 and S7 in FIG. 3). S9). In FIGS. 7A to 7C to FIGS. 13A to 13C, one packet P is represented as one box (), and the number in each box represents the packet number.

[0138] In addition, in FIGS. 7A to 7C to FIGS. 12A to 12C, the ring buffer area 17 in each of the node calls to 4 is represented as a line. However, in the packet input/output processing, the right and left ends of each ring buffer area 17 are connected in a ring. A package is provided for replay processing. Every time a package is released to the next-level node N, the oldest package among the packages stored in the ring buffer area 17 is replaced with a new package released from the upper-level node N. In this way, the above-mentioned FIFO buffer memory is realized.

[0139] The downward hollow arrow used to display the ring buffer area 17 in the server S in FIGS. 7A to 7C to FIGS. 13A to 13C indicates when the packet represented by the box pointed to by the arrow at its left end is sent from the server S Time sequence. The upward solid arrow displayed for the ring buffer area 17 in each node N in FIGS. 8A to 8C to FIGS. 13A to 13C indicates that the image data stored in the position pointed by the arrow in each packet is displayed at this time. In addition, the "-" displayed in the display (frame) of the packet in the ring buffer area 17 of either server S or node N means that there is no data.

CN 1998199 Β

The data is stored in the ring buffer area 17 corresponding to the packet.

[0140] Now suppose that in each of the examples shown in FIGS. 7A and 7B to FIGS. 13A and 13B, the specific value "t" in each node N (specifically, the insufficient storage amount in the ring buffer area 17 The lower limit "t") corresponds to three packets.

[0141] First, before the package is released, the packages to be released are stored in the ring buffer area 17 as shown in FIG. 7A in the order of release. These packets are to be distributed to the nodes N to Shan connected to the lower layer according to the numbers. Specifically, as shown in FIG. 7A, when the packet P is output from the server S_oAt the point in time, the package will be P. (With package number "0") sent to the connection server S and the node called and N₂On line L. After that, the P being sent_oIt is stored in the ring buffer area 17 in each node N, and is relayed by the node N located on a lower level after the time shown in FIGS. 7B and 7C has passed.

[0142] When the package P_oIs published to the node N connected to the bottom of the network system NS₇At the time of calling 4, then as shown in FIG. 8A, the sending of the package P to be released after the package Po is started.

[0143] For example, when the image data of three packets is stored in the node N connected on the bottom of the network system NS as shown in FIG. 8B by repeating the above-mentioned operation₇To N₁₄When, then from package P. To start the node call to N₁₄Image playback processing in the.

[0144] The packet P in each of the node calls to 4 is used as shown in FIG. 8C. To complete the replay processing of a packet, at nodes N to N₁₄Start to use the replay processing of the next package Pi, and publish the fourth package P from the server S₄O

[0145] By repeating the above processing, packets as image data are sequentially issued and replayed while being stored in the ring buffer area 17 in a FIFO manner (or erased from the ring buffer area 17). Assume that in each of the node calls to call 4 in the first embodiment, as shown in FIG. 9A, the packets stored just before the one immediately before the currently replayed packet are sequentially updated.

[0146] As shown in FIGS. 9B and 9C, after publishing all the packets to be published in the server S, the ring buffer area 17 in the blood of the node becomes empty in turn, and all the packets in the network system NS are completed. Post processing.

[0147] Next, the node N in the network system NS will be described with reference to FIGS. 10A and 10B to FIGS. 13A to 13C.₄The processing performed in a case where its relay function is stopped due to, for example, a power failure and withdrawing from the network system NS itself.

[0148] As shown in FIG. 10A, it is assumed that when the process of replaying, for example, the image data included in the package Pi is completed and the package P is sent from the server s at the same time₅To node Νι to N₁₄Each time, node N₄Due to the above reasons, it withdraws from the network system NS (refer to Figure 4D). After exiting, the distribution of the package P5 to the lower nodes% and N is naturally disconnected. . In this case, as shown in FIG. 10B, the current node N] is executed. Then connect to the node called as the upper node N, and the node N₉Reconnect to the reconnected N₁₀(Refer to step S1 in Figure 4D), and republish the package P that has not been released before the release sequence₅o Result, at node N₉To N₁₀The ring buffer area 17 included in each of them is connected with other nodes N to N₈And nodes N to N₁₄Compare and rewrite the timing, only one packet is delayed. As a result, there are only two packets (packet P3 and Bar) that have not been subjected to the playback processing as shown in FIG. 10B, and so-called data delay has occurred (No in step S41 in FIG. 5B). After that, until the release of all packages is completed, at node N₉And N₁₀Continue to node N₉To N₁₀Replay in the call from other nodes to N₈And node N_πThe state of the blood is only delayed by one package.

[0149] Therefore, in the first embodiment, as shown by the hatched arrows in FIGS. 11A and 11B, in the nodes% and %, it is possible to read data from the ring buffer area 17 after the time when the data delay occurs. The speed is slower than in other nodes Ν to

Ν₈And node N_πTo call the speed in 4 (No in step S41 in FIG. 5B and step S50 in FIG. 6). Finally, as shown in FIG. 11C, the number of packets to be reproduced is restored to 3 as a specific value (in the case of FIG. 11C, the number of packets is

CN 1998199 Β

To Jie 6).

[0150] From the node N in FIG. 11C or 12A₉The state and node N of the middle ring buffer area 17₈The comparison between the states of the middle ring buffer area 17 clearly shows that the number of the packet that has not been reproduced in each ring buffer area 17 is restored to 3. The packet to be replayed in node% is packet P5, but the packet to be replayed in node% is packet P-due to the upper node N₄Withdraw from the network system NS, when comparing the replay timing on the time base of a series of packets P, at node N₄The nodes below% and N]. A so-called replay delay occurred in, (that is, the replay timing of the same content in node N deviated). So, for example at node N₉Users and nodes Ν₈In the case of a user playing a so-called online game, a playback delay in these two nodes will disturb the normal use state of node N.

[0151] As a result, in the network control processing of the first embodiment, an empty packet is issued from the server S to the network system NS except for the node N whose playback delay has occurred due to the withdrawal of the upper node N from the network system NS. Therefore, the content playback timings of all nodes N included in the network system NS are synchronized with each other.

[0152] In the first embodiment, as shown in FIG. 12A, in the content stored in the ring buffer area 17 of the server S before being published, empty packets are removed at a predetermined timing or interval in anticipation of future playback delays. Ρ. . Insert into other packages. [0153] As shown in FIGS. 12A and 12B, for example, the node% and the call. In the event of a replay delay in the event, the empty package will be published to nodes other than N₉And Ν₁₀All nodes outside, that is, the node is called to Ν₈And node N_πTo Ν], refer to Figure 5A). As a result, as shown in FIG. 12C, the timing of replaying the published packet in all the nodes included in the network system NS becomes the same at the nodes N to N. The playback delay in is eliminated.

[0154] With reference to FIGS. 13A to 13C, a detailed description will be given of the connection to the node N₉And Ν₁₀The change of the packet transmission/reception time in the node N of the upper layer and the lower layer, and the execution of a series of network control processing. Figures 13A to 13C show that they are connected to node N₉And Ν₁₀The packet sending/receiving status in the node N of the upper layer and the lower layer. The horizontal axis uses its left end as the content release start time to indicate the release time and the vertical axis represents the small package number in the released package (specifically, the lower end of the vertical axis and the package P_οThe small and medium package number is "0" corresponding to the small package).

[0155] In the case where the relay function of the node Shan connected to the upper layer node in FIGS. 7A to 7C to FIGS. 12A to 12C is operating normally, the node% or N]. Repeat the packet from the upper level node N₄The operation of relaying to the next-level node while receiving packets from the node. Although in FIGS. 7A to 7C to FIGS. 12A to 12C, the next level node N is not connected to the node% or N]. , But in order to fully explain the function of each node N included in the network system NS, assume that another node N is connected to the node Ng or N in FIGS. 13A to 13C.₁₀The next level.

[0156] In parallel with the relay function, the node% or N]. The package P to be released from the node is stored in the ring buffer area 17 in the order of release (refer to FIGS. 8A to 8C and FIGS. 9A to 9C). At the point in time when three packets P are stored, the node% is called. The output unit 5 performs processing such as decoding and displaying the data stored in the package P (hereinafter, the processing includes display processing when the data is image data and sound output processing when the data is sound data). In the example shown in FIG. 13A, the first packet P is shown in FIGS. 7A to 7C to FIGS. 12A to 12C. Via node N₄Publish from server S and store it in node N₉Or N₁₀The time in the ring buffer area 17 in the middle corresponds to the left end of the horizontal axis. From the time on the left end, the subsequent package pieces and P2 are similarly issued and stored in the ring buffer area 17. From that time until the beginning of the package P. The time period for the decoding and display of the data stored in the data storage is the time T shown in FIG. 13A._0ΟThe amount of storage in the ring buffer area 17 (the amount of three packets in the first embodiment) corresponds to the amount indicated by the label "M" in FIGS. 13A and 13B.

[0157] After the processing of decoding and displaying the data in each packet P is completed, the part of the ring buffer area 17 where the packet P has been recorded is overwritten and written by the subsequent packet P issued after the packet. In the example shown in Figure 12A,

CN 1998199 Β

From the package P. The processing of decoding and displaying the data in the packet begins, until the data in the subsequent packet is stored (overwritten) in the ring buffer area 17 and the packet has been stored._οIn the part, time T is required.

[0158] Next, as described above with reference to FIGS. 10A and 10B, at nodes% and N]. In the case where the upper-level node Shan exits from the network system NS and continues to stop the function, for the period from the stop to the completion of the reconstruction of the topology as shown in Figure 10B and the restart of the release of the package P, no package Ρ is published to node Ν₉And Ν_10ΟTherefore, in the non-publishing period, although time has passed, no package P is released from the upper node N. At node Ν₉And Ν₁₀Here, as shown in FIG. 13B, the graph showing the publishing status of the package P from the upper node and the relay status of the package P to the lower node includes a flat part corresponding to the non-publishing period. As a result, the amount of storage in the ring buffer area 17 is reduced. The gradient of the graph before and after the non-release period is the same (that is, the release speed and the relay speed of the package P).

[0159] Next, in the first embodiment, when there is a non-posting period, the speed of reading data from the ring buffer area 17 is reduced and underflow is prevented from occurring in the ring buffer area 17 (refer to FIG. 5B Steps S43 and S44 in). Therefore, from the point of time when the non-posting period is started, the speed of decoding and displaying data, etc., is reduced, as shown by the thick solid line in FIG. 13B. When the storage amount in the ring buffer area 17 becomes three packs or more, the speed of decoding and display, etc. is reset to the initial speed as shown in FIG. 13B. The storage amount in the ring buffer area 17 at the start of publishing is "M" in a similar manner to the case shown in 13A, and temporarily becomes smaller than "M" due to the occurrence of a non-publishing period. However, the speed of reading data from the ring buffer area 17 becomes slower, so that after the time T3 in which the data reading speed is slow has passed, the storage amount becomes the initial storage amount "M" again.

[0160] In FIG. 13B, in the non-posting period, the overwrite write processing in the ring buffer area 17 is stopped.

[0161] As shown in FIGS. 12A to 12C, after the node exits, a replay delay occurs in the nodes% and N10 that are different from other nodes' call to% and node's to 4. In the first embodiment, as shown in Figures 12A to 12C, the package Poo is mixed in the content in advance and only published to other nodes called N₈And node N_πTo Ν_14ο

[0162] Therefore, when the slave node% and N]. When looking at the position, as shown in Figure 13C, the empty packet P will be present. . Publish to other nodes% to% and other nodes Ν_π Μ Ν₁₄At the point in time (shown as "empty data" in FIG. 13C), the empty packet P is passed as described in step S37 in FIG. 5A. . To publish content. As a result, in the case of viewing continuous content stored in the initial server S, the node N₉And Ν₁₀The continuity of the packet numbers in is interrupted. As a result, the small packet number (indicated by the label "PK" in FIG. 13C) corresponding to the empty packet Poo is delivered to the empty packet P as shown in FIG. 13C. . Increase the release amount at the release time point, and similarly relay it to the lower node No. Therefore, the packet number provided for processing such as decoding and display is similarly transferred and changed in an increased state.

[0163] As described above, in the network control processing according to the first embodiment, when the relay function in the node N located on the upper layer side in the network system NS is stopped, the content data stored in the lower node N The consumption rate is controlled to be slower than the rate before the stop. As a result, when the distribution of content data is interrupted, it is possible to prevent interruption processing or serious problems in the lower node N due to the consumption of content data at a speed similar to that before the stop without placing a burden on the network system NS itself. Processing delay.

[0164] Therefore, even in the case where the relay function is stopped in any one of the nodes N included in the network system NS, the stopped relay can be restarted without exerting a large influence on the processing in the lower node N Features. Therefore, it is possible to reliably distribute content data while improving the reliability of the network system NS itself.

[0165] Since the content data is image data and the content data consumption speed is reduced by making the reduction speed of the image data storage amount in the ring buffer area 17 lower than the reduction speed before the relay function is stopped, it is possible to reduce the content data consumption speed without performing complicated speed In the case of control processing, the consumption speed is reduced with a simple structure.

CN 1998199 Β

[0166] In addition, by repeatedly outputting the same still image for forming moving image data multiple times, the reduction speed of the storage amount in the ring buffer area 17 is reduced so that the consumption speed can be reliably reduced by simple processing.

[0167] In addition, when the stop of the relay function is detected, the consumption speed is controlled to 4/5 of the consumption speed at the maximum value before the relay function is stopped, so that the processing in the node N can not be greatly affected. In the case of reducing the consumption speed. The reduction in the consumption speed is not limited to 4/5, and ultimately may be a reduction in the reading speed that cannot be perceived by the audience (users of node N).

[0168] Since the relay function of all nodes N except for the node under the node N whose relay function is stopped is reset before the relay function is stopped, the empty packet P is issued._οο, Therefore, the playback time base in other nodes N gradually delays. As a result, it is possible to eliminate the replay on the replay time base that occurs between the node N and all other nodes N due to the decrease in the time consumed per unit time in the node N under the node N where the relay function is stopped. The offset of the position.

[0169] II. Revision of the first embodiment The revision of the first embodiment will now be described with reference to FIGS. 14A and 14B.

[0170] In the foregoing first embodiment, as a method of reducing the consumption speed of data stored in the ring buffer area 17, a method of reducing the speed itself of reading data from the ring buffer area 17 is used. In addition to this method, the following two methods can be used as a method to reduce the consumption speed.

[0171] As a first revision, not the speed of reading data from the ring buffer area 17 itself, but the decoding speed of the output unit 5 of the data read from the ring buffer area 17 can be reduced. More specifically, in the processing of step S43 in FIG. 5B, the data decoding speed in the output unit 5 is set to be reduced to about 4/5 of the speed at the maximum value in the normal state. For example, when the relay function in the upper node N is not stopped, it is sufficient to set the frame rate when decoding the still image data used to form the moving image data to 30 frames per second, and when the ring buffer area When the storage capacity in 17 decreases, setting the frame rate to 29 frames per second is sufficient. The more the amount of storage in the ring buffer area 17 decreases, the lower the frame rate is set.

[0172] In the case where the frame rate is changed in the processing of step S43, as the processing in step S44 to be executed subsequently, as shown in FIG. 14A, first read from the ring buffer area 17 as image information at a normal reading speedThe data(Step S53). The read data is decoded in the output unit 5 by using the frame rate set in step S43 (step S54), and the decoded data is displayed on a display or the like not shown in the figure (step S52 ). In the case of sound information, data is read from the ring buffer area 17 at a reading speed similar to that in the case of image information (step S53), and the read data is decoded at the set frame rate , And generate decoded data from speakers, etc. not shown in the figure (steps S54 and S52).

[0173] As described above, according to the first revision, the data consumption speed in the ring buffer area 17 is reduced by making the data decoding speed in processing in the node N slower than the decoding speed before the relay function is stopped. Therefore, the consumption rate can be reliably reduced by simple processing.

[0174] Next, as a second embodiment, when the content data is moving image data, it is not from the reading speed itself of the ring buffer area 17 and the decoding speed in the output unit 5, but the value of the decoded data. The display speed in the output unit 5 can be reduced. More specifically, in the process of step S43 in FIG. 5B, the display time of each of the plurality of pieces of still image data used to form the moving image data decoded in the output unit 5 is increased to be in the normal state. It is about 5/4 of the maximum display time in the state. Specifically, when the relay function in the upper function in the upper node N is not stopped, the display time of the still image data corresponding to a still image (that is, the display time of each still image) is set 1/30 second is sufficient, and when the storage volume in the ring buffer area 17 drops

When it is low, setting the display time to 1/29 second is sufficient. As the storage amount in the ring buffer area 17 decreases, it is sufficient to make the display time longer.

[0175] In the case where the display time as described above is changed in the processing of step S43, as the processing in step S44 to be executed subsequently, as shown in FIG. 14B, the data is read from the ring buffer area 17 at the normal reading speed. Data of image information (step S53). The read data is decoded in the output unit 5 by using the normal frame rate (step S55). After that, the decoded data is displayed on a display or the like not shown in the figure with the display time per still image set in step S43 (step S56).

[0176] As described above, according to the second revision, the ring buffer area 17 is reduced by making the display time of the still image corresponding to the still image data used to form the moving image data longer than the display time before the relay function is stopped. The rate of consumption in. Therefore, the consumption speed can be reduced more reliably by simple processing.

[0177] It is also possible to record the degree corresponding to the flowcharts of FIGS. 3 to 6 and FIG. 14 on an information recording medium such as a floppy disk or a hard disk, or obtain and record the program via the Internet or the like, and read it by a general-purpose computer. The program is fetched and executed, thereby causing the computer to function as the CPU 1 according to the first embodiment.

[0178] (III) Second Embodiment (A) Embodiment First, a second embodiment according to the present invention will be described with reference to FIGS. 15 to 22A and 22B.

15 is a block diagram showing a schematic structure of a network system according to the second embodiment. Fig. 16 is a block diagram showing the general structure of nodes included in the network system. 17A and 17B show the detailed structure of the node. Figure 18 is a flowchart showing the conventional publishing operation in the network system. FIG. 19 is a block diagram showing the schematic structure of the network system after the relay function of some nodes is stopped. 20A, 20B and 20C are flowcharts showing the operation in the lower node when the relay function is stopped. FIG. 21 shows the operation of the lower node when the relay function is stopped. 22A and 22B show operations in the revised network system according to the second embodiment.

[0180] As shown in FIG. 15, the network system NT according to the first embodiment is formed by using a tree structure with nodes 0-1 as a distribution device as vertices. The network system NT includes nodes and 1-3 as the nodes constituting the first level, nodes 2-1 and 2-2 as the nodes constituting the second level, nodes 3-1, 3-2, 3-3 and 3- 4 is used as the node constituting the third level, and nodes 4-1, 4-2, and 4-3 are used as the node constituting the fourth level. The nodes are connected to each other via a line L that is a wired circuit or a wireless circuit so as to be able to transmit/receive information to each other using the node 0-1 as a vertex. As a specific example of the node, the node 0-1 at the highest level corresponds to, for example, a server as a content distribution source, and all nodes except for the node 0-1 are, for example, personal computers at home. The node can be a home set-top box or router.

[0181] In the network system NT shown in FIG. 15, the content distributed from the node 0-1 and desired by the node user is distributed via the line L through other nodes included in a hierarchy higher than the node. Obviously, the content is published in a digital state in units of so-called packages. In addition, a package is released, which is assigned with a consecutive package number indicating the playback or storage order of the buffer memory 104 (refer to FIG. 16) and the like.

[0182] The specific structure of each node is described below with reference to FIG. 16 and FIGS. 17A and 17B. Lets talk about as shown below

15 shows the detailed structure of the node 3-2 in the network system NT. The detailed structure of each of the other nodes is the same as that of node 3-2.

[0183] As shown in FIG. 16, the node 3-2 has a CPU 100 as a retrieval device, a connection device, a publishing relay device, a publishing control device, and a switching device; a decoder 102; a table memory 103; a buffer memory 104; a broadband interface 105; timer

106 ; Speaker 107; and CRT (Cathode Ray Tube) 108 as a display device. CPU 100, decoder 102, table memory

CN 1998199 Β

The memory 103, the buffer memory 104, the broadband interface 105, and the timer 106 are connected to each other via a bus 109.

[0184] Next, the operation will be described. The broadband connection 105 is directly connected to the line L. As shown in FIG. 15, the broadband interface 105 in the node 3-2 is connected to the broadband interface 105 included in the node 2-1 on the upper level via the line L, and is also connected to the next level via the line L The broadband interface 105 included in the nodes 4-1 to 4-3. The transmission/reception of content and the transmission/reception of control information required for content transmission/reception can be performed in the node in units of packets.

[0185] In the table memory 103 as a nonvolatile memory, a topology table as shown in FIG. 17A is stored. In the topology table T, in addition to the layer information indicating the level, it also includes the node number and the IP (Internet Protocol) address in the network system N, where the node number is used to store the topology table T according to the table memory 103 The position of the node in the network system NT is used to identify other nodes located on a higher or lower level. Since the topology table T shown in FIG. 17A corresponds to the node 3-2, it includes the layer of the node 1-1 on the level (layer 2) two levels higher than the node 3-2 as shown in FIG. 15 Information, the node number of node 1T and the IP address of node 1T (for example, "100. 100. 10. 10").

[0186] Similarly, the topology table T includes: the layer information of the node 2-1 located on the upper level (layer 1) of the node 3-2, the node number of the node 2-1, and the IP of the node 2-1 Address (for example, "100. 100. 10. 21"); the layer information of node 4-1, the node number of node 4-1, and node 4- on the layer below node 3-2 (layer T) 1's IP address (for example, "100. 100. 10. 30"); the layer information of node 4-2 and the node number of node 4-2 on the layer below node 3-2 (layer-1) And the IP address of node 4-2 (for example, "100. 100. 10. 31"); and the layer information of node 4-3 and node 4 on the layer below node 3-2 (layer-1) -3's node number and node 4-3's IP address (for example, "100. 100. 10. 32");

[0187] Referring to FIG. 16, the buffer memory 104 as a volatile memory is a ring buffer memory in the so-called FIFO (First In First Out) form. The buffer memory 104 only stores data corresponding to the published content according to the publishing order with a preset recording capacity, reads the data according to the stored order under the control of the CPU 100, and transfers the read data via the bus 109 Output to the decoder 102.

[0188] The operation of the buffer memory 104 when the network system NT is in a normal state will be described below with reference to FIG. 17B.

[0189] As described above, the buffer memory 104 is configured in a ring shape. For example, as conceptually shown in the upper part of FIG. 17B, content data is input clockwise via the broadband interface 105, and as conceptually shown in the lower part of FIG. 17B, the data is output clockwise to the decoder 102. In the example shown in FIG. 17B, the data in the hatched portion is actually stored content data. In parallel with outputting data to the decoder 102 as indicated by the arrow in the lower part of FIG. 17B, new data is input from the wideband interface 105 and stored in the direction indicated by the arrow in the upper part of FIG. 17B. In the normal state, the amount of data input and the amount of data output in the buffer memory 104 are the same per unit time, and the storage amount in the buffer memory 104 remains constant (more specifically, half the amount of storage of the buffer memory 104)

[0190] On the other hand, when the arrival time of a packet as data fluctuates in a situation where the topology in the network system NT changes, the fluctuation can be absorbed by the buffer memory 104. In the case where the decoder 102 decodes the predetermined amount of data all at once or when the topology is disconnected for some reason, the buffer memory 104 functions to always store the predetermined amount of data so that the buffer memory 104 does not change. Is empty, and the playback processing in the decoder 102 is not interrupted.

[0191] Referring again to FIG. 16, the timer 106 performs counting for detecting the stop of the relay function in any node in the network system NT, as described below.

[0192] The decoder 102 decodes the content data output from the buffer memory 104 via the bus 109,

CN 1998199 Β

The image in the data is output to the CRT 108 that displays the image, and the sound in the data is output via the speaker 107.

[0193] Next, the new node is connected to the network system NT. By using Fig. 18, the operations started and executed in the normal state including the new node and the published content are described all at once. FIG. 18 is a flowchart showing an example of a case where the node 3-1 is newly connected to the node 2-1 in the network system NT to obtain the network system NT as shown in FIG. 15 and the content is published to the node 3-1 . In the embodiment described below, it is assumed that the total storage capacity of the buffer memory 104 is 64 kilobytes corresponding to the data amount of 16 packets.

[0194] When the node 3-1 newly enters the network system NT by being physically connected to the node 2-1, first the CPU 100 in the node 3-1 sets the buffer memory 104 in the node 3-1 (below in each In the figure, the buffer memory 104 itself may be appropriately referred to as a "ring buffer"). The CPU 100 in the node 3 sets the speed parameter (shown as "speed Reg" in the figure) representing the playback speed in the decoder 102 to be the same as the normal playback speed, and sets the speed parameter of the node as the data supply source The IP address is set to the node 2-1 on the upper layer (when the node 3-1 is connected to the node 2-1, the node 3-1 is used to obtain the IP address of the node 2-1). In order to calculate the address used to store data in the buffer memory 104, the CPU 100 counts the input counter (16 bits) used to count the number of bytes of the input data issued from the node 2-1 and indicates that the data is reproduced by the decoder 102. Each of the output counters (16 bits) of the put data amount is initialized to "zero" (step S101).

[0195] The address used to store the data input to the buffer memory 104 is "the head address in the buffer memory 104 and the value of the input counter", and the address used to read the data to the decoder 102 is "buffer memory The head address in 104 + the value of the output counter".

[0196] After completing the necessary initialization processing, next, the CPU 100 in the node 3-1 refers to the topology table τ in the table memory 103 and sends a start message for requesting the start of content data transmission to the node 2-1 (step S102 ).

[0197] Next, the CPU 1 in the node 2-1 for receiving the start message (step S115) determines whether the received message is a start message (step S116). Since the currently received message is the start message (Y in step S116), the CPU 1 returns the packet number of the data currently replayed by the node 2-1 as the start packet number to the node 3-1 (step S117).

[0198] When the node 3-1 obtains the starting packet number from the node 2-1, the node 3-1 stores it in the area of the corresponding packet number on the buffer memory 104 (step S103) and will be used to request actual content data The data request message is sent to the node 2-1 (step S104). The data request message includes information indicating the packet number of the data to be sent to the node 3-1 and the publishing speed (parameter "speed Reg" in the node 2-1) [0199] The data request message has been received (step S115) The node 2-1 determines again whether the received message is a start message (step S116). Since the currently received message is a data request message (N in step S116), the node 2-1 determines whether the received message is a message requesting the topology table T (step S118) [0200] Perform the determination operation in step S118 (And the process of sending the topology table T from the node 2-1 in the case of "Y" in step S118), thinking that when the relay function of any node in the network system NT stops, it is connected to its relay function Another node below the stopped node gets the topology of the network system NT to prepare. This operation is meaningless processing in the processing between the current node 3-1 and the node 2-1 (therefore, naturally, the processing of sending the topology table T from the node 2-1 (step S119) is not performed).

[0201] Since the currently received message in the determination of step S118 is a data request message instead of a topology request message (N in step S118), it is next determined whether the received message is a data request message (step S120). Since the currently received message is a data request message (Y in step S120), one packet of data is sent to node 3-1 at the playback speed specified in the data request message (currently IX) (via line L) (Step S121). When in step

When it is determined in S120 that the message received from node 3-1 at this time is not any one of the start message, topology table request message, and data request message (N in step S120), the preset error message is returned to node 3. -1 (Step S122).

[0202] In parallel with these operations, the node 3-1 is set to start the count in the timer 106 in the node 3-1 in synchronization with the output of the data request message, and when the count becomes "0", the timing is generated The processor interrupts the instruction (step S105). The node 3-1 waits until a predetermined time elapses in the counting of the timer 106 (step S106). When the data from the node 2-1 does not arrive within the predetermined time (N in step S106), the timer interruption instruction is executed, and the reconnection process is executed on the node 2-1.

[0203] On the other hand, when the data from the node 2-1 arrives within a predetermined time (Y in step S106), it is determined whether the arrived data is correctly transmitted (step S107). If the data is sent correctly (Y in step S107), the execution of the timer interrupt command is prohibited (step S108), and the amount of one packet of the received data is stored in the buffer memory 104. In connection with this operation, the value of the input counter and the packet number stored in the buffer memory 104 are only updated by one packet, and besides, the speed parameter is increased only by the current acceleration (set in the initialization process in step S101) (step S109 ) Ο

[0204] It is determined whether the amount of data stored in the buffer memory 104 and not output to the decoder 102 has become eight packets (that is, the storage amount of half of the buffer memory 104) (step S110). When the amount is less than eight packets (Y in step S110), the node 3-1 returns to step S104 to receive the next packet and repeat the above processing. On the other hand, when the amount of data stored in the buffer memory 104 becomes eight packets (N in step S110), the speed parameter and acceleration are reset to the initial values (step Sill), and the data is output to the decoding The device 102 (step S112), and returns to the processing in step S104 to continue receiving subsequent data from the node 2-1.

[0205] By repeating the processing in steps S104 to S110, the speed issued from the node 2-1 is increased at the acceleration (0.2 times the initial value) at this time until the amount of data corresponding to eight packets is stored In the buffer memory 104, the buffer memory 104 is charged at a high speed until the data reception becomes an error (Y in step S107). On the other hand, when an error occurs in data reception (Y in step S107), the subsequent release speed is reduced only by 0.5 times the acceleration (step S113) to reliably perform data transmission, and then continue release.

[0206] In the processing of step S112, when the amount of data of eight packets is stored in the buffer memory 104 and the data is output to the decoder 102, the decoder 102 calculates according to the value of the output counter in the buffer memory 104 at this time. Get the address of the data output from the buffer memory 104 (step S125), decode and replay the data with only one packet (step S126), and increase the value of the output counter in the buffer memory 104 with only one packet (step S126) S127). The processing is repeated only by the amount of eight packets, thereby executing the data playback processing.

[0207] As described above, by performing the data distribution of the node 2-1 and the replay processing in the node 3-1, the data replay processing in the node 3-1 is performed while keeping the amount of data in the buffer memory 104 constant. .

[0208] Next, referring to FIGS. 19 to 21, the processing performed in the case where the content relay function in the node 2-1 is stopped due to the following reasons, such as in the network system NT shown in FIG. 15 The power switch is cut off

[0209] In the network system NT according to the second embodiment, when the relay function in the node 2-1 is stopped due to the above-mentioned reason, as shown in FIG. 19, the node 3- which is connected to the initial node 2-1 1 and 3-2 automatically perform a topology reconstruction operation for connecting to the node 1T located one level above the node 2-1 via the line L, and continue the distribution of the content.

[0210] First, the general operation in the nodes 3-1 and 3-2 when the relay function in the node 2-1 is stopped as shown in FIG. 19 will be described.

[0211] When the relay function in the node 2-1 as shown in FIG. 19 is stopped, even at step S106 as shown in FIG. 18

CN 1998199 Β

After a predetermined time has passed in the processing, the CPU 100 in each of the nodes 3-1 and 3-2 on the next level (because the publishing in the normal state is continued, the processing described with reference to FIG. 18 is repeated) also cannot The node 2-1 receives the data. When the data cannot be received even after a predetermined time has passed (N in step S106), the CPU 100 in each of the nodes 3-1 and 3-2 recognizes the The relay function is stopped and the topology is disconnected.

[0212] After that, when the topology disconnection is recognized, the topology table T stored in the table memory 103 is referred to, and the topology is reconstructed to connect the nodes 3-1 and 3-2 to the high The node Ι-L on the two hierarchical levels is deleted. The node 2-1 is deleted from the topology table T in the table memory 103 in each of the nodes 3-1 and 3-2, and the node 1-1 is removed at the same time. The layer information is updated to "1". After that, each of the nodes 3-1 and 3-2 sends a topology request message to the node 1T (refer to steps S118 and S119 in FIG. 18), and obtains a new topology table T from the node 1-1. According to the obtained topology table T, it is identified that the node located on the upper level of node 1-1 is node 0-1, and at the same time, the IP address and node number of node 0-1 whose layer information is "2" are added To the topology table T (in each of the nodes 3-1 and 3-2) to complete the reconstruction of the topology.

[0213] After that, each of the nodes 3-1 and 3-2 requests the node 1-1 to send the packet following the packet whose packet number was obtained at that time (refer to step S104 in FIG. 18), and the node 1-1 receives Request and send subsequent data to nodes 3-1 and 3-2 as new destinations.

[0214] The changes in the buffer memory 104 in the node 1T and each of the nodes 3-1 and 3-2 during the reconstruction of the topology are described below with reference to FIGS. 20A to 20C.

[0215] As described above, when the relay function of the node 2-1 is stopped, the nodes 3-1 and 3-2 located on the next level of the node 2-1 cannot receive data from the node 2-1, so The amount of data in the buffer memory 104 in each of the nodes 3-1 and 3-2 is reduced. At the same time, the decoders 102 in the nodes 3-1 and 3-2 continue to read the data for playback processing from the buffer memory 104. Therefore, as shown in FIG. 20A, the data in each buffer memory 104 continues to decrease.

[0216] On the other hand, the node 1-1 located on the upper layer of the node 2-1 cannot transmit data to the node located on the lower layer, so that the amount of data in the buffer memory 104 in the node continues to increase, such as Shown in Figure 20B.

[0217] After the topology is reconstructed, the nodes 3-1 and 3-2 newly connected to the level below the node 1T must restore their buffer memory 104 to the normal state as soon as possible (that is, the data is stored half way status). If it cannot continue to receive data from the upper node and the decoder 102 in nodes 3-1 and 3-2 continues to read data from the buffer memory 104, finally the buffer memory 104 becomes empty and the nodes 3-1 and 3-2 The playback processing in is interrupted.

[0218] As a result, in the second embodiment, the nodes 3-1 and 3-2 that are trying to achieve early recovery of the storage amount in their buffer memory 104 send a control signal (command) for increasing the issuance speed to the new Connected node 1-1. For example, the node 1-1 for receiving instructions publishes data to the buffer memory 104 in the nodes 3-1 and 3-2 at a publishing speed twice the normal playback processing speed in the nodes 3-1 and 3-2 , As shown in Figure 20C. Through this operation, the amount of storage in the buffer memory 104 is restored to a normal state almost at the same time as the time required to reconstruct the topology.

[0219] As a mode of increasing the posting speed, more specifically, the maximum value of the posting speed is determined by the transmission processing capacity of the CPU 100 in each node related to the so-called bandwidth (frequency band) of the lines L and L. Test into account the redundancy in general, the above-described normal playback speed is about twice the processing speed is suitable.

[0220] As another mode of changing the publishing speed, for example, instead of doubling the publishing speed immediately, the node 1-1 can continuously increase the publishing speed at a predetermined rate until the following control is sent from the nodes 3-1 and 3-2 The control signal is used to notify the fact that the storage amount in the buffer memory 104 in each of the nodes 3-1 and 3-2 has reached a predetermined amount. Since the limit of the increase in publishing speed depends on the output capacity of node 1-1 and the connection between nodes 3-1 and 3-2

CN 1998199 Β

Therefore, the necessary messages are exchanged between the node and another controller to grasp whether the packet from the node at the higher level has reached the node at the lower level. Increase the publishing speed to the highest allowable speed, and transmit packets at a gradually increasing speed. In this way, the amount of storage in each buffer memory 104 can be restored in the shortest time. The node 3-2 sends the packet to the lower nodes 4-1 and 4-2 at the same speed as the speed at which the packet is received from the node 1-1.

[0221] Next, the flowchart 21 is used to describe all at once when the relay function of the node 2-1 in the network system NT according to the second embodiment is stopped and the nodes 3-1 and 3-which are located on a lower level are stopped. 2 The operation performed when connecting to the node 1-1 again to reconstruct the topology of the network system NT as shown in FIG. 19. Next, the reconnection operation in the node 3-1 of the two nodes 3-1 and 3-2 connected to the node 2-1 will be described. In addition, the reconnection operation in node 3-2 is similarly performed.

[0222] Until the relay function of the node 2-1 is stopped, the processing shown in FIG. 18 is repeated in the node 3-1. When the data from the node 2-1 has not arrived even after the predetermined time has passed in the processing of step S106 as shown in FIG. 18, the node 3-1 determines the relay in the node 2-1 located on the higher level The function stops, and the tree structure reconnection starts.

[0223] Specifically, as shown in FIG. 21, when the data from the node 2-1 cannot be received within a predetermined time, a timer interrupt command is generated in the node 3-1 (refer to step S105 in FIG. 18). ), from the topology table T in the table memory 103, obtain the IP address of the node 1T located one level above the node 2-1, that is, the IP address of the node 1T whose layer information is "2" from the node 3-1 ( Step S130), and delete the information in the topology table T related to the node 2-1. After that, the node 3-1 sends a topology table request message for acquiring the topology table T in the node 1-1 to the node 1T indicated by the IP address (step S131).

[0224] During the topology reconstruction process, in the node 1-1 located at a higher level, the operations performed in the node 2-1 shown in FIG. 4 are performed (steps S115 to S122 in FIG. 18). ) The same operation (steps S135 to S142 in Fig. 21). The type of the message sent from the node 3-1 is determined (steps S136, S137, and S140) and processing according to the determined message type (S139, S138, S141, and S142) is executed.

[0225] In the node 101 that has performed such processing, when the topology table request message is received from the node 3T (step S135) and it is determined that the message is the topology table request message (Y in step S137), the node 1-1 will The topology table T stored in the table memory 103 is sent to the node 3-1 (step S138) ο

[0226] The node 3-1 for receiving the topology table T (step S132) updates the existing topology table T through the topology table T acquired from the node 1-1 (step S133). More specifically, the node 3-1 identifies the node 0-1 and adds the IP address and the node number of the node 0-1.

[0227] After that, in order to restore the storage amount in the buffer memory 104 in the node 3-1 at a high speed, the speed parameter is set to "1", the acceleration is set to 0.2, and the data request message is sent from the node 3. -1 is sent to node 1-1 (step S134). After that, the node 3-1 returns to step S1105 as shown in FIG. 18 and repeats the operations in steps S105 to S112 as a normal state.

[0228] Until the amount of one packet of data corresponding to the data request message is received from the node 1-1, the data in the buffer memory 104 is continuously decoded and continuously reduced (refer to FIG. 20A). When the data starts to be released from the node 1-1 (steps S140 and S141), the data is released while gradually increasing the release speed until the amount of data corresponding to eight packets is stored in the buffer memory 104 (step S109 in FIG. 18) . After storing the amount of data corresponding to the eight packets, the buffer memory 104 in the node 3-1 starts the function in the normal state (refer to FIG. 17B).

CN 1998199 Β

[0229] As described above, in the process of controlling the connection mode in the network system NT according to the second embodiment, when the relay function in any one node is stopped, another node capable of relaying content is retrieved. When continuing content distribution via another retrieved node, the distribution speed is made higher than the speed before the relay function is stopped. Therefore, it is possible to continue content processing in nodes belonging to the next level of the node whose relay function has stopped.

[0230] Since the publishing continues while gradually increasing the speed of content publishing via another node that is newly connected to the upper limit speed that is the maximum value of the publishing speed, nodes belonging to a lower level can acquire necessary content more quickly. It is also possible to detect the amount of unreplayed data stored in the node's buffer after the relay function is stopped and the topology is reconstructed, and control the publishing speed to reach the highest speed based on the detected amount of unreplayed data time.

[0231] (B) Modification of the second embodiment A modification of the second embodiment will be described with reference to FIG. 22. Figure 22 is a flowchart showing the operation of this revision.

[0232] In the above-mentioned second embodiment, when the relay function of the node 2-1 is stopped, another node (node 1T) located on a higher level is retrieved and the distribution of content is resumed from this node. In the following revisions, multiple new nodes are retrieved and content publishing continues from multiple nodes at the same time.

[0233] In the case where the node 3-1 according to the revision cannot receive data from the node 2-1 within a predetermined time as shown in FIG. 22A, steps S130 to S133 shown in FIG. 21 are performed to update the node 3- Topology table T in 1.

[0234] While receiving the data issued from the node 2-1, determine the storage amount (the number of packets) in the buffer memory 104 at this time and the required storage amount (eight packets) in the buffer memory 104 in the normal state. Whether the difference between is greater than the two packets (that is, whether further data storage is necessary) (step S145). When the difference is greater than two packets (that is, when further data storage is necessary, Υ in step S145), a new data request message is sent to node 1-1, the packet number to be received is updated, and the data The request message is sent to the node 0-1 located on the upper level of the node i-ι (step S146). It is determined that each one comes from the data of one packet of the nodes 1-1 and 0-1, that is, the data of a total of two packets Whether the data has been received (step S147). When the two packets of data have not been received (N in step S147), the node 3-1 waits until the two packets of data are received. On the other hand, when the data of two packets have been received (Y in step S147), the data of the two packets are stored in the buffer memory 104, the input counter only updates two packets, and the packet number to be received is updated ( Step S148), and node 3-1 returns to step S145o

[0235] On the other hand, when it is determined in step S145 that the difference is not greater than two packets (that is, further data storage is unnecessary, and it is N in step S145), it is determined whether the difference is greater than one packet (that is, , Whether it is unnecessary to supplement the data urgently) (step S149). When the difference is not greater than one packet (that is, the current storage amount in the buffer memory 104 is in a normal state) (N in step S149), the node 3-1 returns to step Sill in FIG. 4, and continues to receive normally deal with.

[0236] On the other hand, when the difference is greater than one packet (that is, the current storage amount in the buffer memory 104 is close to the normal state and it is not necessary to urgently supplement data) (Y in step S149), the node 3-1 changes the new The data request message is sent to the node 1T (step S150) and waits until a predetermined time elapses in the count of the timer 6 in the node 3-1 (step S151). If the data from the node 2-1 does not arrive within the predetermined time (N in step S151), the timer interrupt instruction is executed, and the data connection process is executed again on the node i-1.

[0237] On the other hand, when the data from the node 1-1 arrives within a predetermined time (Y in step S151), the data amount of one packet received is stored in the buffer memory 104, and the value of the counter is input And stored in the buffer memory

The number of packets in 104 is updated by the amount of only one packet (step S152), and the node 3-1 returns to step S145.

[0238] When it is determined in step S147 that the data from node 0-1 cannot be received (Y in step S147), also

CN 1998199 Β

You can ask another node connected to node 0-1 (node 1-2 or 1-3 in the case of Figure 15), send a data request message to another node, and check whether it can publish data.

[0239] Through the above processing, as shown in FIG. 22B, data is input from a plurality of routers to the buffer memory 104 in the node 3-1, and the necessary storage amount can be quickly restored.

[0240] In a manner similar to the node 3-1, the node 3-2 receives data from multiple routers and restores the necessary storage amount. The nodes 4-1 and 4-2 below the node 3-2 which is the next layer of the node 2-1 whose relay function has stopped also receive data from multiple routers, and quickly restore the necessary storage capacity. At the node where the relay function has stopped

The nodes 4-1 and 4-2 below the node 3-2 of the next layer of 2-1 may not receive data from multiple routers, while the node 3-2 may be higher than the normal state in the second embodiment. [0241] In the above description, by connecting nodes located on different levels to nodes for publishing data, multiple paths to nodes are formed . As a result, even in a case where the possibility of occurrence of a failure such as a stop of the relay function varies among different levels, sufficient redundancy can be reliably ensured.

[0242] (IV) Third Embodiment A third embodiment as another embodiment according to the present invention will now be described with reference to FIGS. 23 to 26.

[0243] FIGS. 23 and 24 are block diagrams, each showing a schematic structure of the network system according to the third embodiment, and FIGS. 25 and 26 are flowcharts showing the operation of the network system.

[0244] In the above-mentioned second embodiment, the case where the relay function in the node stops during the reception of the content distributed via only one line L has been described. In the third embodiment described below, two paths are provided for one node in advance; the main path and the sub path, and the published content is received.

[0245] As shown in FIG. 23, in a manner similar to the second embodiment, the network system NT2 according to the third embodiment is formed by a tree structure whose vertex is the node 0-1 as a publishing device. The network system NT2 includes nodes and 1-3 as the nodes constituting the first level, nodes 2-1 and 2-2 as the nodes constituting the second level, node 3-1,

3-2, 3-3, and 3-4 are used as the nodes constituting the third level, and nodes 4-1 and 4-2 are used as the nodes constituting the fourth level. The nodes are connected to each other via the main line LM as a wired circuit or a wireless circuit to be able to transmit/receive information to each other using the node 0-1 as a vertex. The specific example of each node is similar to that in the second embodiment.

[0246] In addition, in the network system NT2 according to the third embodiment, each node belonging to the second or lower hierarchical level is connected to the node of the upper layer of the upper node to which the main line LM is connected by using the sub-line LS . In the normal publishing state, only the content published via the main line LM is provided to the playback processing in each node. The content distributed via the sub-line LS is temporarily received without being used for playback processing, and then distributed to another node located on the next level. Moreover, it is sufficient to receive the content distributed from the sub-line LS and distribute the content of the main line to the lower node.

[0247] The detailed structure of the node according to the third embodiment is different from the node according to the second embodiment (see FIG. 2) in that two table memories 103, two buffer memories 104, and two timers 106 are provided. , Each of them is one for the main line LM and the other for the sub-line LS. In a manner similar to the second embodiment, a CPU 100, a decoder 102, a broadband interface 105, a speaker 107, and a CRT 108 are provided. In addition, in the topology table T stored in each table memory of a node, unlike the second embodiment, an identifier is added to indicate whether the topology table is used for the topology table T of the main line LM or used for Topology table of sub-line LS To

[0248] For example, when the relay function in the node 2-1 belonging to the network system NT2 in the state shown in FIG. Each of 3-2 automatically connects the main line LM and

CN 1998199 Β

The sub-line LS is again connected to the main line LM, and the sub-line LS, as shown in FIG. 24, and continues to receive the published content. In this case, in FIG. 23, the node 3-1 is connected to the node 2-1 via the main line LM, and it is connected to the node 0-1 via the sub-line LS. Relay in the node 2-1 After the function stops, the node 3-1 connects the main line LM to the node 1T, and the connection of the sub-line LS to the node 0-1 remains unchanged. On the other hand, in FIG. 23, the node 3-2 is connected to the node 2-1 via the main line LM, and it is connected to the node 1-L via the sub-line LS. The relay function in the node 2-1 is stopped After that, the node 3-2 connects the main line LM to the node 3-1 on the same level, and the connection of the sub-line LS to the node 1-1 remains unchanged. In this way, the rules are strictly followed so that the main line LM is connected to the nearest node, and the sub-line LS is connected to a node located on a level higher than the node to which the main line LM is connected.

[0249] For the nodes 4-1 and 4-2 belonging to the fourth level, the node to which the sub-line LS is connected in FIG. 23 loses the relay function, so the main line LM remains unchanged and only the sub-line LS, is connected To the new node (in the case of Figure 24, the sub-line LS^,The connected node is node 3-1) ο

[0250] Next, in the case where the relay function in the node 2-1 is actually stopped, it will be described in the nodes 3-1, 3-2, 4-1, and

The topology reconstruction operation performed in 4-2.

[0251] When the relay function in the node 2-1 is stopped as shown in FIG. 24, first, in the count time of the timer 106 described in the second embodiment, the node 3 of the next layer of the node 2-1 -1 and 3-2 start to connect to nodes located on the higher level of node 2-1. During the counting period of the timer 106, in the connected router, the node 3-1 stores the data released from the node 0-1 in the buffer memory 4, and the node 3-2 stores the data released from the node 1-1 In the buffer memory 4, and distribute the data to the lower nodes 4-1 and 4-2. After the topology is reconstructed, the data distributed from the main line LM is stored in the buffer memory 4 and distributed from the sub-line LS The data is published to the lower node.

[0252] The node 3-1 or 3-2 (the node 3-1 in the case of FIG. 24) having a higher speed is connected to the node 1-1. The node 3-1 connects the new main line LM to the node 1T, in addition, maintains the connection to the node 0-1 via the original sub-line LS, and receives the content published from the node 0-1 (backup release).

[0253] On the other hand, the node 3-2 that cannot be connected to the node 1-1 is connected to the node 3-1 currently belonging to the lower layer of the node 1-1 via the main line LM, while maintaining the connection with the node 1-1. In this case, the original sub-line LSo, the node 3-2 can be directly connected to the node 0-1. However, many nodes are currently connected to the node 0-1, so the main line LM is actually connected to the node 3- 1 ο

[0254] For the nodes 4-1 and 4-2 located on the lower level, the original sub-line LS is disconnected. As a result, the nodes 4-1 and 4-2 inquire about the node 3-2 that belongs to the sub-line LS, and the node 3-2 above the node to be newly connected to. As a result, it can be recognized that a new master is formed between the nodes 3-2 and 3-1 The line LM, so that the node 4-1 or 4-2 that has inquired about the node 3-2 first forms a new sub-line LS' to the node 3-1.

[0255] Next, referring to FIG. 25, operations performed when a new node is connected to the network system NT2 according to the third embodiment and the distribution of content is started and executed in a normal state including the new node will be described all at once. FIG. 25 is a flowchart showing a situation where the node 3-1 is newly connected to the node 2-1 in the network system NT2 to obtain the network system NT2 as shown in FIG. 23, and the content is released to the node 3-1 example of.

[0256] In at least the following nodes among the nodes in the network system NT2 according to the third embodiment, the processing in the flowchart shown in the upper right part is always performed, and the above-mentioned node is another node via the main line LM or The node to which the sub-line LS is connected at a lower level.

[0257] Specifically, whenever a message is sent from a node at a lower level (step S170), the type of the message is sequentially determined (step S171, S173, S175, S177, or S179).

CN 1998199 Β

[0258] When the received message is a start message (Y in step S171), the packet number of the data currently being replayed by the node is sent back as the start packet number (step S117).

[0259] When the received message is a topology data table request message (Y in step S173), in the topology table T currently provided for the node, according to the topology data table request message sent back to the node connected to The topology table T of the line type (main line or sub line) is sent back (step S174).

[0260] When the received message is a data request message (Y in step S175), one packet of data is sent at the playback speed specified in the data request message (step S176).

[0261] When the received message is support request information for a new node that inquires about the possibility of newly using the sub-line LS to publish data (Y in step S177), it is determined according to the association with another node that is publishing data at this time Is it possible to use sub-line L for publishing. If possible, reply with an "Allow" message. On the other hand, if it is not possible, a "prohibited" message is returned (step S178).

[0262] When the received message is a mandatory connection request message for expressing the mandatory release of data using the new sub-line LS (Y in step S179), in a manner similar to the case of step S177, according to the current situation The association of another node that publishes data determines whether it is possible to use the sub-line LS for publishing. If possible, reply with an "Allow" message. On the other hand, if it is not possible, stop data release. The posting is to the following node (step S180), the node being the layer information in the topology table T according to the type of line connected to the node that has sent the mandatory connection request information in the topology table currently provided to the node The node that is "-1".

[0263] When the received message is not any of the above messages (N in step S179), a preset error message is sent back to the node that has sent the message (step S181).

[0264] In each node that continuously performs the above processing, the decoder 102 decodes the data from the node whose layer information is "1" in the topology table T corresponding to the main line LM and outputs the decoded data. On the other hand, data from a node whose layer information is "1" in the topology table T corresponding to the sub-line LS is released to another node located on a lower level without being decoded.

[0265] In the network system NT2 operating in this state, when the node 3-1 newly enters the network system NT2 by physically connecting to the node 2-1, first, the CPU 100 in the new node 3-1 sets the node 3 The two buffer memories 104 in -1 (used for the main line LM and the sub-line LS), set the IP address of the node indicating the data supply source, and set the node 0-1 on the upper level to the sub-line LS Up (when node 3-1 is connected to node 2-1, node 3-1 obtains the IP addresses of the two nodes from node 2-1). In order to calculate the address at which data is stored in each buffer memory 104, the input counter 1 (16 bits), the input counter 2 (16 bits), and the output counter 1 (16 bits) are initialized to "0". The input counter 1 counts the number of bytes of received data issued from the node 0-1 via the main line LM. The input counter 2 counts the number of bytes of received data issued from the node 0-1 via the sub-line LS. The output counter shows the amount of data reproduced by the decoder 102. In addition, at this time, a buffer address indicating "ring buffer 1" (in FIGS. 25 and 26, appropriately expressed as "ring buffer ADR") is set in the buffer memory 104 corresponding to the main line LM (step S160) .

[0266] After completing the necessary initialization processing, the CPU 100 in the node 3-1 refers to the topology table T in the table memory 3, and sends a start message for requesting the nodes 2-1 and 0-1 to start content data Transmission (step S161).

[0267] Next, the CPU 100 in the nodes 2-1 and 0-1 that have received the start message sends the packet number of the data currently replayed by the nodes 2-1 and 0-1 to the node as the start packet number 3-1 (Step S172).

[0268] The node 3-1 obtains the start packet number from the nodes 2-1 and 0-1, and stores it in the buffer memory corresponding to the line

CN 1998199 Β

The corresponding packet number area on the device 104 (step S162), and a data request message for requesting actual content data is sent to the nodes 2-1 and 0-1 (step S163). The data request information includes the packet number information of the data to be sent from nodes 3-1 and 2-1 (replay processing) and node 0-1 (used to immediately publish data to a lower level).

[0269] The nodes 2-1 and 0-1 that have received the data request message ("Y" in step S170 and S175) at the playback speed specified in the data request message (via the main line LM and sub Line LS) Send the data amount of one packet to node 3-1 (step S176) ο

[0270] In parallel with this operation, node 3-1 is set to output a data request message, and at the same time, counting is started by the timer in node 3-1, and a timer interrupt instruction is independently generated when the count becomes "0" (Step S164). The node 3-1 waits until a predetermined time elapses in the count of each timer 106 (step S165). When the data from the nodes 2-1 and 0-1 does not arrive within the predetermined time (N in step S165), the node 3-1 executes the timer interrupt instruction, and performs the connection to the nodes 2-1 and 0-1 again deal with.

[0271] On the other hand, when the data from the nodes 2-1 and 0-1 arrive within a predetermined time (Y in step S165), the node 3-1 prohibits execution of the timer interrupt instruction (step S166), and sets The amount corresponding to one packet of received data is stored in the buffer memory 104. Along with this processing, the values of the input counters 1 and 2 and the packet number stored in the buffer memory 104 are only updated by the amount of one packet.

[0272] The node 3-1 determines whether the amount of data stored in each buffer memory 104 and not yet output to the decoder 102 becomes four packets (that is, 1/4 of the storage amount of the buffer memory 104) (step S168). When the amount is less than four packets (Y in step S168), the node 3-1 returns to step S163 to receive subsequent packets from the nodes 2-1 and 0-1, and repeats the above-mentioned processing. On the other hand, when the amount of data stored in each buffer memory 104 becomes four packets (N in step S168), only the data acquired via the main line LM is output to the decoder 102 (step S169), And it returns to the processing in step S163 to receive subsequent data from the nodes 2-1 and 0-1.

[0273] In the process of step S169, when the data amount of the four packets is stored in the buffer memory 104 corresponding to the main line LM and output to the decoder 102, the decoder 102 only passes the four packets Repeat the following operations to perform data replay processing. In these operations, the decoder 102 obtains the buffer address of the buffer memory 104 and the address of the data output from the buffer memory 104 according to the value of the output counter in the buffer memory 104 indicated by the buffer address (step S182). The data is decoded and replayed in one packet (step S183), and the value of the output counter in the buffer memory 104 is increased by only one packet (step S184).

[0274] By performing the data distribution of the slave node 2-1, the replay processing in the node 3-1, and the distribution of the slave node 0-1 using the sub-line LS as described above, the data reproduction in the node 3-1 is performed Processing while keeping the amount of data in each buffer memory 104 unchanged.

[0275] Next, referring to FIG. 26, it will be described in detail that the content relay function in the node 2-1 stops the processing performed due to reasons such as the power switch being turned off in the network system NT2 as shown in FIG. 23.

[0276] In the network system NT2 according to the third embodiment, when the relay function in the node 2-1 is stopped due to the above-mentioned reason, as shown in FIG. 24, the node 3-1 connected to the original node 2-1 And 3-2 and nodes 4-1 and 4-2 automatically perform topology reconstruction operations, and continue the distribution of content through the topology in the pattern as shown in FIG. 24.

[0277] When the relay function in the node 2-1 as shown in FIG. 25 is stopped, even after a predetermined time has elapsed in the processing of step S165 as shown in FIG. 25, the node 3-1 located on the next level And the CPU 100 in each of 3-2 (because the distribution in the normal state as shown in FIG. 25 is continued, the processing described with reference to FIG. 25 is repeated) also cannot receive data from the node 2-1. When the data cannot be received even after the predetermined time has passed (N in step S165), the node

CN 1998199 Β

The CPU 100 in each of 3-1 and 3-2 recognizes that the relay function in the node 2-1 located on the upper level is stopped and the topology is disconnected.

[0278] After that, when the topology disconnection is recognized, the topology table T stored in the table memory 103 is referred to to execute as shown in FIG.

26 shows the interrupt instruction, and as shown in Figure 24 to reconstruct the topology.

[0279] Specifically, as shown in FIG. 26, when the interrupt instruction is executed, it is determined whether the timer 106 that has passed the predetermined time is the timer 106 corresponding to the main line LM (indicated as "timing" in FIG.Device1") (Step S190). When it is the timer 106 corresponding to the main line LM (Y in step S190), the value of the buffer address is changed to the value used to indicate the buffer memory 104 corresponding to the sub-line LS, so as to compare the value of the buffer memory 104 corresponding to the sub-line LS. The corresponding buffer memory 104 is set to output data to the buffer memory 104 of the decoder 102 (step S191). Next, in the main line LM, the topology table request message corresponding to the main line LM is sent to the node (node 1-1 in the case of FIG. 24) whose layer information value is "2" (steps S192 and S194), and obtain the topology table T corresponding to the main line LM from the topology table T in the node (step S195).

[0280] By using the obtained topology table T, the topology table T in the nodes 3-1 and 3-2 corresponding to the main line LM is updated (step S196).

[0281] Next, the new node support request message queries the node whether the sub-line LS can be connected to the node whose layer information is "2" in the updated topology table T (node 0-1 in the case of FIG. 24) (Step S197).

[0282] Check whether the reply of the new node support request message is "allowed" (step S198). If it is "permitted" (Y in step S198), the IP address of the node to which the sub-line LS is connected before the relay function is stopped is rewritten to the IP address of the node to which the main line LM is currently connected, and the sub-line LM is currently connected to The IP address of the node to which the line LS is newly connected (node 0-1 in the case of FIG. 24) is used as the IP of the node whose layer information is "2" in the topology table T corresponding to the current (new) subline LS The address is written into the topology table T (step S199) ο

[0283] On the other hand, when it is determined in step S198 that the reply of the new node support request message is "prohibited" (N in step S198), the mandatory connection request message is sent to the initial connection of the sub-line LS. Node 1-1 on the next level of node 0-1 (step S200) ο

[0284] When the node 1-1 receives the mandatory connection message and the sub-line LS can be connected in the node 1-1, the connection is allowed in the node 1-1, and the topology table T in the node 3-1 is Update (step S201). When it is not possible, the distribution of data from node 0-1 to node 1-1 is stopped. The node 1-1 to which the data transmission is stopped performs the above-mentioned processing, thereby reconstructing the topology structure.

[0285] On the other hand, when it is determined in step S190 that the interrupt instruction (N in step S190) is executed in the timer 106 corresponding to the sub-line LS, the relay function of the node 2-1 at the upper level The main line LM operates. Therefore, the topology table request message corresponding to the main line LM is sent to the node 2-1 (steps S193 and S194), the topology table T in the node 2-1 is obtained (step S195), and the processing of step S196 and subsequent steps is performed .

[0286] As described above, in the network system NT2 of the third embodiment, the main line LM and the sub-line LS are formed to connect a plurality of nodes to one node. The data distributed via the main line LM is used for playback processing in this one node or the like and distributed to other nodes belonging to a lower hierarchy. On the other hand, the data distributed via the sub-line LS is used to distribute to other nodes belonging to a lower level. By connecting multiple lines to each node, when preparing to stop the relay function in any node, redundancy can be increased, and it is possible to prevent the replay processing etc. in nodes belonging to lower levels from being stopped.

[0287] When the relay function in a node belonging to a higher level is stopped on the main line LM to a node, the data distributed via the sub-line LS is switched to be used for replay processing and the like, so as to be in the one node The replay processing and other categories are not

CN 1998199 Β

Was interrupted.

[0288] In addition, when data posted via the sub-line LS is switched for use in playback processing or the like, any node belonging to a higher hierarchy is retrieved, and a new node is formed by connecting to the retrieved node. The sub-line LS. Therefore, even when the sub-line LS is used as the main line LM, a new sub-line LS is formed, and redundancy can be ensured and maintained.

[0289] In addition, when the relay function in a node belonging to a higher level stops on a sub-line LS to a node, any node belonging to a higher level is retrieved, and is retrieved by connecting to the To form a new sub-line LS. Therefore, even when the relay function in the node on the sub-line LS is stopped, a new sub-line LS is formed, and redundancy can be ensured and maintained.

[0290] Since a node of a higher level is retrieved so that another node on the level to which the one node belongs is included in the new main line LM, the supply of posted information can be retrieved via the node at the same level.

[0291] In each of the foregoing embodiments, as a method for detecting the stop of the relay function in any node, in addition to the above method, the node can periodically detect whether there is a response to a higher or lower level node . For example, when the relay function of the node 2-1 itself in each embodiment is stopped, the stop may be notified to the node 101 at a higher level, or the nodes 3-1 and 3-2 at a lower level.

[0292] The series of connection mode control processing described above may also be performed under the control of another server device located outside the network system NT or NT2. In this case, the server device has the connection mode information of the nodes in the network system NT or NT2. By sending the query from the nodes 3-1 and 3-2 to the server device, the node 1-1 located on the upper level of the node 2-1 whose relay function is stopped is identified.

[0293] In addition, the decoder 102, CRT 108, etc. for replaying content may also be configured to be connected via another network

Replay equipment at locations other than the T node, etc.

[0294] In addition, it is also possible to record the program corresponding to the flowcharts of FIGS. 18, 21, 22, 25, and 26 on an information recording medium such as a floppy disk or hard disk, or obtain and record the program via the Internet or the like, and The program is read and executed by an ordinary computer, so that the computer functions as the CPU 100 according to these embodiments.

Industrial applicability

[0295] As described above, the present application can be applied to the field of publishing content by using a network system having a tree structure. In particular, when the present invention is applied to the real-time broadcast content distribution field such as movies, music, etc., which is not convenient for distribution interruption, obvious effects can be obtained.

CN 1998199 Β