KR100449492B1 - Method for jumping in multicast video on demand system - Google Patents

Abstract

The present invention relates to a method of jumping in a situation where a client can simultaneously receive stream data on two multicasting channels. Conventionally, in a batch-based video-on-demand system that sets a fixed time interval and processes a requested service during the time, a simple method of finding and joining a group of channels transmitting the closest jump position has been used. This method has a disadvantage in that it requires as much waiting time as necessary to join a channel group. The present invention has the effect of improving the utility of the system by enabling the user to interact with the video data by supporting a jump method in a multicasting environment. In addition, unlike the conventional method, it enables the jump without waiting time and saves the network bandwidth by using the data stored in the disk.

Description

Jump method in video-on-demand systems using multicast channels to send stream data {METHOD FOR JUMPING IN MULTICAST VIDEO ON DEMAND SYSTEM}

The present invention relates to a jump method in an on-demand video system for transmitting stream data using a multicasting channel, and more particularly, to a jumping method in a situation in which a client can receive stream data simultaneously on two multicasting channels. will be.

In order to increase the number of users that can be supported while being the most important element in an on-demand video system, there is a lot of research on a technique of allowing multiple clients to share the same stream.

1 is a block diagram illustrating service provision of an on-demand video system using multicasting, in which the first to nth clients C1 to CN in the client group 101 share the same channel with each other so as to share the network bandwidth by the number of shared channels. Can save. The first to nth clients C1 to CN receive the video moving picture file 104 through the video server 103 and the network manager 102.

Batching is the simplest method of multicasting, in which a user's request for a particular video is stored for a period of time and then processed into a single multicasting channel. This approach incurs an initial latency delay because each client's service period must wait for the next batching cycle.

To solve this problem, the existing multicasting streams are shared and stored temporarily on disk, and a separate channel is created to receive and play an initial data stream that cannot be shared. When the reproduction of the initial data is finished, the reproduction of the data stored in the disc is performed.

2 is a schematic diagram illustrating a patching scheme. When clients A and B request a service at a time interval for the same video, a multicasting channel Ch1 is generated for the first client A requested. When the client A makes a session request from the client B while playing the video at time t1, the S2 portion of the multicasting channel Ch1 can be shared. Patching is performed by creating a separate channel Ch2 for S1 that cannot be shared in Ch1. The playback order of the client B is to play the S1 patched on the channel Ch2 first, and when the playback of the S1 is finished, play the S2 portion shared on the Ch1 and stored on the disc. In the client B, data S1 corresponding to time t1 is always stored on the disk.

At this time, the channel Ch1 is called a regular channel and is expressed as MCr. Ch2 is called a patching channel and is expressed as MCp. The interval in which the client performs patching is called a patching interval, and the time length of the patching interval is called a patching length. The patching interval of client B is S1 and the patching length is t1.

Patching can reduce network bandwidth as much as shareable multicasting streams and minimize the initial latency of clients. It is also an effective way to save network bandwidth on the server, even for less popular video.

Jump is used when a client moves the video playback position to an arbitrary position during video playback, and the jump method in an on-demand video system enhances the utility of the system by enabling user interaction. Jumping in a batch-based video-on-demand system that sets a constant time interval and processes the requested service during that time is simply solved by simply changing the channel group of the stream to which it is sent.

The simplest way to support jumps in the patching method is to search for the fastest shareable regular channel, as in the batching method, and share it when the regular channel starts sending the jump position. Supporting the jump method using only regular channels has the advantage that no additional use of channels is required, but the waiting time for the service becomes too long. Fast service response is important for VCR operations, and therefore requires an approach that minimizes latency, even allowing for additional use of the channel.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described drawbacks, and provides a method for efficiently handling jumps while minimizing the use of a server's multicasting channel in an on-demand video system that transmits stream data using the multicasting channel. There is a purpose.

1 is a block diagram illustrating service provision of an on-demand video system using multicasting;

2 is a schematic diagram conceptually illustrating a patching technique;

3 is a block diagram showing a service provision of an on-demand video system for transmitting stream data using a multicasting channel used in the present invention;

4 is a conceptual diagram of a jump when a patching process is completed in a jump method in an on-demand video system that transmits stream data using a multicasting channel according to the present invention;

5 is a conceptual diagram of a jump when having a patching channel in a jump method in an on-demand video system that transmits stream data using a multicasting channel according to the present invention;

6 is a flowchart illustrating a step-by-step jump method when a patching channel is completed in a jump method in an on-demand video system for transmitting stream data using a multicasting channel according to the present invention;

7 is a flowchart illustrating a step-by-step jump method when a patching channel is included in a jump method in an on-demand video system for transmitting stream data using a multicasting channel according to the present invention.

<Description of the symbols for the main parts of the drawings>

201: Client 202: Regular Channel

203: patching channel 204: control channel

205: Video Server 206: Video Information Manager

207: Channel Manager 208: Session Manager

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

3 is a block diagram illustrating a service providing of an on-demand video system that transmits stream data using a multicasting channel used in the present invention. The on-demand video system includes a client 201 and a video server 205. Communication channels between the client 201 and the video server 205 include a control channel 204, a regular channel 202, a patching channel 203, and the like. The control channel 204 is used for transmitting control messages between the video server 205 and the client 201, and the regular channel 202 and the patching channel 203 are used for transmitting actual video data. The video server 205 consists of a video information manager 206, a channel manager 207, and a session manager 208. The video information manager 206 manages video information supported by the video on demand system. The channel manager 207 and the session manager 208 manage channel and session information generated by the video server 205.

The video information manager 206 manages information on video data such as Table 1 to be serviced.

VID ID of the video λi Arrival rate of video i (request / min) Li Length of video i (min) Ti Patchable range of video i (min)

In patching, the client 201 receives streams simultaneously on up to two channels. Client 201 can be shared if at least one regular channel 202 exists for the same video. However, as time goes by, the shareable regular stream decreases and the length of the patching stream grows, so continuous allocation of the patching channel 203 is rather inefficient in terms of network channel usage. Thus, even if patching is possible, it is advantageous to newly allocate the canonical channel 202 so that later requesting clients share more canonical streams.

The patchable range Ti, which minimizes the use of network bandwidth for a particular video i, is represented as a function with an optimized value depending on the arrival rate λi for the video and the video size Li. That is, the patchable range has a smaller value as the arrival rate increases, and the value becomes larger as the video size increases. Patchable time Ti also represents the maximum amount of data to be stored on the client disk. Since the client 201 cannot patch data beyond the disk storage space, if the size of the disk is smaller than Ti, the Ti value becomes the size B of the disk. The patchable range of the regular channel 202 in video is as follows.

Ti = (sqrt (2Li * λi)-1) / λi (if Ti <B)

Ti = B (else)

Information managed by the channel manager 207 in the video server 205 is shown in Table 2.

MCID Multicast Channel ID MCvid Video ID being serviced on the multicast channel MCtype Type of transmission, with regular channel (R) and patching channel (P) values MCstate Indicates the status of the transmission, has values for transmission (S) and pause (P) MCdelay Number of minutes that a multicasting transmission stopped MCstart Absolute time when multicasting started MCbegin Relative transmission start position (min) in video i 0 <= MCbegin <Li MCend Relative transmission end position (min) in video i 0 <MCend <= Li, MCend = Li for regular channel MCsn The number of sessions belonging to the channel

The current transmission position of the channel or the current reproduction position of the client 201 uses a position time value indicating how many time intervals a specific point in time of the video is located after the beginning of the video. MCbegin and MCend, which are managed by the channel, represent the start and end positions of video i as time values, respectively. In other words, if MCbegin = 5 and MCend = 15, the 15-minute position is transmitted from the 5-minute position of video i. Multicast channels may be temporarily suspended by VCR operations. MCdelay is the total time at which transmission was stopped. Therefore, if the current time is Tcurr and the start time of the multicast channel is MCstart, the current transmission position of the channel is Tcurr-MCstart-MCdelay + MCbegin.

The session information managed by the session manager 208 has regular channel, patching channel, and patching length values as shown in Table 3.

SID Session id MCr Regular stream channel ID MCp Patching Stream Channel ID Plength Patching Length of Session B Buffer size of the client

The jump method largely depends on whether the patching process receives data at the regular channel 202 and the patching channel 203 at the same time, or whether the reception at the patching channel 203 is completed and is being received only at the regular channel 202. In addition, each case is divided according to whether it is a jump in the patchable interval range or a jump out of the range, and the operation scenario is different in each case.

When a patching process is completed in a jump method in an on-demand video system that transmits stream data using a multicasting channel according to the present invention, there are four cases of jumping as shown in FIG. 4.

In a jump method in a video system using on-demand transmission of multi-cast channels according to the present invention, jumps may have five cases as shown in FIG. 5.

6 is a flowchart illustrating a step-by-step jump method when a patching channel is finished in a jump method in an on-demand video system that transmits stream data using a multicasting channel according to the present invention.

First, the client 201 sends the session ID and jump position to the video server 205 for the jump (step 301).

The video server 205 checks if it is a jump within the patchable interval range (step 302).

The video server 205 checks whether the jump is possible using the data stored in the disk of the client 201 when the jump is within the patchable interval range as a result of the check in step 302 (step 303). Since the data of patching time size is shared on the regular channel and stored on the disk, it is possible to jump to the disc data if the jump position is between the current playback position and the transmission position of the regular channel.

The video server 205 adjusts the patching length without changing the channel if it is possible to jump using the data stored in the disk of the client 201 as a result of the inspection of step 303 (step 304).

The video server 205 generates a new patching channel when the jump is impossible using the data stored in the disk of the client 201 because the jump position is smaller than the current playback position as a result of the check in step 303 (step 305). The transmission start position of the generated patching channel becomes a jump position, and the transmission end position becomes a current playback position.

The video server 205 deletes the patching channel by ignoring the data stored in the disk during the patching process without sharing the current normal channel any more when the jump of the patching interval is not performed as a result of the inspection of step 302 (step 306). ).

The video server 205 checks whether the current session has a regular channel uniquely used (step 307).

The video server 205 deletes the regular channel if the current session has a regular channel uniquely used as a result of the inspection of step 307.

The video server 205 searches for a regular channel in which a jump position exists in a patchable section if the current session does not have a regular channel used after the execution of step 308 or as a result of the inspection in step 307. It is checked if it can be shared (step 309).

The video server 205 generates a new patching channel if the regular channel can be shared as a result of the check in step 309 (step 310). The transmission start position of the patching channel becomes the jump position, and the transmission end position becomes the current transmission position of the regular channel. The patching length becomes "transmission position or jump position of the found regular channel".

The video server 205 generates a new normal channel if the normal channel cannot be shared as a result of the check in step 309 (step 311).

The client 201 selectively receives a regular channel, a patching channel, a patching length, and a waiting time from the video server 205 each time step 304, step 305, step 310, and step 311 are performed (step 312).

7 is a flowchart illustrating a step-by-step jump method when a patching channel is included in a jump method in an on-demand video system that transmits stream data using a multicasting channel according to the present invention.

First, client 201 sends a session ID and jump location to video server 205 for a jump (step 401).

The video server 205 checks whether it is a patchable interval range (step 402).

The video server 205 checks whether the jump is possible using the data stored in the disk of the client 201 when the inspection result of the step 402 is a patchable interval range (step 403). The patching channel stores the time-sized data in the regular channel and stores it on the disk.

The video server 205 adjusts the patching length when the jump operation can be performed using only the data stored in the disk as a result of the inspection in step 403 (step 404).

The video server 205 deletes the patching channel in this case since no more patching is needed (step 405).

The video server 205 changes the transmission start position of the patching channel to the jump position when the jump operation is impossible as the disk storage data as a result of the check in step 403 (step 406).

Video server 205 adjusts the patching length (step 407). Do not change the transmission end position.

The video server 205 no longer shares the current regular channel if the jump is outside the patchable interval range as a result of the check of step 402, and ignores the data stored on the disk during the patching process, and the current session is used only. Check if there is a regular channel (step 408).

The video server 205 deletes the normal channel if the current session has a regular channel uniquely used as a result of the check in step 408 (step 409).

The video server 205 searches for a regular channel having a jump position in a patchable section if the current session does not have a regular channel used after the execution of step 409 or as a result of the check in step 408. It is checked if it can be shared (step 410).

The video server 205 generates a new patching channel when the test result of step 410 enables sharing of other regular channels (step 411). The transmission start position of the patching channel becomes the jump position, and the transmission end position becomes the current transmission position of the regular channel. The patching length becomes "transmission position or jump position of the found regular channel".

The video server 205 generates a new normal channel if the normal channel cannot be shared as a result of the check in step 410 (step 412).

The client 201 selectively receives a regular channel, a patching channel, a patching length, and a waiting time from the video server 205 whenever step 405, step 407, step 411, and step 412 are performed (step 413).

As described above, the present invention supports a jump operation to an arbitrary position in a patching environment in which a client can simultaneously receive video data in two channels, thereby enabling the user to interact with the video data, thereby increasing the utility of the system. Giving effect. It supports latency-free jump operations and reduces network bandwidth by using data stored on disk.

Claims (12)

In the on-demand video system having a video server that is connected to the client with a regular channel, a patching channel, and a control channel to provide a multicasting channel to the client, when the patching channel is finished, A first step of the client sending a session ID and a jump position to the video server for jumping; A second step of checking whether the video server is a jump within a patchable interval range; A third step of checking whether the video server can jump using the data stored in the disk of the client when the video server is a jump within the patchable interval range as a result of the checking of the second step; A fourth step of adjusting, by the video server, a patching length if a jump is possible using data stored in the disk of the client as a result of the inspection in the third step; A fifth step of generating, by the video server, a new channel if the jump is outside the patchable interval range as a result of the inspection of the second step; And Using a multicasting channel comprising a sixth step of selectively receiving a regular channel, a patching channel, a patching length, a waiting time from the video server each time the fourth and fifth steps are performed respectively. Jump method in video-on-demand systems that transmit stream data. The method as claimed in claim 1, wherein the video server generates a new patching channel when the jump position is less than the current playback position and jumping is impossible using the data stored in the disk of the client. The method of claim 1, further comprising the step of performing. The method of claim 1, The fifth step, A 51 st step of deleting, by the video server, the patching channel without deleting the current normal channel any more if the jump is outside the patchable interval range as a result of the checking of the second step; A fifty-second step of checking whether the video server has a regular channel uniquely used by the current session; A 53rd step of deleting the normal channel when the video server has a regular channel uniquely used by the video server as a result of the check in the 52nd step; If the video server does not have a regular channel used solely for the current session after the execution of the 53rd step or the test of the 52nd step, the existing server searches for a regular channel having a jump position in a patchable section. Step 54, checking whether the regular channel can be shared; And And the fifty-fifth step of creating a new patching channel and performing the sixth step when the video server checks that the regular channel can be shared as a result of the check in the fifty-fourth step. The method of claim 3, wherein And if the video server checks that the regular channel cannot be shared as a result of the check in the 54th step, creating a new normal channel and performing the sixth step. The method of claim 2, wherein the transmission start position of the generated patching channel is a jump position. 2. The method of claim 1, wherein the patching length adjusted in the fourth step is " transmission position or jump position of the found regular channel. &Quot; A jump method in the case of having a patching channel in an on-demand video system having a video server connected to a client with a regular channel, a patching channel, and a control channel to provide a multicasting channel to the client, A first step of the client sending a session ID and a jump position to the video server for jumping; A second step of checking whether the video server is in a patchable interval range; A third step of checking whether the video server can jump using the data stored in the disk of the client when the video server has the patchable interval range as a result of the checking of the second step; A fourth step of adjusting, by the video server, a patching length if a jump operation can be performed using only data stored on a disk as a result of the inspection of the third step; A fifth step of the video server deleting a patching channel; A sixth step of generating, by the video server, a new channel when the check result of the second step is a jump outside the patchable interval range; And Using a multicasting channel comprising a seventh step of selectively receiving a regular channel, a patching channel, a patching length, and a waiting time from the video server whenever the fifth and sixth steps are performed respectively. Jump method in video-on-demand systems that transmit stream data. The method of claim 7, wherein An eleventh step of changing, by the video server, a transfer start position of a patching channel to a jump position when a jump operation is impossible as the disk storage data as a result of the inspection in the third step; And And a twelfth step of the video server adjusting the patching length and performing the sixth step. The method of claim 7, wherein The sixth step, When the video server checks that the jump is outside the patchable interval range, the video server no longer shares the current regular channel, ignores the data stored on the disk during the patching process, and uses the current session only. A sixty-first step of checking whether a regular channel is present; A sixty-second step of deleting, by the video server, the regular channel if the current session has a regular channel uniquely used as a result of the checking in the sixty-first step; If the video server does not have a regular channel that the current session uses after the execution of step 62 or as a result of the check of step 61, the existing server searches for a regular channel having a jump position in a patchable section. Step 63, checking whether the regular channel can be shared; And And a sixty sixth step of creating a new patching channel and performing the sixth step if the video server can share another regular channel as a result of the check in the 63rd step. The method of claim 9, And if the video server checks that the regular channel cannot be shared as a result of the check in the 63rd step, generating a new normal channel and performing the sixth step. 9. The method of claim 8, wherein the transmission start position of the patching channel is a jump position. 8. The method of claim 7, wherein the patching length adjusted in the fourth step is " transmission position or jump position of the found regular channel. &Quot;