CN1976444A - Digital medium channel switching method and system - Google Patents

Abstract

The invention provides a digital media channel switching method and a digital media channel switching system, which are used to solve the problem in the prior art that user channel switching time is too long. The method of the invention includes: setting a multicast replication point on the access device of the user end, and performing buffer processing on the multicast data, and sending the buffer processed multicast data to the terminal when the channel is switched. The present invention transfers the jitter buffer processing performed by the terminal to the user-side access device to complete, so that the multicast data is immediately de-jittered on the user-side access device, because the user-side access device is the system device closest to the terminal , so the requirements for jitter removal processing can be reduced at the terminal, and the low watermark of the terminal buffer can be set lower, which can reduce the duration of T1 during the user switching process, thereby reducing the channel switching time.

Description

A digital media channel switching method and system

technical field

The invention relates to the communication field, in particular to a digital media channel switching method and a digital media channel switching system.

Background technique

IPTV, or Interactive Network Television, is a technology that utilizes broadband networks and integrates technologies such as the Internet, multimedia, and communications to provide home users with a variety of interactive services including clear and smooth video programs. This technology is mainly provided by telecom operators. Video services can be easily bundled and integrated with voice, data and mobile nomadic services provided by telecom operators to form a user-centered integrated service system, which shows great potential. market potential. The IPTV technical system architecture is divided into several parts such as the head-end service system, the bearer transmission network, and the terminal, each of which performs different functions.

As shown in Figure 1, it is a schematic diagram of a common IPTV network architecture, which is mainly divided into a head-end service system, a bearer network, user-side access devices, and terminals. The head-end service system provides content service and management, the bearer network completes service transmission, the user end access device provides user access and sends service content, and the terminal provides service content playback. Among them, the terminal part includes TV, personal computer (PC) and set-top box (STB), etc. Currently, there are many terminal applications using STB+TV. Among them, the STB is mainly used as a receiving and processing device for digital video signals, and interacts with the network to realize IPTV service functions, and the TV completes the display work of digital video.

IPTV services mainly include VOD/NVOD (video on demand/quasi video on demand), BTV (live TV), Time-shift TV (time-shift TV), PVR/nPVR (personal video storage/personal network video storage and other main services, of which BTV Also commonly known as live TV, programs on the same channel are sent to all users who watch this channel. The head-end service system uses multicast technology to simultaneously push the same channel program (media stream) to all users who choose this channel through the bearer network.

When the BTV service adopts the multicast technology to push channel programs, two key problems need to be solved: the selection of the multicast control point and the selection of the multicast replication point.

The multicast control point completes the judgment of the user's channel authority, which is usually implemented on the user's access device. When the user selects a channel, the terminal device (set-top box device, etc.) sends an Internet Group Management Protocol (IGMP) request to join the corresponding channel. After receiving the request, the multicast control point judges whether the user has the right to watch the program according to the user's order information.

The multicast replication point replicates the multicast data flow to the port according to whether the port has a multicast reception request and the corresponding user rights. Below the multicast replication point, live broadcast programs are transmitted to users through unicast, so the closer the multicast replication point is to the user equipment, the more network bandwidth can be saved. Usually, the multicast replication point is set on the user end access device (such as digital subscriber line access multiplexer (DSLAM) or broadband access server (BRAS), etc.), and the devices above the multicast replication point support multicast, and the user end The access device needs to support IGMP snooping, IGMP proxy and other protocols to realize the controllable forwarding of multicast data.

During the process of receiving BTV programs, the user needs to change the channel. In the process of changing from one channel to another, there is inevitably a time delay problem caused by changing channels. As shown in Figure 2, it is a schematic diagram of the process for the user to change the channel. It can be seen from the figure that the user needs to go through the following steps to change the channel:

Step 201, the user sends a channel switching request to the DSLAM, requesting to leave the current channel;

Step 202, the DSLAM stops the signaling control of the video stream being played, and leaves the multicast group;

Step 203, the user sends a request to join a new channel to the DSLAM;

Step 204, the DSLAM sends new channel content to the terminal, and the user switches to the new channel to receive the multicast data stream;

Step 205: After the terminal buffer reaches the low water mark, the data stream is dispatched to the decoding buffer, and the decoder waits for the first important frame (I frame) to output the video code stream, complete the channel change, and normally receive the content sent by the new channel.

It can be seen from the above process that the main factors causing the delay in the channel switching process are as follows:

1. The time T1 for scheduling the data flow to the decoding buffer after the terminal jitter buffer reaches the low watermark;

2. The time T2 it takes for the terminal decoder to decode and output the video code stream after obtaining the first I frame.

The above times T1 and T2 can be controlled by some schemes.

The jitter buffer is a buffer set in the terminal to compensate for the impact caused by network jitter. It is used to cache data packets for a period of time before forwarding data packets to smooth the transmission of data packets, and compensate for packet jitter, loss, delay and other conditions. adverse effects caused. The T1 value can be reduced by adjusting the settings of the jitter buffer on the terminal, and the T1 value can be reduced to a certain extent by setting the high-watermark and low-watermark of the jitter buffer. However, the downside of adjusting the low watermark and high watermark of the jitter cache is that it will have other effects on successive generations. Since the size of the jitter buffer is generally an integer multiple of the estimated arrival time of the data packet, if the buffer is set too small, it may not be able to overcome the impact of network packet loss and jitter, and reduce the quality of service; if it is set too large, it will cause Excessive delay will also cause bad user experience.

In the prior art, in order to reduce the delay when users change channels, the multicast replication point can choose to join the multicast group statically, that is, for a certain channel, no matter whether there is a terminal joining the multicast group, the multicast replication point will join the multicast group in advance If there are users in the group, the multicast data will be sent directly to the users, and if there are no users, the multicast data of the channel will be discarded. The main purpose of statically joining a multicast group is to speed up the speed of users joining the multicast group, and will not wait for a long time because the user requesting channel switching is the first user to switch to the channel, thereby ensuring the delay and delay of the BTV service flow. The problem of switching time is actually to exchange bandwidth for time, which speeds up the channel switching speed and saves the transmission delay of programs from the live broadcast source to the multicast replication point.

In addition, during channel switching, the channel switching process is further accelerated by optimizing the IGMP protocol (that is, by implementing multicast fast leave).

The channel switching process after the above optimization still has a switching transmission delay of 2 to 3s. The main bottleneck lies in the buffer settings in the terminal to deal with network jitter and the time spent in searching for I frames in the terminal. For BTV users, it is still is unbearable.

Contents of the invention

Embodiments of the present invention provide a digital media channel switching method and a digital media channel switching system, which are used to solve the problem in the prior art that user channel switching time is too long.

A digital media channel switching method, comprising:

Set the multicast replication point on the client access device;

After the client access device receives the multicast data sent by the head-end service system through the bearer network, it caches the multicast data;

When the channel is switched, the buffered and processed multicast data is sent to the terminal.

A digital media channel switching system, including a head-end service system that provides service sources, a user-side access device that provides access for terminals, the user-side access device sets a multicast replication point, and the user-side access device Also set with:

A cache unit, which caches the multicast data and removes network jitter;

When the terminal sends a channel switching request, the client access device sends the multicast data stored in the buffer unit to the terminal.

In the embodiment of the present invention, the jitter buffer processing performed by the terminal is forwarded to the user-end access device to complete, so that the multicast data is immediately de-jittered on the user-side access device. Since the user-side access device is the closest to the terminal System equipment, so the requirements for dejittering processing can be reduced at the terminal, and the low watermark of the terminal buffer can be set lower, which can reduce the duration of T1 during the user switching process, thereby reducing the terminal switching time.

Description of drawings

FIG. 1 is a schematic diagram of an IPTV network architecture in the prior art;

FIG. 2 is a schematic flow diagram of a user changing channels in the prior art;

3 is a schematic structural diagram of a digital media channel switching system according to an embodiment of the present invention;

4 is a schematic structural diagram of a digital media channel switching system according to Embodiment 2 of the present invention;

5 is a schematic flow diagram of a user changing a channel according to Embodiment 1 of the present invention;

FIG. 6 is a schematic flow diagram of a user changing a channel according to Embodiment 2 of the present invention.

Detailed ways

The specific implementation manner of the present invention will be described below in conjunction with the accompanying drawings.

In the embodiment of the present invention, the jitter buffer originally located at the terminal is placed on the user-side access device, so that the multicast data is de-jittered on the user-side access device. Since the user-side access device is the system closest to the terminal equipment, so the requirements for dejittering processing can be reduced at the terminal, and the low watermark of the terminal buffer can be set lower, which can reduce the T1 value during the user switching process, thereby reducing the terminal switching time.

On this basis, the embodiment of the present invention also pre-obtains the I frame information of the cached multicast data on the user terminal access device, thereby reducing the T2 value during the user switching process.

Embodiment one:

As shown in FIG. 3 , it is a schematic diagram of a digital media channel switching structure in Embodiment 1 of the present invention. In this figure, a DSLAM is an access device at a user end and provides access services for a terminal. A multicast replication point is set on the DSLAM, and the multicast replication point is added to the multicast group in a static joining manner. In this system, multicast transmission is carried out from the head-end system to the DSLAM, multicast data is copied and distributed at the DSLAM, and unicast transmission is performed from the DSLAM to the terminal. Embodiment 1 uses the STB as the terminal to illustrate, and it is also applicable to other terminals such as PCs. In the first embodiment, a DSLAM is used as the client access device, which is also applicable to other client access devices such as BRAS.

It can be seen from FIG. 3 that in Embodiment 1, a first buffer unit is provided on the user end access device DSLAM, and the multicast data is buffered through the first buffer unit.

The first cache unit set by the DSLAM caches the multicast data to remove network jitter. The first cache unit can store 1-2s multicast data frames to cope with the impact of network jitter on the multicast data flow. Since the group on the DSLAM The broadcast replication point has statically joined the multicast group, and the first buffer unit has been filled with the multicast data of the channel program when the user applies for the multicast service, because the first buffer unit has been added to the DSLAM to resist network jitter, and the DSLAM is already the closest The user's operator's equipment, so the requirements for removing network jitter can be reduced in the STB accordingly, and the cache low watermark of the STB can be set relatively small, such as a few frames, which can greatly reduce the T1 value.

On the basis of the above solution, an important frame information determining unit may be further set up for determining I frame information. When the first buffer unit buffers the multicast data, the important frame information determination unit determines the I frame information, so that when the multicast data buffered by the first buffer unit is sent to the terminal, the I frame can be quickly found, the specific scheme I frame start detection can be carried out at the ingress port of the first buffer unit, and the address pointer of the I frame is stored. When a user applies for channel switching, and the DSLAM starts to send a new channel data stream to the terminal, the first buffer unit directly reads from the This pointer starts to read the cached multicast data and transmits it to the terminal through unicast, because it is transmitted from the I frame, and the terminal can determine the I frame information when receiving the transmitted data, so it saves the terminal to search I frame time, thus speeding up the channel switching, can greatly reduce the T2 value.

As shown in Figure 4, it is a schematic flow diagram corresponding to the scheme of Embodiment 1. It can be seen from the figure that it mainly includes:

Step 401, the multicast replication point statically joins the multicast group, and caches the multicast data;

The multicast replication point located on the access device at the user end joins the multicast group in a static way, and immediately copies and caches the multicast data sent by the network. If the user with the activation permission requests the multicast data, the cached multicast data is sent to the user, otherwise the cached multicast data will be discarded immediately.

Step 402, determine I frame information;

The DSLAM receives the multicast data sent by the head-end service system through the bearer network, and determines the I frame information when the DSLAM device buffers the multicast data. This step can be performed before the buffered multicast data enters the first buffer unit. The I frame information is detected at the entrance of the unit, and the I frame address pointer is stored according to the detection result. This solution is convenient for quickly searching for the I frame when the multicast data buffered by the first buffer unit is sent to the terminal.

403. The user initiates a channel switching request;

When the user wants to switch channels, he sends a channel switching request to the DSLAM, activates the corresponding process, and the DSLAM stops sending the multicast data being played to the user, and controls the process of leaving and rejoining the user multicast group.

404. The user sends a new channel request;

The user sends a new channel request to the DSLAM, requesting to receive multicast data sent in the new channel.

405. Send the buffered multicast data to the STB according to the I frame information;

When the user applies for new channel content and the DSLAM starts to send multicast data of the new channel to the terminal, according to the multicast data cached by the first buffer unit, the cached multicast data can be read directly from the I frame pointer in the cache And transmit to the terminal, because it is transmitted from the I frame, when the terminal receives the data, the first frame is the I frame, so it can save the time for the terminal to search for the I frame, thereby speeding up the channel switching.

406. Schedule the data stream to the decoding buffer, and the decoding buffer outputs the video code stream.

The terminal decodes and outputs the data stream of the new channel whose first frame is an I frame according to the received first frame through a small buffer, and then outputs it to the TV for program playing.

In Embodiment 1, the first buffering process is performed on the user terminal access device, and the buffering and dejittering processing of the multicast data is performed, so that the multicast data is immediately dejittered on the user terminal access device. The input device is the system device closest to the terminal, so the requirements for dejittering processing can be reduced at the terminal, and the terminal cache low watermark can be set lower, which can reduce the T1 value during the user switching process, thereby reducing the terminal switching time.

Embodiment two

As shown in FIG. 5 , it is a schematic structural diagram of a digital media channel switching system in Embodiment 2 of the present invention. In this figure, a DSLAM is an access device for a user end and provides access services for a terminal. Set the multicast replication point on the DSLAM and join the multicast group statically. In this system, multicast transmission is performed from the head-end system to the DSLAM, multicast data is copied and distributed at the DSLAM, and unicast transmission is performed from the DSLAM to the terminal. Embodiment 2 uses the STB as the terminal to illustrate, and it is also applicable to other terminals such as PCs. In the first embodiment, a DSLAM is used as the client access device, which is also applicable to other client access devices such as BRAS.

It can be seen from FIG. 5 that in the second embodiment, a first cache unit and a second cache unit are provided on the user end access device DSLAM, and the multicast data is cached through the first cache unit to perform jitter removal processing, and the second cache unit is used to The cache unit performs I frame information processing on the multicast data cached and processed by the first cache unit.

In the solution of Embodiment 2, the first cache unit set by the DSLAM performs cache processing on the multicast data, and the multicast data frames of 1 to 2 s can be stored in the first cache unit to deal with the impact of network jitter on the multicast data stream , and because the DSLAM has statically joined the multicast group, the first buffer unit is already full of channel program media stream data when the user applies for the multicast service. Because the first cache unit has been added to the DSLAM to resist network jitter, and the DSLAM is already the operator's equipment closest to the user, so the corresponding requirements for jitter processing in the STB can be reduced accordingly, and the STB cache low watermark can be set relatively small , such as a few frames, which can greatly reduce the T1 value.

In the solution of the second embodiment, a second buffer unit is also provided for quickly searching for I frames when switching channels, and the multicast data buffered by the second buffer unit may be a duplicate copy of the multicast data buffered by the first buffer unit.

At the entry end of the second buffer unit, an important frame start detection unit is set to detect the multicast data I frame buffered and processed by the second buffer unit. Through this important frame start detection unit, the I frame start detection can be carried out at the ingress port of the second buffer unit, the I frame information is obtained, and the I frame address pointer is stored. When a user applies for channel switching, the DSLAM starts to send new channel data When streaming to the STB, the second buffer unit directly reads the channel program data from the pointer and transmits it to the STB, because it is transmitted from the I frame. After the STB receives the relevant data, the first frame is the I frame. So it saves the time of STB looking for I frame, thus speeding up the channel switching.

The second cache unit can also be finely set according to system requirements, and can be set as multiple independent storage units to cache and process data. In this way, when multiple users initiate channel switching requests at the same time, each storage unit can be used for different The user who initiates the switching request handles it, which can avoid problems caused by insufficient system resources when multiple users initiate switching requests at the same time. The storage units 1-n support multiple users to quickly search for I frames when they switch channels concurrently. They can be obtained by duplicating multiple copies of the multicast data cached by the first buffer unit. Several frame capacities can be selected to ensure that there are always valid pointers available. The number of storage units depends on the system requirements. If it is set too many, it will cause waste of system resources, and if it is set too few, it will not be able to meet the switching needs of users.

As shown in Figure 6, it is a schematic flow chart corresponding to the scheme of Embodiment 2. It can be seen from the figure that it mainly includes:

Step 601, the multicast replication point statically joins the multicast group, and the first cache unit caches the multicast data;

The multicast replication point located at the user’s access device joins the multicast group statically. The DSLAM receives the multicast data sent by the head-end service system through the bearer network, and the first buffer unit immediately caches the multicast data sent by the network. If the user requests the multicast data, it will be sent to the user, otherwise, the cached multicast data will be discarded immediately.

Step 602, the second cache unit caches the multicast data cached by the first cache unit, and determines the I frame information;

The second cache unit caches the multicast data cached by the first cache unit again, and simultaneously determines the I frame information,

This step can be carried out before the buffered multicast data enters each storage unit of the second buffer unit, and I frame information detection is performed at the entrance of the storage unit, so that when the multicast data buffered by the storage unit is sent to the terminal, it can be quickly Look for I-frames.

In this step, the I-frame address pointer is also stored according to the detection result.

603. The user initiates a channel switching request;

When the user wants to switch channels, he sends a channel switching request to the DSLAM, activates the corresponding process, and the DSLAM stops the multicast data being played to the user, and controls the process of leaving and rejoining the multicast group.

604. The user sends a new channel request;

The user sends a new channel request to the DSLAM, requesting to receive multicast data sent in the new channel.

605. The second cache unit sends the cached multicast data to the STB according to the I frame information;

When the user applies for new channel content and the DSLAM starts to send the new channel data stream to the terminal, according to the multicast data cached by each storage unit in the second cache unit, the cached group can be read directly from the cached I frame pointer. Broadcast data and transmit it to the terminal, because it is transmitted from the I frame, when the data is transmitted to the STB, the STB already knows the I frame information, so it saves the time for the terminal to search for the I frame, thereby speeding up the channel switching.

In this step, if there are multiple STBs sending channel requests in parallel, storage unit 1 in the second cache unit—storage unit n can be processed in parallel, and the cached data are sent to the corresponding STBs respectively, and each storage unit can correspond to One user, which saves channel switching time.

606. Schedule the data stream to the decoding buffer, and the decoding buffer outputs the video code stream.

The terminal decodes and outputs the data stream of the new channel whose first frame is an I frame according to the received first frame through a small buffer, and then outputs it to the TV for program playing.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and equivalent technologies thereof, the present invention also intends to include these modifications and variations.

Claims (12)

1, a kind of digital medium channel switching method is characterized in that,

At the user side access device multicast replication point is set;

The user side access device carries out caching process to multicast packet after receiving the multicast packet of head end operation system transmission through bearer network;

When channel switches, the multicast packet of described caching process is sent to terminal.

2, the method for claim 1 is characterized in that, multicast packet is carried out the caching process step, also comprises determining important frame information step.

3, the method for claim 1 is characterized in that, described multicast packet is carried out the caching process step, further comprises:

At the user side access device multicast packet is carried out first caching process, remove network jitter;

At the user side access device multicast packet is carried out second caching process, obtain data cached important frame information.

4, method as claimed in claim 3 is characterized in that, described second caching process, a plurality of second caching process multicast packets of parallel generation.

5, as claim 2,3 or 4 described methods, it is characterized in that, also comprise the step of the initial detection of important frame.

As claim 2,3 or 4 described methods, it is characterized in that 6, described important frame information is important frame address pointer.

As claim 1,2,3 or 4 described methods, it is characterized in that 7, described multicast replication point takes static adding mode to add multicast group.

8, a kind of digital medium channel switched system, comprise the head end of service source operation system is provided that for terminal provides the user side access device of access, described user side access device is provided with the multicast replication point, it is characterized in that described user side access device also is provided with:

Buffer unit carries out caching process to multicast packet;

When terminal sent channel switch request, described user side access device sent to terminal with the multicast packet of described buffer unit storage.

9, system as claimed in claim 8 is characterized in that, described user side access device also comprises:

Important frame information determining unit is used for determining the multicast packet important frame information of buffer unit buffer memory.

10, system as claimed in claim 8 is characterized in that,

Described buffer unit further comprises:

First buffer unit is used for multicast packet is carried out first caching process, to remove network jitter;

Second buffer unit is used for the multicast packet of the first buffer unit caching process is carried out second caching process, and obtains data cached important frame information.

11, system as claimed in claim 10 is characterized in that, described second buffer unit comprises a plurality of memory cell, is used for parallel a plurality of caching process multicast packets that produce.

As claim 10 or 11 described systems, it is characterized in that 12, described second buffer unit also is provided with:

The initial detecting unit of important frame is used to detect the multicast packet important frame of the second buffer unit caching process.

Images