CN101459852B - Method and device for predicting video service occurrence delay - Google Patents

Abstract

The present invention discloses a method and device for predicting the time delay of video services. The method includes: predicting the sequence number of the video frame being played by the receiving end; acquiring the sequence number of the video frame waiting to be sent by the sending end; if the receiving end The sequence number of the video frame being played by the end is greater than or equal to the sequence number of the video frame waiting to be sent by the sending end, and it is judged that the video frame waiting to be sent by the sending end will be delayed. The device includes: a first prediction unit, used to predict the sequence number of the video frame being played by the receiving end; a first obtaining unit, used to obtain the sequence number of the video frame waiting to be sent by the sending end; a first delay judging unit, If the sequence number of the video frame predicted by the first prediction unit is greater than or equal to the sequence number of the video frame obtained by the first obtaining unit, the video frame obtained by the first obtaining unit will be delayed. The method and device can predict whether the video frames waiting to be sent will be delayed at the network side.

Description

Method and device for predicting video service delay

technical field

The invention relates to the communication field, in particular to a method and device for predicting time delay of video services.

Background technique

The video service is to download data from the server to the mobile terminal for playback in real time. Its biggest feature is that the user does not have to wait for the entire file to be downloaded, but can use it only after a few seconds or tens of seconds of start-up delay.

The transmission protocols used by video services include RTP (Real-time Transport Protocol, Real-time Transport Protocol), RTCP (RTP Control Protocol, RTP Control Protocol) and RTSP (Real-Time Streaming Protocol, real-time streaming media protocol). RTSP is architecturally on top of RTP and RTCP. It uses TCP or RTP for data transfer.

In the prior art, the sending end divides a video frame into several RTP packets, and then sends these RTP packets to the wireless network with SDU (Service Data Unit, Service Data Unit) as the carrier for transmission. As shown in FIG. 1 , it is a schematic diagram of subpacketizing and bearing video frames. The receiving end receives the SDU to obtain the RTP packet, then combines the RTP packets belonging to the same video frame into a video frame, and finally plays the combined video frames frame by frame in sequence.

Frequency hopping and mosaic phenomenon often occur during video playback, mainly caused by packet loss, including delay and packet loss. Delay loss means that the terminal plays video frames in order, and each frame has a fixed playback time. If the video data arrives at the terminal later than its playback time, it is called delay. The data is useless and will be discarded by the terminal. , that is, delay packet loss occurs.

In the prior art, it is not possible to predict whether the video frames waiting to be sent will be delayed on the network side.

Contents of the invention

The embodiment of the present invention discloses a method and device capable of predicting whether a video service is delayed on the network side, so as to improve the efficiency of video service transmission. The scheme is as follows:

The embodiment of the present invention discloses a method for predicting video service delay, including:

Predict the sequence number of the video frame being played by the receiving end;

Obtain the sequence number of the video frame that the sender is waiting to send;

If the sequence number of the video frame being played by the receiving end is greater than or equal to the sequence number of the video frame waiting to be sent by the sending end, it is determined that the video frame waiting to be sent by the sending end will be delayed.

The embodiment of the present invention also discloses a method for predicting the time delay of video services, including:

Predict the sequence number of the SDU of the video frame being played by the receiving end;

Obtain the sequence number of the SDU that the sender is waiting to send;

If the sequence number of the SDU of the video frame being played by the receiving end is greater than or equal to the sequence number of the SDU waiting to be sent by the sending end, it is determined that the SDU waiting to be sent by the sending end will be delayed.

The embodiment of the present invention also discloses a device for predicting the time delay of video services, including:

The first prediction unit is used to predict the sequence number of the video frame being played by the receiving end;

The first obtaining unit is used to obtain the sequence number of the video frame waiting to be sent by the sending end;

The first delay judging unit is configured to judge the video acquired by the first acquiring unit if the serial number of the video frame predicted by the first predicting unit is greater than or equal to the serial number of the video frame acquired by the first acquiring unit Frames are delayed.

The embodiment of the present invention also discloses a device for predicting the time delay of video services, including:

The second prediction unit is used to predict the sequence number of the SDU of the video frame being played by the receiving end;

The second obtaining unit is used to obtain the sequence number of the SDU waiting to be sent by the sending end;

The third delay judging unit is configured to judge that the SDU acquired by the second acquiring unit will occur if the sequence number of the SDU predicted by the second predicting unit is greater than or equal to the sequence number of the SDU acquired by the second acquiring unit delay.

The method and device provided by the embodiments of the present invention can predict whether a video frame will be delayed on the network side, so that the network side can take corresponding measures to improve the efficiency of network transmission.

Description of drawings

Fig. 1 is the corresponding relation of video frame, RTP and SDU in the prior art;

FIG. 2 is a flowchart of a method for predicting video service delay provided by Embodiment 1 of the present invention;

FIG. 3 is a flowchart of a method for predicting video service delay provided by Embodiment 2 of the present invention;

FIG. 4 is a structural diagram of a device for predicting video service delay provided by Embodiment 3 of the present invention;

FIG. 5 is a structural diagram of an apparatus for predicting video service delay provided by Embodiment 4 of the present invention;

Fig. 6 is a data structure diagram of the first recording table and the second recording table used in the embodiment of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, several embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

Embodiment one

Referring to FIG. 2 , it is a schematic diagram of a method for predicting video service delay provided by Embodiment 1 of the present invention, including the following steps:

Step 201, predict the sequence number of the video frame being played by the receiving end.

That is to estimate the serial number of the video frame being played at the receiving end at the current moment. Generally, the timestamp of the video frame is used as the serial number of the video frame.

Predict the sequence number of the video frame being played by the receiving end. The specific implementation method, for example, obtain the playback duration and playback speed of the receiving end, and then estimate the sequence number of the video frame being played by the receiving end at the current moment.

Or, obtain the network feedback information RTCP packet fed back by the receiving end, the RTCP packet contains one or more fields carrying the sequence number of the SDU to be decoded by the receiving end, and use the RTCP packet to estimate the sequence of the video frame being played by the receiving end Number.

When the application at the receiving end starts an RTP session, two ports are used: one for RTP to transmit business data, and one for RTCP for transmission control. During video service transmission, RTCP will periodically report network feedback information to the sender, so that the sender can take corresponding measures, such as transmission reliability control, flow control, or congestion control.

The specific implementation method, for example, when the network side of the sending end receives the network feedback information RTCP packet from the receiving end, it parses out the sequence number of the next RTP packet to be decoded from the network feedback information (that is, the sequence number of the SDU ), through this serial number, the serial number of the video frame corresponding to the SDU packet can be obtained, and the video frame is the next video frame to be played at the receiving end.

The sequence number of the next video frame to be played can be obtained as follows. For example, a first record table (see FIG. 6 ) is designed to record the correspondence between the sequence number of the RTP packet (ie, the sequence number of the SDU) and the sequence number of the video frame to which the SDU belongs. After obtaining the sequence number of the next SDU to be decoded, the sequence number of the video frame to which the SDU belongs can be retrieved from the first record table, and the sequence number of the video frame is the sequence number of the next video frame to be played.

Then use the sequence number of the next video frame to be played to estimate the sequence number of the currently playing video frame at the receiving end.

In general, the sequence numbers of adjacent video frames differ by one timestamp. For example, if the frame rate is 10 frames/s, it means that there are 10 frames per unit time. If the timestamp of the frame starts from 0, then it can be calculated that the timestamp corresponding to the video frame is 0, 0.1, 0.2... ...0.9 1.0. If the serial number of the video frame is 11.0, the serial number of the previous adjacent video frame is 10.9, and the difference between the serial numbers is 1 timestamp, which is 0.1. The timestamp is generally equal to the inverse of the frame rate, but there may be some deviation.

Therefore, the estimated value can be set as a time stamp, and the estimated value can be subtracted from the sequence number of the next video frame to be played to obtain the video frame being played by the receiving end. If the network delay is severe, the receiving end may skip the next video frame to be played. In this case, the estimated value can be set to zero or a negative number. The absolute value of the estimated value is equal to the integer multiple of the timestamp, and then used The estimated value is subtracted from the sequence number of the next video frame to be played to obtain the sequence number of the video frame being played at the receiving end.

The estimated value can also be set as a coefficient related to the network delay condition. The worse the network delay, the smaller the estimated value should be. Then subtract the estimated value from the next video frame to be played at the receiving end to get the video being played at the receiving end The sequence number of the frame.

The network side can also use other methods to predict the video frame currently played by the receiving end. For example, according to the size of the sequence number of the video frame, all the sequence numbers of the video frames that have been successfully sent by the network side are recorded in the record table. When the next After the sequence number of the video frame to be played, the video frame closest to the sequence number of the next video frame to be played is retrieved from the recording table.

It should be noted that the video frame being played by the receiving end mentioned here is only a prediction, and does not represent the video frame actually played by the receiving end, nor is it limited to the video frame actually played by the receiving end. It is absolutely possible to use other video frames as the currently playing video frame. For example, the video frame corresponding to the next RTP to be decoded can be directly used as the currently played video frame, or the serial number of the video frame can be directly subtracted by a certain value as the video frame being played by the receiving end.

Step 202, acquire the sequence number of the video frame waiting to be sent by the sending end.

Video frames are transmitted in the form of SDUs. As the sending end, the network side also needs to obtain SDUs from other network elements, and there may be multiple SDUs waiting to be sent at the sending end. The sending end may select an SDU from the SDU queue waiting to be sent, and obtain the video frame to which the SDU belongs, and the video frame is the video frame waiting to be sent by the sending end.

The specific implementation method is, for example, designing a second recording table (see FIG. 6 ), which only records the SDUs currently to be transmitted, and the corresponding relationship between the SDUs and their associated video frames. Select the SDU with the smallest sequence number from the second record table, and obtain the sequence number of the video frame to which the SDU belongs from the second record table. After the SDU is successfully sent, it is deleted from the second record table.

Further, step 202 may only acquire the sequence number of the video frame corresponding to the SDU that needs to be retransmitted. Errors may occur during the transmission of the SDU, which needs to be retransmitted. If the retransmission exceeds a certain number of times, it is prone to delay. This function can be realized by using a second record table, for example, only SDUs that need to be retransmitted are recorded in the second record table, so as to further improve efficiency.

The execution order of steps 201 and 202 can be interchanged.

Step 203, judging whether the video frame waiting to be sent by the sending end will be delayed.

If the sequence number of the video frame being played by the receiving end is greater than or equal to the sequence number of the video frame that the sending end is waiting to send, that is, the sequence number of the video frame predicted in step 201 is greater than or equal to the sequence number of the video frame obtained in step 202, the judgment step 202 The acquired video frames will be delayed.

The inventors found that, usually, as the receiving end of the user, the video frame with a small serial number will be played first in time, and the video frame with a large serial number will be played later, and the serial number itself implies the information of the playing time.

Therefore, if the sequence number of the video frame being played by the receiving end is greater than or equal to the sequence number of the video frame waiting to be sent by the sending end, the playback time of the video frame waiting to be sent by the sending end has passed, even if the video frame waiting to be sent arrives At the receiving end, the result is a delay; conversely, if the video frame waiting to be sent by the sending end does not experience delay, the video frame with a sequence number larger than the video frame waiting to be sent by the sending end may not be delayed.

Therefore, Embodiment 1 may further include the step of determining in step 203 that the video frame waiting to be sent by the sending end will be delayed, and obtaining the sequence numbers of other video frames waiting to be sent by the sending end, if the video frame waiting to be sent by the sending end There will be no time delay in the video frame, and no time delay will occur in judging the video frames whose sequence number is larger than the video frame waiting to be sent by the sending end among the other video frames.

Its implementation, for example, can use the second recording table in step 202 to sort the second recording table according to the sequence number of the video frame from small to large, if the video frame waiting to be sent by the sending end will not be delayed, then In the second recording table, the video frames below the video frames waiting to be sent by the sending end will not be delayed, so that the efficiency can be greatly improved.

In the prior art, it is not possible to predict whether the unsent video frames will be delayed on the network side. This embodiment provides a prediction method, which can facilitate the network side to predict whether the video frames waiting to be sent will be delayed, so as to take corresponding measures. measures to improve network efficiency.

Embodiment two

As shown in Figure 3, it is a schematic diagram of a method for predicting video service delay provided by Embodiment 2 of the present invention, including the following steps:

Step 301, predict the sequence number of the SDU of the video frame being played by the receiving end.

That is, the sequence number of the SDU of the video frame being played at the receiving end is estimated at the current moment. Its specific implementation method, one is to estimate the serial number of the video frame being played by the receiving end at the current moment through the playback duration and playback speed of the receiving end, and then obtain the serial number of an SDU of the video frame.

Another method is to calculate the sequence number of the SDU of the video frame being played by the receiving end through the network feedback information. For example, a field may be added to the network feedback information to carry the sequence number of the next SDU to be decoded at the receiving end. With this serial number, you can get the serial number of the video frame corresponding to the SDU, use the serial number of the video frame to guess the serial number of the video frame being played by the receiving end, and then select an SDU of the video frame being played serial number. Specifically, the sequence number of the SDU of the video frame being played by the receiving end can be predicted through the following steps.

The first step is to obtain current network feedback information from the receiving end.

That is to obtain the current network feedback information RTCP packet from the receiving end, and one or more fields in the RTCP packet carry the sequence number of the SDU to be decoded by the receiving end.

The second step is to analyze the network feedback information.

That is, from the network feedback information RTCP packet, parse the sequence number of the next SDU to be decoded at the receiving end.

The third step is to obtain the sequence number of the next video frame to be decoded at the receiving end.

For example, a first recording table may be designed to record the correspondence between the sequence number of the RTP packet (that is, the sequence number of the SDU) and the sequence number of the video frame to which the SDU belongs. After obtaining the sequence number of the next SDU to be decoded, the sequence number of the video frame to which the SDU belongs can be retrieved from the first record table, and the sequence number of the video frame is the sequence number of the next video frame to be played.

The fourth step is to estimate the sequence number of the video frame currently played by the receiving end.

That is, the sequence number of the video frame currently being played at the receiving end can be estimated by subtracting an estimated value from the sequence number of the next video frame to be played obtained in the third step. The estimated value can be set as a timestamp; if the network delay is relatively severe, the receiving end may skip the next video frame to be played. In this case, the estimated value can be set to zero or a negative number, and the absolute value of the estimated value The value is equal to an integer multiple of the timestamp.

The fifth step is to obtain an SDU belonging to the currently played video frame.

One video frame corresponds to multiple SDUs, and one SDU can be selected from the SDUs belonging to the video frame being played by the receiving end, for example, the serial number of the SDU with the largest serial number in the video frame being played is selected as the video frame being played The serial number of the SDU.

Step 302, obtaining the sequence number of the SDU waiting to be sent by the sending end;

There may be multiple SDUs waiting to be sent on the network side, and the sender can select an SDU from the queue of SDUs waiting to be sent.

For example, a second record table is used to record the sequence numbers of all SDUs waiting to be transmitted by the sending end, and the SDU with the smallest sequence number is selected as the SDU waiting to be sent by the sending end. After the SDU is sent successfully, it is about to be deleted from the second record table.

Further, step 302 may only acquire the sequence numbers of the SDUs that need to be retransmitted, because the retransmitted SDUs are more likely to be delayed. This function can be implemented by using the second record table, for example, only the sequence numbers of the SDUs that need to be retransmitted can be recorded in the second record table, thereby improving efficiency.

The execution order of steps 301 and 302 can be interchanged.

Step 303, judging whether the SDU waiting to be sent by the sending end will be delayed.

If the sequence number of the SDU of the video frame being played by the receiving end is greater than or equal to the sequence number of the SDU waiting to be sent by the transmitting end, that is, the sequence number of the SDU predicted in step 301 is greater than or equal to the sequence number of the SDU obtained in step 302, It is judged that the SDU waiting to be sent by the sending end will be delayed.

The inventors found that, in general, the serial numbers of SDUs are sorted from small to large. If the video frame corresponding to the SDU with a large serial number has been played, it means that the video frame corresponding to the SDU with a small serial number is The playback moment has passed, that is, there will be a delay if the SDU reaches the receiving end.

Further, the present embodiment may also include the following steps. After judging in step 303 that the SDU acquired in step 302 will be delayed, the sequence numbers of other SDUs waiting to be sent by the sending end are obtained, and the sequence numbers in the other SDUs are judged to be higher than those in step 302. The acquired SDU with a large SDU will not be delayed.

In the prior art, it is impossible to predict whether the SDU of the unsent video frame will be delayed on the network side. This embodiment provides a method to predict whether the SDU of the video frame waiting to be sent by the sending end will be delayed, so that It is convenient for the network side to make processing in advance.

Embodiment three

As shown in FIG. 4, it is a schematic diagram of a device for predicting video service delay provided by Embodiment 3 of the present invention, including:

The first prediction unit 401 is configured to predict the sequence number of the video frame being played by the receiving end.

The receiving end will feed back the network information RTCP packet. In order to obtain the serial number of the video frame being played by the receiving end, it is necessary to carry the serial number of the next SDU to be decoded by the receiving end in the RTCP packet. Through this serial number, the receiving end can be calculated. The sequence number of the video frame currently being played by the client.

For a specific implementation, for example, the first prediction unit 401 includes a feedback information acquisition unit and a calculation unit.

The feedback information acquisition unit is configured to acquire network feedback information currently from the receiving end, and the network feedback information includes one or more fields, which are used to carry the sequence number of the SDU to be decoded by the receiving end.

The calculation unit is used to analyze the serial number of the next RTP packet (ie, the serial number of the SDU) that needs to be decoded from the network feedback information after the feedback information acquisition unit receives the network feedback information RTCP from the receiving end, through With this serial number, you can get the serial number of the video frame to which the SDU packet belongs. This video frame is the next video frame to be played at the receiving end, and then use the serial number of the next video frame to be played to estimate the receiving end. The sequence number of the currently playing video frame. Specifically, it can be estimated through the following subunits:

The first recording table unit is used to record the correspondence between the serial number of the RTP packet (that is, the serial number of the SDU) and the serial number of the video frame to which the SDU belongs.

The parsing subunit is configured to parse the network feedback information obtained by the feedback information obtaining unit to obtain the sequence number of the next SDU to be decoded at the receiving end.

After obtaining the sequence number of the next SDU to be decoded, the estimation subunit can retrieve from the first record table unit the video frame to which the SDU parsed by the analysis subunit belongs, and the sequence number of this video frame is the next The sequence number of the video frame being played. With the sequence number of the next video frame to be played, the estimation subunit subtracts an estimated value from the sequence number of the video frame to estimate the sequence number of the video frame currently being played at the receiving end. As for the estimated value, you can set the estimated value as a timestamp; if the network delay is severe, the receiving end may skip the next video frame to be played. In this case, you can set the estimated value to zero or a negative number. The absolute value of the estimate is equal to one or more timestamps.

The currently played video frame is only used as a criterion for predicting the possibility of delay in the video frame to be sent. Various determination methods are possible, and are not limited to actual played video frames.

The first obtaining unit 402 is configured to obtain the sequence number of the video frame waiting to be sent by the sending end.

Video frames are transmitted in the form of SDUs. As the network side of the sending end, there may be multiple SDUs waiting to be sent. The sending end may select an SDU from the SDU queue waiting to be sent, and obtain the video frame to which the SDU belongs, and the video frame is the video frame waiting to be sent by the sending end.

For example, the first acquisition unit 402 may include a second recording table unit, which is used to record all SDUs that need to be sent to the receiver at the sending end, and the correspondence between the SDU and the associated video frame; the first acquisition unit 402 also includes a The retrieval subunit is used to select the SDU with the smallest sequence number from the second record table unit, and obtain the sequence number of the video frame to which the SDU belongs from the second record table, and after the SDU is successfully sent, send it from the second Deleted from the record table.

The delay judging unit 403 is configured to judge whether the video frame waiting to be sent by the sending end is delayed.

If the sequence number of the video frame being played by the receiving end predicted by the first prediction unit 401 is greater than or equal to the sequence number of the video frame that the sending end is waiting to send acquired by the first acquiring unit 402, then it is judged that the sending end acquired by the first acquiring unit 402 is waiting The video frames sent will be delayed.

The inventors found that, usually, as the receiving end of the user, the video frame with a small serial number will be played first in time, and the video frame with a large serial number will be played later, and the serial number itself implies the information of the playing time. Therefore, if the sequence number of the video frame being played by the receiving end is greater than the sequence number of the video frame waiting to be sent by the sending end, the playback time of the video frame waiting to be sent by the sending end has passed, even if the video frame waiting to be sent arrives at the receiving end end, the result is delay.

This embodiment may further include a second delay judging unit for acquiring other video frames (not including the video frames acquired by the first acquiring unit 402) waiting to be sent by the sending end, if the first delay judging unit 403 judges the first The video frames acquired by the acquiring unit 402 will not be delayed, and it is judged that the video frames with sequence numbers larger than those acquired by the first acquiring unit 402 among the other video frames will not be delayed.

In the prior art, the network side cannot predict whether there will be a delay in the video frames that have not yet been sent. This embodiment provides a device that can predict whether the video frames waiting to be sent by the sending end will be delayed, so that it is convenient for the network side to make advance decisions. deal with.

Embodiment four

As shown in FIG. 5, it is a schematic diagram of a device for predicting video service delay provided by Embodiment 4 of the present invention, including:

The second prediction unit 501 is configured to predict the sequence number of the SDU of the video frame being played by the receiving end.

Predict the video frame being played at the receiving end, and obtain the sequence number of an SDU of the video frame being played.

The next SDU sequence number to be decoded at the receiving end can be carried through the network feedback information. With this serial number, the serial number of the video frame corresponding to the SDU can be obtained. The serial number of the video frame can be used to guess the serial number of the video frame being played by the receiving end, and then select an SDU of a video frame being played. serial number.

For a specific implementation, for example, the second prediction unit may include the following subunits: a first record table unit, a network feedback information acquisition unit, an analysis subunit, an estimation subunit, and a query word unit.

The first recording table unit is used to record the correspondence between the serial number of the RTP packet (that is, the serial number of the SDU) and the serial number of the video frame to which the SDU belongs.

A network feedback information obtaining unit, configured to obtain current network feedback information from the receiving end, where the network feedback information includes the sequence number of the next SDU to be decoded by the receiving end.

The parsing subunit is configured to parse the network feedback information obtained by the network feedback information obtaining unit to obtain the sequence number of the next SDU to be decoded at the receiving end.

The estimation subunit is used to obtain the sequence number of the next SDU to be decoded from the parsing subunit, and then retrieve the sequence number of the video frame to which the SDU belongs from the first recording table unit. This video frame sequence number is the sequence number of the next SDU to be decoded. The sequence number of the video frame being played. Subtract an estimated value from the sequence number of the video frame to estimate the sequence number of the video frame being played at the receiving end. As for the size of the estimated value, it can be set as a timestamp. If the network delay is severe, the receiving end may skip the next video frame to be played. In this case, the estimated value can be set to zero or a negative number. The absolute value of the value is equal to one or more timestamps.

The query subunit is used to retrieve the SDU belonging to the video frame being played from the first record table unit. Since one video frame corresponds to multiple SDUs, the sequence number of the SDU with the largest sequence number in the video frame being played can be selected .

The second obtaining unit 502 is configured to obtain the sequence number of the SDU waiting to be sent by the sending end.

There may be multiple SDUs waiting to be sent on the network side, and the second obtaining unit 502 may select an SDU from the queue of SDUs waiting to be sent.

For example, a second record table is used to record the sequence numbers of all SDUs waiting to be transmitted by the sending end, and the SDU with the smallest sequence number is selected as the SDU waiting to be sent by the sending end. After the SDU is sent successfully, it is about to be deleted from the second record table.

The third delay judging unit 503 is configured to judge whether the SDU waiting to be sent by the sending end will be delayed.

If the sequence number predicted by the second prediction unit 501 is greater than or equal to the sequence number obtained by the second obtaining unit 502, the SDU obtained by the second obtaining unit 502 will be delayed.

The inventors found that, in general, the serial numbers of SDUs are sorted from small to large. If the video frame corresponding to the SDU with a large serial number has been played, it means that the video frame corresponding to the SDU with a small serial number is The playback moment has passed, that is, there will be a delay if the SDU reaches the receiving end.

Further, this embodiment may also include a fourth time delay judging unit, which is used to obtain other SDUs waiting to be sent by the sending end. If the SDU obtained by the second obtaining unit 502 does not have a time delay, it is judged that the sequence numbers in other SDUs are higher than the first SDU. The SDU with a larger SDU acquired by the second acquisition unit 502 will not be delayed.

In the prior art, the network side cannot predict whether the SDUs that have not been sent will be delayed. This embodiment provides a device that can predict whether the SDUs waiting to be sent by the sending end will be delayed, so that the network side can make processing in advance.

The above-mentioned embodiments of the invention can be realized by software, and the corresponding software can be stored in a readable storage medium, such as a hard disk, an optical disk or a floppy disk of a computer.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

Claims (12)

1. A method for predicting video service time delay, characterized in that the method comprises: Predict the sequence number of the video frame being played by the receiving end; Obtain the sequence number of the video frame that the sender is waiting to send; If the sequence number of the video frame being played by the receiving end is greater than or equal to the sequence number of the video frame waiting to be sent by the sending end, it is determined that the video frame waiting to be sent by the sending end will be delayed; Wherein, said predicting the sequence number of the video frame being played by the receiving end includes obtaining current network feedback information from the receiving end, and the network feedback information includes the sequence number of the next SDU to be decoded by the receiving end, using the Estimate the sequence number of the video frame being played by the receiving end based on the network feedback information. 2. The method according to claim 1, wherein the step of using the network feedback information to predict the sequence number of the video frame being played by the receiving end includes: Analyzing the sequence number of the next SDU to be decoded at the receiving end from the network feedback information; Obtain the sequence number of the video frame to which the next SDU to be decoded at the receiving end belongs from the first recording table, and the first recording table records the correspondence between the SDU and the video frame to which the SDU belongs; The result obtained by subtracting an estimated value from the sequence number of the video frame to which the next SDU to be decoded at the receiving end belongs is used as the sequence number of the video frame being played by the receiving end. 3. The method according to claim 1, wherein the step of obtaining the sequence number of the video frame that the sending end waits to send comprises: Obtain the SDU with the smallest sequence number from the second recording table, the second recording table records the sequence number of the SDU queue that the sending end is currently waiting to send and the video frame to which the SDU waiting to send belongs; Acquire the sequence number of the video frame to which the SDU with the smallest sequence number belongs from the second record table, and use it as the sequence number of the video frame waiting to be sent by the sending end. 4. The method of claim 1, further comprising: Acquiring the sequence numbers of other video frames waiting to be sent by the sending end, if the video frames waiting to be sent by the sending end will not be delayed, judging that the sequence numbers of the other video frames are higher than those of the video frames waiting to be sent by the sending end Large video frames also do not suffer from lag. 5. A method for predicting video service time delay, characterized in that the method comprises: Predict the sequence number of the SDU of the video frame being played by the receiving end; Obtain the sequence number of the SDU waiting to be sent by the sender; If the sequence number of the SDU of the video frame being played by the receiving end is greater than or equal to the sequence number of the SDU waiting to be sent by the sending end, it is judged that the SDU waiting to be sent by the sending end will be delayed; Wherein, the sequence number of the SDU of the video frame being played by the prediction receiving end includes: Acquire current network feedback information from the receiving end, where the network feedback information includes the sequence number of the next SDU to be decoded at the receiving end; Analyzing the network feedback information to obtain the sequence number of the next SDU to be decoded at the receiving end; Obtain the sequence number of the video frame to which the next SDU to be decoded at the receiving end belongs from the first recording table, and the first recording table records the correspondence between the SDU and the video frame to which the SDU belongs; The result obtained by subtracting an estimated value from the sequence number of the video frame to which the next SDU to be decoded at the receiving end is used as the sequence number of the currently played video frame; Retrieving the SDUs belonging to the currently played video frame from the first record table, and taking the sequence number of the SDU with the largest sequence number. 6. The method of claim 5, further comprising: Obtain the sequence numbers of other SDUs that the sender is waiting to send. If the SDUs that the sender is waiting to send will not be delayed, it is judged that the SDUs in the other SDUs that have a sequence number larger than the SDU that the sender is waiting to send will not Latency occurs. 7. A device for predicting the time delay of video services, characterized in that the device comprises: The first prediction unit is used to predict the sequence number of the video frame being played by the receiving end; The first obtaining unit is used to obtain the sequence number of the video frame waiting to be sent by the sending end; The first delay judging unit is configured to judge the video acquired by the first acquiring unit if the serial number of the video frame predicted by the first predicting unit is greater than or equal to the serial number of the video frame acquired by the first acquiring unit Frames will be delayed; Wherein, the first prediction unit includes: a feedback information acquiring unit, configured to acquire network feedback information currently from the receiving end, the network feedback information including the sequence number of the next SDU to be decoded at the receiving end; A calculating unit, configured to estimate the sequence number of the video frame being played by the receiving end according to the network feedback information acquired by the feedback information acquiring unit. 8. The device according to claim 7, wherein the computing unit comprises: The first recording table unit is used to record the corresponding relationship between the SDU and the video frame; The parsing subunit is configured to parse the network feedback information obtained by the feedback information obtaining unit to obtain the sequence number of the next SDU to be decoded at the receiving end; The estimation subunit is used to obtain the sequence number of the video frame to which the SDU analyzed by the analysis subunit belongs from the first recording table unit, and subtract an estimated value as the video being played by the receiving end The sequence number of the frame. 9. The device according to claim 7, wherein the first obtaining unit comprises: The second recording table unit is used to record the SDU queue that the sending end is waiting to send and the sequence number of the video frame to which the SDU waiting to send belongs; The retrieval subunit is configured to obtain, from the second recording table unit, the sequence number of the video frame to which the SDU with the smallest sequence number belongs, as the sequence number of the video frame waiting to be sent by the sending end. 10. The device of claim 9, further comprising: The second delay judging unit is used to acquire other video frames waiting to be sent by the sending end, if the first delay judging unit judges that the video frames acquired by the first acquiring unit will not be delayed, judge the other video frames A video frame whose sequence number is larger than the sequence number obtained by the first obtaining unit will not be delayed. 11. A device for predicting video service delay, characterized in that the device comprises: The second prediction unit is used to predict the sequence number of the SDU of the video frame being played by the receiving end; The second obtaining unit is used to obtain the sequence number of the SDU waiting to be sent by the sending end; The third delay judging unit is configured to judge that the SDU predicted by the second acquiring unit will occur if the sequence number of the SDU predicted by the second predicting unit is greater than or equal to the sequence number of the SDU acquired by the second acquiring unit delay; Wherein, the second prediction unit includes: The first recording table unit is used to record the corresponding relationship between the SDU and the video frame; A network feedback information acquisition unit, configured to acquire current network feedback information from the receiving end, the network feedback information including the sequence number of the next SDU to be decoded at the receiving end; The parsing subunit is configured to parse the network feedback information obtained by the network feedback information obtaining unit to obtain the sequence number of the next SDU to be decoded at the receiving end; The estimation subunit is used to acquire the sequence number of the video frame to which the SDU analyzed by the parsing subunit belongs from the first recording table unit, and subtract an estimated value to obtain the sequence number of the video frame being played by the receiving end ; The query subunit is configured to retrieve the SDUs belonging to the video frame calculated by the estimation subunit from the first recording table unit, and obtain the sequence number of the SDU with the largest sequence number. 12. The device of claim 11, further comprising: The fourth delay judging unit is used to obtain other SDUs waiting to be sent by the sending end. If the third delay judging unit judges that the SDU obtained by the second obtaining unit will not be delayed, judge the sequence in the other SDUs The SDU with a number larger than the SDU obtained by the second obtaining unit will not be delayed.

Images