CN103747283B - Video fragment download method - Google Patents

Video fragment download method Download PDF

Info

Publication number
CN103747283B
CN103747283B CN201310723717.4A CN201310723717A CN103747283B CN 103747283 B CN103747283 B CN 103747283B CN 201310723717 A CN201310723717 A CN 201310723717A CN 103747283 B CN103747283 B CN 103747283B
Authority
CN
China
Prior art keywords
video
burst
download
threshold time
code check
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN201310723717.4A
Other languages
Chinese (zh)
Other versions
CN103747283A (en
Inventor
辛永辉
李杨
林涛
慈松
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Acoustics CAS
Original Assignee
Institute of Acoustics CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Institute of Acoustics CAS filed Critical Institute of Acoustics CAS
Priority to CN201310723717.4A priority Critical patent/CN103747283B/en
Publication of CN103747283A publication Critical patent/CN103747283A/en
Application granted granted Critical
Publication of CN103747283B publication Critical patent/CN103747283B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention relates to a video fragment download method which comprises the following steps: after a video server receives a video fragment download request sent by a client, a current video fragment is downloaded for the client; if the current video fragment is downloaded within a first threshold time, a second threshold time for downloading the next video fragment is calculated; if the current video fragment hasn't been downloaded when the first threshold time arrives, first code rate of the current video fragment is lowered to second code rate; if the current video fragment which has been adjusted to the second code rate is downloaded within the first threshold time, a third threshold time for downloading the next video fragment is calculated; and if the current video fragment which has been adjusted to the second code rate hasn't been downloaded within the first threshold time, the second code rate is adjusted as the lowest code rate. By a predict interrupt mechanism and a step-by-step lowering method, fluency of the video that the client watches is guaranteed. Meanwhile, network caching resources are fully utilized, and video quality is improved.

Description

The method for down loading of video slicing
Technical field
The present invention relates to video technique, particularly to a kind of method for down loading of video slicing.
Background technology
Current the Internet experiencings people and turns to the communication pattern centered on content towards the point to point link pattern of main frame Become, compared with the position of content, user more concerned be content itself.In order to adapt to the Internet this change it is proposed that with Centered on content, network (content centric network, ccn) is the next generation network of representative, content oriented main body, and Non-content position, the router in this network architecture all has caching function, to provide for the follow-up request to identical content Service, inherently changes the transmission mode of traditional ip network end-to-end.And media file accounts for absolutely in current internet traffic Major part, in order to preferably provide multimedia service, the dynamic self-adapting Streaming Media (dynamic based on http in a network Adaptive streaming over http, dash) there is deployment simply, the advantages of impact by fire wall.In dash Adaptive strategy has simple ratebased mode, also has the adaptive algorithm (has) of sublevel segmentation, they are to a certain degree On can improve the viewing experience of user, but be simply implanted in ccn network and some problems occurred.Ccn's is slow Depositing function can make user obtain content from intermediate node or server, simply easily caused to network speed according to speed of download Erroneous judgement, leads to client buffer area underflow even to be interrupted, needs to judge as early as possible and process in time;May be cached with ccn simultaneously Abundant video resource, needs to improve the video quality of user's viewing using effective method.
Content of the invention
The purpose of the present invention is based on http, for realize to the network bandwidth erroneous judgement after handle it in time it is ensured that That plays is smoothed out, and makes full use of the resource of cache node caching simultaneously, proposes the method that code check substep is lowered, lifts video Quality.
For achieving the above object, the invention provides a kind of method for down loading of video slicing.The method comprises the following steps:
Video server receives the download request of the video slicing that client sends;
Described video server to described client downloads current video burst, if under completing within the first threshold time Carry described current video burst, calculate the Second Threshold time of next video slicing download;
If reaching the first threshold time, not completing to download described current video burst, then described current video being divided It is adjusted to second code rate under first code check of piece;If complete to download to be adjusted to described in the second code check within the first threshold time worked as Front video slicing, calculates the 3rd threshold time that next video slicing described is downloaded;If not complete within the first threshold time Become to download the described current video burst being adjusted to the second code check, then the second code check is adjusted to lowest bit rate.
Further, described video server also includes to before described client downloads current video burst, according to upper The size of the actual download time of one video slicing and a described upper video slicing calculates the lower contained network of a described upper video slicing Speed, is calculated described first code check according to the download speed of a described upper video slicing using adaptive algorithm.
Further, methods described also includes, the download speed according to a described upper video slicing and described current video The size of burst calculates and estimates download time, and estimates download time and be calculated the first threshold time according to described.
Further, the download request of the video slicing that described video server reception client sends specifically includes: regards Frequency server receives the download request of the video slicing that client sends according to mpd file, and described mpd file includes http service The media slicing of device storage, uri, demultiplexing, decoded information.
Further, it is adjusted to second code rate under described the first code check by described current video burst to specifically include: by institute The first code check stating current video burst is lowered according to cache resources substep.
It is an advantage of the invention that carrying out timer interrupt in advance for the erroneous judgement of the network bandwidth, it is to avoid blindly download it is ensured that That plays is smoothed out;Make full use of the resource of cache node caching, lower the code check of video slicing, lifting user's viewing in advance The total quality of video.
Brief description
Fig. 1 is a kind of method for down loading schematic flow sheet of video slicing of the embodiment of the present invention;
Fig. 2 is a kind of method for down loading application process schematic diagram of video slicing of the embodiment of the present invention.
Specific embodiment
Below by drawings and Examples, technical scheme is described in further detail.
Fig. 1 is a kind of method for down loading schematic flow sheet of video slicing of the embodiment of the present invention, as shown in figure 1, the method Comprise the steps:
Step 101, video server receives the download request of the video slicing that client sends;
Step 102, video server to client downloads current video burst, if under completing within the first threshold time Carry current video burst, calculate the Second Threshold time of next video slicing download;
Step 103, if reaching the first threshold time, not completing to download current video burst, then dividing current video It is adjusted to second code rate under first code check of piece;If complete to download within the first threshold time be adjusted to the second code check work as forward sight Frequency division piece, calculates the 3rd threshold time of next video slicing download;If do not complete to download within the first threshold time adjusted Whole be the second code check current video burst, then the second code check is adjusted to lowest bit rate.
If the code check of a upper video slicing is r1, the actual download time is t1, a length of τ during burst, a length of 2 during acquiescence burst Second, then the download speed going up a video slicing is:
v 1 ( k , n ) = r 1 τ t 1 - - - ( 1 )
Wherein r1τ is the size of a upper video slicing, utilizes adaptive algorithm according to the download speed of a upper video slicing () obtains the first code check r of current video burst taking ratedbased as a example2.
t pk 2 = r 2 τ v 1 ( k , n ) - - - ( 2 )
v 2 ( k , n ) = r 2 τ t 2 - - - ( 3 )
Wherein r2τ is the size of current video burst, t2For the actual download time of current video burst, according to formula (2) Obtain the threshold time t of current video burstPk2,, and download speed v obtaining current video burst according to formula (3)2(k, N), obtain the code check r of next video slicing using adaptive algorithm according to formula (3)3.
t pk 3 = r 3 τ v 2 ( k , n ) - - - ( 4 )
r2It is the first code check for current video burst, and current video burst completed to download within the first threshold time When, draw the threshold time of next video slicing according to the download speed of current video burst, i.e. Second Threshold time tpk3.
The first code check r when current video burst2Under be adjusted to second code rate r21When, the download speed meeting of current video burst It is changed into:
v 21 ( k , n ) = r 21 τ t 21 - - - ( 5 )
Wherein t21It is adjusted to second code rate r under current video burst21When download time.
Draw the code check r of next video slicing according to formula (5) using adaptive algorithm31, thus showing that next video divides The threshold time of piece, i.e. the 3rd threshold time tpk31.
t pk 31 = r 31 τ v 21 ( k , n ) - - - ( 6 )
In order to make full use of channel width and ensure that user watches smoothness and the high-quality of video, the present invention follows following four Individual principle: (1) avoids playing interrupting as far as possible;(2) provide high-quality video segment in the case of bandwidth permission as far as possible;(3) Video switching times are few, because from the angle of user, the video ratio being always held at low bit- rate is always between high low bit- rate The viewing experience that the video being continually changing brings will be got well;(4) start-up study is little, selects to play certain video to this from user The time that first section of video commences play out is as far as possible short.
In this four principles, first principle should have highest priority, because causing worst Consumer's Experience It is stagnation or the interruption of video playback, also imply that the fluency in order to ensure client video quality, to a certain extent The quality of video can be sacrificed.Second and the 3rd principle need to be weighed, the best method making full use of bandwidth is each Select the video of the highest code check of current network support, but the switching times that this can make video in the network of change are significantly Increase.Simultaneously second and the 4th principle be also required to be compromised, because startup time delay will be reduced require client downloads the One video segment time is the shortest, so can only ask minimum code check for the first time, cause the wave of bandwidth to a certain extent Take, final judging process is above balance in four principles and selects.
The present invention follows above four principles and proposes specific embodiment, and Fig. 2 is that one kind of the embodiment of the present invention regards The method for down loading application process schematic diagram of frequency division piece, as shown in Fig. 2 this implementation process is:
Step 201, makes t=0, and k=1, t=0 are initial time, and k=1 is that to number be first point for the burst of request video Piece;
In order to ensure the minimum delay starts, by video the first burst s1Code check r1It is set as minimum code rate, i.e. r1=rmin
Wait code check is r1Video slicing s1Download, if code check is r1Video slicing s1After the completion of download, go to Step 202, otherwise goes to step 203;
Step 202, code check is r1Video slicing s1After the completion of download, burst k Jia 1 automatically, i.e. k=k+1, and calculates Video slicing s1Download speed v (k, n);
Video slicing s is drawn according to adaptive algorithm1Next burst skCode check rk,skEstimate download time tpk, this In k=2, go to step 204;
Step 203, video slicing is waited, and request is downloaded again;
Step 204, is r to code checkkVideo slicing skSend request, concurrently set count-down device tCk,Work as skTo have downloaded Cheng Hou, and count-down device tckMore than 0 and less than tpkWhen, go to step 206;
If described count-down device tckEqual to 0, go to step 205;
Step 205, in advance by video slicing skCode check rkLower, go to step 207;
Step 206, timer interrupt, carry out video slicing skNext fragment request;
Step 207, first by video slicing skCode check rkIt is transferred to rk', rk' it is less than code check rkMaximum, in the event of the Secondary interruption, goes to step 208;
Step 208, by video slicing skCode check rkIt is transferred to code check rkLowest bit rate rk=rkmin.
In Fig. 2, involved symbol description is as shown in table 1:
Table 1
When the speed of download of known front n fragment, we can be given by certain adaptive algorithm and will descend slide glass Section skCode check rk.This based on download mode estimate that the method for link bandwidth is not fee from the erroneous judgement to network speed, especially when When having cache node in network.Assume that the speed of download calculating is v (k, n), the burst code obtaining according to adaptive algorithm Rate is rkIt is known that burst duration τ, fragment skEstimate download time.After sending the request to this code check it is contemplated that network not If intellectual completes it is believed that can download in the time, illustrate that network link is close with expection, can continue further fragments are initiated Request;Otherwise it is meant that link circuit condition is than expected bad luck, or because nodes do not cache this high code check fragment, In order to prevent the wait of longer time futile effort, also avoid relief area that the interruption play occurs because of empty grade, we terminate in advance This is asked and starts to send request to this fragment low bit- rate.
For the phenomenon of waiting-timeout appearing above, we should ask the file of this fragment low bit- rate, but with regard to relatively low The number of times of the setting of code check and downward does not do specific restriction.Because it is contemplated that user watches the seriality of video, that is, user sees Fragment s of certain video completekIt is more likely that also having seen fragment sk+1, in such network, it is cached with fragment sk+1Probability very Greatly;It is also contemplated that the impact of network-caching, same video, the average quality of the client's viewing asked afterwards ratio is above Good, also disclose fragment s for a video simultaneouslyk, the people seeing how, also can cache various code rate in network.In conjunction with , lowering fragment s at upper 2 pointskCode check rkWhen, the various code rate resource substep that may cache in network can be made full use of and lower Code check, is adjusted to r under first timek=rk', request is downloaded.If there is no rk, but be likely to have and compare rkOne or two grades low Code check, if this fragment skInterrupt for second, stagnate for avoiding playing, r should be adjusted to down immediatelyk=rkmin.
Above-described specific embodiment, has been carried out to the purpose of the present invention, technical scheme and beneficial effect further Describe in detail, be should be understood that the specific embodiment that the foregoing is only the present invention, be not intended to limit the present invention Protection domain, all any modification, equivalent substitution and improvement within the spirit and principles in the present invention, done etc., all should comprise Within protection scope of the present invention.

Claims (5)

1. a kind of method for down loading of video slicing is it is characterised in that comprise the following steps:
Video server receives the download request of the video slicing that client sends;
Described video server is to described client downloads current video burst, if completing to download institute within the first threshold time State current video burst, calculate the Second Threshold time of next video slicing download;
The computing formula of described first threshold time is:
v 1 ( k , n ) = r 1 τ t 1
t pk 2 = r 2 τ v 1 ( k , n )
Wherein v1(k, n) is the speed of download of a upper burst, r1For the code check of a upper video slicing, τ is burst duration, t1For reality Border download time, r2For the first code check of current video burst, tpk2For the described first threshold time, k is the volume of video slicing Number, n is front n (n < k) individual burst;
The computing formula of described Second Threshold time is:
v 2 ( k , n ) = r 2 &tau; t 2
t pk 3 = r 3 &tau; v 2 ( k , n )
Wherein tpk3For described Second Threshold time, r3For the resolution code check of next video, v2(k, n) is the upper of current video burst Pass network speed;
If reaching the first threshold time, do not complete to download described current video burst, then by described current video burst It is adjusted to second code rate under first code check;If completing to download to be adjusted to described in the second code check within the first threshold time and working as forward sight Frequency division piece, calculates the 3rd threshold time that next video slicing described is downloaded;If under not completing within the first threshold time Carry the described current video burst being adjusted to the second code check, then the second code check is adjusted to lowest bit rate;
The computing formula of described 3rd threshold time is:
v 21 ( k , n ) = r 21 &tau; t 21
t pk 31 = r 31 &tau; v 21 ( k , 1 n )
Wherein v21Download speed for current video burst;r21For the second code check;r31For the code check of next video slicing, tpk31For Described 3rd threshold time.
2. the method for down loading of video slicing according to claim 1 is it is characterised in that described video server is to described visitor Family end also includes before downloading current video burst, and the actual download time according to a upper video slicing and a described upper video divide The size of piece calculates the download speed of a described upper video slicing, and the network speed according to a described upper video slicing utilizes self adaptation to calculate Method is calculated described first code check.
3. the method for down loading of video slicing according to claim 2 is it is characterised in that methods described also includes, according to institute The size of the download speed and described current video burst of stating a video slicing calculates estimates download time, and according to described pre- Meter sends Time Calculation and obtains the first threshold time.
4. the method for down loading of video slicing according to claim 1 is it is characterised in that described video server receives client The download request of the video slicing that end sends specifically includes: video server receives the video that client sends according to mpd file The download request of burst, described mpd file include http-server storage media slicing, uri, demultiplexing, decoded information.
5. video slicing according to claim 1 method for down loading it is characterised in that described by described current video burst The first code check under be adjusted to second code rate and specifically include: by the first code check of described current video burst according to cache resources substep Lower.
CN201310723717.4A 2013-12-24 2013-12-24 Video fragment download method Expired - Fee Related CN103747283B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310723717.4A CN103747283B (en) 2013-12-24 2013-12-24 Video fragment download method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310723717.4A CN103747283B (en) 2013-12-24 2013-12-24 Video fragment download method

Publications (2)

Publication Number Publication Date
CN103747283A CN103747283A (en) 2014-04-23
CN103747283B true CN103747283B (en) 2017-02-01

Family

ID=50504265

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310723717.4A Expired - Fee Related CN103747283B (en) 2013-12-24 2013-12-24 Video fragment download method

Country Status (1)

Country Link
CN (1) CN103747283B (en)

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105263037B (en) * 2014-07-18 2018-07-03 中国科学院声学研究所 A kind of adaptive fluidizing method based on client-cache
CN105744299A (en) * 2014-12-08 2016-07-06 深圳Tcl数字技术有限公司 Dynamic code rate switching method, server and system
CN104581229B (en) * 2015-01-16 2018-08-03 京东方科技集团股份有限公司 A kind of transmission of flow media data device, method and system
CN104581340B (en) * 2015-01-16 2018-02-16 京东方科技集团股份有限公司 Client, stream medium data method of reseptance and stream medium data transmission system
CN106331717B (en) * 2015-06-30 2019-05-07 成都鼎桥通信技术有限公司 Video code rate self-adapting regulation method and sending ending equipment
WO2017063189A1 (en) * 2015-10-16 2017-04-20 Qualcomm Incorporated Deadline signaling for streaming of media data
CN105578201B (en) * 2015-12-23 2018-11-13 武汉市烽视威科技有限公司 A kind of system and method for realizing HLS live streaming code check number dynamic adjustment
CN109644296A (en) * 2016-10-10 2019-04-16 华为技术有限公司 A kind of video stream transmission method, relevant device and system
WO2018068262A1 (en) * 2016-10-13 2018-04-19 华为技术有限公司 Video code rate acquisition method, apparatus, and network side device
CN109218847B (en) * 2017-06-30 2022-03-04 中兴通讯股份有限公司 Download control method and device and multimedia terminal
CN107809647A (en) * 2017-10-24 2018-03-16 深圳市创维软件有限公司 Code check switching method, device and the storage medium of video playback
CN109547810A (en) * 2018-12-28 2019-03-29 北京奇艺世纪科技有限公司 A kind of live broadcasting method and device
CN110149534B (en) * 2019-06-12 2021-06-08 深圳市大数据研究院 Decision tree-based adaptive video stream transcoding method and device
CN111147332B (en) * 2019-12-29 2022-04-29 北京浪潮数据技术有限公司 Method, system and related device for determining uploading progress of cloud backup of storage system
CN115002557B (en) * 2022-05-23 2024-01-30 北京字跳网络技术有限公司 Network speed prediction method, device, equipment and storage medium
CN116320507B (en) * 2023-05-26 2023-08-08 湖南快乐阳光互动娱乐传媒有限公司 Video data downloading method and device

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103024593A (en) * 2012-12-26 2013-04-03 人民搜索网络股份公司 Online VOD (video on demand) acceleration system and online VOD playing method
CN103428107A (en) * 2012-05-14 2013-12-04 中国科学院声学研究所 Self-adaptive bitstream switching method and system based on cache underflow probability estimation

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7818445B2 (en) * 2008-10-15 2010-10-19 Patentvc Ltd. Methods and devices for obtaining a broadcast-like streaming content

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103428107A (en) * 2012-05-14 2013-12-04 中国科学院声学研究所 Self-adaptive bitstream switching method and system based on cache underflow probability estimation
CN103024593A (en) * 2012-12-26 2013-04-03 人民搜索网络股份公司 Online VOD (video on demand) acceleration system and online VOD playing method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
《A Rate Adaptation Solution for Distance Education System over HTTP Streaming》;Bin Li et al.;《2013 IEEE 10th International Conference on High Performance Computing and Communications&2013 IEEE International Conference onEmbedded and Ubiquitous Computing》;20131115;2385-2389 *
《一种实时自适应HTTP流化码流切换算法》;张辉帅;《计算机工程》;20130228;第39卷(第2期);1-5 *

Also Published As

Publication number Publication date
CN103747283A (en) 2014-04-23

Similar Documents

Publication Publication Date Title
CN103747283B (en) Video fragment download method
CN104254109B (en) User equipment, base station, streaming media self-adapting Transmission system and method
CN106664255A (en) Data flow control method
CN111357318B (en) Method and device for synchronizing between different data packet flows
SG171628A1 (en) Packet based communications
Cherif et al. DASH fast start using HTTP/2
CN109245936A (en) Control method, service quality control entity and the communication equipment of network transfer delay
CN108965775A (en) Method of adjustment, device and the storage medium of data packetloss processing strategie
RU2018132348A (en) RETURNING DATA ON PACKAGE NETWORKS
CN110691037A (en) Service delivery in a communication network
Liu et al. Segment duration for rate adaptation of adaptive HTTP streaming
CN104661112B (en) Video stream file transmission method and device based on scalable selection window
Wamser et al. Utilizing buffered YouTube playtime for QoE-oriented scheduling in OFDMA networks
Hwang et al. HAVS: Hybrid adaptive video streaming for mobile devices
Najm et al. Improvement of SCTP congestion control in the LTE-A network
Evensen et al. Adaptive media streaming to mobile devices: challenges, enhancements, and recommendations
CN105472404B (en) A kind of data processing method in video traffic
WO2018077042A1 (en) Downlink power control method and apparatus for emtc, and base station
CN107483990A (en) Dynamic code rate adjusting method, device and the Transmission system of a kind of streaming media
CN104580006B (en) A kind of mobile network&#39;s method of controlling transmission rate, apparatus and system
CN105490964B (en) A kind for the treatment of method and apparatus of service quality
CN104471911B (en) A kind of method, system and the equipment of transmission services data
Ko et al. A handover-aware seamless video streaming scheme in heterogeneous wireless networks
EP3000198B1 (en) Method and apparatus for optimization of video transmissions
CN110048800B (en) MCS (modulation and coding scheme) grade adjusting method and base station

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20170201

Termination date: 20191224