CN104320424A - Streaming media fragmentation downloading method and device - Google Patents
Streaming media fragmentation downloading method and device Download PDFInfo
- Publication number
- CN104320424A CN104320424A CN201410670352.8A CN201410670352A CN104320424A CN 104320424 A CN104320424 A CN 104320424A CN 201410670352 A CN201410670352 A CN 201410670352A CN 104320424 A CN104320424 A CN 104320424A
- Authority
- CN
- China
- Prior art keywords
- code check
- code
- grade
- group
- place
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/266—Channel or content management, e.g. generation and management of keys and entitlement messages in a conditional access system, merging a VOD unicast channel into a multicast channel
- H04N21/2662—Controlling the complexity of the video stream, e.g. by scaling the resolution or bitrate of the video stream based on the client capabilities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client
- H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
- H04N21/637—Control signals issued by the client directed to the server or network components
- H04N21/6377—Control signals issued by the client directed to the server or network components directed to server
- H04N21/6379—Control signals issued by the client directed to the server or network components directed to server directed to encoder, e.g. for requesting a lower encoding rate
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/433—Content storage operation, e.g. storage operation in response to a pause request, caching operations
- H04N21/4335—Housekeeping operations, e.g. prioritizing content for deletion because of storage space restrictions
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Information Transfer Between Computers (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a streaming media fragmentation downloading method. The method includes the steps that when a terminal device determines that a current cache duration is less than a first threshold value and no fragmentation downloaded in process exists at present, the optimal code rate is decided through an MDP; in terms of grades, a code rate set where the optimal code rate exists is compared with a code rate set where a code rate at which a fragmentation is downloaded before; the code rate at which a next fragmentation is downloaded is determined according to the comparative result. Based on the same inventive concept, the invention further provides a streaming media fragmentation downloading device through which smooth broadcasting of streaming media can be achieved under a dynamic network environment.
Description
Technical field
The present invention relates to communication technical field, particularly a kind of Streaming Media burst method for down loading and device.
Background technology
Current, Internet video operator, when providing media content, not only needs to take into account all types of terminal equipment, also needs to adapt to the dynamic change of same terminal equipment in ability and bandwidth, the viewing experience that guarantee user is level and smooth simultaneously.
In the face of the challenge that terminal unit ability and bandwidth change, adaptive stream media technology is arisen at the historic moment, and the operation principle of adaptive stream media technology is: first source file is cut into multiple burst according to multiple code check by content, is stored on server; Then when content is play, according to the network bandwidth and the media processing capacity of terminal equipment, the content file of the suitable code check of Dynamic Selection is play.
Provide Qos to serve as much as possible based on existing packet network, therefore precisely predict that the network bandwidth is more difficult by tuning, it is prediction instantaneous bandwidth, or use the average bandwidth of more difficult reaction network bandwidth change, cause the code check selected many times can not reflect network environment in the long term.
Under ultra high-definition (Ultra High Definition, UHD) environment, code check is very high, and network bandwidth requirement is higher, and bandwidth variation also can be comparatively large, and the video flowing code check provided is understood more naturally, and how level and smooth switching particularly important.The existing bandwidth according to prediction directly switches when may occur that the network bandwidth changes greatly (such as: switch from SD (SD) to UHD), code check directly switches, the picture difference of user awareness is too large, experience bad, particularly occur that shake often may occur the situations such as buffer memory; Switchover policy step by step, when code check is more, switch speed is comparatively slow, the experience of entire effect user high-quality video;
In addition, existing adaptive decision-making model under UHD environment, multi code Rate of Chinese character makes policy space larger, causes switch speed comparatively slow, and time larger for the burst code check grade difference predicted, do not do corresponding smooth optimization, Consumer's Experience is good not.
Summary of the invention
In view of this, the application provides a kind of Streaming Media burst method for down loading and device, can not realize the problem of the level and smooth broadcasting of Streaming Media under solving dynamic network environment.
For solving the problems of the technologies described above, the technical scheme of the application is achieved in that
A kind of Streaming Media burst method for down loading, the method comprises:
Terminal equipment receives and stores the code check group grade that content service provider issues, and the code check that each code check group is corresponding;
Described terminal equipment determination current cache duration is less than first threshold, and current when there is not the burst downloaded, and goes out optimal code rates by MDP strategic decision-making; And by the code check group at described optimal code rates place, compare with the grade of the code check group at the code check place of a front downloading slicing;
When determining the code check group at described optimal code rates place, during grade higher than the code check group at the code check place of a front downloading slicing, if first timer does not exist, then starting first timer, and using the code check of a front downloading slicing to download next burst;
When determining the grade of code check group at described optimal code rates place, during grade lower than the code check group at the code check place of a front downloading slicing, each code check is traveled through from high to low in the code check group that the code check group grade at the code check place than a front downloading slicing is low, described optimal code rates is greater than if traverse, and meet the code check that current cache residual time length is greater than the 3rd threshold value, then use the code check traversed to download next burst; Otherwise, use described optimal code rates to download next burst;
When determining the grade of code check group at described optimal code rates place, time identical with the grade of the code check group at the code check place of a front downloading slicing, described optimal code rates is used to download next burst.
A kind of Streaming Media burst download apparatus, this device comprises: receiving element, memory cell, decision package and processing unit;
Described receiving element, for receiving the code check group grade that content service provider issues, and the code check that each code check group is corresponding;
Described memory cell, for the code check group grade received by described receiving element, and code check corresponding to each code check group stores;
Described decision package, for when determining that current cache duration is less than first threshold, and current when there is not the burst downloaded, go out optimal code rates by MDP strategic decision-making;
Described processing unit, for the code check group in the described memory cell at optimal code rates place that described decision package decision-making gone out, compares with the grade of the code check group at the code check place of a front downloading slicing; When determining the code check group at described optimal code rates place, during grade higher than the code check group at the code check place of a front downloading slicing, if first timer does not exist, then starting first timer, and using the code check of a front downloading slicing to download next burst; When determining the grade of code check group at described optimal code rates place, during grade lower than the code check group at the code check place of a front downloading slicing, each code check is traveled through from high to low in the code check group that the code check group grade at the code check place than a front downloading slicing is low, described optimal code rates is greater than if traverse, and meet the code check that current cache residual time length is greater than the 3rd threshold value, then use the code check traversed to download next burst; Otherwise, use described optimal code rates to download next burst; When determining the grade of code check group at described optimal code rates place, time identical with the grade of the code check group at the code check place of a front downloading slicing, described optimal code rates is used to download next burst.
From technical scheme above, in the application, terminal equipment determination current cache duration is less than first threshold, and current when there is not the burst downloaded, and goes out optimal code rates by MDP strategic decision-making; And by the code check group at described optimal code rates place, compare with the grade of the code check group at the code check place of a front downloading slicing; And determine according to comparative result the code check downloading next burst, the level and smooth broadcasting of Streaming Media can be realized under dynamic network environment.
Accompanying drawing explanation
Fig. 1 is Streaming Media burst method for down loading schematic flow sheet in the embodiment of the present application;
Fig. 2 starts high to Low code check group switching flow schematic diagram in the embodiment of the present application;
Fig. 3 is the apparatus structure schematic diagram being applied to above-mentioned technology in the embodiment of the present application.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with accompanying drawing and according to embodiment, technical scheme of the present invention is described in detail.
Terminal equipment needs level and smooth playing stream media, due to the impact such as bandwidth, buffer memory, needs a kind of adaptive stream media burst method for down loading to realize.Hereafter for convenience Streaming Media burst be directly called burst.
In the embodiment of the present application, terminal equipment can receive and store the code check group grade that content service provider issues, and the code check that each code check group is corresponding; Also code check group grade can be configured in this locality, and the code check that each code check group is corresponding.
The code check that each code check group is corresponding, if continuous print code check, also can use range of code rates to represent.All code checks that the code check group that grade is high comprises all are greater than all code checks that the low code check group of code check group grade comprises.If code check component is three groups: UHD code check group (grade is 1), high definition (HD) code check group (grade is 2) and SD code check group (grade is 3).The grade of the less correspondence of grade point is higher, and namely the grade of UHD code check group is higher than the grade of HD code check group, and the grade of HD code check group is higher than the grade of SD code check group.All code checks in UHD code check group are all greater than the code check that in HD code check group, code check is maximum, and all code checks in HD code check group are all greater than the code check that in HD code check group, code check is maximum.
Be Streaming Media burst method for down loading schematic flow sheet in the embodiment of the present application see Fig. 1, Fig. 1.Concrete steps are:
Step 101, terminal equipment determination current cache duration is less than first threshold, and current when there is not the burst downloaded, and goes out optimal code rates by Markovian decision process (Markov Decision Process, MDP) strategic decision-making.
Current in this step do not exist the burst downloaded, and is the last burst downloaded and downloaded, and do not start the download of next burst.When determining that current cache duration is less than first threshold, determining currently whether there is the burst downloaded, if so, then waiting for that this burst has been downloaded, carrying out MDP decision-making; Otherwise, directly carry out MDP decision-making.
When terminal equipment determination current cache duration is not less than first threshold, start second timer, during second timer time-out, use the code check of a front downloading slicing to download next burst.Before second timer time-out, terminal equipment is in wait state, does not download next burst.
This terminal equipment goes out optimal code rates by MDP strategic decision-making, specifically comprises:
Supposing that the code check in each code check group adds up to N, is then each Data Rate Distribution code check grade from 1 to N, wherein, N be greater than 0 integer.
(1), can also buffer memory duration and by code check grade corresponding for the code check of a front downloading slicing in the buffer by the front burst once downloaded, respectively as t0 and q0 in the initial condition S (t0, q0) in MDP strategy.
(2) when, supposing to download next burst, corresponding state is S (j, y), then at initial condition S (t0, q0) under, determine to download action collection corresponding to the code check of next burst according to probability transfer matrix and evaluation function value, a corresponding a kind of action of code check, any one decision-making can enter different state possibilities, i.e. the corresponding multiple possibility of same code check grade.
Wherein, using current bandwidth as the bandwidth parameter in the cumulative distribution function (Cumulative Distribution Function, CDF) of calculating probability transfer matrix.Suppose fixed storage M burst in buffer memory, each point of leaf length is T second, and point discretizations such as n per second, and the duration of burst buffer memory is in the buffer not more than M × T × n, then have 0≤i≤M × T × n, 0≤j≤M × T × n; When the grade then selecting the code check of next burst of download corresponding is q, transfer matrix can be defined as:
Wherein,
Wherein, the burst that i once downloads before being can also buffer memory duration t0, x be the code check grade q0 that the code check of a front downloading slicing is corresponding in the buffer; S (q) is the burst size of code check q, and F () represents the CDF of bandwidth Network Based, 0≤i≤(M-1) * T*n represent buffer memory less than; Otherwise then represent that buffer memory is full.
During the application's specific implementation, can the rewards and punishments value corresponding under different conditions of pre-configured each code check, also can provide a function and under different conditions, calculate corresponding rewards and punishments value for each code check; The rewards and punishments value no matter obtained by which kind of mode, as evaluation function value, is together obtained to use with probability transfer matrix and uses the code check selected to download action value corresponding to next burst by obtaining.
Provide a kind of mode calculating rewards and punishments value below, when initial condition is S (t0, q0), during the code check grade q selecting the code check of next burst of download corresponding, corresponding evaluation function E
q(t0, q0) is:
E
q(t0,q0)＝r(q)-d(t0,q)-p(t0,q)
Wherein,
The award score of code check q is chosen in r (q) representative, and this value is pre-configured, the award score that namely pre-configured each code check is corresponding; The punishment in burst time limit is missed in d (t0, q) representative, and C can be set to constant to finely tune model; P (t0, q) represents the punishment that code check grade switches, and this penalty value is pre-configured, i.e. pre-configured penalty value when switching to another one code check downloading slicing by a code check.
(3), according to state transition probability matrix and evaluation function value determine the action collection of each code check under different conditions, carry out MDP strategic decision-making, decision-making goes out optimal code rates.
In order to reduce tactful branch, in the iterative process of decision-making, introducing policy information entropy, for arbitrary iteration decision path branch, when policy information entropy is greater than default entropy threshold after iteration preset times, stopping the iteration of described decision path branch.
Bandwidth parameter in the CDF of calculating probability transfer matrix is current bandwidth.In order to the change of dynamic response network environment, this terminal equipment is according to available network bandwidth value in cycle detection Preset Time, adopt Poisson distribution analog network Bandwidth Model to obtain current network bandwidth distribute and upgrade CDF, thus can probability transfer matrix be dynamically updated.
Step 102, this terminal equipment, by the code check group at described optimal code rates place, compares with the grade of the code check group at the code check place of a front downloading slicing.
Step 103, when the code check group at described optimal code rates place, during grade higher than the code check group at the code check place of a front downloading slicing, if first timer does not exist, then this terminal equipment starts first timer, and uses the code check of a front downloading slicing to download next burst.Process ends.
If first timer exists, and determine that first timer is super, then the first timer of time-out deleted by terminal equipment, and determine whether current cache duration is greater than Second Threshold, if so, use and download next burst than lowest bit rate in the higher leveled code check group of the code check group at the code check place of a front downloading slicing; Otherwise, use the code check of a front downloading slicing to download next burst;
If first timer exists, and determine that first timer is not overtime, then use the code check of a front downloading slicing to download next burst.
When the grade of the code check group at described optimal code rates place, time identical with the grade of the code check group at the code check place of a front downloading slicing, described optimal code rates is used to download next burst.
Step 104, when the grade of the code check group at described optimal code rates place, during grade lower than the code check group at the code check place of a front downloading slicing, each code check is traveled through from high to low in the code check group that the code check group grade at the code check place than a front downloading slicing is low, described optimal code rates is greater than if traverse, and meet the code check that current cache residual time length is greater than the 3rd threshold value, then use the code check traversed to download next burst; Otherwise, use described optimal code rates to download next burst.Process ends.
In this step, before traveling through each code check from high to low described in execution in the code check group that the code check group grade at the code check place than a front downloading slicing is low, also need to determine whether current cache duration is less than the 3rd threshold value, if so, use described optimal code rates to download next burst; Otherwise, described in performing, in the code check group that the code check group grade at the code check place than a front downloading slicing is low, travel through each code check and subsequent step from high to low.
Described current cache residual time length in this step is, the difference of the duration of current cache duration and the many consumption of next burst of download;
The duration of the many consumption of next burst of described download is, the difference of the current code check that traverses and described optimal code rates, with the product of the burst duration of next burst described, then with the business of average bandwidth in Preset Time.
This step is code check group switching flow from high to low.Be specifically in the embodiment of the present application, start high to Low code check group switching flow schematic diagram see Fig. 2, Fig. 2.Concrete steps are:
Step 201, when the grade of the code check group at described optimal code rates place, lower than the grade of the code check group at the code check place of a front downloading slicing, and when current cache duration is not more than the 3rd threshold value, using the code check group at the code check place of a front downloading slicing as current code check group.
Step 202, the code check group that grade is lower than current code check group selected by this terminal equipment.
Step 203, this terminal equipment travels through the code check in this code check group from high to low.
Step 204, this terminal equipment determines whether the current code check traversed is not more than described optimal code rates, if so, performs step 207; Otherwise, perform step 205.
Step 205, whether this terminal equipment determination current residual buffer memory is greater than the 3rd threshold value, if so, performs step 208; Otherwise, perform step 206.
Step 206, the code check whether do not traversed in addition in this code check group determined by this terminal equipment, if so, performs step 203; Otherwise, perform step 202.
Step 207, this terminal equipment uses described optimal code rates to download next burst.Process ends
Step 208, this terminal equipment uses the current code check traversed to download next burst.
Step 105, when the grade of the code check group at described optimal code rates place, time identical with the grade of the code check group at the code check place of a front downloading slicing, uses described optimal code rates to download next burst.
Based on same inventive concept, a kind of Streaming Media burst download apparatus is also proposed in the embodiment of the present application.It is the apparatus structure schematic diagram being applied to above-mentioned technology in the embodiment of the present application see Fig. 3, Fig. 3.This device comprises: receiving element 301, memory cell 302, decision package 303 and processing unit 304;
Receiving element 301, for receiving the code check group grade that content service provider issues, and the code check that each code check group is corresponding;
Memory cell 302, for the code check group grade received by receiving element 301, and code check corresponding to each code check group stores;
Decision package 303, for when determining that current cache duration is less than first threshold, and current when there is not the burst downloaded, go out optimal code rates by MDP strategic decision-making;
Processing unit 304, for code check group in the memory cell 302 at optimal code rates place that decision package 303 decision-making gone out, compares with the grade of the code check group in the memory cell 302 at the code check place of a front downloading slicing; When determining the code check group at described optimal code rates place, during grade higher than the code check group at the code check place of a front downloading slicing, if first timer does not exist, then starting first timer, and using the code check of a front downloading slicing to download next burst; When determining the grade of code check group at described optimal code rates place, during grade lower than the code check group at the code check place of a front downloading slicing, each code check is traveled through from high to low in the code check group that the code check group grade at the code check place than a front downloading slicing is low, described optimal code rates is greater than if traverse, and meet the code check that current cache residual time length is greater than the 3rd threshold value, then use the code check traversed to download next burst; Otherwise, use described optimal code rates to download next burst; When determining the grade of code check group at described optimal code rates place, time identical with the grade of the code check group at the code check place of a front downloading slicing, described optimal code rates is used to download next burst.
Preferably,
Processing unit 304, be further used for the code check group when determining described optimal code rates place, during grade higher than the code check group at the code check place of a front downloading slicing, if first timer exists, and determine that first timer is overtime, delete the first timer of time-out, and determine whether current cache duration is greater than Second Threshold, if so, use and download next burst than lowest bit rate in the higher leveled code check group of the code check group at the code check place of a front downloading slicing; Otherwise, use the code check of a front downloading slicing to download next burst; If first timer exists, and determine that first timer is not overtime, then use the code check of a front downloading slicing to download next burst.
Preferably,
Processing unit 304, when being further used for determining that current cache duration is not less than first threshold, starts second timer, during second timer time-out, uses the code check of a front downloading slicing to download next burst.
Preferably,
Processing unit 304, be further used for the grade when the code check group determining described optimal code rates place, during grade lower than the code check group at the code check place of a front downloading slicing, determine whether current cache duration is less than the 3rd threshold value further, if so, described optimal code rates is used to download next burst; Otherwise, described in performing, in the code check group that the code check group grade at the code check place than a front downloading slicing is low, travel through each code check and subsequent step from high to low.
Preferably,
Decision package 303, for configuring code check grade corresponding to each code check, code check grade is not less than 1 and is not more than N, and N is the number of code check; Configure the rewards and punishments value that each code check is corresponding under different conditions, as evaluation function value; Using current bandwidth distribution as the bandwidth parameter in the cumulative distribution function CDF of calculating probability transfer matrix; When going out optimal code rates by MDP strategic decision-making, can also buffer memory duration and by code check grade corresponding for the code check of a front downloading slicing in the buffer by the front burst once downloaded, respectively as t0 and q0 in the initial condition S (t0, q0) in MDP strategy; And determine the action collection of each code check under different conditions according to state transition probability matrix and evaluation function value, carry out MDP strategic decision-making, decision-making goes out optimal code rates corresponding to optimal code rates grade.
Preferably,
Decision package 303, is further used for introducing policy information entropy in the iterative process of decision-making, for arbitrary iteration decision path branch, when policy information entropy is greater than default entropy threshold after iteration preset times, stops the iteration of described decision path branch.
Preferably,
Processing unit 304, is further used for according to available network bandwidth value in cycle detection Preset Time, adopts Poisson distribution analog network Bandwidth Model to obtain current network bandwidth and distributes and upgrade CDF.
The unit of above-described embodiment can be integrated in one, and also can be separated deployment; A unit can be merged into, also can split into multiple subelement further.
In sum, the application is less than first threshold by terminal equipment determination current cache duration, and current when there is not the burst downloaded, and goes out optimal code rates by MDP strategic decision-making; And by the code check group at described optimal code rates place, compare with the grade of the code check group at the code check place of a front downloading slicing; And determine according to comparative result the code check downloading next burst, the level and smooth broadcasting of Streaming Media can be realized under dynamic network environment.
Use MDP model more can adapt to change of network environment in the long run in the application's specific embodiment, comentropy is used for representing the size of amount of information, entropy is larger, amount of information is less, therefore in certain tactful branch of decision-making, introduce the comentropy sum that policy information entropy represents node on this path, this value exceedes default entropy threshold (amount of information is too small), stops the decision-making of this branch, thus minimizing policy space, realize high-speed decision and switching.In addition, to the multi code Rate of Chinese character that may exist under UHD environment from SD, HD and UHD, divide into groups by providing code check (can support that user configures, should be less with group code rate difference, switch according to model decision and do not affect Consumer's Experience), switch between the group for model decision and carry out taking over seamlessly of Lasy, specifically: when switching to grade height code check group by grade is low, by waiting for a period of time, and before being switched to startup waiting timer, the lowest bit rate of the high ranking score group of code check grouping selects burst, reaches the object removed network jitter and take over seamlessly; For when switching from grade is high to grade low bit-rate group, next burst is downloaded according to high code rate in code check group in the middle of current media buffering data selection, high-quality video be provided and take over seamlessly, being switched by the transition between group, finally allow user obtain better viewing experience.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (14)
1. a Streaming Media burst method for down loading, is characterized in that, the method comprises:
Terminal equipment receives and stores the code check group grade that content service provider issues, and the code check that each code check group is corresponding;
Described terminal equipment determination current cache duration is less than first threshold, and current when there is not the burst downloaded, and goes out optimal code rates by Markovian decision process MDP strategic decision-making; And by the code check group at described optimal code rates place, compare with the grade of the code check group at the code check place of a front downloading slicing;
When determining the code check group at described optimal code rates place, during grade higher than the code check group at the code check place of a front downloading slicing, if first timer does not exist, then starting first timer, and using the code check of a front downloading slicing to download next burst;
When determining the grade of code check group at described optimal code rates place, during grade lower than the code check group at the code check place of a front downloading slicing, each code check is traveled through from high to low in the code check group that the code check group grade at the code check place than a front downloading slicing is low, described optimal code rates is greater than if traverse, and meet the code check that current cache residual time length is greater than the 3rd threshold value, then use the code check traversed to download next burst; Otherwise, use described optimal code rates to download next burst;
When determining the grade of code check group at described optimal code rates place, time identical with the grade of the code check group at the code check place of a front downloading slicing, described optimal code rates is used to download next burst.
2. method according to claim 1, is characterized in that, described when determine the code check group at described optimal code rates place, and during grade higher than the code check group at the code check place of a front downloading slicing, described method comprises further:
If first timer exists, and determine that first timer is overtime, delete the first timer of time-out, and determine whether current cache duration is greater than Second Threshold, if so, use and download next burst than lowest bit rate in the higher leveled code check group of the code check group at the code check place of a front downloading slicing; Otherwise, use the code check of a front downloading slicing to download next burst;
If first timer exists, and determine that first timer is not overtime, then use the code check of a front downloading slicing to download next burst.
3. method according to claim 1, is characterized in that, described method comprises further:
When described terminal equipment determination current cache duration is not less than first threshold, start second timer, during second timer time-out, use the code check of a front downloading slicing to download next burst.
4. method according to claim 1, is characterized in that, the described grade of code check group when determine described optimal code rates place, and during grade lower than the code check group at the code check place of a front downloading slicing, described method comprises further:
Determine whether current cache duration is less than the 3rd threshold value, if so, use described optimal code rates to download next burst; Otherwise, described in performing, in the code check group that the code check group grade at the code check place than a front downloading slicing is low, travel through each code check and subsequent step from high to low.
5. the method according to claim 1-4 any one, is characterized in that, describedly goes out optimal code rates by MDP strategic decision-making, comprising:
Configure the code check grade that each code check is corresponding, code check grade is not less than 1 and is not more than N, and N is the number of code check; Configure the rewards and punishments value that each code check is corresponding under different conditions, as evaluation function value; Using current bandwidth distribution as the bandwidth parameter in the cumulative distribution function CDF of calculating probability transfer matrix;
Can also buffer memory duration and by code check grade corresponding for the code check of a front downloading slicing in the buffer by the front burst once downloaded, respectively as t0 and q0 in the initial condition S (t0, q0) in MDP strategy; And determine the action collection of each code check under different conditions according to state transition probability matrix and evaluation function value, carry out MDP strategic decision-making, decision-making goes out optimal code rates corresponding to optimal code rates grade.
6. method according to claim 5, is characterized in that, described method comprises further:
In the iterative process of decision-making, introduce policy information entropy, for arbitrary iteration decision path branch, when policy information entropy is greater than default entropy threshold after iteration preset times, stop the iteration of described decision path branch.
7. method according to claim 5, is characterized in that, described method comprises further:
According to available network bandwidth value in cycle detection Preset Time, adopt Poisson distribution analog network Bandwidth Model to obtain current network bandwidth and distribute and upgrade CDF.
8. a Streaming Media burst download apparatus, is characterized in that, this device comprises: receiving element, memory cell, decision package and processing unit;
Described receiving element, for receiving the code check group grade that content service provider issues, and the code check that each code check group is corresponding;
Described memory cell, for the code check group grade received by described receiving element, and code check corresponding to each code check group stores;
Described decision package, for when determining that current cache duration is less than first threshold, and current when there is not the burst downloaded, go out optimal code rates by Markovian decision process MDP strategic decision-making;
Described processing unit, for the code check group in the described memory cell at optimal code rates place that described decision package decision-making gone out, compares with the grade of the code check group at the code check place of a front downloading slicing; When determining the code check group at described optimal code rates place, during grade higher than the code check group at the code check place of a front downloading slicing, if first timer does not exist, then starting first timer, and using the code check of a front downloading slicing to download next burst; When determining the grade of code check group at described optimal code rates place, during grade lower than the code check group at the code check place of a front downloading slicing, each code check is traveled through from high to low in the code check group that the code check group grade at the code check place than a front downloading slicing is low, described optimal code rates is greater than if traverse, and meet the code check that current cache residual time length is greater than the 3rd threshold value, then use the code check traversed to download next burst; Otherwise, use described optimal code rates to download next burst; When determining the grade of code check group at described optimal code rates place, time identical with the grade of the code check group at the code check place of a front downloading slicing, described optimal code rates is used to download next burst.
9. device according to claim 8, is characterized in that,
Described processing unit, be further used for the code check group when determining described optimal code rates place, during grade higher than the code check group at the code check place of a front downloading slicing, if first timer exists, and determine that first timer is overtime, delete the first timer of time-out, and determine whether current cache duration is greater than Second Threshold, if so, use and download next burst than lowest bit rate in the higher leveled code check group of the code check group at the code check place of a front downloading slicing; Otherwise, use the code check of a front downloading slicing to download next burst; If first timer exists, and determine that first timer is not overtime, then use the code check of a front downloading slicing to download next burst.
10. device according to claim 8, is characterized in that,
Described processing unit, when being further used for determining that current cache duration is not less than first threshold, starts second timer, during second timer time-out, uses the code check of a front downloading slicing to download next burst.
11. devices according to claim 8, is characterized in that,
Described processing unit, be further used for the grade when the code check group determining described optimal code rates place, during grade lower than the code check group at the code check place of a front downloading slicing, determine whether current cache duration is less than the 3rd threshold value further, if so, described optimal code rates is used to download next burst; Otherwise, described in performing, in the code check group that the code check group grade at the code check place than a front downloading slicing is low, travel through each code check and subsequent step from high to low.
12. devices according to Claim 8 described in-11 any one, is characterized in that,
Described decision package, for configuring code check grade corresponding to each code check, code check grade is not less than 1 and is not more than N, and N is the number of code check; Configure the rewards and punishments value that each code check is corresponding under different conditions, as evaluation function value; Using current bandwidth distribution as the bandwidth parameter in the cumulative distribution function CDF of calculating probability transfer matrix; When going out optimal code rates by MDP strategic decision-making, can also buffer memory duration and by code check grade corresponding for the code check of a front downloading slicing in the buffer by the front burst once downloaded, respectively as t0 and q0 in the initial condition S (t0, q0) in MDP strategy; And determine the action collection of each code check under different conditions according to state transition probability matrix and evaluation function value, carry out MDP strategic decision-making, decision-making goes out optimal code rates corresponding to optimal code rates grade.
13. devices according to claim 12, is characterized in that,
Described decision package, is further used for introducing policy information entropy in the iterative process of decision-making, for arbitrary iteration decision path branch, when policy information entropy is greater than default entropy threshold after iteration preset times, stops the iteration of described decision path branch.
14. devices according to claim 12, is characterized in that,
Described processing unit, is further used for according to available network bandwidth value in cycle detection Preset Time, and adopts Poisson distribution analog network Bandwidth Model acquisition current network bandwidth distribute and upgrade CDF.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410670352.8A CN104320424B (en) | 2014-11-20 | 2014-11-20 | A kind of Streaming Media burst method for down loading and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410670352.8A CN104320424B (en) | 2014-11-20 | 2014-11-20 | A kind of Streaming Media burst method for down loading and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104320424A true CN104320424A (en) | 2015-01-28 |
| CN104320424B CN104320424B (en) | 2017-06-16 |
Family
ID=52375593
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410670352.8A Active CN104320424B (en) | 2014-11-20 | 2014-11-20 | A kind of Streaming Media burst method for down loading and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104320424B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105611328A (en) * | 2015-12-25 | 2016-05-25 | 深圳Tcl新技术有限公司 | Video data accelerated downloading method and device based on HLS (HTTP Live Streaming)stream media |
| CN106791930A (en) * | 2017-01-04 | 2017-05-31 | 北京百度网讯科技有限公司 | A kind of method for processing video frequency and device |
| CN108668146A (en) * | 2017-03-27 | 2018-10-16 | 华为技术有限公司 | A kind of method and apparatus of adjustment streaming media bit rate |
| CN109714631A (en) * | 2019-02-26 | 2019-05-03 | 华南理工大学 | One kind being based on HTTP video flowing dynamic self-adaptation bit-rate selection method |
| CN109769140A (en) * | 2018-12-20 | 2019-05-17 | 南京杰迈视讯科技有限公司 | A kind of network video smoothness control method for playing back based on stream media technology |
| CN110022498A (en) * | 2018-01-08 | 2019-07-16 | 深圳市中兴微电子技术有限公司 | A kind of method and device for realizing code rate switching |
| CN110662093A (en) * | 2019-10-17 | 2020-01-07 | 北京达佳互联信息技术有限公司 | Initial code rate determination method and device, electronic equipment and storage medium |
| CN110933516A (en) * | 2018-09-19 | 2020-03-27 | 华为技术有限公司 | Multimedia live broadcast method, device and equipment |
| CN112565906A (en) * | 2020-11-05 | 2021-03-26 | 深圳市麦谷科技有限公司 | On-demand video processing method and system |
| CN113473248A (en) * | 2020-03-30 | 2021-10-01 | 中国电信股份有限公司 | Streaming media transmission method, device and computer readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070204056A1 (en) * | 2006-02-28 | 2007-08-30 | Sharp Laboratories Of America, Inc. | Systems and methods for reducing the effects of variations on the playback of streaming media |
| CN101478564A (en) * | 2008-12-31 | 2009-07-08 | 西安交通大学 | Adaptive hierarchical transmission structure design method for P2P stream media network |
| CN103379379A (en) * | 2012-04-13 | 2013-10-30 | 腾讯科技（深圳）有限公司 | Streaming media buffer play method and apparatus |
-
2014
- 2014-11-20 CN CN201410670352.8A patent/CN104320424B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070204056A1 (en) * | 2006-02-28 | 2007-08-30 | Sharp Laboratories Of America, Inc. | Systems and methods for reducing the effects of variations on the playback of streaming media |
| CN101478564A (en) * | 2008-12-31 | 2009-07-08 | 西安交通大学 | Adaptive hierarchical transmission structure design method for P2P stream media network |
| CN103379379A (en) * | 2012-04-13 | 2013-10-30 | 腾讯科技（深圳）有限公司 | Streaming media buffer play method and apparatus |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017107378A1 (en) * | 2015-12-25 | 2017-06-29 | 深圳Tcl新技术有限公司 | Accelerated video data downloading method and device based on hls stream media |
| CN105611328B (en) * | 2015-12-25 | 2019-01-01 | 深圳Tcl新技术有限公司 | Video data based on HLS Streaming Media accelerates method for down loading and device |
| CN105611328A (en) * | 2015-12-25 | 2016-05-25 | 深圳Tcl新技术有限公司 | Video data accelerated downloading method and device based on HLS (HTTP Live Streaming)stream media |
| CN106791930B (en) * | 2017-01-04 | 2020-03-31 | 北京百度网讯科技有限公司 | Video processing method and device |
| CN106791930A (en) * | 2017-01-04 | 2017-05-31 | 北京百度网讯科技有限公司 | A kind of method for processing video frequency and device |
| CN108668146A (en) * | 2017-03-27 | 2018-10-16 | 华为技术有限公司 | A kind of method and apparatus of adjustment streaming media bit rate |
| CN108668146B (en) * | 2017-03-27 | 2021-07-16 | 华为技术有限公司 | Method and equipment for adjusting streaming media code rate |
| CN110022498B (en) * | 2018-01-08 | 2021-04-27 | 深圳市中兴微电子技术有限公司 | Method and device for realizing code rate switching |
| CN110022498A (en) * | 2018-01-08 | 2019-07-16 | 深圳市中兴微电子技术有限公司 | A kind of method and device for realizing code rate switching |
| CN110933516A (en) * | 2018-09-19 | 2020-03-27 | 华为技术有限公司 | Multimedia live broadcast method, device and equipment |
| CN110933516B (en) * | 2018-09-19 | 2022-06-10 | 华为技术有限公司 | Multimedia live broadcasting method, device and equipment |
| CN109769140A (en) * | 2018-12-20 | 2019-05-17 | 南京杰迈视讯科技有限公司 | A kind of network video smoothness control method for playing back based on stream media technology |
| CN109714631A (en) * | 2019-02-26 | 2019-05-03 | 华南理工大学 | One kind being based on HTTP video flowing dynamic self-adaptation bit-rate selection method |
| CN110662093A (en) * | 2019-10-17 | 2020-01-07 | 北京达佳互联信息技术有限公司 | Initial code rate determination method and device, electronic equipment and storage medium |
| CN113473248A (en) * | 2020-03-30 | 2021-10-01 | 中国电信股份有限公司 | Streaming media transmission method, device and computer readable storage medium |
| CN112565906A (en) * | 2020-11-05 | 2021-03-26 | 深圳市麦谷科技有限公司 | On-demand video processing method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104320424B (en) | 2017-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104320424A (en) | Streaming media fragmentation downloading method and device | |
| CN103370709A (en) | A cache manager for segmented multimedia and corresponding method for cache management | |
| US9402058B2 (en) | Content delivery system, content delivery method and content delivery program | |
| US20190014166A1 (en) | Buffering in HTTP Streaming Client | |
| CN108833996A (en) | Service node selection, update and code rate adaptive approach in distributed DASH system | |
| CN103905820A (en) | Client side video quality self-adaption method and system based on SVC | |
| CN102088620A (en) | Method for downloading media file in content distribution network and client | |
| KR20130101585A (en) | Variable bit video streams for adaptive streaming | |
| CN104205771A (en) | Controlling HTTP streaming between source and receiver over multiple TCP connections | |
| CN103380610A (en) | Media switching unit and method | |
| CN104205770A (en) | Improved DASH client and receiver with request cancellation capabilities | |
| CN102137139A (en) | Method and device for selecting cache replacement strategy, proxy server and system | |
| CN103843358A (en) | Contents distribution system, cache server, and contents distribution method | |
| US20170048558A1 (en) | Controlling the Transmission of a Video Data Stream over a Network to a Network User Device | |
| US20200092202A1 (en) | High-quality adaptive bitrate video through multiple links | |
| CN105847979A (en) | Method and device of downloading and providing video files | |
| CN103828378A (en) | Content delivering system, cache server, and content delivering method | |
| CN104170315B (en) | For the method and video gateway of bandwidth allocation | |
| CN108124192B (en) | HLS streaming media playing method and system | |
| CN113099259B (en) | Multicast transmission method and device of streaming media, multicast server and medium | |
| CN107273310A (en) | A kind of read method of multi-medium data, device, medium and equipment | |
| CN109040018A (en) | Data matching method, device, display terminal and readable storage medium storing program for executing | |
| CN106937392B (en) | Wireless resource scheduling method based on user experience quality in DASH service | |
| CN108768886A (en) | A kind of SaaS data access increased quality method | |
| CN106134202A (en) | MMT assets and there is the distortion signaling of enhancing of ISOBMFF of MMT QOS descriptor of the improvement comprising multiple QOE operating point |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |