CN110191362A - Data transmission method and device, storage medium and electronic equipment - Google Patents
Data transmission method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN110191362A CN110191362A CN201910456765.9A CN201910456765A CN110191362A CN 110191362 A CN110191362 A CN 110191362A CN 201910456765 A CN201910456765 A CN 201910456765A CN 110191362 A CN110191362 A CN 110191362A
- Authority
- CN
- China
- Prior art keywords
- client
- destination client
- status information
- bit rate
- destination
- 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
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
- H04L65/80—Responding to QoS
-
- 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
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/238—Interfacing the downstream path of the transmission network, e.g. adapting the transmission rate of a video stream to network bandwidth; Processing of multiplex streams
-
- 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
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/24—Monitoring of processes or resources, e.g. monitoring of server load, available bandwidth, upstream requests
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The present invention provides a kind of data transmission methods, comprising: when receiving bit-rate allocation instruction, deterministic bit rate distribution instructs the status information of corresponding destination client；The communications status information of the communication link where destination client is obtained, and determines the status information current in remaining each client in communication link；The status information and communications status information that status information, each client of foundation destination client are current generate global state information；Status information and global state information according to destination client generate state vector；State vector is input in the neural network model constructed in advance, to determine bit rate tutorial message corresponding with the destination client；The bit rate tutorial message is sent to destination client.The bit rate tutorial message of destination client, the QoE demand for meeting different clients that can be optimized are determined by planning as a whole the state between destination client and each client, the communications status information of communication link.
Description
Technical field
The present invention relates to the communications field, in particular to a kind of data transmission method and device, storage medium and electronic equipment.
Background technique
With the development of Information technology, Internet content sources become increasingly abundant, wherein network video is increasingly becoming internet
The major way of content share, people can obtain miscellaneous video information by network, in video flow and number of users
In the ever-increasing situation of amount, guarantee that the user experience QoE (Qality of Experience) of video user becomes one
Item research hotspot.
QoE is most important for the design of transmission of video in internet, and main indexes are embodied in user and watch view
The factors such as the bit rate and video cardton event size of frequency；Existing a large amount of client-based bit rate adaptive algorithms
(Adaptive Bitrate Algorithm, ABR), but these methods carry out bit rate decision according only to user's oneself state,
If multi-user competes same bottleneck bandwidth, first it is larger may to measure network bandwidth resources by the user of addition network, and then asks
HD video is sought, so as to measure network bandwidth resources smaller by the user that network is added afterwards, the lower video of clarity can only be requested,
Cause the unfairness of user QoE.
Summary of the invention
Technical problem to be solved by the invention is to provide a kind of data transmission method, can according to global state information and
The status information of destination client generates bit rate tutorial message, makes destination client according to the bit rate tutorial message to data
Server request data, the QoE demand for meeting different clients that can be optimized effectively promote the fairness of QoE between user.
The present invention also provides a kind of data transmission devices, to guarantee the realization and application of the above method in practice.
A kind of data transmission method, comprising:
When receiving bit-rate allocation instruction, determine that the bit-rate allocation instructs the state of corresponding destination client
Information；
The communications status information of the communication link where the destination client is obtained, and determination is currently at the communication
The status information of each client in the first client set in link；The first client collection is combined into the communication chain
The current set of all clients in addition to the destination client in road；
Shape according to each client in the status information of the destination client, the first client set
State information and the communications status information generate global state information；
It is generated and the destination client pair according to the status information of the destination client and the global state information
The state vector answered；
The state vector is input in the neural network model constructed in advance, with the determining and destination client pair
The bit rate tutorial message answered；
The bit rate tutorial message is sent to destination client, refers to the destination client according to the bit rate
It leads information and sends data transfer request to data server.
Above-mentioned method optionally when receiving bit-rate allocation instruction, determines that the bit-rate allocation instruction corresponds to
Destination client status information, comprising:
Obtain the state recording for including in the bit-rate allocation instruction；
Determine the data transport priority and facility information of the destination client；
The destination client is generated according to the state recording, the data transport priority and the facility information
Status information.
Above-mentioned method, optionally, the determination are currently in the first client set in the communication link
The status information of each client, comprising:
The state storage table pre-established is traversed, to obtain each client in the first client set
State recording；
State recording according to each client in the first client set determines first client
The status information of each client in set.
Above-mentioned method is optionally generated according to the status information of the destination client and the global state information
State vector corresponding with the destination client, comprising:
The status information of the destination client and the global state information are forwarded to and the destination client pair
The progress of work answered, to obtain state vector corresponding with the destination client.
Above-mentioned method, it is optionally, described that the state vector is input in the neural network model constructed in advance, with
Determine bit rate tutorial message corresponding with the destination client, comprising:
Determine that the bit-rate allocation instructs the specification information of corresponding target data, the target data is the target
Client is currently to send data of the data transfer request to obtain to data server；
The output set of the neural network model is determined according to the specification information；
When the state vector is input to the neural network model, the neural network model is obtained in the output
The bit rate tutorial message corresponding with the state vector determined in set.
Above-mentioned method, it is optionally, described that the state vector is input in the neural network model constructed in advance, with
Determine bit rate tutorial message corresponding with the destination client, comprising:
The state vector is input in the neural network model, generate the state vector with it is pre-set defeated
The corresponding probability value of each initial bit rate tutorial message in set out；
Sequence based on each probability value from large to small is chosen in each probability value and meets preset condition
Destination probability value；
Obtain initial bit rate information corresponding with the destination probability value in the output set；
By the corresponding initial bit rate information of the destination probability value, it is determined as bit corresponding with the destination client
Rate tutorial message.
Above-mentioned method, optionally, it is described the bit rate tutorial message is sent to destination client after, also wrap
It includes:
Receive the downloading end signal corresponding with the bit rate tutorial message of the destination client feedback；
Reward value corresponding with the bit rate tutorial message is determined according to the downloading end signal；
The parameter of the neural network is adjusted based on the reward value.
A kind of data transmission device, comprising:
Receiving unit when for receiving bit-rate allocation instruction, determines that the bit-rate allocation instructs corresponding target
The status information of client；
Acquiring unit, for obtaining the communications status information of the communication link where the destination client, and determination is worked as
The status information of each client in preceding the first client set in the communication link；The first client collection
The currently set of all clients in addition to the destination client is combined into the communication link；
First generation unit, for according to the destination client status information, in the first client set
The status information of each client and the communications status information generate global state information；
Second generation unit, for according to the destination client status information and the global state information generate with
The corresponding state vector in the destination client；
Determination unit, for the state vector to be input in the neural network model constructed in advance, with determining and institute
State the corresponding bit rate tutorial message in destination client；
Allocation unit, for the bit rate tutorial message to be sent to destination client, make the destination client according to
Data transfer request is sent to data server according to the bit rate tutorial message.
A kind of storage medium, the storage medium include the instruction of storage, wherein in described instruction operation described in control
Equipment where storage medium executes above-mentioned data transmission method.
A kind of electronic equipment, including memory and one perhaps one of them or one of more than one instruction with
Upper instruction is stored in memory, and is configured to execute above-mentioned transmission side data by one or more than one processor
Method.
Via above scheme it is found that the present invention provides a kind of data transmission methods, comprising: when receiving bit-rate allocation
When instruction, determine that the bit-rate allocation instructs the status information of corresponding destination client；Obtain the destination client institute
Communication link communications status information, and determine each in the first client set for being currently in the communication link
The status information of a client；The first client collection is combined into the communication link currently in addition to the destination client
All clients set；According to each institute in the status information of the destination client, the first client set
The status information and the communications status information for stating client generate global state information；State according to the destination client
Information and the global state information generate state vector corresponding with the destination client；The state vector is input to
In the neural network model constructed in advance, to determine bit rate tutorial message corresponding with the destination client；By the ratio
Special rate tutorial message is sent to destination client, makes the destination client according to the bit rate tutorial message to data service
Device sends data transfer request.By plan as a whole state in destination client and the first client set between each client,
The communications status information of communication link determines the bit rate tutorial message of destination client, and what can be optimized meets different clients
The QoE demand at end effectively promotes the fairness of QoE between user.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without any creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is a kind of method flow diagram of data transmission method provided by the invention；
Fig. 2 is a kind of another method flow diagram of data transmission method provided by the invention；
Fig. 3 is an a kind of exemplary diagram of data transmission method provided by the invention；
Fig. 4 is a kind of another exemplary diagram of data transmission method provided by the invention；
Fig. 5 is a kind of structural schematic diagram of data transmission device provided by the invention；
Fig. 6 is the structural schematic diagram of a kind of electronic equipment provided by the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The present invention can be used in numerous general or special purpose computing device environment or configurations.Such as: personal computer, service
Device computer, handheld device or portable device, laptop device, multi-processor device including any of the above devices or devices
Distributed computing environment etc..
The embodiment of the invention provides a kind of data transmission method, this method can be applied in multiple systems platform, be held
Row main body can be the processor of intelligent Edge Server, which can be terminal or various movements
Equipment, the intelligence Edge Server may be in data communication system, which may include: data service
Each client in device, the intelligence Edge Server, destination client and the first client set；Data server passes through logical
Letter link is connected with each client in destination client and the first client set；The intelligence Edge Server is placed in logical
Believe in link, is connected respectively with each client in data server, destination client and the first client set；It is described
The method flow diagram of method is as shown in Figure 1, specifically include:
S101: when receiving bit-rate allocation instruction, determine that the bit-rate allocation instructs corresponding destination client
Status information.
In method provided in an embodiment of the present invention, when receiving bit-rate allocation instruction, determine that the bit-rate allocation instructs
Corresponding client, and the client is determined as destination client.
In method provided in an embodiment of the present invention, which has been previously obtained the processor of intelligent Edge Server
Assignment process.Optionally, which can be ABR process.
In method provided in an embodiment of the present invention, bit-rate allocation instruction can be destination client transmission, can also
To be that data server is sent.
In method provided in an embodiment of the present invention, the processor of intelligent Edge Server can repeatedly be received and the target
The corresponding bit-rate allocation instruction of client.
In method provided in an embodiment of the present invention, bit-rate allocation instruction includes destination client mesh currently to be requested
Mark the quality of the target data of the Data Identification of data, the data buffer storage size of current goal client and a upper request.
S102: the communications status information of the communication link where the destination client is obtained, and determination is currently at institute
State the status information of each client in the first client set in communication link.
It should be noted that the first client collection be combined into the communication link currently except the destination client with
The set of outer all clients；
In method provided in an embodiment of the present invention, which is each in destination client and the first client set
The shared communication link of a client, the processor of the intelligence Edge Server can acquire the transmission of data in the communication link
Rate, to obtain the communications status information of the communication link.The mode of acquisition transmission rate can be real-time acquisition, be also possible to
It is acquired by preset time interval.
S103: according to each client in the status information of the destination client, the first client set
The status information at end and the communications status information generate global state information.
In method provided in an embodiment of the present invention, global state information may include destination client and the first client collection
The data of the difference of the QoE value of each client in conjunction, destination client and each client in the first client set
Data transport priority difference in cache information, destination client and first object client set between each client,
The one of the above or a variety of such as the broadband utilization rate of average buffer value and communication link of destination client and each client.
S104: it is generated and the target customer according to the status information of the destination client and the global state information
Hold corresponding state vector.
Method provided in an embodiment of the present invention, different moments receive bit-rate allocation corresponding with destination client and refer to
Enable, the global state information at each moment and the status information of destination client are not necessarily identical, and then generate state to
It measures also not necessarily identical.
S105: the state vector is input in the neural network model constructed in advance, with determining objective with the target
The corresponding bit rate tutorial message in family end.
In method provided in an embodiment of the present invention, which is the neural network model through over-fitting.
In method provided in an embodiment of the present invention, the neural network model will it is corresponding with the state vector output as a result,
It is forwarded in the progress of work corresponding with destination client.
In method provided in an embodiment of the present invention, optionally, the output, currently to be requested according to the neural network model
The mark and target data address of target data generate bit rate tutorial message.The output of neural network model can be number of targets
According to data requirement, such as the bit rate of target data.
S106: the bit rate tutorial message is sent to destination client, makes the destination client according to the ratio
Special rate tutorial message sends data transfer request to data server.
In method provided in an embodiment of the present invention, which can characterize destination client and currently be assigned
Obtained video bitrate, i.e. video definition.
Method provided in an embodiment of the present invention, by planning as a whole in communication link between destination client and each client
State, the status information of communication link are destination client deterministic bit rate tutorial message, and the satisfaction that can not only be optimized is not
With the QoE demand of client, the fairness of user QoE is improved, moreover it is possible to effectively the competition between different clients be avoided to cause
The waste of the communication resource.
In method provided in an embodiment of the present invention, on the basis of above-mentioned implementation process, specifically, when receiving bit rate
When distribution instruction, determine that the bit-rate allocation instructs the process of the status information of corresponding destination client, as shown in Fig. 2,
May include:
S201: the state recording for including in the bit-rate allocation instruction is obtained.
In method provided in an embodiment of the present invention, the state recording for including in bit-rate allocation instruction may include target visitor
The data requirement of family end current data cache size, the data obtained.
Method provided in an embodiment of the present invention, if bit-rate allocation instruction is the corresponding first bit rate in destination client
Distribution instruction, the state recording of bit-rate allocation instruction at this time are sky, i.e. current data cache size and the data obtained
Data requirement be 0, wherein the instruction of first bit-rate allocation can be the bit-rate allocation instruction that data server is sent, first
A bit-rate allocation instruction may include the data description file of intention data, and the intention data is by target data and multiple data
Block composition, if the intention data is video data, which is media description file (Media
Presentation Description, MPD), timestamp, the system of each video block of video data are contained in MPD file
The information such as one Resource Locator (Uniform Resource Location, URL), video resolution and bit rate, each bit
Rate distribution instruction can correspond to the bit rate tutorial message for obtaining one or more video blocks.Preferably, each bit-rate allocation
Instruction can correspond to the bit rate tutorial message for obtaining a video block.Data in the instruction of first bit-rate allocation are described into text
Part is stored.
In method provided in an embodiment of the present invention, if bit-rate allocation instruction is the corresponding non-first ratio in destination client
Special rate distribution instruction, then bit-rate allocation instruction can send for destination client.
It should be noted that judging whether to be stored with and be somebody's turn to do after getting the state recording of bit-rate allocation instruction
The corresponding state recording in destination client, if so, then deleting destination client corresponding states record, and by the bit
State recording in rate distribution instruction is stored as the new state recording in destination client；If nothing, this instructs the bit-rate allocation
In state recording be stored as the state recording of the destination client.
S202: the data transport priority and facility information of the destination client are determined.
In method provided in an embodiment of the present invention, obtained in the storage region pre-established corresponding with the destination client
Data transport priority and facility information.
Optionally, the storing process of the corresponding data transport priority in the destination client and facility information can be with are as follows: when
When session request is established to data server transmission by communication link in destination client, the processor of intelligent Edge Server is logical
Cross the communication link and copy this and establish session request, and according to this establish session request be the destination client establish work into
Journey；The facility information established in session request is obtained, and whether judge that this is established in session request includes destination client
Data transport priority, if comprising storing the data transport priority；If not including, sends and requests to data server,
To obtain the data transport priority of the destination client；The data transport priority is stored.
Wherein, the IP for establishing the URL for the data that session request may include destination client request, the destination client
Address, MAC Address, facility information, operating system and wait it is above-mentioned one or more, facility information can be destination client screen
Curtain resolution ratio.
S203: the state recording, the data transport priority and the facility information are generated into the target customer
The status information at end.
In method provided in an embodiment of the present invention, optionally, the status information of the destination client may include target
Data requirement, current data cache size, current value of utility and facility information of data etc..
In method provided in an embodiment of the present invention, the calculating process of QoE value is as follows:
Include 3 indexs in QoE value: the quality of data of target data, the delay time of acquisition data and the quality of data are cut
It changes；Indicate the bit rate of the target data k of client i downloading,Bit rate is mapped to the quality of target data；Indicate the Caton time caused by downloading target data k, μ is the penalty coefficient to Caton；Table
Show the switching of the quality of data, whereinIt is the quality of the previous target data of user i downloading.According to the calculating
Journey can get the QoE value of destination client and each client.
In method provided in an embodiment of the present invention, in order to realize the differentiated service of different data transmission priority user,
Value of utility is proposed, which generated based on QoE value and data transport priority, specific calculating process are as follows:
viWhat is indicated is the data transport priority of client i；The Priority Service higher grade of client i, viValue get over
Greatly；It can get the value of utility of destination client and each client according to the calculating process.
Method provided in an embodiment of the present invention, in order to improve using various equipment obtain target data fairness, and
Realize the service of differentiation between client with different data transmission priority, ABR process can by destination client i and
The value of utility difference of each client in first client set is as input, the specific calculating process of the value of utility difference
Are as follows:
Indicate how much higher the value of utility of user i is than other users,Indicate that the value of utility of user i compares other users
It is how much low；N indicates to share the destination client of the communication link and the total quantity of each client.
In method provided in an embodiment of the present invention, on the basis of above-mentioned implementation process, specifically, the determination is worked as
The process of the status information of each client in preceding the first client set in the communication link, comprising:
The state storage table pre-established is traversed, to obtain each client in the first client set
State recording；
State recording according to each client in the first client set determines first client
The status information of each client in set.
In method provided in an embodiment of the present invention, each client that the progress of work has been established of intelligent edge server storage
State recording.
In method provided in an embodiment of the present invention, the current status information of each client in the first client set can
With the data requirement etc. of the size of data comprising client current cache data and the last data obtained.
In method provided in an embodiment of the present invention, on the basis of above-mentioned implementation process, specifically, according to the target visitor
The status information at family end and the global state information generate state vector corresponding with the destination client, comprising:
The status information of the destination client and the global state information are forwarded to and the destination client pair
The progress of work answered, to obtain state vector corresponding with the destination client.
In providing method of the embodiment of the present invention, the status information of destination client and global state information are forwarded to and mesh
The corresponding progress of work of client is marked, the state identified with the progress of work or with destination client mark can be obtained
Vector pre-establishes neural network model according to progress of work calling, and the state vector is input to the neural network mould
In type.
In method provided in an embodiment of the present invention, on the basis of above-mentioned implementation process, specifically, described by the state
Vector is input in the neural network model constructed in advance, to determine bit rate tutorial message corresponding with the destination client
Process, may include:
Determine that the bit-rate allocation instructs the specification information of corresponding target data, the target data is the target
Client is currently to send data of the data transfer request to obtain to data server；
The output set of the neural network model is determined according to the specification information；
When the state vector is input to the neural network model, the neural network model is obtained in the output
The bit rate tutorial message corresponding with the state vector determined in set.
Wherein, it is described obtain the neural network model in the output set determine it is corresponding with the state vector
Bit rate tutorial message, it will be appreciated that be to determine the classification type of the neural network model according to the output set, make nerve net
Network model chooses the maximum classification type of probability value in each classification type.
In method provided in an embodiment of the present invention, on the basis of above-mentioned implementation process, specifically, described by the state
Vector is input in the neural network model constructed in advance, to determine bit rate tutorial message corresponding with the destination client
Process, may include:
The state vector is input in the neural network model, initial ratio corresponding with the state vector is generated
Special rate tutorial message；
The initial bit rate tutorial message is matched with pre-set output set, obtains the bits of original
The matching degree of each bit rate tutorial message in rate tutorial message and the output set；
Bit rate guidance letter corresponding with the destination client is determined in the output set according to the matching degree
Breath.
In method provided in an embodiment of the present invention, which can be the corresponding data requirement set of target data,
If the target data be video data or music data, the output set may include 396bit/s, 1000bit/s,
2500bit/s and 4200bit/s etc. are specifically determined by MPD file, i.e., by the number of the target data stored in data server
It is determined according to specification.
It should be noted that the degree of closeness of matching degree characterization numerical value, the matching degree such as 1000bit/s and 396bit/s are big
In the matching degree of 1000bit/s and 2500bit/s；It is matched by being chosen in output set with the initial bit rate tutorial message
Spend maximum bit rate tutorial message.Optionally, can also refer to be determined in output set no more than the initial bit rate
The sub- output set of information is led, and determining and maximum ratio of initial bit rate tutorial message matching degree in the sub- output set
Special rate tutorial message.Specific executive mode can be adjusted by technical staff.
In method provided in an embodiment of the present invention, on the basis of above-mentioned implementation process, it is preferred that described by the state
Vector is input in the neural network model constructed in advance, to determine bit rate guidance letter corresponding with the destination client
Breath, comprising:
The state vector is input in the neural network model, generate the state vector with it is pre-set defeated
The corresponding probability value of each initial bit rate tutorial message in set out；
Sequence based on each probability value from large to small is chosen in each probability value and meets preset condition
Destination probability value；
Obtain initial bit rate information corresponding with the destination probability value in the output set；
By the corresponding initial bit rate information of the destination probability value, it is determined as bit corresponding with the destination client
Rate tutorial message.
In method provided in an embodiment of the present invention, the maximum probability value of numerical value is determined as destination probability value.
In method provided in an embodiment of the present invention, the classification function which uses can be softmax points
Class function.
In method provided in an embodiment of the present invention, for the neural network by carrying out tagsort to state vector, determining should
State vector probability value corresponding with each initial bit rate tutorial message in each pre-set output set；It is logical
Selection probability value maximum initial bit rate information is crossed as the corresponding bit rate tutorial message in destination client.
In method provided in an embodiment of the present invention, the training method of the neural network model can be to utilize Mininet etc.
Network simulator establishes multiple virtual machines in local server and simulates multiple client, intelligent Edge Server and video respectively
Server, user virtual machine will simulate the specific behavior of user in the access network using client behavioral data to the nerve net
Network model carries out pre-training.
In method provided in an embodiment of the present invention, on the basis of above-mentioned implementation process, specifically, described by the bit
Rate tutorial message is sent to after destination client, further includes:
Receive the downloading end signal corresponding with the bit rate tutorial message of the destination client feedback；
Reward value corresponding with the bit rate tutorial message is determined according to the downloading end signal；
The parameter of the neural network is adjusted based on the reward value.
In method provided in an embodiment of the present invention, which is that target data downloading is completed in destination client
When send.
In method provided in an embodiment of the present invention, the calculation formula of reward value is as follows:
WhereinIt is the duration for downloading target data k；T' is the time that user i sends request of data,It is t'
The data buffer storage size at moment.Reward value includes three indexs: value of utility, fairness and the punishment to the Caton time.The reward
The first item expression of value calculation formula is wished to provide the user with high usage value.The Section 2 and Section 3 of the reward value calculation formula
Indicate the fairness of the value of utility of each client.When all clients show fair behavior, the Section 2 of calculation formula and
Section 3 is equal to zero.Since when value of utility is lower than other people, client is more concerned with unjust phenomenon, therefore, usually there is η1<η2。
Last of reward value calculation formula is the punishment heavier to Caton event because user to the tolerance of Caton event compared with
Low, this can be used for accelerans network in the training of initial stage.
Method provided in an embodiment of the present invention can using in data transmission system, as shown in figure 3, the data transmission system
It may include: data server 301, intelligent Edge Server 302, destination client 303 and at least one client 304；Number
It is connected 303 with destination client 302 and each client by communication link according to server 301；The intelligence Edge Server
It is placed in 302 in communication link, is connected respectively with data server 301, destination client 303 and each client 304.
The intelligence Edge Server can be divided into request process layer, state measurement layer, Status register layer, joint QoE optimization
Layer；Wherein request process layer is used to acquire the data in communication link, such as request, data server transmission that client is sent
Data etc.；State measurement layer be used for according to request process layer acquisition data determine the current status information in destination client and
Global state information；Status information and with the client corresponding global state of the Status register layer for storage state client
Information；Joint QoE optimization layer is used to according to the status information of destination client and its corresponding global state information be target visitor
Family end generates bit rate tutorial message, and the bit rate tutorial message is sent to destination client.
Intelligent Edge Server can be exemplified below there are many deployment way: 1) be deployed in Internet service
The edge side of provider (Internet Service Provider, ISP) or outlet side (such as campus network or enterprise for accessing network
The outlet of industry net) directly deployment provide computing capability server；2) QoE optimization module will be combined as virtual functions
(Virtual Function) or application department are deployed in the edge data center of ISP；3) by crowd raise P2P in the way of, will
Joint QoE optimization module is deployed in the node for being ready open computing capability, and content supplier can be by certain incentive mechanism
The node of open computing capability provides reward.
Data transmission method provided in an embodiment of the present invention can be applied to a variety of field of data transmission, such as in transmission of video
In field, it can be carried out below using data server as video server based on the method that the embodiments of the present invention provide
For example:
As shown in figure 4, being Video transmission system exemplary diagram provided in an embodiment of the present invention, Video transmission system, comprising: view
Frequency server 401, intelligent Edge Server 402 and client set 403；
It include destination client and the first client set in the client set 403；
Video server 401 is provided with various video, and every kind of video is all cut into the equal multiple video blocks of duration, often
A video block is correspondingly arranged on corresponding serial number, and each video block is encoded as a variety of bit rates；Intelligent Edge Server 402
When receiving bit-rate allocation instruction, meeting instruct the status information of corresponding destination client, first according to the bit-rate allocation
The current communication information in the status information and communication link of each client in client set generates bit rate guidance letter
It ceases and the bit rate tutorial message is sent to destination client corresponding with bit-rate allocation instruction, make the destination client
The video block of corresponding bit rate is requested to video server 401 according to the bit rate tutorial message.The state of destination client
Information includes the resolution ratio of the bit rate of foradownloaded video block, current video caching, current value of utility and video equipment；It is global
Status information includes current communications link information, QoE difference value and video cache mean value.
When customer end A wants the first video of request, session request first is established to video server transmission, this establishes session
The information relevant to client such as URL, User IP, device type and operating system in request comprising first video.It will
It establishes session request and video server is sent to by communication link, in transmission process, this establishes session request by intelligence
It can be replicated by intelligent Edge Server when Edge Server；
The processor of intelligent Edge Server is got when establishing session request, and establishing session request according to this is client
A establishes ABR process, the i.e. progress of work；And the screen resolution information that session request obtains customer end A is established by this；Judgement
Whether this establishes comprising data transport priority in session request, if comprising obtaining the number from described establish in session request
According to transmission priority, if not including, the data transport priority of request customer end A is sent to data server；
Data server customer in response end A's establishes session request, sends the instruction comprising MPD file to customer end A, should
When bit-rate allocation instruction is by intelligent Edge Server, the backup that the processor of intelligent Edge Server obtains the instruction refers to
It enables, and the backup instruction is determined as bit-rate allocation instruction, which is determined as destination client for customer end A, and really
The status information for the client that sets the goal；The communications status information of the communication link where destination client is obtained, and is determined
The current status information of remaining each client in the communication link；State according to the destination client is believed
The current status information of breath, each client and the communications status information generate global state information；According to the mesh
The status information and the global state information for marking client generate state vector corresponding with the destination client；It will be described
State vector is input in the neural network model constructed in advance, to determine that bit rate corresponding with the destination client instructs
Information；The bit rate tutorial message is sent to destination client, instructs the destination client according to the bit rate
Information sends data transfer request to data server.Contained in the data transfer request video block to be requested serial number,
The cache size of current video, the video block quality of a upper request, wherein video block quality is the bit rate by video block
It determines.The processor copies the data transfer request for updating destination client current state information.
After data server receives request, video block corresponding with the request is returned to destination client；Target visitor
After the downloading of the video block is completed at family end, destination client judges whether current video cache size is greater than preset caching threshold
Value；
If more than preset cache threshold, then according to current video buffer status and the last video block having requested that
Video quality sends downloading end signal to intelligent Edge Server, makes the update processor target of the intelligent Edge Server
The status information of client；And judge whether current video cache size is less than preset re-request threshold value, it is default when being less than
Re-request threshold value when, according to the video quality of current video buffer status and the last video block having requested that intelligence
The distribution instruction of Edge Server Transmit Bit Rate；
If being less than preset cache threshold, according to current video buffer status and the last video block having requested that
Video quality sends downloading end signal to intelligent Edge Server, at this point, including bit-rate allocation in the downloading end signal
Instruction.
Optionally, cache threshold can be set to 30s, request threshold value to can be set to 15s again, when user's buffered foot
After the video of enough durations, existing caching has been enough to play the video that the user sees certain time, and user can be by bandwidth at this time
Resource leave for caching it is less, see that the user of Caton will occur for video." largest buffered threshold value " is set as 30s, i.e., user's is slow
Video block will no longer be requested after being greater than 30s by depositing." re-requesting cache threshold " is set as 15s, when the caching of user is less than or equal to
When 15s, user is to the distribution instruction of intelligent Edge Server Transmit Bit Rate to obtain new bit rate tutorial message.
The derivatization process of above-mentioned each concrete implementation mode and each implementation, all falls in the scope of protection of the present invention.
Corresponding with method described in Fig. 1, the embodiment of the invention also provides a kind of data transmission devices, for Fig. 1
The specific implementation of middle method, data transmission device provided in an embodiment of the present invention can be set with application computer terminal or various movements
In standby, structural schematic diagram is as shown in figure 5, specifically include:
Receiving unit 501 when for receiving bit-rate allocation instruction, determines that the bit-rate allocation instructs corresponding mesh
Mark the status information of client；
Acquiring unit 502 for obtaining the communications status information of the communication link where the destination client, and determines
It is currently at the status information of each client in the first client set in the communication link；First client
Collection is combined into the communication link the currently set of all clients in addition to the destination client；
First generation unit 503, for according to the destination client status information, in the first client set
Each client status information and the communications status information generate global state information；
Second generation unit 504, for the status information and global state information life according to the destination client
At state vector corresponding with the destination client；
Determination unit 505, for the state vector to be input in the neural network model constructed in advance, with determine with
The corresponding bit rate tutorial message in the destination client；
Allocation unit 506 makes the destination client for the bit rate tutorial message to be sent to destination client
Data transfer request is sent to data server according to the bit rate tutorial message.
In one embodiment of the invention, aforementioned schemes are based on, receiving unit 501 is configured that
Obtain the state recording for including in the bit-rate allocation instruction；
Determine the data transport priority and facility information of the destination client；
The destination client is generated according to the state recording, the data transport priority and the facility information
Status information.
In one embodiment of the invention, aforementioned schemes are based on, determine that remaining is each in the communication link
The current state information acquisition unit 502 of client is configured that
The determination is in the status information that remaining each client is current in the communication link, comprising:
The state storage table pre-established is traversed, to obtain each client in the first client set
State recording；
State recording according to each client in the first client set determines first client
The status information of each client in set.
In one embodiment of the invention, aforementioned schemes are based on, the second generation unit 504 is configured that
The status information of the destination client and the global state information are forwarded to and the destination client pair
The progress of work answered, to obtain state vector corresponding with the destination client.
In one embodiment of the invention, aforementioned schemes are based on, determination unit 505 is configured that
Determine that the bit-rate allocation instructs the specification information of corresponding target data, the target data is the target
Client is currently to send data of the data transfer request to obtain to data server；
The output set of the neural network model is determined according to the specification information；
When the state vector is input to the neural network model, the neural network model is obtained in the output
The bit rate tutorial message corresponding with the state vector determined in set.
In one embodiment of the invention, aforementioned schemes are based on, determination unit 505 is configured that
The state vector is input in the neural network model, generate the state vector with it is pre-set defeated
The corresponding probability value of each initial bit rate tutorial message in set out；
Sequence based on each probability value from large to small is chosen in each probability value and meets preset condition
Destination probability value；
Obtain initial bit rate information corresponding with the destination probability value in the output set；
By the corresponding initial bit rate information of the destination probability value, it is determined as bit corresponding with the destination client
Rate tutorial message.
In one embodiment of the invention, aforementioned schemes, the data transmission device are based on further include:
Adjustment unit, for receiving the downloading corresponding with the bit rate tutorial message of the destination client feedback
End signal；Reward value corresponding with the bit rate tutorial message is determined according to the downloading end signal；Based on the prize
Encourage the parameter that value adjusts the neural network.
Data transmission device provided in an embodiment of the present invention determines the bit rate when receiving bit-rate allocation instruction
Distribution instructs the status information of corresponding destination client；Obtain the communications status of the communication link where the destination client
Information, and determine the status information current in remaining each client in the communication link；According to the target customer
The current status information of the status information at end, each client and the communications status information generate global state information；
State corresponding with the destination client is generated according to the status information of the destination client and the global state information
Vector；The state vector is input in the neural network model constructed in advance, it is corresponding with the destination client with determination
Bit rate tutorial message；The bit rate tutorial message is sent to destination client, makes the destination client according to institute
It states bit rate tutorial message and sends data transfer request to data server.By plan as a whole destination client and each client it
Between state, the communications status information of communication link determine the bit rate tutorial message of destination client, what can be optimized expires
The QoE demand of sufficient different clients.
The embodiment of the invention also provides a kind of storage medium, the storage medium includes the instruction of storage, wherein in institute
It states the equipment where controlling the storage medium when instruction operation and executes above-mentioned data transmission method.
The embodiment of the invention also provides a kind of electronic equipment, structural schematic diagram is as shown in fig. 6, specifically include memory
601 and one perhaps more than one 602 one of them or more than one instruction of instruction 602 be stored in memory 601
In, and be configured to by one or more than one processor 603 execute the one or more instruction 602 carry out with
Lower operation:
When receiving bit-rate allocation instruction, determine that the bit-rate allocation instructs the state of corresponding destination client
Information；
The communications status information of the communication link where the destination client is obtained, and determination is currently at the communication
The status information of each client in the first client set in link；The first client collection is combined into the communication chain
The current set of all clients in addition to the destination client in road；
Shape according to each client in the status information of the destination client, the first client set
State information and the communications status information generate global state information；
It is generated and the destination client pair according to the status information of the destination client and the global state information
The state vector answered；
The state vector is input in the neural network model constructed in advance, with the determining and destination client pair
The bit rate tutorial message answered；
The bit rate tutorial message is sent to destination client, refers to the destination client according to the bit rate
It leads information and sends data transfer request to data server.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment weight
Point explanation is the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other.
For device class embodiment, since it is basically similar to the method embodiment, so being described relatively simple, related place ginseng
See the part explanation of embodiment of the method.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
For convenience of description, it is divided into various units when description apparatus above with function to describe respectively.Certainly, implementing this
The function of each unit can be realized in the same or multiple software and or hardware when invention.
As seen through the above description of the embodiments, those skilled in the art can be understood that the present invention can
It realizes by means of software and necessary general hardware platform.Based on this understanding, technical solution of the present invention essence
On in other words the part that contributes to existing technology can be embodied in the form of software products, the computer software product
It can store in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that a computer equipment
(can be personal computer, server or the network equipment etc.) executes the certain of each embodiment or embodiment of the invention
Method described in part.
A kind of data transmission method provided by the present invention and device are described in detail above, it is used herein
A specific example illustrates the principle and implementation of the invention, and the above embodiments are only used to help understand originally
The method and its core concept of invention；At the same time, for those skilled in the art, according to the thought of the present invention, specific
There will be changes in embodiment and application range, in conclusion the content of the present specification should not be construed as to of the invention
Limitation.
Claims (10)
1. a kind of method of data transmission characterized by comprising
When receiving bit-rate allocation instruction, determine that the bit-rate allocation instructs the state of corresponding destination client to believe
Breath；
The communications status information of the communication link where the destination client is obtained, and determination is currently at the communication link
In the first client set in each client status information；The first client collection is combined into the communication link
The currently set of all clients in addition to the destination client；
State letter according to each client in the status information of the destination client, the first client set
Breath and the communications status information generate global state information；
It is generated according to the status information of the destination client and the global state information corresponding with the destination client
State vector；
The state vector is input in the neural network model constructed in advance, it is corresponding with the destination client with determination
Bit rate tutorial message；
The bit rate tutorial message is sent to destination client, the destination client is made to instruct to believe according to the bit rate
It ceases to data server and sends data transfer request.
2. the method according to claim 1, wherein determining the ratio when receiving bit-rate allocation instruction
Special rate distribution instructs the status information of corresponding destination client, comprising:
Obtain the state recording for including in the bit-rate allocation instruction；
Determine the data transport priority and facility information of the destination client；
The state of the destination client is generated according to the state recording, the data transport priority and the facility information
Information.
3. the method according to claim 1, wherein the determination is currently at first in the communication link
The status information of each client in client set, comprising:
The state storage table pre-established is traversed, to obtain the state of each client in the first client set
Record；
State recording according to each client in the first client set determines the first client set
In each client status information.
4. the method according to claim 1, wherein according to the status information of the destination client and described complete
Office's status information generates state vector corresponding with the destination client, comprising:
The status information of the destination client and the global state information are forwarded to corresponding with the destination client
The progress of work, to obtain state vector corresponding with the destination client.
5. the method according to claim 1, wherein described be input to the state vector mind constructed in advance
Through in network model, to determine bit rate tutorial message corresponding with the destination client, comprising:
Determine that the bit-rate allocation instructs the specification information of corresponding target data, the target data is the target customer
End is currently to send data of the data transfer request to obtain to data server；
The output set of the neural network model is determined according to the specification information；
When the state vector is input to the neural network model, the neural network model is obtained in the output set
The bit rate tutorial message corresponding with the state vector of middle determination.
6. the method according to claim 1, wherein described be input to the state vector mind constructed in advance
Through in network model, to determine bit rate tutorial message corresponding with the destination client, comprising:
The state vector is input in the neural network model, the state vector is generated and pre-set output collects
The corresponding probability value of each initial bit rate tutorial message in conjunction；
Sequence based on each probability value from large to small chooses the target for meeting preset condition in each probability value
Probability value；
Obtain initial bit rate information corresponding with the destination probability value in the output set；
By the corresponding initial bit rate information of the destination probability value, it is determined as bit rate corresponding with the destination client and refers to
Lead information.
7. the method according to claim 1, wherein described be sent to target visitor for the bit rate tutorial message
After the end of family, further includes:
Receive the downloading end signal corresponding with the bit rate tutorial message of the destination client feedback；
Reward value corresponding with the bit rate tutorial message is determined according to the downloading end signal；
The parameter of the neural network is adjusted based on the reward value.
8. a kind of data transmission device characterized by comprising
Receiving unit when for receiving bit-rate allocation instruction, determines that the bit-rate allocation instructs corresponding target customer
The status information at end；
Acquiring unit for obtaining the communications status information of the communication link where the destination client, and determines current place
The status information of each client in the first client set in the communication link；The first client collection is combined into
The current set of all clients in addition to the destination client in the communication link；
First generation unit, for the status information, each in the first client set according to the destination client
The status information of the client and the communications status information generate global state information；
Second generation unit, for according to the destination client status information and the global state information generate with it is described
The corresponding state vector in destination client；
Determination unit, for the state vector to be input in the neural network model constructed in advance, with the determining and mesh
Mark the corresponding bit rate tutorial message of client；
Allocation unit makes the destination client according to institute for the bit rate tutorial message to be sent to destination client
It states bit rate tutorial message and sends data transfer request to data server.
9. a kind of storage medium, which is characterized in that the storage medium includes the instruction of storage, wherein run in described instruction
When control the equipment where the storage medium and execute data transmission method as described in claim 1~7 any one.
10. a kind of electronic equipment, which is characterized in that including memory and one or more than one instruction, one of them
Perhaps more than one instruction is stored in memory and is configured to be executed by one or more than one processor as right is wanted
Seek data transmission method described in 1~7 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910456765.9A CN110191362B (en) | 2019-05-29 | 2019-05-29 | Data transmission method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910456765.9A CN110191362B (en) | 2019-05-29 | 2019-05-29 | Data transmission method and device, storage medium and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110191362A true CN110191362A (en) | 2019-08-30 |
| CN110191362B CN110191362B (en) | 2021-03-16 |
Family
ID=67718535
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910456765.9A Active CN110191362B (en) | 2019-05-29 | 2019-05-29 | Data transmission method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110191362B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110784760A (en) * | 2019-09-16 | 2020-02-11 | 清华大学 | Video playing method, video player and computer storage medium |
| CN111478965A (en) * | 2020-04-07 | 2020-07-31 | 四川虹美智能科技有限公司 | Method, device and system for processing device shadow |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7400588B2 (en) * | 2003-08-01 | 2008-07-15 | Thomson Licensing | Dynamic rate adaptation using neural networks for transmitting video data |
| CN104581958A (en) * | 2014-12-31 | 2015-04-29 | 重庆邮电大学 | Resource allocation method and device based on rate self-adaption norm |
| CN107360473A (en) * | 2017-07-20 | 2017-11-17 | 中国传媒大学 | A kind of DASH systems of the flow scheduling of the congestion aware based on SDN |
| CN107370676A (en) * | 2017-08-03 | 2017-11-21 | 中山大学 | Fusion QoS and load balancing demand a kind of route selection method |
| CN107547457A (en) * | 2017-09-15 | 2018-01-05 | 重庆大学 | A kind of approach for blind channel equalization based on Modified particle swarm optimization BP neural network |
| US20180039856A1 (en) * | 2016-08-04 | 2018-02-08 | Takayuki Hara | Image analyzing apparatus, image analyzing method, and recording medium |
| CN108965949A (en) * | 2018-07-27 | 2018-12-07 | 清华大学 | Meet the code rate adaptive approach of user individual experience in a kind of video traffic |
| CN108964815A (en) * | 2018-07-30 | 2018-12-07 | 太原理工大学 | A kind of channel selection and bit rate adaptive approach based on BP neural network |
| US20190007690A1 (en) * | 2017-06-30 | 2019-01-03 | Intel Corporation | Encoding video frames using generated region of interest maps |
| CN109218744A (en) * | 2018-10-17 | 2019-01-15 | 华中科技大学 | A kind of adaptive UAV Video of bit rate based on DRL spreads transmission method |
| US20190050710A1 (en) * | 2017-08-14 | 2019-02-14 | Midea Group Co., Ltd. | Adaptive bit-width reduction for neural networks |
| CN109768879A (en) * | 2018-12-14 | 2019-05-17 | 北京爱奇艺科技有限公司 | The determination method, apparatus and server of target service server |
-
2019
- 2019-05-29 CN CN201910456765.9A patent/CN110191362B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7400588B2 (en) * | 2003-08-01 | 2008-07-15 | Thomson Licensing | Dynamic rate adaptation using neural networks for transmitting video data |
| CN104581958A (en) * | 2014-12-31 | 2015-04-29 | 重庆邮电大学 | Resource allocation method and device based on rate self-adaption norm |
| US20180039856A1 (en) * | 2016-08-04 | 2018-02-08 | Takayuki Hara | Image analyzing apparatus, image analyzing method, and recording medium |
| US20190007690A1 (en) * | 2017-06-30 | 2019-01-03 | Intel Corporation | Encoding video frames using generated region of interest maps |
| CN107360473A (en) * | 2017-07-20 | 2017-11-17 | 中国传媒大学 | A kind of DASH systems of the flow scheduling of the congestion aware based on SDN |
| CN107370676A (en) * | 2017-08-03 | 2017-11-21 | 中山大学 | Fusion QoS and load balancing demand a kind of route selection method |
| US20190050710A1 (en) * | 2017-08-14 | 2019-02-14 | Midea Group Co., Ltd. | Adaptive bit-width reduction for neural networks |
| CN107547457A (en) * | 2017-09-15 | 2018-01-05 | 重庆大学 | A kind of approach for blind channel equalization based on Modified particle swarm optimization BP neural network |
| CN108965949A (en) * | 2018-07-27 | 2018-12-07 | 清华大学 | Meet the code rate adaptive approach of user individual experience in a kind of video traffic |
| CN108964815A (en) * | 2018-07-30 | 2018-12-07 | 太原理工大学 | A kind of channel selection and bit rate adaptive approach based on BP neural network |
| CN109218744A (en) * | 2018-10-17 | 2019-01-15 | 华中科技大学 | A kind of adaptive UAV Video of bit rate based on DRL spreads transmission method |
| CN109768879A (en) * | 2018-12-14 | 2019-05-17 | 北京爱奇艺科技有限公司 | The determination method, apparatus and server of target service server |
Non-Patent Citations (3)
| Title |
|---|
| ABDELHAK BENTALEB等: ""SDNDASH: Improving QoE of HTTP Adaptive Streaming Using Software Defined Networking"", 《ACM》 * |
| HONGZI MAO等: ""Neural Adaptive Video Streaming with Pensieve"", 《ACM》 * |
| 刘晓颖: ""VBR MPEG视频源流量的智能集成预测模型"", 《计算机工程与应用》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110784760A (en) * | 2019-09-16 | 2020-02-11 | 清华大学 | Video playing method, video player and computer storage medium |
| CN111478965A (en) * | 2020-04-07 | 2020-07-31 | 四川虹美智能科技有限公司 | Method, device and system for processing device shadow |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110191362B (en) | 2021-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104104973B (en) | A kind of group's Bandwidth Management optimization method for being applied to cloud media system | |
| US8024740B2 (en) | Acquisition system for distributed computing resources | |
| US7451226B1 (en) | Method for grouping content requests by behaviors that relate to an information system's ability to deliver specific service quality objectives | |
| US7925738B2 (en) | Analytical cache performance model for a media server | |
| CN106330997B (en) | A kind of method and system of the content distribution for mobile terminal application | |
| CN106796547A (en) | For the method and system that proxy caching smart object is eliminated | |
| CN108494868A (en) | A kind of load-balancing method under the operation system based on cloud and system | |
| CN109819057A (en) | A kind of load-balancing method and system | |
| Gao et al. | Combining QoS-based service selection with performance prediction | |
| Ma et al. | QAVA: QoE-aware adaptive video bitrate aggregation for HTTP live streaming based on smart edge computing | |
| CN110191362A (en) | Data transmission method and device, storage medium and electronic equipment | |
| CN111935025B (en) | Control method, device, equipment and medium for TCP transmission performance | |
| CN106375471A (en) | Edge node determination method and device | |
| Azzedin et al. | Modeling BitTorrent choking algorithm using game theory | |
| Ma et al. | Learning-based joint QoE optimization for adaptive video streaming based on smart edge | |
| CN107295358A (en) | A kind of 3D Streaming Media storage methods under cloud environment | |
| Shi et al. | CoLEAP: Cooperative learning-based edge scheme with caching and prefetching for DASH video delivery | |
| Bhattacharyya et al. | QFlow: A learning approach to high QoE video streaming at the wireless edge | |
| Aioffi et al. | Dynamic content distribution for mobile enterprise networks | |
| Zhu et al. | Load balancing algorithm for web server based on weighted minimal connections | |
| On | Quality of availability for widely distributed and replicated content stores | |
| Lu | Optimization Simulation of Balanced Distribution of Multimedia Network Modular Teaching Resources | |
| Aioffi et al. | Mobile dynamic content distribution networks | |
| KR102561320B1 (en) | A container replica recommendation system through resource trend prediction and a recommendation method | |
| CN114885215B (en) | Training method of code rate self-adaptive model, video code rate self-adaptive method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |