CN114786137A - Cache-enabled multi-quality video distribution method - Google Patents

Cache-enabled multi-quality video distribution method Download PDF

Info

Publication number
CN114786137A
CN114786137A CN202210424517.8A CN202210424517A CN114786137A CN 114786137 A CN114786137 A CN 114786137A CN 202210424517 A CN202210424517 A CN 202210424517A CN 114786137 A CN114786137 A CN 114786137A
Authority
CN
China
Prior art keywords
user
video
quality
network
users
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
Application number
CN202210424517.8A
Other languages
Chinese (zh)
Other versions
CN114786137B (en
Inventor
吴大鹏
徐瑞鑫
张鸿
李职杜
王汝言
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN202210424517.8A priority Critical patent/CN114786137B/en
Publication of CN114786137A publication Critical patent/CN114786137A/en
Application granted granted Critical
Publication of CN114786137B publication Critical patent/CN114786137B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/06—Selective distribution of broadcast services, e.g. multimedia broadcast multicast service [MBMS]; Services to user groups; One-way selective calling services
    • 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
    • H04N21/231—Content storage operation, e.g. caching movies for short term storage, replicating data over plural servers, prioritizing data for deletion
    • H04N21/23103—Content storage operation, e.g. caching movies for short term storage, replicating data over plural servers, prioritizing data for deletion using load balancing strategies, e.g. by placing or distributing content on different disks, different memories or different servers
    • 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
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/10—Flow control between communication endpoints
    • H04W28/14—Flow control between communication endpoints using intermediate storage
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
    • H04W28/24—Negotiating SLA [Service Level Agreement]; Negotiating QoS [Quality of Service]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Multimedia (AREA)
  • Quality & Reliability (AREA)
  • Databases & Information Systems (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention relates to a cache-enabled multi-quality video distribution method, and belongs to the technical field of wireless communication. The method comprises the steps of formulating user QoE as an optimization target according to video quality requested by a user, content caching positions and channel conditions of the user, and optimizing multicast cooperative beam forming and video quality selection by utilizing an SCA framework; furthermore, under the condition of considering insufficient network resources, users which have great influence on user groups in the network are preferentially removed, and after the existing users can be met, the energy efficiency of the network is maximized, so that the user service under the condition of resource shortage is ensured as far as possible. Compared with the traditional strategy, the invention can effectively improve the user experience and more flexibly manage the wireless resources.

Description

Cache-enabled multi-quality video distribution method
Technical Field
The invention belongs to the technical field of wireless communication, and relates to a cache-enabled multi-quality video distribution method.
Background
Recently, mass video contents generated by services such as short video, high-definition movie, online live broadcast and the like gradually become a main form of 5G mobile traffic. Most of the mobile traffic data is generated by repeated transmission of hot video and tends to obey twenty-eight law. Therefore, a Fog Access Network (F-RAN) incorporating an edge caching technique is proposed to address the enormous traffic pressure in the Network and to optimize the user experience. When a user requests a video, the user can directly acquire content from a local F-AP (fog wireless access point), which provides potential for reducing receiving delay and the like while reducing the burden of a forward link and improving service efficiency. In addition, through a cooperative wireless multicast technology among the F-APs, the network capacity and the user experience of the system can be further improved.
With the improvement of hardware capability of mobile devices, the requirements of users on video definition gradually develop from Standard Definition (SD) and High Definition (HD) to Ultra High Definition (UHD). With Scalable Video Coding (SVC) technology, Video data can be coded into different quality levels and flexibly combined to meet personalized Video requests of different users. However, heterogeneous video quality presents greater challenges to video transmission in different scenes. On the one hand, when there are fewer users in the network, the operator may wish to provide the best experience to the users from a user-centric perspective. Selecting which users and which way to enhance the network experience becomes a key issue. On the other hand, in a user-intensive environment, how to provide services to more users with the highest energy efficiency becomes a pain point in the scenario due to the limitation of communication resources and the power resources of the F-AP itself. In order to achieve the above goal, under a cooperative multicast framework, limited system resources need to be allocated more reasonably according to actual needs.
In a large number of existing researches, a beamforming design and multi-base station cooperative beamforming are performed for a single base station, and limited wireless resources are managed to some extent, but most of the researches do not consider different video qualities, and do not consider influences brought to a user by cache contents acquired from different positions, and only consider contents to be transmitted to the user. On the other hand, most of the scenarios do not consider the requests of the dense users, and often only serve a small number of users, which brings great pressure to the beamforming design in the resource-limited scenario. In addition, most of the studied optimization indexes only aim at the throughput of the network, and the index which is more practical for the user is the experience quality of the user playing the video. How to efficiently transfer the cache content to the device remains an open question.
Therefore, in combination with the multi-F-AP multicast beamforming coordination scenario, the management of limited communication resources by personalized video quality and user requirements is still urgently needed to be further researched.
Disclosure of Invention
In view of the above, the present invention provides a cache-enabled multi-quality video distribution method for solving the problems of uncertain cache content hit, diversified video quality requested by a user, limited system communication resources, etc. in a fog radio access network, according to multi-quality video requests of different users, a multicast cooperative transmission framework in an F-RAN is introduced, a problem model is transformed by using an SCA technique, the limitation of system power is fully considered, and reasonable scheduling is performed for the user, so as to adaptively and rapidly change video transmission and dense user access scenarios, and improve the QoE of the whole network as much as possible.
In order to achieve the purpose, the invention provides the following technical scheme:
a cache-enabled multi-quality video distribution method comprises the steps of formulating user QoE as an optimization target according to video quality requested by a user, content cache position and channel condition of the user, and optimizing multicast cooperative beam forming and video quality selection by utilizing an SCA (sequential Convex Approximation) frame; furthermore, under the condition of considering network resource shortage, users which have great influence on user groups in the network are preferentially removed, and after the existing users can be met, the energy efficiency of the network is maximized, so that the user service under the condition of resource shortage is ensured as much as possible; the self-adaptive mode provides the best experience for users in the dense fog wireless access network.
The method specifically comprises the following steps:
s1: initializing relevant information: collecting a user request video and a user channel condition, dividing multicast groups according to the user request, and constructing a cache scheduling cost model;
s2: planning a system model: establishing a multi-quality video cooperative multicast communication model and a user QoE model, and constructing a problem model aiming at different scenes in a process;
s3: determining network resource availability: using the rate satisfaction xi to measure whether the existing network resource can provide service for all users, if so, performing step S4, otherwise, entering step S5;
s4: network QoE maximization: firstly, aiming at the existing user request, improving the QoE of the network by improving the low-quality video to the high-quality video, and optimizing a cooperative beam forming scheme in the process; performing loop iteration until optimal allocation is achieved, and finally deploying a beam forming scheme and a video quality selection scheme to each F-AP (fog wireless access point);
s5: user scheduling selection: using the user service cost function v to measure the cost for providing the service to the user, and continuously eliminating the users with higher cost until the requirements of all the existing users can be met, then entering step S6;
s6: network energy efficiency maximization: when the existing communication resources are enough to meet the requirements of users, the ratio of the network QoE to the energy consumption is taken as the energy efficiency under the scene, the beam forming scheme is redesigned so as to ensure that the communication resources are saved as much as possible while the requirements of the users are met as much as possible, and finally the beam forming scheme is deployed to each F-AP.
Further, step S1 specifically includes the following steps:
s11: initializing channel information of user u and F-AP k
Figure BDA0003607947510000031
The video quality parameter specifically requested by user u is denoted as lu＝{1,2}，l u1 indicates that the user requests low quality video,/u2 denotes that the user requests high quality video; user u requests a multi-quality video representation as
Figure BDA0003607947510000032
The multicast group is further divided into
Figure BDA0003607947510000033
The requested content and quality versions of different multicast groups g are denoted g, respectivelyfAnd gl(ii) a Assuming that each F-AP has its own inherent power limit Pk(> 0), the set of beamforming vectors can be expressed as:
Figure BDA0003607947510000034
wherein ,
Figure BDA0003607947510000035
representing the set of beamforming vectors, w, at F-AP kk,gRepresents the beamforming vector assigned to multicast group g,
Figure BDA0003607947510000036
representing a beamforming vector dimension of Ak×1，
Figure BDA0003607947510000037
Representing a F-AP set;
s12: the cache content generates different delay costs according to different positions of the cache content in the retrieval process of the F-AP. Considering a limited video content library
Figure BDA0003607947510000038
And assuming that the content server of the BBU pool stores data of all video contents, and the F-AP k is internally provided withThe buffer status of the capacity f is expressed as:
Figure BDA0003607947510000039
furthermore, the extra search delay cost D generated by different cache positionsf(which is determined by the maximum of all F-APs) is expressed as:
Figure BDA00036079475100000310
wherein ,ck,fThe cache state of the content F in the F-AP k is expressed; if the video request is responded to on all the local service F-APs, i.e. ck,fWhen the search time delay is 0, the additional search time delay is 0; if only a part of the F-APs respond to the video request, additional retrieval delay cost d caused by video resource sharing among the F-APs is broughtf(ii) a When all F-AP nodes do not have cache content F, the video needs to be forwarded to each F-AP from the BBU pool in advance; here, further considering the constant routing delay and forwarding delay of the BBU pool to obtain the content from the content server, d is used0Is shown, and d0＞＞df。
Further, step S2 specifically includes the following steps:
s21: establishing a multi-quality video cooperation multicast transmission communication model, and expressing the transmission rate which can be achieved by the user u as follows by using a Shannon formula according to the information in S11:
Figure BDA00036079475100000311
wherein, B represents a bandwidth; signal to interference plus noise ratio
Figure BDA0003607947510000041
Expressed as:
Figure BDA0003607947510000042
wherein ,
Figure BDA0003607947510000043
representing the power of a gaussian white noise interferer,
Figure BDA0003607947510000044
indicates the set of multicast groups, h, to which user u is assigneduRepresenting the channel gain vectors for user u to all F-APs,
Figure BDA0003607947510000045
s22: in view of fairness within the multicast group, the actual communication rates of the users in multicast group g are all consistent with the group communication rate, which is expressed as:
Figure BDA0003607947510000046
wherein ,
Figure BDA0003607947510000047
indicating reception of content f for multicast group glThe intra-group rate of (a) is,
Figure BDA0003607947510000048
indicating that user u in the group receives content flThe intra-group rate of (a);
s23: establishing a user QoE model, wherein two parts, namely the quality level of a video acquired by a user and the playing delay of the video acquired by the user, are mainly considered, and the two parts are specifically represented as follows:
Figure BDA0003607947510000049
wherein ,QoEuThe practical QoE value of a user u is shown, beta shows the change trend of marginal benefit for controlling the playing delay, eta shows a delay satisfaction factor, the closer to 1, the more obvious the influence of the delay on the QoE of the user is, and lELIndicating high qualitySelection parameters of video, Du,maxIndicating the maximum transmission delay of the user currently receiving all video layer data, i.e.
Figure BDA00036079475100000410
Hereinafter, Q is referred to as a network QoE and represents a weighted sum of qoes of all users in the network.
S24: aiming at different communication resource conditions, network QoE maximization, network energy efficiency maximization and user scheduling problems are formulated as follows;
Figure BDA00036079475100000411
Figure BDA00036079475100000412
Figure BDA00036079475100000413
Figure BDA00036079475100000414
Figure BDA00036079475100000415
wherein Q represents the weighted sum of all QoE users in the network; w and l respectively represent a beam forming vector group and a video quality selection parameter obtained by a user; c1Is a fairness constraint within the multicast group; c2A video quality bit rate constraint indicating multicast group rate correspondence,
Figure BDA0003607947510000051
is the play bit rate threshold corresponding to the video quality; c3Representing the power constraints of the F-AP node,
Figure BDA0003607947510000052
represents the square of the 2 norm; c4Indicating that the video quality obtained by each user should not be lower than the originally requested video quality,
Figure BDA0003607947510000053
is the video quality that the user originally requested,
Figure BDA0003607947510000054
representing a set of users;
and further constructing a network energy efficiency maximization problem model as follows:
Figure BDA0003607947510000055
s.t.C1-C3
and finally, constructing a user scheduling model:
Figure BDA0003607947510000056
Figure BDA0003607947510000057
Figure BDA0003607947510000058
wherein z ═ { z ═ z0,...,zu,...zUThe service cost parameter set is used for measuring the cost of the service user, and is specifically represented as:
Figure BDA0003607947510000059
wherein ,Ru,lIndicating the transmission rate achievable by user u requesting video of quality l.
Further, step S3 specifically includes the following steps:
s31: due to problems
Figure BDA00036079475100000510
C in (1)1The quadratic term of middle w appears at the same time
Figure BDA00036079475100000511
The resulting non-convex nature makes the problem difficult to solve. Furthermore, auxiliary variables are introduced
Figure BDA00036079475100000512
And gamma, will be problematic
Figure BDA00036079475100000513
The constraint of (2) is modified to:
Figure BDA00036079475100000514
Figure BDA00036079475100000515
Figure BDA00036079475100000516
s32: constraint C after replacement5Still non-convex, using SCA technique, mix C5The taylor first order expansion on its right side is used instead as follows:
Figure BDA00036079475100000517
i.e. C5Instead of using
Figure BDA00036079475100000518
wherein ,
Figure BDA00036079475100000519
is an approximate substitution function, t represents the number of iterations; initial set of beamforming vectors given compliance constraintsThe local optimal solution can be obtained in an iterative mode;
s33: to give an initial w(0)An optimal rate satisfaction is obtained
Figure BDA0003607947510000061
The problem model of (2) is:
Figure BDA0003607947510000062
Figure BDA0003607947510000063
Figure BDA0003607947510000064
an optimal xi can be obtained by utilizing a CVX solver in an iterative mode;
s34: if ξ < 1 indicates that the existing resources are not enough to meet the requirements of all users, the step S5 is skipped, otherwise, the optimized improvement of the QoE of the network is indicated, and the step S4 is skipped.
Further, step S4 specifically includes the following steps:
s41: due to the coupling of the beamforming design and the video quality selection parameter, firstly, the video quality of a user is enhanced, and a video quality enhancement parameter v is designeduModifying the constraint C1Is composed of
Figure BDA0003607947510000065
Will have the original problem
Figure BDA0003607947510000066
Optimization goal replacement
Figure BDA0003607947510000067
Finally, calculating by using CVX to obtain an enhanced cost set v;
s42: selecting low-cost user enhancement according to the result in v; after the quality of the corresponding video is modified, calculating the optimal beam forming by taking the network QoE as an optimization target again, and iterating until the network QoE is reduced or the resources are insufficient to provide the existing service;
s43: a beamforming scheme is deployed to each F-AP.
Further, step S5 specifically includes the following steps:
s51: first, the original objective function z is transformed due to the user scheduling problemuThrough a0The range after regularization is within 1, and l1The norm calculation will make the variable keep the original scale unchanged. Thus, by applying a function at the objective function zuThe weight value corresponding to the extra structure
Figure BDA0003607947510000068
Differences in the variable scale of the new problem and the original problem can be balanced. Will question
Figure BDA0003607947510000069
Is expressed as
Figure BDA00036079475100000610
wherein
Figure BDA00036079475100000611
τ > 0 is a regularization parameter, which guarantees zuWhen the weight value is 0, the weight value of the corresponding update is not positive infinity;
s52: the iterative computation is carried out on the user service cost set z to obtain z*Performing descending arrangement;
s53: using dichotomy, for z*The user schedule in (1) is selected, whether the rate satisfaction ξ is greater than 1 can be made as a jump-out condition until the end is reached when the most users can be served, and the process goes to step S6.
Further, step S6 specifically includes the following steps:
s61: due to the fractal form in the energy efficiency optimization target function, the Dinkelbach algorithm is adopted to convert the non-convex target function, and a new auxiliary variable lambda is introduced as follows:
Figure BDA0003607947510000071
wherein ,
Figure BDA0003607947510000072
the objective function is a subtraction of the numerator and denominator form as follows:
F(λ)＝Q(w)-λPsum(w)
wherein F (λ) represents the transformed objective function;
s62: further, with the SCA framework, the update can be iterated through the CVX solver until the update is completed
Figure BDA0003607947510000073
Convergence below a minimum threshold;
s63: and deploying the beamforming scheme to each F-AP.
The invention has the beneficial effects that: the invention can provide QoE as good as possible for users by utilizing limited resources by comprehensively considering the multi-quality user requests under the cooperative multicast scene in the F-RAN, and can provide efficient service for more users as far as possible under the condition of insufficient resources. Compared with the traditional strategy, the invention can effectively improve the user experience and manage the wireless resources more flexibly. The method has better effect in the dense network taking the user as the center, and solves the problem that the personalized video requirement of the user is difficult to meet.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a network architecture diagram of a cache-enabled multi-quality video distribution method of the present invention;
FIG. 2 is a diagram of a multi-quality cooperative multicast transmission model of the present invention;
fig. 3 is a flow chart of a multi-quality video distribution method of the present invention.
Detailed Description
The following embodiments of the present invention are provided by way of specific examples, and other advantages and effects of the present invention will be readily apparent to those skilled in the art from the disclosure herein. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Referring to fig. 1 to 3, the present invention provides a cache-enabled multi-quality video distribution method, and for a significant improvement of a video service in a mobile traffic in a fog radio access network scenario, pre-caching content in a base station becomes an effective solution, and reasonable distribution of content also affects user experience but is often ignored. Under a multicast cooperative transmission framework of a fog radio access network (F-RAN), aiming at the difference between video content requested by a user and video quality, two types of problems which take green energy conservation and user experience as optimization targets are respectively designed by combining a cache cost model. Firstly, aiming at maximizing system benefit, the user with better channel condition can improve the overall user quality of experience (QoE) by improving the user video quality; and then aiming at the energy efficiency of a large network, the admission problem of a user needs to be additionally considered when the power is limited, and due to the difficulty of directly solving the problem, an efficient cooperative beam forming scheme is designed by utilizing an approximation method and a Sequential Convex Approximation (SCA) technology to meet the basic requirements of the user, and the efficient cooperative beam forming scheme is aimed at two types of scenesUser selected problem design l1Approximating the model to find a sub-optimal solution to the problem.
The method specifically comprises the following steps:
step 1: initializing relevant information: and collecting user request videos and user channel conditions, dividing multicast groups according to the user requests, and constructing a cache scheduling cost model. The method specifically comprises the following steps:
step 1.1: initializing channel information of users u and F-APk
Figure BDA0003607947510000081
The video quality parameter specifically requested by user u is denoted as l u1,2, when luWhen 1 indicates that the user requests low quality video,/uWhen 2, the user requests high-quality video, and the user u requests multi-quality video is expressed as
Figure BDA0003607947510000082
The multicast group is further divided into
Figure BDA0003607947510000083
The requested content and quality versions of different multicast groups g are denoted g, respectivelyfAnd gl. Assuming that each F-AP has its own inherent power limit Pk> 0, the set of beamforming vectors can be expressed as:
Figure BDA0003607947510000084
step 1.2: the cache content generates different delay costs according to different positions of the cache content in the retrieval process of the F-AP. Consider a limited video content library
Figure BDA0003607947510000085
And assuming that the content server of the BBU pool stores all the data of the video content, the cache state of the content F at the F-AP k is expressed as follows:
Figure BDA0003607947510000086
further, the additional search delay cost generated by different cache locations is DfIt is determined by the largest of all F-APs, which is expressed as:
Figure BDA0003607947510000091
Figure BDA0003607947510000092
expressed as the content F in F-APkuThe cache state of (c). If the video request is responded to at all the local service F-APs, i.e.
Figure BDA0003607947510000093
The extra search delay is now 0. If only a part of the F-APs respond to the video request, additional retrieval delay cost d caused by video resource sharing among the F-APs is broughtf. When all F-AP nodes do not cache the content, the video needs to be forwarded to each F-AP from the BBU pool in advance. Here, we further consider the constant routing delay and forwarding delay of the BBU pool to obtain the content from the content server, using d0Is shown, and d0＞＞df。
And 2, step: planning a system model: and establishing a cooperative multicast communication model and a user QoE model, and constructing a problem model aiming at different scenes in a process. The method specifically comprises the following steps:
step 2.1: establishing a multi-quality video cooperation multicast transmission communication model, and expressing the transmission rate of the user u as follows by using a shannon formula according to the information in S11:
Figure BDA0003607947510000094
where B represents the bandwidth and the signal to interference and noise ratio is expressed as:
Figure BDA0003607947510000095
wherein
Figure BDA0003607947510000096
Representing a Gaussian white noise interference power;
step 2.2: in view of fairness within the multicast group, the actual communication rates of the users in multicast group g are all consistent with the group communication rate, which is expressed as:
Figure BDA0003607947510000097
step 2.3: the user QoE model mainly considers two parts, namely the quality level of the video obtained by the user and the playing time delay of the video obtained by the user, and is specifically represented as
Figure BDA0003607947510000098
Beta is used for controlling marginal benefit variation trend of playing time delay, eta represents a time delay satisfaction factor, the closer to 1, the more obvious influence of the time delay on QoE of a user is, and Du,maxIndicating the maximum transmission delay of all video layer data currently being received by the user, i.e.
Figure BDA0003607947510000099
Hereinafter, Q is referred to as a network QoE and represents a weighted sum of qoes of all users in the network.
Step 2.4: for different communication resource conditions, network QoE maximization, network energy efficiency maximization and user scheduling problems are formulated as follows.
Figure BDA0003607947510000101
Figure BDA0003607947510000102
Figure BDA0003607947510000103
Figure BDA0003607947510000104
Figure BDA0003607947510000105
wherein ,C1Is a fairness constraint within the multicast group; c2A video quality bit rate constraint indicating multicast group rate correspondence,
Figure BDA0003607947510000106
is the play bit rate threshold corresponding to the video quality; c3Corresponding to the power constraints of the F-AP node,
Figure BDA0003607947510000107
represents the square of the 2 norm; c4It is limited that the video quality obtained by each user should not be lower than the originally requested video quality,
Figure BDA0003607947510000108
is the video quality originally requested by the user.
And further constructing a network energy efficiency maximization problem model:
Figure BDA0003607947510000109
s.t.C1-C3
finally, a user scheduling model is designed:
Figure BDA00036079475100001010
Figure BDA00036079475100001011
Figure BDA00036079475100001012
wherein z＝{z0,...,zu,...zUThe service cost parameter is used for measuring the cost of the service user, and is specifically expressed as follows:
Figure BDA00036079475100001013
and step 3: determining network resource availability: and proposing a rate satisfaction degree xi for measuring whether the existing network resources can provide services for all users, if so, carrying out step S4, and otherwise, carrying out step S5. The method specifically comprises the following steps:
step 3.1: due to the problems
Figure BDA00036079475100001014
C in (1)1The quadratic term of middle w appears at the same time
Figure BDA00036079475100001015
The resulting non-convex nature makes the problem difficult to solve. Furthermore, we introduce auxiliary variables
Figure BDA00036079475100001016
And γ, and modifying the original constraint to:
Figure BDA00036079475100001017
Figure BDA0003607947510000111
Figure BDA0003607947510000112
step 3.2: constraint C after replacement5Still non-convex, we will mix C with SCA technique5The taylor first order expansion on its right side is used instead as follows:
Figure BDA0003607947510000113
further mixing C5Is replaced by
Figure BDA0003607947510000114
Giving an initial beamforming vector group which accords with the constraint, and obtaining a local optimal solution of the initial beamforming vector group in an iterative manner;
step 3.3: to give an initial w(0)We propose a method to obtain optimal rate satisfaction
Figure BDA0003607947510000115
The problem model of (2) is as follows:
Figure BDA0003607947510000116
Figure BDA0003607947510000117
Figure BDA0003607947510000118
an optimal xi can be obtained by utilizing a CVX solver in an iteration mode;
step 3.4: if ξ < 1 indicates that the existing resources are not enough to meet all the user requirements, the step S5 is skipped, otherwise, the optimized improvement of the network QoE is indicated, and the step S4 is skipped.
And 4, step 4: network QoE maximization: firstly, aiming at the existing user request, the QoE of the network is improved by improving the low-quality video to the high-quality video, and the cooperative beam forming scheme is optimized in the process. And finally, deploying a beam forming scheme and a video quality selection scheme to each F-AP through loop iteration until the optimal allocation is achieved. The method specifically comprises the following steps:
step 4.1: firstly, because of the coupling of the beam forming design and the video quality selection parameter, the video quality of a user is enhanced, and a video quality enhancement parameter v is provideduModifying constraint C1Is composed of
Figure BDA0003607947510000119
Replacing the original problem optimization target with
Figure BDA00036079475100001110
Finally, obtaining an enhanced cost set v by utilizing CVX calculation;
and 4.2: based on the results in v, a low cost user enhancement is selected. After the quality of the corresponding video is modified, calculating the optimal beam forming by taking the network QoE as an optimization target again, and iterating until the network QoE is reduced or the resources are insufficient to provide the existing service;
step 4.3: a beamforming scheme is deployed to each F-AP.
And 5: user scheduling selection: a user service cost function v is proposed to measure the cost of providing services to the users, and step S6 is entered by continuously eliminating more costly users until the needs of all existing users can be met. The method specifically comprises the following steps:
step 5.1: first, the original objective function z is transformed due to the user scheduling problemuThrough a process of0The range after regularization is within 1, and l1The norm calculation will make the variable keep the original scale unchanged. Therefore, by additionally constructing a corresponding weight on the objective function
Figure BDA0003607947510000121
Differences in the variable scale of the new problem and the original problem can be balanced. Will question
Figure BDA0003607947510000122
Is expressed as
Figure BDA0003607947510000123
wherein
Figure BDA0003607947510000124
τ > 0 is a regularization parameter that guarantees zuWhen the weight value is 0, the corresponding updated weight value is not positive and infinite;
step 5.2: further carrying out iterative computation on the user service cost function set to obtain z*Performing descending arrangement;
step 5.3: using dichotomy, for z*The user schedule in (1) is selected, whether the rate satisfaction ξ can be made larger than 1 is taken as a jump-out condition until the end is reached when the most users can be served, and the process proceeds to step S6.
Step 6: network energy efficiency maximization: when the existing communication resources are enough to meet the requirements of users, the ratio of the network QoE to the energy consumption is proposed as the energy efficiency under the scene, the beam forming scheme is redesigned so as to save the communication resources as much as possible while the requirements of the users are met as much as possible, and finally the beam forming scheme is deployed to each F-AP. The method specifically comprises the following steps:
step 6.1: due to the fractional form in the energy efficiency optimization objective function, a Dinkelbach algorithm is adopted to convert a non-convex objective function, and a new auxiliary variable lambda is introduced as follows:
Figure BDA0003607947510000125
wherein ,
Figure BDA0003607947510000126
the objective function is the subtraction of the numerator denominator form:
F(λ)＝Q(w)-λPsum(w)
step 6.2: furthermore, with the SCA framework, we can iteratively update through the CVX solver until
Figure BDA0003607947510000127
Convergence below a minimum threshold;
step 6.3: and deploying the beamforming scheme to each F-AP.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.

Claims (8)

1. A cache-enabled multi-quality video distribution method is characterized in that a user QoE is formulated as an optimization target according to the video quality requested by a user, a content cache position and the channel condition of the user, and multicast cooperative beam forming and video quality selection are optimized by utilizing an SCA framework; then, under the condition of considering the shortage of network resources, users which have great influence on user groups in the network are preferentially removed, the energy efficiency of the network is maximized after the existing users can be met, and the user service under the condition of resource shortage is ensured.
2. The multi-quality video distribution method according to claim 1, characterized in that the method comprises the following steps:
s1: initializing relevant information: collecting user request videos and user channel conditions, dividing multicast groups according to the user requests, and constructing a cache scheduling cost model;
s2: planning a system model: establishing a multi-quality video cooperation multicast communication model and a user QoE model, and establishing a problem model aiming at different scenes in a process;
s3: determining network resource availability: using the rate satisfaction xi to measure whether the existing network resource can provide service for all users, if so, performing step S4, otherwise, performing step S5;
s4: network QoE maximization: firstly, aiming at the existing user request, improving the QoE of the network by improving the low-quality video to the high-quality video, and optimizing a cooperative beam forming scheme in the process; performing loop iteration until optimal distribution is achieved, and finally deploying a beam forming scheme and a video quality selection scheme to each fog wireless access point F-AP;
s5: user scheduling selection: using the user service cost function v to measure the cost for providing the service to the user, and continuously eliminating the users with higher cost until the requirements of all the existing users can be met, then entering step S6;
s6: network energy efficiency maximization: when the existing communication resources are enough to meet the requirements of users, the ratio of the network QoE to the energy consumption is taken as the energy efficiency under the scene, the beam forming scheme is redesigned, the communication resources are saved while the requirements of the users are met, and finally the beam forming scheme is deployed to each F-AP.
3. The multi-quality video distribution method according to claim 2, wherein the step S1 specifically includes the steps of:
s11: initializing channel information of users u and F-AP k
Figure FDA0003607947500000011
The video quality parameter specifically requested by user u is denoted as lu＝{1,2}，lu1 indicates that the user requests low quality video,/u2 denotes that the user requests high quality video; user u requests a multi-quality video representation as
Figure FDA0003607947500000012
The multicast group is then divided, denoted as
Figure FDA0003607947500000013
Separate representation of requested content and quality versions for different multicast groups gIs gfAnd gl(ii) a Assuming that each F-AP has its own inherent power limit Pk> 0, the set of beamforming vectors is represented as:
Figure FDA0003607947500000014
wherein ,
Figure FDA0003607947500000021
representing the set of beamforming vectors, w, at F-AP kk,gRepresents the beamforming vector assigned to multicast group g,
Figure FDA0003607947500000022
representing a beamforming vector dimension of Ak×1，
Figure FDA0003607947500000023
Representing a F-AP set;
s12: cache content during retrieval of F-AP, a limited video content library is considered
Figure FDA0003607947500000024
And assuming that the content server of the BBU pool stores all the data of the video content, the cache state of the content F at the F-AP k is represented as:
Figure FDA0003607947500000025
then, the extra retrieval delay cost D generated by different cache positions is addedfExpressed as:
Figure FDA0003607947500000026
wherein ,ck,fExpressed as the cache state of the content F in the F-AP k; if the video request is in all booksGet a response on the ground service F-AP, i.e. ck,fWhen the search time delay is 0, the additional search time delay is 0; if only a part of the F-APs respond to the video request, additional retrieval delay cost d caused by video resource sharing among the F-APs is broughtf(ii) a When all F-AP nodes do not have the cache content F, the video needs to be forwarded to each F-AP from a BBU pool in advance; namely, d is used for considering the constant routing delay and the forwarding delay of the BBU pool for obtaining the content from the content server0Is shown, and d0＞＞df。
4. The multi-quality video distribution method according to claim 3, wherein the step S2 specifically comprises the steps of:
s21: establishing a multi-quality video cooperation multicast transmission communication model, and expressing the transmission rate which can be achieved by a user u as follows by utilizing a Shannon formula:
Figure FDA0003607947500000027
wherein B represents a bandwidth; signal to interference plus noise ratio
Figure FDA0003607947500000028
Expressed as:
Figure FDA0003607947500000029
wherein ,
Figure FDA00036079475000000210
representing the power of a gaussian white noise interferer,
Figure FDA00036079475000000211
represents the set of multicast groups to which user u is assigned, huRepresenting the channel gain vectors for user u to all F-APs,
Figure FDA00036079475000000212
s22: in view of fairness within the multicast group, the actual communication rates of the users in multicast group g are consistent with the group communication rate, which is expressed as:
Figure FDA0003607947500000031
wherein ,
Figure FDA0003607947500000032
indicating reception of content f for multicast group glThe inter-group rate of (c) is,
Figure FDA0003607947500000033
indicating that user u in the group receives content flThe intra-group rate of (c);
s23: establishing a user QoE model, considering two parts of the quality level of the video acquired by the user and the playing time delay of the video acquired by the user, and specifically expressing the two parts as follows:
Figure FDA0003607947500000034
wherein ,QoEuExpressing the practical QoE value of a user u, beta expressing the marginal benefit variation trend of controlling the playing time delay, eta expressing the time delay satisfaction factor, and lELSelection parameters representing high quality video, Du,maxRepresenting the maximum transmission delay of all video layer data currently received by a user;
s24: aiming at different communication resource conditions, network QoE maximization, network energy efficiency maximization and user scheduling problems are formulated as follows;
Figure FDA0003607947500000035
Figure FDA0003607947500000036
Figure FDA0003607947500000037
Figure FDA0003607947500000038
Figure FDA0003607947500000039
wherein Q represents the weighted sum of all user QoEs in the network; w and l respectively represent a beam forming vector group and a video quality selection parameter obtained by a user; c1Is a fairness constraint within the multicast group; c2A video quality bit rate constraint indicating multicast group rate correspondence,
Figure FDA00036079475000000310
is the play bit rate threshold corresponding to the video quality; c3Representing the power constraint of the F-AP node,
Figure FDA00036079475000000311
represents the square of the 2 norm; c4Indicating that the video quality obtained by each user should not be lower than the originally requested video quality,
Figure FDA00036079475000000312
is the video quality that the user originally requested,
Figure FDA00036079475000000313
representing a set of users;
then, constructing a network energy efficiency maximization problem model as follows:
Figure FDA00036079475000000314
s.t.C1-C3
and finally, constructing a user scheduling model:
Figure FDA0003607947500000041
Figure FDA0003607947500000042
Figure FDA0003607947500000043
wherein z ═ z0,...,zu,...zUThe service cost parameter set is used for measuring the cost of the service user, and is specifically represented as:
Figure FDA0003607947500000044
wherein ,Ru,lIndicating the transmission rate achievable by user u requesting video of quality/.
5. The multi-quality video distribution method according to claim 4, wherein the step S3 specifically comprises the steps of:
s31: introducing auxiliary variables
Figure FDA0003607947500000045
And gamma, will be problematic
Figure FDA0003607947500000046
Is modified to:
Figure FDA0003607947500000047
Figure FDA0003607947500000048
Figure FDA0003607947500000049
s32: using SCA technique, add C5The taylor first order expansion on its right side is used instead as follows:
Figure FDA00036079475000000410
i.e. C5Is replaced by
Figure FDA00036079475000000411
wherein ,
Figure FDA00036079475000000412
is an approximate substitution function, t represents the number of iterations; giving an initial beamforming vector group which accords with the constraint, namely obtaining a local optimal solution of the initial beamforming vector group in an iteration mode;
s33: obtaining optimal rate satisfaction
Figure FDA00036079475000000413
The problem model of (2) is:
Figure FDA00036079475000000414
Figure FDA00036079475000000415
Figure FDA00036079475000000416
obtaining an optimal xi by utilizing a CVX solver in an iteration mode;
s34: if ξ < 1 indicates that the existing resources are not enough to meet the requirements of all users, the step S5 is skipped, otherwise, the optimized improvement of the QoE of the network is indicated, and the step S4 is skipped.
6. The multi-quality video distribution method according to claim 5, wherein the step S4 specifically comprises the steps of:
s41: firstly, the video quality of a user is enhanced, and a video quality enhancement parameter v is designeduModifying constraint C1Is composed of
Figure FDA0003607947500000051
Will be the original problem
Figure FDA0003607947500000052
Optimization goal replacement
Figure FDA0003607947500000053
Finally, calculating by using CVX to obtain an enhanced cost set v;
s42: selecting a low-cost user enhancement according to the result in v; after the corresponding video quality is modified, calculating the optimal beam forming by taking the network QoE as an optimization target again, and iterating until the network QoE is reduced or the resources are not enough to provide the existing service;
s43: a beamforming scheme is deployed to each F-AP.
7. The multi-quality video distribution method according to claim 6, wherein the step S5 specifically comprises the steps of:
s51: at an objective function zuThe weight value corresponding to the extra structure
Figure FDA0003607947500000054
Will question
Figure FDA0003607947500000055
Is expressed as
Figure FDA0003607947500000056
wherein
Figure FDA0003607947500000057
τ > 0 is a regularization parameter, which guarantees zuWhen the weight value is equal to 0, the corresponding updated weight value is not positive infinity;
s52: iterative computation is carried out on the user service cost set z to obtain z*Performing descending order arrangement;
s53: using dichotomy, for z*The user schedule in (1) is selected, whether the rate satisfaction ξ is greater than 1 can be made as a jump-out condition until the end is reached when the most users can be served, and the process goes to step S6.
8. The multi-quality video distribution method according to claim 7, wherein the step S6 specifically comprises the steps of:
s61: the Dinkelbach algorithm is adopted to convert the non-convex target function, and a new auxiliary variable lambda is introduced as follows:
Figure FDA0003607947500000058
wherein ,
Figure FDA0003607947500000059
the objective function is the subtraction of the numerator denominator form:
F(λ)＝Q(w)-λPsum(w)
wherein F (λ) represents the transformed objective function;
s62: iteratively updating by a CVX solver using an SCA framework until
Figure FDA00036079475000000510
Convergence below a minimum threshold;
s63: and deploying the beamforming scheme to each F-AP.
CN202210424517.8A 2022-04-21 2022-04-21 Cache-enabled multi-quality video distribution method Active CN114786137B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210424517.8A CN114786137B (en) 2022-04-21 2022-04-21 Cache-enabled multi-quality video distribution method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210424517.8A CN114786137B (en) 2022-04-21 2022-04-21 Cache-enabled multi-quality video distribution method

Publications (2)

Publication Number Publication Date
CN114786137A true CN114786137A (en) 2022-07-22
CN114786137B CN114786137B (en) 2023-06-20

Family

ID=82431284

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210424517.8A Active CN114786137B (en) 2022-04-21 2022-04-21 Cache-enabled multi-quality video distribution method

Country Status (1)

Country Link
CN (1) CN114786137B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2493171A1 (en) * 2011-02-25 2012-08-29 Tektronix International Sales GmbH Video data stream evaluation systems and methods
CN105025580A (en) * 2015-08-18 2015-11-04 北京邮电大学 Method and device for allocating wireless resources
EP3099013A1 (en) * 2015-05-29 2016-11-30 Alcatel Lucent Providing a qoe-dependent charging rate
CN109150405A (en) * 2018-09-19 2019-01-04 湖北工业大学 A kind of video multicast transmission method based on the white frequency range of TV
CN110049315A (en) * 2019-04-26 2019-07-23 山西大学 A method of improving live video system user Quality of experience
CN111953407A (en) * 2020-08-24 2020-11-17 西南大学 Unmanned aerial vehicle video relay system and energy consumption minimizing method thereof
US20200404623A1 (en) * 2019-05-01 2020-12-24 Cypress Semiconductor Corporation Devices, systems, and methods for predicting communication channel conditions
CN113114756A (en) * 2021-04-08 2021-07-13 广西师范大学 Video cache updating method for self-adaptive code rate selection in mobile edge calculation
CN113596848A (en) * 2021-07-27 2021-11-02 南京工业大学 Content caching and spectrum allocation combined design method for NOMA wireless video multicast

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2493171A1 (en) * 2011-02-25 2012-08-29 Tektronix International Sales GmbH Video data stream evaluation systems and methods
EP3099013A1 (en) * 2015-05-29 2016-11-30 Alcatel Lucent Providing a qoe-dependent charging rate
CN105025580A (en) * 2015-08-18 2015-11-04 北京邮电大学 Method and device for allocating wireless resources
CN109150405A (en) * 2018-09-19 2019-01-04 湖北工业大学 A kind of video multicast transmission method based on the white frequency range of TV
CN110049315A (en) * 2019-04-26 2019-07-23 山西大学 A method of improving live video system user Quality of experience
US20200404623A1 (en) * 2019-05-01 2020-12-24 Cypress Semiconductor Corporation Devices, systems, and methods for predicting communication channel conditions
CN111953407A (en) * 2020-08-24 2020-11-17 西南大学 Unmanned aerial vehicle video relay system and energy consumption minimizing method thereof
CN113114756A (en) * 2021-04-08 2021-07-13 广西师范大学 Video cache updating method for self-adaptive code rate selection in mobile edge calculation
CN113596848A (en) * 2021-07-27 2021-11-02 南京工业大学 Content caching and spectrum allocation combined design method for NOMA wireless video multicast

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LAMINE AMOUR等: ""QoE-Based Framework to Optimize User Perceived Video Quality"", 《2017 IEEE 42ND CONFERENCE ON LOCAL COMPUTER NETWORKS (LCN)》 *
王汝言等: ""QoE感知的FiWi视频分发机制"", 《通信学报》 *

Also Published As

Publication number Publication date
CN114786137B (en) 2023-06-20

Similar Documents

Publication Publication Date Title
Li et al. QoE-driven mobile edge caching placement for adaptive video streaming
CN111031102B (en) Multi-user, multi-task mobile edge computing system cacheable task migration method
CN113114756B (en) Video cache updating method for self-adaptive code rate selection in mobile edge calculation
CN111935784B (en) Content caching method based on federal learning in fog computing network
CN108601074B (en) Network resource allocation method and device based on heterogeneous joint cache
CN108495340B (en) Network resource allocation method and device based on heterogeneous hybrid cache
CN110312231A (en) Content caching decision and resource allocation joint optimization method based on mobile edge calculations in a kind of car networking
CN111552564A (en) Task unloading and resource optimization method based on edge cache
Chua et al. Resource allocation for mobile metaverse with the Internet of Vehicles over 6G wireless communications: A deep reinforcement learning approach
CN115714814B (en) Edge cache replacement method based on multi-agent reinforcement learning
CN115065678A (en) Multi-intelligent-device task unloading decision method based on deep reinforcement learning
Dai et al. Proactive caching over cloud radio access network with user mobility and video segment popularity awared
CN113993168B (en) Collaborative caching method based on multi-agent reinforcement learning in fog wireless access network
CN112887314B (en) Time delay perception cloud and mist cooperative video distribution method
CN111447506B (en) Streaming media content placement method based on delay and cost balance in cloud edge environment
CN114786137A (en) Cache-enabled multi-quality video distribution method
Li et al. Video caching and scheduling with edge cooperation
CN114786200A (en) Intelligent data caching method based on cooperative sensing
CN115720237A (en) Caching and resource scheduling method for edge network self-adaptive bit rate video
CN112954026A (en) Multi-constraint content collaborative cache optimization method based on edge calculation
Liu et al. Joint mixed-timescale optimization of content caching and delivery policy in NOMA-based vehicular networks
Zhang et al. Cache-enabled dynamic rate allocation via deep self-transfer reinforcement learning
Lin et al. Edge intelligence-based joint caching and transmission for QoE-aware video streaming
CN115134418B (en) Collaborative storage method and system for multi-access edge calculation and electronic equipment
Cao et al. A Cooperative Edge Caching Approach Based on Multi-Agent Deep Reinforcement Learning

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