A video stream transmission method, device, terminal device, and storage medium

Abstract

The present invention discloses a video stream transmission method, apparatus, terminal equipment and storage medium, the method comprising: constructing a video stream transmission model including a cloud server, an edge server and a plurality of mobile devices; wherein the cloud server is connected to the edge server, and the edge server is connected to the plurality of mobile devices; obtaining basic data of the video stream transmission model; based on the basic data, constructing an energy efficiency optimization model, computing power resource constraints and transmission power constraints with the goal of minimizing the video stream transmission cost while ensuring the quality of user experience; solving the energy efficiency optimization model under the constraints of the computing power resource constraints and the transmission power constraints, generating the computing power resources and transmission power allocated to each mobile device at each moment when minimizing the video stream transmission cost while ensuring the quality of user experience; and controlling the video stream transmission model according to the computing power resources and transmission power allocated to each mobile device at each moment to perform video stream transmission.

Description

Technical Field

[0001] The present invention relates to the technical field of streaming media data transmission, and in particular to a video stream transmission method, apparatus, terminal equipment and storage medium. Background Art

[0002] Mobile Edge Computing (MEC) is an emerging technology that provides low-latency and high-bandwidth computing services to mobile devices by deploying computing and storage resources at the edge of the network.

[0003] With the popularity of mobile devices and advancements in mobile networks, more and more users want to watch high-quality real-time video on their mobile devices, such as streaming services, video conferencing, and remote monitoring. However, traditional video streaming faces several challenges, such as high latency, bandwidth limitations, and energy consumption. To address these issues, mobile edge computing (MEC) has been proposed. As a distributed computing model, MEC provides the ability to bring computing and storage resources closer to user devices. MEC deploys edge servers or edge nodes at the edge of mobile networks, pushing computing tasks and data processing to the edge of the network. Compared to traditional cloud computing models, MEC offers lower latency and higher bandwidth. However, existing MEC models fail to consider the trade-off between user quality of experience and energy consumption during communication transmission, failing to minimize the cost of video streaming while ensuring user quality.

[0004] Therefore, minimizing the cost of video streaming while ensuring the quality of user experience is an urgent problem that needs to be solved. Summary of the Invention

[0005] The embodiments of the present invention provide a video stream transmission method, apparatus, terminal device and storage medium, which can effectively solve the problem that the existing technology cannot minimize the video stream transmission cost while ensuring the user experience quality, and reduce the video stream transmission cost while ensuring the user experience quality.

[0006] An embodiment of the present invention provides a video stream transmission method, including:

[0007] Constructing a video stream transmission model; wherein the video stream transmission model includes: a cloud server, an edge server, and a plurality of mobile devices; the cloud server is connected to the edge server, and the edge server is connected to the plurality of mobile devices;

[0008] Obtaining basic data for a video stream transmission model; wherein the basic data includes: a maximum computing power of the edge server, a maximum transmit power of the edge server, a video clip transmission rate, a video clip transmission power, power consumption of components other than the processor in the edge server, clock cycles included in the processor, a channel gain from the edge server to the mobile device, a bit rate of the video clip to be processed by each mobile device at each time, and a bit rate selected by each mobile device for the video clip at each time;

[0009] Based on the basic data, an energy efficiency optimization model is constructed with the goal of minimizing the video streaming transmission cost while ensuring user experience quality; and computing resource constraints and transmission power constraints are constructed based on the basic data;

[0010] Under the constraints of computing resources and transmission power, the energy efficiency optimization model is solved to generate the computing resources and transmission power allocated to each mobile device at each time when minimizing the video streaming transmission cost while ensuring the quality of user experience;

[0011] The video stream transmission model is controlled according to the computing resources and transmission power allocated to each mobile device at each moment to transmit the video stream.

[0012] Furthermore, the energy efficiency optimization model is specifically as follows:

[0013]

[0014]

[0015] Among them, E av (t) represents the average energy consumption required by N users to transmit video streams at time t; QoE av (t) represents the average quality of experience of N users when transmitting video streams at time t; t represents time t; q represents the power consumption of the components in the edge server except the processor; η represents a scaling factor; ρ represents the number of clock cycles contained in a processor; ε n (t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required; represents the transcoding time required for the nth user to transmit a video clip at time t; p n represents the transmit power allocated by the edge server to the nth mobile device; represents the total video segment transmission time of the nth mobile device at time t; P represents the video segment transmission power; represents the transmission time of the video transmitted from the cloud server to the edge server by the nth mobile device at time t; represents the transmission time of video transmission from the edge server to the mobile device at time t for the nth mobile device; ω1, ω2, and ω3 are three non-negative parameters representing the importance of experience quality; Indicates that the mobile device obtains the video clip from the cloud server; Indicates that the mobile device obtains the video clip from the edge server; b(i n (t)) represents the video segment i n (t) size; i n (t) represents the bitrate version selected by the nth mobile device for the video segment at time t; Indicates the video freeze time of the nth mobile device at time t.

[0016] Furthermore, the computing power resource constraints are specifically:

[0017]

[0018] Where F represents the maximum computing power of the edge server; Represents the computing resources allocated to the nth mobile device at time t.

[0019] Furthermore, the transmit power constraint is specifically:

[0020] p n ≤P max ;

[0021] Among them, p n represents the transmission power allocated by the edge server to the nth mobile device; P max Indicates the maximum transmit power of the edge server.

[0022] Based on the above method embodiment, the present invention provides a corresponding device embodiment;

[0023] An embodiment of the present invention provides a video stream transmission device, comprising: a video stream transmission model construction module, a data acquisition module, an energy efficiency optimization model and constraint condition construction module, an energy efficiency optimization model solution module, and a video stream transmission control module;

[0024] The video stream transmission model construction module is used to construct a video stream transmission model; wherein the video stream transmission model includes: a cloud server, an edge server and a plurality of mobile devices; the cloud server is connected to the edge server, and the edge server is connected to the plurality of mobile devices;

[0025] The data acquisition module is configured to acquire basic data of the video stream transmission model; wherein the basic data includes: the maximum computing power of the edge server, the maximum transmission power of the edge server, the video clip transmission rate, the video clip transmission power, the power consumption of components other than the processor in the edge server, the clock cycles of the processor, the channel gain from the edge server to the mobile device, the bit rate of the video clip to be processed by each mobile device at each time, and the bit rate selected by each mobile device for the video clip at each time;

[0026] The energy efficiency optimization model and constraint condition construction module is used to construct an energy efficiency optimization model based on the basic data with the goal of minimizing the video streaming transmission cost while ensuring the user experience quality; and to construct computing resource constraints and transmission power constraints based on the basic data;

[0027] The energy efficiency optimization model solving module is used to solve the energy efficiency optimization model under the constraints of computing power resources and transmission power constraints to generate computing power resources and transmission power allocated to each mobile device at each time when minimizing the video streaming transmission cost while ensuring the quality of user experience;

[0028] The video stream transmission control module is used to control the video stream transmission model to perform video stream transmission according to the computing resources and transmission power allocated to each mobile device at each moment.

[0029] Furthermore, the energy efficiency optimization model is specifically as follows:

[0030]

[0031] Among them, E av (t) represents the average energy consumption required by N users to transmit video streams at time t; QoE av (t) represents the average quality of experience of N users when transmitting video streams at time t; t represents time t; q represents the power consumption of the components in the edge server except the processor; η represents a scaling factor; ρ represents the number of clock cycles contained in a processor; ε n (t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required; represents the transcoding time required for the nth user to transmit a video clip at time t; p n represents the transmit power allocated by the edge server to the nth mobile device; represents the total video segment transmission time of the nth mobile device at time t; P represents the video segment transmission power; represents the transmission time of the video transmitted from the cloud server to the edge server by the nth mobile device at time t; represents the transmission time of video transmission from the edge server to the mobile device at time t for the nth mobile device; ω1, ω2, and ω3 are three non-negative parameters representing the importance of experience quality; Indicates that the mobile device obtains the video clip from the cloud server; Indicates that the mobile device obtains the video clip from the edge server; b(i n (t)) represents the video segment i n (t) size; i n (t) represents the bitrate version selected by the nth mobile device for the video segment at time t; Indicates the video freeze time of the nth mobile device at time t.

[0032] Furthermore, the computing power resource constraints are specifically:

[0033]

[0034] Where F represents the maximum computing power of the edge server; Represents the computing resources allocated to the nth mobile device at time t.

[0035] Furthermore, the transmit power constraint is specifically:

[0036] p n ≤P max ;

[0037] Among them, p n represents the transmission power allocated by the edge server to the nth mobile device; P max Indicates the maximum transmit power of the edge server.

[0038] Another embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements a video stream transmission method described in the above embodiment of the invention when executing the computer program.

[0039] Another embodiment of the present invention provides a storage medium, which includes a stored computer program. When the computer program is executed, the device where the storage medium is located is controlled to execute the video stream transmission method described in the above embodiment of the invention.

[0040] The following beneficial effects are achieved by implementing the present invention:

[0041] The present invention provides a video stream transmission method, apparatus, terminal equipment and storage medium, wherein the transmission method constructs a video stream transmission model comprising a cloud server, an edge server and several mobile devices; on the basis of the video stream transmission model, basic data is obtained, and an energy efficiency optimization model is constructed based on the obtained basic data with the goal of minimizing the video stream transmission cost while ensuring the quality of user experience; computing power resource constraints and transmission power constraints are constructed based on the obtained basic data; the energy efficiency optimization model is then solved under the constructed constraint conditions to generate computing power resources and transmission power allocated to each mobile device when minimizing the video stream transmission cost while ensuring the quality of user experience; the video stream transmission model is controlled according to the determined computing power resources and transmission power allocated to each mobile device for video stream transmission; by constructing the energy efficiency optimization model to determine the computing power resources and transmission power allocated to each mobile device at each moment in the video stream transmission model when minimizing the video stream transmission cost while ensuring the quality of user experience, the problem is solved that the existing technology cannot minimize the video stream transmission cost while ensuring the quality of user experience, and the video stream transmission cost is reduced while ensuring the quality of user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 The figure is a flow chart of a video stream transmission method provided by one embodiment of the present invention.

[0043] Figure 2 It is a structural diagram of a video stream transmission model provided by an embodiment of the present invention.

[0044] Figure 3 It is a structural diagram of a video stream transmission device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0046] To better understand the content of this invention, it is necessary to explain that Mobile Edge Computing (MEC) pushes computing tasks and data processing to the edge of the network by deploying edge servers or edge nodes at the edge of the mobile network. Compared with the traditional cloud computing model, it has lower latency and higher bandwidth. In video streaming, mobile edge computing can realize the following technologies and applications:

[0047] (1) Edge caching: Mobile edge servers can cache popular video content, allowing users to obtain the required video streams faster and reducing the delay and bandwidth consumption of transmitting videos from remote servers.

[0048] (2) Real-time encoding and decoding: Mobile edge servers can provide real-time video encoding and decoding services, compressing and decompressing video streams to adapt to the bandwidth and processing power of mobile devices. This allows users to watch high-quality videos smoothly on mobile devices.

[0049] (3) Network optimization: Mobile edge computing can optimize the network and provide more efficient video transmission and network resource management. For example, through intelligent network selection and routing algorithms, video stream transmission can be transferred from congested network nodes to edge servers, improving the quality and stability of video transmission.

[0050] (4) Resource sharing and collaborative computing: Mobile edge computing can achieve resource sharing and collaborative computing, combining the computing and storage resources of multiple mobile devices to form a distributed computing network. Through collaborative computing, video streaming can better utilize the computing power and storage capacity of edge nodes, providing a better user experience.

[0051] like Figure 1 FIG. 1 is a video stream transmission method provided by an embodiment of the present invention, comprising:

[0052] Step S1: Constructing a video stream transmission model; wherein the video stream transmission model includes: a cloud server, an edge server, and a plurality of mobile devices; the cloud server is connected to the edge server, and the edge server is connected to the plurality of mobile devices;

[0053] Step S2: Obtain basic data of the video stream transmission model; wherein the basic data includes: the maximum computing power of the edge server, the maximum transmission power of the edge server, the video clip transmission rate, the video clip transmission power, the power consumption of the components of the edge server other than the processor, the clock cycle of the processor, the channel gain from the edge server to the mobile device, the bit rate of the video clip to be processed by each mobile device at each time, and the bit rate selected by each mobile device for the video clip at each time;

[0054] Step S3: Based on the basic data, an energy efficiency optimization model is constructed with the goal of minimizing the video streaming transmission cost while ensuring the quality of user experience; and computing resource constraints and transmission power constraints are constructed based on the basic data;

[0055] Step S4: Under the constraints of computing resources and transmission power, the energy efficiency optimization model is solved to generate computing resources and transmission power allocated to each mobile device at each time point to minimize the video streaming transmission cost while ensuring user experience quality;

[0056] Step S5: Control the video stream transmission model according to the computing resources and transmission power allocated to each mobile device at each moment to perform video stream transmission.

[0057] For step S1, construct Figure 2 The video stream transmission model shown mainly includes a cloud server, an edge server connected to the cloud server, and a number of mobile devices connected to the edge server, each mobile device being used by a corresponding user.

[0058] Preferably, the video stream transmission model is constructed based on a mobile edge computing network, which also includes a buffer model, an adaptive video quality module, an edge cache module, a video transcoding module, and a video transmission module. In the downlink scenario of video transmission in a mobile network of a base station (BS), there are N mobile users, that is, corresponding to N mobile devices. When each user requests a video from the base station (BS), it requests the video from the edge server or cloud server. The buffer model in the video stream transmission model is used to include videos that the user has not viewed. In the buffer model, each video that the user has not viewed is divided into multiple segments, each segment lasting d seconds and having K different bit rate versions. In this embodiment, b(k) is used to represent the bit rate, and each video segment is arranged in the buffer model in descending order of the bit rate version, where k∈1,2,3,…,K. The bit rate of the original segment is the highest bit rate. In this embodiment, it is assumed that all original segments are downloaded from the remote content data center and cached in the virtual device.

[0059] In step S2, basic data for constructing an energy efficiency optimization model and constraints is obtained based on the constructed video stream transmission model. The basic data includes: the maximum computing power of the edge server, the maximum transmission power of the edge server, the video clip transmission rate, the video clip transmission power, the power consumption of components other than the processor in the edge server, the clock cycles of the processor, the channel gain from the edge server to the mobile device, the bit rate of the video clip to be processed by each mobile device at each time, and the bit rate selected by each mobile device for the video clip at each time.

[0060] For step S3, based on the basic data obtained, an energy efficiency optimization model is constructed, as well as computing resource constraints and transmission power constraints, with the goal of minimizing the video stream transmission cost while ensuring the quality of user experience.

[0061] In a preferred embodiment, the energy efficiency optimization model is specifically:

[0062]

[0063]

[0064] Among them, E av (t) represents the average energy consumption required by N users to transmit video streams at time t; QoE av (t) represents the average quality of experience of N users when transmitting video streams at time t; t represents time t; q represents the power consumption of the components in the edge server except the processor; η represents a scaling factor; ρ represents the number of clock cycles contained in a processor; ε n (t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required; represents the transcoding time required for the nth user to transmit a video clip at time t; p n represents the transmit power allocated by the edge server to the nth mobile device; represents the total video segment transmission time of the nth mobile device at time t; P represents the video segment transmission power; represents the transmission time of the video transmitted from the cloud server to the edge server by the nth mobile device at time t; represents the transmission time of video transmission from the edge server to the mobile device at time t for the nth mobile device; ω1, ω2, and ω3 are three non-negative parameters representing the importance of experience quality; Indicates that the mobile device obtains the video clip from the cloud server; Indicates that the mobile device obtains the video clip from the edge server; b(i n (t)) represents the video segment i n (t) size; i n (t) represents the bitrate version selected by the nth mobile device for the video segment at time t; Indicates the video freeze time of the nth mobile device at time t.

[0065] In a preferred embodiment, the computing resource constraints are specifically:

[0066]

[0067] Where F represents the maximum computing power of the edge server; The computing resources allocated to the nth mobile device at any given moment.

[0068] In a preferred embodiment, the transmit power constraint is specifically:

[0069] p n ≤P max ;

[0070] Among them, p n represents the transmission power allocated by the edge server to each mobile device; P max Indicates the maximum transmit power of the edge server.

[0071] Specifically, the above energy efficiency optimization model and corresponding constraints are obtained after considering the following situations.

[0072] When a user requests a video that freezes, that is, when a sudden drop in the quality of the video or video segment transmission or transcoding is detected, resulting in a choppy playback, before playing the video segment, the mobile device (MD) pre-fetches the segment that the user wants to play into the buffer of the buffer model and arranges the video segments in descending order of bit rate version. Since the buffer can contain video segments of different bit rates, the length of each video segment to be buffered is used to measure the size of the buffer; the length of the video segment is in seconds. represents the buffer size of the nth user at time t. Due to the lack of responsive resources, the buffer may be empty. This causes the video to play unsmoothly, which is defined as the video freeze phenomenon in the prefetch buffer model. The video freeze time can be expressed as:

[0073]

[0074] in, Indicates the time when the video freeze occurs for the nth user at time t; represents the transcoding time required for the nth user to transmit a video clip at time t; represents the transmission time required for the nth user to transmit the video clip at time t; Indicates the buffer size of the nth user at the tth moment.

[0075] The dynamic buffer, that is, the buffer size at the next moment depends on the buffer size at the current moment, the transcoding time required to transmit the video segment at the current moment, and the transmission time required to transmit the video segment at the current moment. The dynamic buffer is represented as follows:

[0076]

[0077] Among them, Γ n (t+1) represents the buffer size of the nth user at the t+1th time.

[0078] l n (t) represents the subscript index of the video segment to be processed by the nth mobile device at time t, i n(t) represents the subscript index of the bit rate version selected by the nth mobile device for the video segment at time t. Then the size of the video segment to be processed by the nth mobile device at time t is:

[0079] y n (t) = b(i n (t))d;

[0080] Among them, y n (t) represents the size of the video segment to be processed by the nth mobile device at time t; i n (t) represents the subscript index of the bit rate version selected by the nth mobile device for the video segment at time t; b(i n (t)) represents the bit rate selected by the nth mobile device for the video segment at time t;

[0081] The size of the original video clip is:

[0082] y′ n (t) = b(1)d;

[0083] Among them, y′ n (t) represents the size of the original video segment of the nth mobile device at time t; b(1) is the first video segment, that is, the bit rate of the original video segment.

[0084] Furthermore, the cache status of each video segment needs to be determined.

[0085] c j,k Indicates the cache status of the k-th bit rate version of the video segment selected by the j-th segment. When c j,k =1, it means that the video segment corresponding to the k-th bit rate version selected by the j-th segment has been cached on the edge server and no transcoding operation is required. j,k =0 means that the video segment corresponding to the k-th bit rate version selected by the j-th segment is not cached on the edge server and needs to be transcoded.

[0086] When transcoding is required, When the bit rate version of the original video clip is transcoded into the bit rate version of the subscript i n (t) bit rate version of the video clip; with ε n (t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required, that is, the computing power resource constraints when transcoding is required can be obtained:

[0087]

[0088] Among them, ε n(t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required; F represents the maximum computing power of the edge server.

[0089] The required transcoding time is:

[0090]

[0091] in, represents the transcoding time required for the nth user to transmit the video segment at time t; ξ represents the amount of data that the edge server can process in each processor (CPU) cycle; y′ n (t) represents the size of the original video clip of the nth mobile device at time t; ε n (t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required;

[0092] Assuming that there are ρ clock cycles in one CPU cycle, the power of the edge server during transcoding operation can be expressed as:

[0093]

[0094] Where q represents the power consumption of components other than the CPU, which refers to the power consumption of non-critical components other than the CPU involved in the calculation, such as memory, hard disk, etc., and is a given fixed value; η represents a scaling factor; ρ represents the number of clock cycles in a CPU cycle; ε n (t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required;

[0095] Therefore, the energy consumption of the transcoding operation at time t can be obtained as follows:

[0096]

[0097] in, It represents the energy consumption of transcoding on the nth mobile device at time t.

[0098] When no transcoding operation is required, that is, when When , no transcoding energy consumption is required, and the corresponding transcoding energy consumption is 0.

[0099] Taking into account both transcoding and non-transcoding situations, the computing power resources allocated to the nth mobile device at time t can be expressed as:

[0100]

[0101] in, represents the computing resources allocated to the nth mobile device at time t;

[0102] The transcoding time required for the nth user to transmit a video clip at time t is expressed as:

[0103]

[0104] in, It represents the transcoding time required for the nth user to transmit a video clip at time t.

[0105] After determining the energy consumption and time required for transcoding, we need to further determine the transmission time and energy consumption required for transmission in the video streaming model. This process is mainly divided into two parts: one is the transmission time and energy consumption required for transmission from the cloud server to the edge server, and the other is the transmission time and energy consumption required for transmission from the edge server to each mobile device.

[0106] The transmission time from the cloud server to the edge server is expressed as:

[0107]

[0108] in, represents the transmission time of the video from the cloud server to the edge server when the nth mobile device transmits it at time t; n (t) represents the size of the video segment to be processed by the nth mobile device at time t; r represents the transmission rate of the video segment, which is a fixed value and a constant.

[0109] The transmission energy consumption from the cloud server to the edge server is expressed as:

[0110]

[0111] in, It represents the energy consumption of the nth mobile device when transmitting video from the cloud server to the edge server at time t; P represents the transmission power of the video clip, which is a fixed value and a constant.

[0112] According to Shannon's theorem, the transmission rate from the edge server to the mobile device can be expressed as:

[0113]

[0114] Among them, R n (t) represents the transmission rate of the video from the edge server to the mobile device at time t; B represents the bandwidth allocated to each user during the transmission process; p n represents the transmission power allocated by the edge server to the nth mobile device; h nrepresents the channel gain from the edge server to the nth mobile device; N0 represents the power spectral density of additive white Gaussian noise.

[0115] Then, the transmission time from the edge server to the mobile device is expressed as:

[0116]

[0117] in, represents the transmission time of the video from the edge server to the mobile device at time t for the nth mobile device;

[0118] The energy consumption from the edge server to the mobile device is expressed as:

[0119]

[0120] in, represents the energy consumption of the nth mobile device when transmitting video from the edge server to the mobile device at time t.

[0121] g j,k Indicates the cloud edge selection state. When g j,k =1 means obtaining video clips from the cloud server. j,k = 0, which means the video clip is transcoded and transmitted from the edge server. The total transmission time can be expressed as:

[0122]

[0123] in, It represents the total transmission time of the video clip by the nth mobile device at time t.

[0124] According to the above derivation, the average energy consumption required for N users to transmit video streams is:

[0125]

[0126] Among them, E av (t) represents the average energy consumption required by N users when transmitting video streams at time t. The average quality of experience (QoE) of N users when transmitting video streams is expressed as:

[0127]

[0128] Among them, QoE av (t) represents the average quality of experience of N users when transmitting video streams at time t.

[0129] Based on the average quality of experience of N users when transmitting video streams at time t and the average energy consumption required by N users when transmitting video streams at time t, an energy efficiency optimization model can be constructed with the goal of minimizing the video streaming cost while ensuring the quality of user experience:

[0130]

[0131] Computing resource constraints for the energy efficiency optimization model:

[0132]

[0133] Transmit power constraints for the energy efficiency optimization model:

[0134] p n ≤P max ;

[0135] In step S4, the energy efficiency optimization model is solved using a self-learning strategy using deep reinforcement learning (DDPG). This method uses the DDPG model to solve the energy efficiency optimization model based on the randomness of task requests and the dynamic changes of the MEC network. The MEC network is the video streaming transmission model constructed above, which includes cloud servers, edge servers, and several mobile devices.

[0136] The DDPG model includes: system state space Action Space System reward function

[0137] System state space include:

[0138] s(t)=[h1(t),…,h N (t),Γ1(t),…,Γ N (t),b(i1(t-1)),…,b(i N (t-1))];

[0139] Among them, h n (t) represents the channel gain of the communication link of the nth user at time t; Γ n (t) represents the remaining buffer size of the nth user at time t; b(i n (t-1)) represents the bit rate of the video segment at time t-1.

[0140] System state space include:

[0141]

[0142] System appreciation function include:

[0143]

[0144] Define the long-term return R(t) as:

[0145]

[0146] Where γ represents the discount factor, It is the decay value of future rewards, indicating the immediate reward of the system. The larger the γ is, the more emphasis is placed on the experience gained from past training during the update, and vice versa, the more emphasis is placed on current gains.

[0147] Each state-action pair corresponds to an action value function (also called Q-function). Users can evaluate and improve the task offloading strategy based on the Q value. The Q-function is defined as: taking action a(t) in the current state s(t) and obtaining the cumulative reward Q according to a certain strategy π π (s,a), the basic equation is expressed as follows:

[0148]

[0149] in, represents the expectation, and γ represents the discount factor.

[0150] After obtaining the above-mentioned system state, system reward function, and action value function, deep reinforcement learning (DDPG) is used to solve the energy efficiency optimization model to generate the computing resources and transmission power allocated to each mobile device at each moment while minimizing the video streaming transmission cost while ensuring the quality of user experience.

[0151] In step S5, the video stream transmission model is controlled according to the computing resources and transmission power allocated to each mobile device at each moment to perform video stream transmission.

[0152] Based on the above method embodiments, the present invention provides corresponding device embodiments.

[0153] like Figure 3 As shown, an embodiment of the present invention provides a video stream transmission device, comprising: a video stream transmission model construction module, a data acquisition module, an energy efficiency optimization model and constraint condition construction module, an energy efficiency optimization model solving module, and a video stream transmission control module;

[0154] The video stream transmission model construction module is used to construct a video stream transmission model; wherein the video stream transmission model includes: a cloud server, an edge server and a plurality of mobile devices; the cloud server is connected to the edge server, and the edge server is connected to the plurality of mobile devices;

[0155] The data acquisition module is configured to acquire basic data of the video stream transmission model; wherein the basic data includes: the maximum computing power of the edge server, the maximum transmission power of the edge server, the video clip transmission rate, the video clip transmission power, the power consumption of components other than the processor in the edge server, the clock cycles of the processor, the channel gain from the edge server to the mobile device, the bit rate of the video clip to be processed by each mobile device at each time, and the bit rate selected by each mobile device for the video clip at each time;

[0156] The energy efficiency optimization model and constraint condition construction module is used to construct an energy efficiency optimization model based on the basic data with the goal of minimizing the video streaming transmission cost while ensuring the user experience quality; and to construct computing resource constraints and transmission power constraints based on the basic data;

[0157] The energy efficiency optimization model solving module is used to solve the energy efficiency optimization model under the constraints of computing power resources and transmission power constraints to generate computing power resources and transmission power allocated to each mobile device at each time when minimizing the video streaming transmission cost while ensuring the quality of user experience;

[0158] The video stream transmission control module is used to control the video stream transmission model to perform video stream transmission according to the computing resources and transmission power allocated to each mobile device at each moment.

[0159] Furthermore, the energy efficiency optimization model is specifically as follows:

[0160]

[0161] Among them, E av (t) represents the average energy consumption required by N users to transmit video streams at time t; QoE av (t) represents the average quality of experience of N users when transmitting video streams at time t; t represents time t; q represents the power consumption of the components in the edge server except the processor; η represents a scaling factor; ρ represents the number of clock cycles contained in a processor; ε n (t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required; represents the transcoding time required for the nth user to transmit a video clip at time t; p n represents the transmit power allocated by the edge server to the nth mobile device; represents the total video segment transmission time of the nth mobile device at time t; P represents the video segment transmission power; represents the transmission time of the video transmitted from the cloud server to the edge server by the nth mobile device at time t; represents the transmission time of video transmission from the edge server to the mobile device at time t for the nth mobile device; ω1, ω2, and ω3 are three non-negative parameters representing the importance of experience quality; Indicates that the mobile device obtains the video clip from the cloud server; Indicates that the mobile device obtains the video clip from the edge server; b(i n (t)) represents the video segment i n (t) size; i n (t) represents the bitrate version selected by the nth mobile device for the video segment at time t; Indicates the video freeze time of the nth mobile device at time t.

[0162] Furthermore, the computing power resource constraints are specifically:

[0163]

[0164] Where F represents the maximum computing power of the edge server; The computing resources allocated to the nth mobile device at any given moment.

[0165] Furthermore, the transmit power constraint is specifically:

[0166] p n ≤P max ;

[0167] Among them, p n represents the transmission power allocated by the edge server to each mobile device; P max Indicates the maximum transmit power of the edge server.

[0168] It should be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive effort.

[0169] Those skilled in the art will clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0170] Based on the above method embodiment, the present invention provides a corresponding terminal device embodiment.

[0171] An embodiment of the present invention provides a terminal device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, a video stream transmission method described in any one of the present invention is implemented.

[0172] The terminal device may be a computing device such as a desktop computer, a notebook computer, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0173] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, connecting various parts of the entire terminal device using various interfaces and lines.

[0174] The memory can be used to store the computer program, and the processor realizes various functions of the terminal device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created based on the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0175] Based on the above method embodiment, the present invention provides a corresponding storage medium embodiment.

[0176] An embodiment of the present invention provides a storage medium, which includes a stored computer program. When the computer program is executed, the device where the storage medium is located is controlled to execute any one of the video stream transmission methods described in the present invention.

[0177] The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. The computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunications signal, and a software distribution medium.

[0178] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A video stream transmission method, characterized in that: include: Constructing a video stream transmission model; wherein the video stream transmission model includes: a cloud server, an edge server, and a plurality of mobile devices; the cloud server is connected to the edge server, and the edge server is connected to the plurality of mobile devices; Obtaining basic data for a video stream transmission model; wherein the basic data includes: a maximum computing power of the edge server, a maximum transmit power of the edge server, a video clip transmission rate, a video clip transmission power, power consumption of components other than the processor in the edge server, clock cycles included in the processor, a channel gain from the edge server to the mobile device, a bit rate of the video clip to be processed by each mobile device at each time, and a bit rate selected by each mobile device for the video clip at each time; Based on the basic data, an energy efficiency optimization model is constructed with the goal of minimizing the video streaming transmission cost while ensuring user experience quality; and computing resource constraints and transmission power constraints are constructed based on the basic data; Under the constraints of computing resources and transmission power, the energy efficiency optimization model is solved to generate the computing resources and transmission power allocated to each mobile device at each time when minimizing the video streaming transmission cost while ensuring the quality of user experience; The video stream transmission model is controlled according to the computing resources and transmission power allocated to each mobile device at each moment to transmit the video stream.

2. A video stream transmission method according to claim 1, characterized in that: The energy efficiency optimization model is specifically: Among them, E av (t) represents the average energy consumption required by N users to transmit video streams at time t; QoE av (t) represents the average quality of experience of N users when transmitting video streams at time t; t represents time t; q represents the power consumption of the components in the edge server except the processor; η represents a scaling factor; ρ represents the number of clock cycles contained in a processor; ε n (t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required; represents the transcoding time required for the nth user to transmit a video clip at time t; p n represents the transmit power allocated by the edge server to the nth mobile device; represents the total video segment transmission time of the nth mobile device at time t; P represents the video segment transmission power; represents the transmission time of the video transmitted from the cloud server to the edge server by the nth mobile device at time t; represents the transmission time of video transmission from the edge server to the mobile device at time t for the nth mobile device; ω1, ω2, and ω3 are three non-negative parameters representing the importance of experience quality; Indicates that the mobile device obtains the video clip from the cloud server; Indicates that the mobile device obtains the video clip from the edge server; b(i n (t)) represents the video segment i n (t) size; i n (t) represents the bitrate version selected by the nth mobile device for the video segment at time t; Indicates the video freeze time of the nth mobile device at time t.

3. A video stream transmission method according to claim 1, characterized in that: The computing power resource constraints are specifically: Where F represents the maximum computing power of the edge server; Represents the computing resources allocated to the nth mobile device at time t.

4. A video stream transmission method according to claim 1, characterized in that: The transmit power constraint is specifically: p n ≤P max ; Among them, p n represents the transmission power allocated by the edge server to the nth mobile device; P max Indicates the maximum transmit power of the edge server.

5. A video stream transmission device, characterized in that: include: Video stream transmission model construction module, data acquisition module, energy efficiency optimization model and constraint condition construction module, energy efficiency optimization model solution module and video stream transmission control module; The video stream transmission model construction module is used to construct a video stream transmission model; wherein the video stream transmission model includes: a cloud server, an edge server and a plurality of mobile devices; the cloud server is connected to the edge server, and the edge server is connected to the plurality of mobile devices; The data acquisition module is configured to acquire basic data of the video stream transmission model; wherein the basic data includes: the maximum computing power of the edge server, the maximum transmission power of the edge server, the video clip transmission rate, the video clip transmission power, the power consumption of components other than the processor in the edge server, the clock cycles of the processor, the channel gain from the edge server to the mobile device, the bit rate of the video clip to be processed by each mobile device at each time, and the bit rate selected by each mobile device for the video clip at each time; The energy efficiency optimization model and constraint condition construction module is used to construct an energy efficiency optimization model based on the basic data with the goal of minimizing the video streaming transmission cost while ensuring the user experience quality; and to construct computing resource constraints and transmission power constraints based on the basic data; The energy efficiency optimization model solving module is used to solve the energy efficiency optimization model under the constraints of computing power resources and transmission power constraints to generate computing power resources and transmission power allocated to each mobile device at each time when minimizing the video streaming transmission cost while ensuring the quality of user experience; The video stream transmission control module is used to control the video stream transmission model to perform video stream transmission according to the computing resources and transmission power allocated to each mobile device at each moment.

6. The video stream transmission device according to claim 5, wherein: The energy efficiency optimization model is specifically: Among them, E av (t) represents the average energy consumption required by N users to transmit video streams at time t; QoE av (t) represents the average quality of experience of N users when transmitting video streams at time t; t represents time t; q represents the power consumption of the components in the edge server except the processor; η represents a scaling factor; ρ represents the number of clock cycles contained in a processor; ε n (t) represents the computing power resources allocated to the nth mobile device at time t when transcoding is required; represents the transcoding time required for the nth user to transmit a video clip at time t; p n represents the transmit power allocated by the edge server to the nth mobile device; represents the total video segment transmission time of the nth mobile device at time t; P represents the video segment transmission power; represents the transmission time of the video transmitted from the cloud server to the edge server by the nth mobile device at time t; represents the transmission time of video transmission from the edge server to the mobile device at time t for the nth mobile device; ω1, ω2, and ω3 are three non-negative parameters representing the importance of experience quality; Indicates that the mobile device obtains the video clip from the cloud server; Indicates that the mobile device obtains the video clip from the edge server; b(i n (t)) represents the video segment i n (t) size; i n (t) represents the bitrate version selected by the nth mobile device for the video segment at time t; Indicates the video freeze time of the nth mobile device at time t.

7. The video stream transmission device according to claim 5, wherein: The computing power resource constraints are specifically: Where F represents the maximum computing power of the edge server; Represents the computing resources allocated to the nth mobile device at time t.

8. The video stream transmission device according to claim 5, wherein: The transmit power constraint is specifically: p n ≤P max ; Among them, p n represents the transmission power allocated by the edge server to the nth mobile device; P max Indicates the maximum transmit power of the edge server.

9. A terminal device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements a video stream transmission method according to any one of claims 1 to 4 when executing the computer program.

10. A computer-readable storage medium, characterized in that The storage medium includes a stored computer program, wherein when the computer program is executed, the device where the storage medium is located is controlled to execute a video stream transmission method according to any one of claims 1 to 4.

Images