A method for transmitting streaming media over a satellite network

Abstract

This invention relates to the field of communications, specifically to a satellite network streaming media transmission method. The method includes receiving an encoded scalable video stream; uniformly dividing the transmission process into time slots; establishing a global virtual queue storing energy thresholds for N time slots and a virtual queue storing satellite uplink and downlink transmission rates for N time slots; establishing a virtual queue for each user storing video playback rate and video downlink transmission rate; at the beginning of each time slot, determining the optimal satellite transmission power, the optimal number of video layers for each user, and the allocated time resources for each user; updating the queue length and battery energy based on the optimal transmission power and optimal number of video layers; and transmitting the video stream to each user based on the optimal number of video layers, optimal transmission power, and allocated time resources. This invention satisfies rate constraints at both the satellite and user ends, avoids satellite buffer overflow, and prevents video playback interruption at the user end, making it a highly efficient video transmission solution.

Description

Technical Field

[0001] This invention relates to the field of communications, and more specifically to a method for transmitting streaming media over a satellite network. Background Technology

[0002] Compared to traditional terrestrial cell networks, low Earth orbit satellite networks can provide a wider coverage area, reducing handover frequency for a large number of users. Furthermore, they offer more flexible access methods to meet emergency communication needs. Due to the widespread use of satellite communication networks, providing real-time and high-quality video streaming via satellite networks is a promising approach. In satellite networks, video streams are uploaded from ground stations to communication satellites and then downloaded from the satellites to mobile user equipment. However, addressing fairness among users in satellite communication networks is a significant challenge due to varying channel quality.

[0003] As part of the H.264 / Advanced Video Coding (AVC) extension, Scalable Video Coding (SVC) can effectively provide satisfactory real-time video transmission under various conditions. SVC technology is well-suited for applications requiring flexible video streaming, breaking down a single-stream video into multiple streams, including a base layer and several enhancement layers. It allows for flexible adjustment of the number of enhancement layers to provide differentiated services to users with different network conditions and channel qualities when transmitting video streams over satellite networks. However, selecting the appropriate number of video layers in real-time is a significant challenge, considering fairness. Furthermore, satellites are typically powered by solar energy, serving as an energy harvesting application. Decisions regarding energy allocation and selecting appropriate power for video transmission are also necessary.

[0004] Existing technologies generally coordinate the control of video rate and energy to meet data-related latency constraints in multi-node wireless networks. For energy-harvested assisted wireless video transmission, the large deviation principle is also used to estimate the probability of energy shortage in video layer selection during single-user transmission. Furthermore, for multi-user energy-harvested assisted wireless communication systems, satellites collect energy and transmit SVC video streams to multiple users, considering fairness among users and fluctuations in video quality during power allocation and video layer selection to ensure smooth playback.

[0005] Existing video transmission optimization technologies mainly consider the downlink network conditions, ignoring the uplink. Furthermore, existing video transmission optimization technologies do not take into account that the buffer on the satellite is also limited, and buffer overflow may cause playback interruption. Summary of the Invention

[0006] To address the above problems, this invention provides a satellite network streaming media transmission method.

[0007] The method includes:

[0008] Step 1: Receive the encoded scalable video stream;

[0009] Step 2: Divide a single transmission process into N time slots, each time slot having a length of T.

[0010] Step 3: Establish a function table relating the average subjective opinion score to the number of video layers, and define the video quality obtained by user k in the nth time slot as q. k,n Define system utility Q n ;

[0011] Step 4: Establish a global virtual queue H to store the energy thresholds for N time slots. n Establish a global virtual queue Y that stores the uplink and downlink transmission rates of N time slots for satellites. n For each user k, a virtual queue z is established to store the video playback rate and downlink transmission rate of user k in the nth time slot. k,n k∈{1,…,K}, where K is the number of users;

[0012] Step 5: At the beginning of each time slot, based on the satellite battery energy and the virtual queue H... n The decision determines the satellite's optimal transmission power;

[0013] Step 6: Based on the user's video playback rate and the virtual queue z k,n Calculate the optimal number of video layers for each user in the queue;

[0014] Step 7, based on the time slot length T and the virtual queue Y n Virtual queue z k,n Calculate the time resources allocated to each user based on the channel conditions of each user;

[0015] Step 8: Update the queue length and battery energy based on the optimal transmission power and optimal video layer number;

[0016] Step 9: Transmit the video stream to each user based on the optimal number of video layers, optimal transmission power, and allocated time resources selected by each user in each time slot.

[0017] Furthermore, step four specifically includes:

[0018] The virtual queue is constructed as follows:

[0019] H n+1 ={H n +θ-E n+1} + ;

[0020] z k,n+1 ={z k,n +r k,n -c k,n} + ;

[0021]

[0022] Where θ represents the energy threshold required for sustainable video transmission, E n+1 r represents the remaining energy of the satellite battery in the (n+1)th time slot. k,n c represents the video playback rate for user k. k,n S represents the video transmission rate of user k. n Represents the uplink transmission rate, {} + This represents the larger value obtained by comparing the expression within the parentheses with 0.

[0023] Furthermore, the relationship between the average subjective opinion score and the video layer number function mentioned in step three is as follows:

[0024] Average subjective opinion score User experience <![CDATA[Total number of received video layers l k,n > 5 Excellent 5 4 good 4 3 generally 3 2 Difference 2 1 Very bad 1

[0025] The video quality q obtained by user k in the nth time slot k,n =λ*MOS, where λ is a predefined constant;

[0026] System utility

[0027] Furthermore, in step five, at the beginning of each time slot, based on the satellite battery energy and the virtual queue H... n The process of determining the optimal transmission power for a satellite includes:

[0028] Energy from satellite batteries and virtual queue H n Calculate the optimal transmission power of the communication satellite at the beginning of each time slot.

[0029]

[0030] Among them, E n This represents the remaining energy of the satellite battery in the nth time slot.

[0031] Furthermore, step six specifically includes:

[0032] Define the optimal number of video layers l for user k in the nth time slot. k,n The decision model is as follows:

[0033] Minimize z k,n r k,n-VQ n / K;

[0034] Subject to l k,n ∈{0,1,…,L};

[0035] Where V is a preset constant and L is a predefined maximum number of video layers.

[0036] Preferably, the maximum number of video layers L is 5.

[0037] Furthermore, step seven specifically includes:

[0038] Calculate channel capacity G k,n :

[0039]

[0040] Where W represents the channel bandwidth, γ k,n N represents the channel gain, and N0 represents the power spectral density.

[0041] According to G k,n Calculate the proportion of time resources (a) allocated to user k in time slot n. k,n :

[0042]

[0043] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0044] This invention designs a more efficient video streaming algorithm that considers both uplink and downlink quality to determine the number of video layers, time resource allocation, and power allocation, thereby avoiding satellite buffer overflow and achieving the best user experience. Attached Figure Description

[0045] Figure 1 A structural diagram of a satellite network SVC video transmission system provided in an embodiment of the present invention;

[0046] Figure 2 The flowchart of the DLSA algorithm provided in the embodiment of the present invention. Detailed Implementation

[0047] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Before describing the technical solutions of each embodiment of the present invention in detail, the terms and terms involved will be explained. In this specification, components with the same name or the same reference numerals represent similar or the same structures and are limited to illustrative purposes.

[0048] Satellites have greatly facilitated communication networks, and SVC technology is widely used in satellite video transmission to improve video service quality. However, network fluctuations during transmission, satellite energy harvesting efficiency, and fairness among multiple users pose significant challenges to satellite decision-making. This invention applies Lyapunov optimization theory to model the SVC video transmission system of a satellite network, solve the optimization problem, and proposes a Dynamic Layer Selection Algorithm (DLSA).

[0049] This algorithm addresses three issues to achieve the best user experience: the number of video layers, time resource allocation, and power allocation. At the beginning of each time slot, it first determines the transmission power and the time resource allocation for each user. Then, it iterates through the number of video layers for each user to select the optimal number of video layers for transmission, repeating this process continuously.

[0050] The following section introduces the system architecture for video transmission using SVC technology in satellite networks, and establishes an energy model and a scalable video streaming model on this system architecture. Finally, a joint optimization problem is proposed.

[0051] like Figure 1 As shown, the SVC video transmission system in the satellite network includes an SVC video server, ground stations, communication satellites, and users. When the SVC video server receives a user request, it encodes the video and transmits the encoded video stream to the communication satellite via a ground base station. The communication satellite receives the video stream from the SVC video server and forwards it to the corresponding user. The communication satellite is powered by solar panels, and its batteries are charged by energy harvesting elements that collect energy from the surrounding solar radiation. The user terminal receives the video stream from the satellite, decodes it, and plays the video.

[0052] This invention assumes that the SVC video stream has L layers, and the number of enhancement layers is L-1. It employs Time Division Multiple Access (TDMA) technology, with a wireless channel bandwidth of W. To achieve real-time video streaming service, SVC video packets queued on the satellite must be transmitted to all users immediately in the next time slot. If the battery has insufficient power, video packets will be discarded, causing playback interruption.

[0053] A single transmission process is divided into multiple time slots, each with a length of T. At the beginning of each time slot, based on the different wireless channel conditions for user k, the amount of energy stored in the battery, and the uplink and downlink transmission rates, the number of video layers to be transmitted is determined. K represents the number of users.

[0054] set up Let C represent the channel gain of the link between the satellite and user k in time slot n, assuming that the fading conditions remain constant during this period, but may vary with time slot changes. Since video data is transmitted to the user in a time-switched manner, this system model uses an additive white Gaussian noise channel model, where C represents the amount of data C transmitted to user k in time slot n. k,n Must obey Shannon's formula:

[0055]

[0056] Among them, t k,n Let N be the transmission time of user k in time slot n, and N0 represent the power spectral density of additive white Gaussian noise for user k. γ represents the transmission power of the nth time slot. k,n This represents channel gain.

[0057] Since the communication satellite uses time division multiple access (TDMA) to transmit video data in this system model, it needs to allocate time resources within a time slot to all users. Here, it is assumed that the communication satellite allocates time resources to users at the beginning of each time slot. Due to the time slot duration constraint, the following time constraints are obtained:

[0058]

[0059] 1. Energy Model

[0060] like Figure 1 As shown, energy is obtained from the environment and stored in a container with a maximum capacity of E. max In the storage battery. Assume and These represent the energy consumed and the energy gained during video transmission in the nth time slot, respectively. Assume the gained energy... It can be considered as an independent and identically distributed variable. At the beginning of time slot n, the remaining energy of the battery is E. n Assume the energy captured in the nth time slot... It can only be used in the next time slot. Based on this, the formula for updating battery energy is:

[0061]

[0062] Communication satellites determine their transmission power at the beginning of each time slot and maintain it within that slot; however, the transmission power may differ between time slots. Therefore, the energy used for video transmission is... in Indicates transmission power.

[0063] Define the long-term time average of battery energy. for:

[0064]

[0065] When playing video, user terminals often choose a lower video quality compared to when playback is interrupted. For communication satellites, transmitting video requires a certain amount of battery power to maintain transmission; otherwise, transmission will be interrupted, resulting in no video reception for user terminals and potentially interrupting playback. Satellites should maintain a certain amount of energy for video transmission to support uninterrupted video streaming. To achieve this, the time-averaged energy... Constrained by a predefined energy threshold θ:

[0066]

[0067] Because high-quality video requires more video layers, it's more likely to deplete the battery, leaving no energy reserves. If the battery reaches zero, transmission will be interrupted, affecting video playback on the user's device, and ultimately severely degrading the user's Quality of Experience (QoE). Therefore... Sustainable video transmission needs to be constrained by θ.

[0068] 2. Scalable video streaming model

[0069] The video stream encoded using SVC technology in this chapter has L layers. Different numbers of video layers correspond to different video qualities. Once the number of video layers selected for transmission by the communication satellite is determined, the video quality transmitted in that time slot is also determined. To ensure uninterrupted video playback, at least one layer, the base layer, must be transmitted.

[0070] Video quality is difficult to evaluate subjectively and requires quantitative assessment. This invention uses the Mean Opinion Score (MOS) for quantitative evaluation. For simplicity, higher video quality also represents a higher user experience quality (QoE). The number of video layers received by user k in time slot n is defined as... Then the video quality of the video received by user k in time slot n is q. k,n Here, q k,n It can be expressed as a function of the MOS value:

[0071] q k,n =λ*MOS;

[0072] Where λ is a predefined constant.

[0073] The higher the number of video layers, the higher the MOS value of the video. This invention uses a 5-point MOS value to evaluate the video, which represents different QoEs. Therefore, the number of video stream layers L is set to 5. Thus, the functional relationship between QoE (MOS value) and the number of video layers is shown in Table 1:

[0074] Table 1. Relationship between QoE (MOS) and Video Layer Number

[0075] Average subjective opinion score User experience <![CDATA[Total number of received video layers l k,n > 5 Excellent 5 4 good 4 3 generally 3 2 Difference 2 1 Very bad 1

[0076] The number of video layers transmitted by a communication satellite may vary between time slots, but remains constant within a time slot. The communication satellite determines the number of video layers to transmit at the beginning of each time slot based on feedback from the network status and channel quality to improve the overall user experience. Let l be the number of video layers transmitted by the communication satellite in time slot n. k,n The corresponding amount of video data sent is used This is represented as follows: Assuming all the transmitted video data is received by the user, then user k has a fixed video playback rate r within time slot n. k,n for:

[0077]

[0078] The fixed video transmission rate c of the communication satellite within time slot n k,n for:

[0079]

[0080] Among them, C k,n This refers to the amount of data transmitted to user k in time slot n, as mentioned earlier.

[0081] For the user terminal, the video player has a playback buffer, which stores some data to prevent playback interruptions due to untimely data reception. To allow the playback buffer to buffer video data, the amount of video data entering the buffer needs to exceed the amount consumed; this translates to a transmission rate exceeding the playback rate. Therefore, the user terminal video playback rate constraint can be derived as follows:

[0082]

[0083] in, This represents the long-term average video playback rate, corresponding to The long-term average video transmission rate can be calculated using the following formula:

[0084]

[0085]

[0086] Where E[] represents taking the expected value of the expression within [], r k,t c represents the fixed video playback rate for user k in the t-th time slot. k,tThis represents the fixed video transmission rate of the communication satellite in the t-th time slot.

[0087] Furthermore, considering the limited storage space on the satellite, the average uplink time-to-sunlight transmission rate from the video server to the satellite should be less than the average downlink time-to-sunlight transmission rate from the satellite to the user; otherwise, it would lead to video buffer overflow on the satellite. This ensures that the storage space on the satellite is always sufficient, and the cached video will not be lost. Therefore, the following communication satellite uplink and downlink transmission rate constraints hold true:

[0088]

[0089] in, This represents the average uplink transmission rate from the video server to the satellite.

[0090] 3. Joint optimization

[0091] The optimization objective of this system model is to achieve the best experience quality for all users, therefore the system utility Q is defined. n The average MOS value for all users in time slot n is defined as follows:

[0092]

[0093] Where, q k,n Let q be the video quality received by user k in time slot n, as mentioned earlier. k,n Therefore, we can define long-run average utility. for:

[0094]

[0095] Among them, Q t This represents the average MOS value of all users within the t-th time slot.

[0096] Finally, the optimization problem is derived. This problem needs to satisfy constraints such as energy threshold, time resource constraints, communication satellite uplink and downlink transmission rates, and user-end video playback rate. The optimization objective is to maximize the time-averaged system utility under the constraints of the following six formulas.

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103] It can be seen that the objective function of the above 6 formulas for the optimization problem is time-averaged, and the constraints are also time-averaged, making it difficult to solve directly. This optimization problem can be solved using Lyapunov stochastic optimization theory.

[0104] Algorithm Design

[0105] This invention uses Lyapunov stochastic optimization theory, firstly by applying the three time-averaged constraints of the original optimization problem. The problem is transformed into a queue stability constraint, which, while ensuring queue stability, becomes a new optimization problem. The next step is to solve this new optimization problem, which can be decomposed into three parts: transmission power allocation, video layer selection, and time resource allocation. These sub-problems are independent of each other, and solutions are provided for each. Finally, based on these solutions, a video transmission algorithm is summarized.

[0106] 1. Lyapunov stochastic optimization model

[0107] The time averaging constraint is difficult to solve directly. It can be transformed into the time averaging constraint of the energy threshold, the time averaging constraint of the uplink and downlink transmission rates of communication satellites, and the time averaging constraint of the video playback rate by constructing K+2 virtual queues. The virtual queue is given by the following formula:

[0108] H n+1 ={H n +θ-E n+1} + ;

[0109] z k,n+1 ={z k,n +r k,n -c k,n} + ;

[0110]

[0111] Where θ represents the energy threshold required for sustainable video transmission, E n+1 r represents the remaining energy of the satellite battery in the (n+1)th time slot. k,n c represents the video playback rate for user k. k,n S represents the video transmission rate of user k. n Represents the uplink transmission rate, {} + This represents the larger value obtained by comparing the expression within the parentheses with 0.

[0112] Using Lyapunov stochastic optimization theory, to ensure that the time-averaged constraints regarding the energy threshold, the uplink and downlink transmission rates of the communication satellite, and the video playback rate at the user end hold, it is necessary to guarantee that the virtual queue H... n , z k,n and Y n It has a stable average rate.

[0113] make To represent a concatenated vector, the Lyapunov function is defined as follows:

[0114]

[0115] Therefore, the corresponding Lyapunov drift can be obtained in time slot n, as shown in the following equation:

[0116]

[0117] To make H n ,z k,n and Y n Since the system is rate-stable, we need to minimize the Lyapunov drift to satisfy the time-averaging constraint. The optimization objective is to maximize system utility. Therefore, according to Lyapunov optimization theory, adding a penalty term for system utility to the Lyapunov drift yields the "drift plus penalty" (DPP). DPP is defined as follows:

[0118] DPP=Δ(Ψ n )-VE[Q n |Ψ n ];

[0119] Minimizing DPP allows the Lyapunov drift E[F(Ψ) to be minimized. n+1 )-F(Ψ n )|Ψ n The goal is to minimize the DPP while maximizing the system's utility. Therefore, solving the original optimization problem requires finding a solution that minimizes the DPP.

[0120] To find the minimum value of DPP, we can first find the upper bound of DPP, and then minimize this upper bound. The following is the derivation of the upper bound of DPP. Let H... n+1 ={H n +θ-E n+1} + Squaring both sides of the corresponding inequality, we get formula A:

[0121]

[0122] For z k,n+1 With Y n+1Performing the same process yields formulas B and C:

[0123]

[0124]

[0125] Substituting formulas A, B, and C into the DPP calculation formula, we obtain formula D:

[0126]

[0127] in and They are a pair of constants because E n H n Y n and S n It is known that it will begin in the nth time slot.

[0128] Minimizing formula D allows us to find the minimum value of DPP, thus solving the original optimization problem. Therefore, the original optimization problem can be transformed into formula E:

[0129]

[0130]

[0131] in,

[0132] 2. Strategies for Video Layer Selection and Resource Allocation

[0133] To solve the new optimization problem, the objective function can be decomposed into three parts, resulting in three sub-problems: transmission power allocation, video layer selection, and time resource allocation. These sub-problems are independent of each other, and solutions are provided for each. Finally, based on these solutions, a video transmission algorithm is summarized.

[0134] 1) Power Allocation

[0135] The optimal choice of power allocation can be obtained by minimizing the first two terms of the objective function in the optimization problem formula E. The transmission power optimization problem is as follows:

[0136]

[0137] in, The energy used for video transmission, as mentioned earlier. This represents the expected value of the captured energy.

[0138] The above expression is a quadratic function, with the independent variable being... The minimum value, or optimal power distribution, can be obtained from the properties of a quadratic function.

[0139]

[0140] 2) Video layer selection

[0141] To derive the strategy for selecting the number of video layers, we need to minimize the terms related to the number of video layers in the objective function of problem formula E, i.e., the middle two terms. For the K users in the video transmission system model, they are independent of each other. Therefore, the optimal number of video layers l transmitted by the communication satellite to user k in the nth time slot is... k,n That is, the solution to the following problems (formulas F and G):

[0142] Minimize z k,n r k,n -VQ n / K;

[0143] Subject to l k,n ∈{0,1,…,L};

[0144] In formula F, the system utility Q n and video playback speed r k,n The number of video layers is determined by the formula F. The key to solving the problem is to choose the appropriate number of video layers. k,n Under normal circumstances, the number of layers in an SVC video stream is not very large. For this system model, the number of video layers L is set to 5, which is finite. Therefore, formula F can be solved simply by using a traversal method. In terms of algorithm complexity, it is only on the order of O(KL) for the entire system, which is not a significant burden on the processing capabilities of communication satellites.

[0145] 3) Time resource allocation

[0146] To derive the optimal time resource allocation strategy, we need to obtain the time resource-related terms in the objective function of formula E, specifically the last two terms. This can be achieved by solving the following optimization problem (formula H):

[0147]

[0148]

[0149] For simplicity, let c k,n =a k,n G k,n a k,n G represents the proportion of time resources allocated to user k in time slot n. k,n For channel capacity, It can be obtained

[0150] Solving the optimization formula H is not difficult; simply select the user with the best channel quality and transmit video data only to them throughout the entire time slot. This maximizes H, but it would prevent users with poor channel quality from receiving video data, increasing user unfairness. Therefore, to ensure user fairness, a suboptimal solution for time resource allocation is proposed here.

[0151] For the user, the virtual queue length z k,n and Y n Larger channels should be allocated more time resources, and channels with better quality should also be allocated more time resources. Therefore, (z) will be allocated more time resources here. k,n +Y n )G k,n User weighting is used for time resource allocation. The proportion of time resources that the satellite needs to allocate to users is:

[0152]

[0153] In particular, when In this scenario, time resources need to be evenly distributed among each user. In terms of algorithmic complexity, it is only on the order of O(K) for the entire system. This is not a significant burden on the processing capacity of the communication satellite, which can make decisions quickly.

[0154] 3. Algorithm DLSA

[0155] The DLSA algorithm flowchart is as follows: Figure 2 As shown. At the start of video transmission, the virtual queue Y is initialized. n and H n The initial values ​​Y1 and H1 are 0. Starting at time slot n, for user k, the virtual queue z is first initialized. k,n initial value z k,1 If the value is 0, the optimal transmission power P is obtained based on the previous transmission power allocation strategy. Then, for each user, all possible video layers are traversed to select and calculate h. k,l h here k,l The definition is as follows:

[0156] h k,l =z k,n r k,n -VQ n / K;

[0157] Compare h k,l This leads to the selection of the minimum number of tiers, l. This is achieved by calculating the time resource allocation ratio 'a' for each user. k,n At the end of each time slot, the length of each virtual queue and the battery power are updated. This process is repeated for each time slot until the transmission is complete.

[0158] This invention studies the SVC video stream transmission problem in satellite networks and proposes the DLSA algorithm. This algorithm can select the number of video layers, allocate time resources, and allocate power during video transmission, greatly improving the user experience. The DLSA algorithm, by applying Lyapunov optimization theory, maximizes the long-term average system utility under constraints of energy threshold, time resources, and video rate, achieving the best user experience quality. Based on this, the maximization problem is decomposed into three independent sub-problems: power allocation, video layer selection, and time resource allocation, and solved separately. Compared with existing solutions, this algorithm considers both uplink and downlink quality in selecting the number of video layers, allocating time resources, and allocating power, satisfying rate constraints at both the satellite and user ends, avoiding satellite buffer overflow, and preventing video playback interruption at the user end, making it a highly efficient video transmission solution.

[0159] The above-described embodiments are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for transmitting streaming media over a satellite network, characterized in that, Includes the following steps: Step 1: Receive the encoded scalable video stream; Step 2: Divide a single transmission process into N time slots, each time slot having a length of T. Step 3: Establish a function table relating the average subjective opinion score to the number of video layers, and define the video quality obtained by user k in the nth time slot as q. k,n Define system utility Q n ; Step 4: Establish a global virtual queue H to store the energy thresholds for N time slots. n Establish a global virtual queue Y that stores the uplink and downlink transmission rates of N time slots for satellites. n For each user k, a virtual queue z is established to store the video playback rate and downlink transmission rate of user k in the nth time slot. k,n k∈{1,…,K}, where K is the number of users; Step 5: At the beginning of each time slot, based on the satellite battery energy and the virtual queue H... n The decision determines the satellite's optimal transmission power; Step 6: Based on the user's video playback rate and the virtual queue z k,n Calculate the optimal number of video layers for each user in the queue; Step 7, based on the time slot length T and the virtual queue Y n Virtual queue z k,n Calculate the time resources allocated to each user based on the channel conditions of each user; Step 8: Update the queue length and battery energy based on the optimal transmission power and optimal video layer number; Step 9: Transmit the video stream to each user based on the optimal number of video layers, optimal transmission power, and allocated time resources selected by each user in each time slot.

2. The satellite network streaming media transmission method as described in claim 1, characterized in that, Step four specifically includes: The virtual queue is constructed as follows: H n+1 ={H n +θ-E n+1 } + ; z k，n+1 ={z k，n +r k，n -c k，n } + ; Where θ represents the energy threshold required for sustainable video transmission, E n+1 r represents the remaining energy of the satellite battery in the (n+1)th time slot. k,n c represents the video playback rate for user k. k,n S represents the video transmission rate of user k. n Represents the uplink transmission rate, {} + This represents the larger value obtained by comparing the expression within the parentheses with 0.

3. The satellite network streaming media transmission method as described in claim 1, characterized in that, The relationship between the average subjective opinion score and the video layer number function described in step three is as follows: The video quality q obtained by user k in the nth time slot k,n =λ*MOS, where λ is a predefined constant; System utility 4. The satellite network streaming media transmission method as described in claim 1, characterized in that, In step five, at the beginning of each time slot, based on the satellite battery energy and the virtual queue H... n The process of determining the optimal transmission power for a satellite includes: Energy from satellite batteries and virtual queue H n Calculate the optimal transmission power of the communication satellite at the beginning of each time slot. Among them, E n This represents the remaining energy of the satellite battery in the nth time slot.

5. The satellite network streaming media transmission method as described in claim 1, characterized in that, Step six specifically includes: Define the optimal number of video layers l for user k in the nth time slot. k,n The decision model is as follows: Minimize z k,n r k,n VQ n / K; Subject to l k，n ∈{0,1,…,L}; Where V is a preset constant and L is a predefined maximum number of video layers.

6. The satellite network streaming media transmission method as described in claim 5, characterized in that, The maximum number of video layers L is 5.

7. The satellite network streaming media transmission method as described in claim 1, characterized in that, Step seven specifically includes: Calculate channel capacity G k, : Where W represents the channel bandwidth, γ k, N represents the channel gain, and N0 represents the power spectral density. According to G k, Calculate the proportion of time resources (a) allocated to user k in time slot n. k, :

Images