Photovoltaic sensing carbon effect optimization method based on base station dormancy and user association

Abstract

The invention discloses a photovoltaic sensing carbon efficiency optimization method based on base station dormancy and user association, and belongs to the field of wireless communication, the method establishes a carbon emission efficiency model of a hybrid energy network and a cellular information network, and the optimization objective of the model is to maximize the carbon emission efficiency of the hybrid energy supply cellular network. On the premise of considering photovoltaic function fluctuation and load flow fluctuation, matching of energy flow and information flow can be completed, and the problems of mismatching of resources and services, low resource utilization rate, high carbon emission and the like are solved; the carbon emission of the network is measured by using an emission factor method, and the carbon dioxide equivalent is taken as a basic unit for measuring the greenhouse effect, so that the improvement of the total energy consumption and intensity regulation and control can be assisted, the mains supply consumption can be controlled, and a total carbon emission and intensity double-control system is gradually turned to.

Description

Technical Field

[0001] The present invention belongs to the field of wireless communications, and more specifically, relates to a photovoltaic-aware carbon efficiency optimization method based on base station sleep and user association. Background Art

[0002] Due to 5G communication's high speed (10Gbps), low latency (1ms), high number of connections (1 million / km²), and high reliability, 5G has been deeply integrated into various vertical industries, with widespread and booming applications. However, the rapid development of global digitalization and information technology, coupled with the increase in user traffic and digital infrastructure, has also made the Information and Communications Technology (ICT) industry a notable energy-intensive and carbon-intensive sector. Research indicates that the ICT industry currently directly contributes approximately 10% to global greenhouse gas emissions, with mobile communication networks contributing over 15%.

[0003] Energy conservation and carbon reduction in mobile communications can be summarized into two aspects: "increasing revenue" and "reducing expenditure." "Increasing revenue" refers to the large-scale introduction of green energy (such as solar and wind energy) to power communications equipment. "Reducing expenditure" refers to the use of intelligent algorithms for rational allocation, minimizing the waste of information resources and energy. The application of "5G base stations + photovoltaics + energy storage" has become an important way to save energy and reduce carbon emissions in digital infrastructure, but this application presents certain challenges. On the one hand, photovoltaic power supply has the disadvantages of intermittency, volatility, and low energy flow density. Using photovoltaic energy alone will cause power supply fluctuations in base stations, seriously affecting their normal operation. On the other hand, the load flow of base stations also has typical temporal and spatial non-uniformity, resulting in dynamic power consumption fluctuations in base stations. This leads to a mismatch between energy flow and information flow, resulting in challenges such as resource and service mismatch, low resource utilization, difficulty in ensuring service quality, high energy consumption, and high carbon emissions. Summary of the Invention

[0004] In response to the above-mentioned defects or improvement needs of the existing technology, the present invention provides a photovoltaic-aware carbon efficiency optimization method based on base station sleep and user association. Considering the cellular network scenario with mixed power supply of photovoltaic and traditional energy, starting from user association and base station sleep, according to the differences in green energy supply of each base station, reasonable base station resource management is carried out to save energy and reduce carbon emissions.

[0005] To achieve the above objectives, according to a first aspect of the present invention, a photovoltaic-aware carbon efficiency optimization method based on base station sleep and user association is provided, which is applied to a hybrid energy supply cellular network. Each base station in the network is preferentially powered by photovoltaics. When photovoltaic energy supply is insufficient, it switches to traditional energy supply. When photovoltaic power is excessive, excess energy is stored in a battery energy storage system. The base stations include macro base stations and micro base stations. The macro base stations are always in an operating state, and the micro base stations can selectively sleep. The method includes:

[0006] Under preset constraints, by maximizing the carbon emission efficiency of the hybrid energy supply cellular network, a connection relationship between base stations and users in the hybrid energy supply cellular network and a dormancy strategy of the base stations are obtained;

[0007] The carbon emission efficiency is the ratio of the total downlink data volume of the hybrid energy cellular network to the total carbon emissions of all base stations; base station B i Carbon emissions E at time slot t c,i (t)E c,i (t) = EF grid ·max{P bs,i (t)-P pv,i (t),0}·Δt+EF pv ·min{P pv,i (t),P bs,i (t)}·Δt; where EF grid EF pv are the carbon emission factors of traditional energy and photovoltaic energy respectively, P pv,i (t), P bs,i (t) are base station B i The photovoltaic function power and the power consumed in time slot t are calculated, and Δt is the time interval for calculating energy consumption. The preset constraints include decision variable value constraints, user-base station connection constraints, base station resource block number upper limit constraints, and user service quality lower limit constraints. The user-base station connection constraints include: users can only connect to base stations in working state, and users can only connect to one base station.

[0008] According to a second aspect of the present invention, there is provided an electronic device comprising: a computer-readable storage medium and a processor;

[0009] The computer-readable storage medium is used to store executable instructions;

[0010] The processor is configured to read the executable instructions stored in the computer-readable storage medium and execute the method according to the first aspect.

[0011] According to a third aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute the method according to the first aspect.

[0012] According to a fourth aspect of the present invention, there is provided a computer program product comprising a computer program or instructions, which implement the method according to the first aspect when executed by a processor.

[0013] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0014] 1. The method provided by the present invention takes into account two-layer base stations in terms of base station types - macro base stations and micro base stations, which can significantly optimize network performance and enhance network efficiency through spectrum reuse and layered interference management. Its flexible deployment characteristics (such as rapid installation on street lights and building exterior walls) reduce the construction cost of urban dense areas, while making up for the signal blind spots in indoor and marginal areas, improving anti-interference capabilities and user experience; a carbon emission efficiency model of a hybrid energy network and a cellular information network is established. The optimization goal of the model is to maximize the carbon emission efficiency of the hybrid energy-supply cellular network. Under the premise of considering the volatility of photovoltaic functions and load flow fluctuations, the energy flow and information flow are matched, solving the problems of resource and service mismatch, low resource utilization, and high carbon emissions; the emission factor method is used to measure the carbon emissions of the network, and carbon dioxide equivalent is used as the basic unit for measuring the greenhouse effect, which can help improve the total energy consumption and intensity regulation, control city electricity consumption, and gradually shift to a "dual control" system of total carbon emissions and intensity.

[0015] 2. The method provided by the present invention uses user online rate to characterize user service quality to meet the user's service quality requirements. It can directly reflect whether the user can use the network service normally, is highly correlated with the user's actual experience, and can also reflect user availability. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a hybrid energy supply cellular network structure provided by an embodiment of the present invention;

[0017] Figure 2 A schematic diagram of the objective function solution process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0019] An embodiment of the present invention provides a photovoltaic-aware carbon efficiency optimization method based on base station sleep and user association, which is applied to a hybrid energy supply cellular network. Each base station in the network is preferentially powered by photovoltaics. When photovoltaic energy supply is insufficient, it switches to traditional energy supply. When photovoltaic power is excessive, excess energy is stored in a battery energy storage system. The base stations include macro base stations and micro base stations. The macro base stations are always in an active state, and the micro base stations can selectively sleep. The method includes:

[0020] Under preset constraints, by maximizing the carbon emission efficiency of the hybrid energy supply cellular network, a connection relationship between base stations and users in the hybrid energy supply cellular network and a dormancy strategy of the base stations are obtained;

[0021] The carbon emission efficiency is the ratio of the total downlink data volume of the hybrid energy cellular network to the total carbon emissions of all base stations. For any base station, its carbon emissions at any time are the sum of the carbon emissions of traditional energy and photovoltaic energy at that time. The calculation formula is: E c,i (t) = EF grid ·max{P bs,i (t)-P pv,i (t),0}·Δt+EF pv ·min{P pv,i (t),P bs,i (t)}·Δt; where EF grid EF pv They are respectively the carbon emission factors of traditional energy and photovoltaic energy, max{P bs,i (t)-P pv,i (t),0}·Δt is the energy consumption of traditional energy;,min{P pv,i (t),P bs,i (t)}·Δt represents the photovoltaic energy consumption, P pv,i (t), P bs,i (t) are base station B i The photovoltaic function power and the power consumed in time slot t, Δt is the time interval for calculating energy consumption;

[0022] The preset constraints include decision variable value constraints, user and base station connection constraints, base station resource block number upper limit constraints and user service quality lower limit constraints; the user and base station connection constraints include: the user can only connect to a base station in a working state, and the user can only connect to one base station.

[0023] The method provided by the present invention considers a two-layer heterogeneous network, such as Figure 1 As shown, it is assumed that there are several macro base stations and micro base stations in the area, represented as in is a macro base station, N m is the number of macro base stations. For micro base stations, N b is the total number of base stations, and the number of micro base stations is N b -N m There are several users in the area at the same time, denoted as U(t), and the number of users at time t is recorded as N u (t), the connection relationship between base stations and users in the network uses the association matrix N u (t) is the total number of users at time t), where x i,j (t) is base station B i With user U j The connection relationship has the following meanings:

[0024]

[0025] At the same time, the base station can selectively sleep. The base station sleep vector Y(t) can be expressed as where y i (t) is base station B i Working / sleeping state,

[0026]

[0027] Because macro base stations ensure regional coverage quality and remain operational at all times, yi(t) = 1, (i = 0, 1, ..., Nm-1). Micro base stations can selectively go dormant. When some micro base stations are dormant, users cannot access the dormant micro base stations and can only access the macro base station or other active micro base stations.

[0028] Each base station is equipped with a photovoltaic system and a battery energy storage system (main power source), with one-way access to traditional energy (backup power source). The base station is primarily powered by photovoltaics. When the photovoltaic power supply is insufficient, it switches to traditional energy. When the photovoltaic power supply is excessive, the excess energy is stored in the battery energy storage system.

[0029] An Intelligent User Association System (IUAS) needs to be deployed within the macro base station. The IUAS is responsible for sensing the service status of each base station in the network and making and implementing user association scheduling decisions based on the energy supply of each base station.

[0030] Considering that base stations typically utilize frequency division multiple access (FDMA), the following illustrates the upper limit on the number of base station resource blocks (RBs) using both macro and micro base stations. It is understood that the upper limit on the number of RBs is established similarly for base stations utilizing resource division schemes such as non-orthogonal multiple access (NOMA), sparse code division multiple access (SCMA), and multi-user shared access (MUSA), and is not further detailed in this article.

[0031] Macro base stations and micro base stations divide the base station's transmission power and bandwidth resources by resource blocks. It is assumed that the transmission power P allocated to each resource block of the base station is r and bandwidth B r The communication resources between the base station and the user are the resource blocks allocated by the base station to the user. The number of resource blocks required for the user to access the base station can be expressed as Among them RB i,j (t) is user U j Access base station B i Number of resource blocks required:

[0032]

[0033] Among them, P i,j (t) is user U j Access base station B i The allocated transmit power, P r is the transmission power of a single resource block. Therefore, the total number of resource blocks occupied by each base station is RB oc,i (t) is the total number of resource blocks occupied by the i-th base station, and the specific expression is:

[0034]

[0035] Where ⊙ represents the Hadamard product of two matrices.

[0036] Because base station resources are limited, the following conditions must be met:

[0037]

[0038] in, is the total number of resource blocks of the macro base station. The total number of resource blocks of each macro base station is equal. The total number of resource blocks of the micro base station is equal to the total number of resource blocks of each micro base station. The total number of resource blocks can be obtained by dividing the maximum transmit power of the base station by the transmit power of a single resource block P r Calculation yields:

[0039] RB max =P max / P r

[0040] in, is the maximum transmit power of the macro base station, is the maximum transmit power of the micro base station.

[0041] Since the downlink traffic in cellular network communication is much larger than the uplink traffic, the present invention only considers the downlink of the cellular network. i,j (t) is:

[0042]

[0043] Among them, h i,j (t) is the small-scale fading between the user and its communication base station (obeying the exponential distribution h~Exp(1) with parameter 1), and the channel matrix is l i,j (t) is the distance between the user and its communication base station (the distance matrix is ), α is the path loss exponent, σ 2 is the additive noise power consumption. According to Shannon’s formula, when user U j Access base station B i When the access information receiving rate R i,j (t) is

[0044] R i,j (t) = B r ·RB i,j (t)·log2(1+SINR i,j (t))

[0045] Among them, B r is the bandwidth of a single resource block. The information reception rate is available The total downlink rate of the network C(t) is

[0046]

[0047] The total amount of network downlink data DT(t) is

[0048]

[0049] In network communications, it is necessary to ensure the quality of user service. User service quality can be characterized by parameters such as throughput, latency, and user online rate. Considering that parameters such as throughput and latency may be one-sided metrics, the user online rate directly indicates whether users can connect to the network and use services normally. Based on this, the method provided by the present invention preferably uses the user online rate to characterize user service quality.

[0050] It can be considered that the signal-to-interference-noise ratio of the information received by the user must meet the threshold requirement. The signal-to-interference-noise ratio threshold requirement for all users is expressed as where θ SINR,j (t) is user U j The signal-to-interference-and-noise ratio requirement.

[0051] The user online rate is the ratio of the number of users who meet the signal-to-interference-noise ratio threshold requirement to the total number of users, using δ ue (t) means,

[0052]

[0053] Among them, N[SINR i,j (t)≥θ SINR,j (t)] represents the number of users that meet the signal-to-interference-noise ratio threshold requirement, N u (t) represents the total number of users. Users only access base stations that can meet the signal-to-interference-noise ratio requirements. X(t) is a dimension of N b ×N u The user association matrix, Is a whole set of 1, dimension N u × 1 column vector. Therefore For an N b ×1 column vector, which represents the number of users that meet the signal-to-interference-and-noise ratio threshold requirements and access each base station. is a 1×N b row vector, so represents the total number of users that meet the signal-to-interference-noise ratio threshold requirement, so therefore:

[0054]

[0055] In cellular networks, the user online rate must meet the network user online rate requirements, namely:

[0056] δ ue (t)≥δ ue,req

[0057] Among them, δ ue,req Online rate requirements for network users.

[0058] Establish an energy consumption carbon emission model. When the base station is working normally, the total power of the base station can be expressed as the sum of dynamic power and static power. When the base station is dormant, the total power of the base station only includes dormant static power. Use the following power model:

[0059]

[0060] Among them, P r is the power value allocated to a single resource block, RB oc,i (t) is base station B i The number of resource blocks occupied, η is the power amplifier efficiency, P work is the working static power, P sleep is the dormant static power, P work >P sleep , base station power It can be expressed as:

[0061]

[0062] in, and They are the static power and sleep power of macro / micro base stations respectively.

[0063] This paper uses the emission factor method to measure a network's carbon emissions. According to ITU-T standards, the carbon emission factor is a conversion factor for carbon emissions of electricity. It represents the greenhouse gas emissions per unit of electricity and is influenced by the energy mix. The emission factor method uses the formula: system carbon emissions = energy consumed by the system to produce carbon emissions × carbon emission factor.

[0064] At the same time, because CO2 is the most common greenhouse gas produced by human activities, the International Organization for Standardization stipulates in ISO 14064-1 that carbon dioxide equivalent (CO 2e ) is the basic unit for measuring greenhouse effect. Therefore, the carbon emissions of base stations can be Indicates that E c,i (t) is base station B i of carbon emissions,

[0065] E c,i (t) = EF grid ·max{P bs,i (t)-P pv,i (t),0}·Δt+EF pv ·min{P pv,i (t),P bs,i (t)}·Δt

[0066] Among them, E c,i (t) is base station B i Carbon emissions at time slot t, P pv,i (t) is base station B i The available energy in time slot t, that is, base station B i Photovoltaic functional efficiency at time slot t, EF grid is the carbon emission factor of traditional energy, EF pv is the photovoltaic carbon emission factor, η pv <<η grid .

[0067] The total carbon emissions E(t) of the cellular network are:

[0068]

[0069] Carbon emission efficiency is defined as the ratio of the total downlink data volume of the cellular network to the total carbon emissions. Carbon efficiency can be calculated by the following formula:

[0070]

[0071] Among them, CE(t) is the carbon emission efficiency, DT(t) is the total data volume of the network downlink, and E(t) is the total carbon emission.

[0072] The total data volume DT(t) is expressed as:

[0073]

[0074] The total data volume DT(t) is a function with X(t) as the independent variable, so it can also be expressed as:

[0075]

[0076] The expression of total carbon emissions E(t) is:

[0077]

[0078] The photovoltaic power supply power P of the system pv (t) is:

[0079] P pv (t) = η pv ·T s (t)⊙S pv

[0080] in, is the solar radiation intensity received by the system, are the photovoltaic panel areas of macro / micro base stations respectively, Base station power P bs (t) is a function with X(t) and Y(t) as independent variables, so it can also be expressed as:

[0081]

[0082] The optimization problem can be expressed as follows: For cellular information networks, the user association status and sleep status of base stations should be adjusted without affecting user service quality. For hybrid energy networks, carbon emission efficiency should be maximized by managing the battery charge and discharge status within base stations and sharing photovoltaic energy between base stations. The network's carbon emission model can be described as follows:

[0083]

[0084]

[0085] δ ue (t)≥δ ue,req (f)

[0086] Among them, constraints (a) and (b) are constraints on the values of associated variables and dormant variables, constraint (c) requires that users can only connect to base stations in working state, constraint (d) requires that users can only connect to one base station, constraint (e) limits the number of base station resource blocks, and constraint (f) is a user service quality constraint, requiring that the user online rate be greater than the online rate requirement.

[0087] The above models include cellular network model, energy consumption and carbon emission model, photovoltaic power supply model and battery energy storage model.

[0088] Solving the above optimization model can obtain the user association with the highest carbon emission efficiency and base station sleep strategy

[0089] The above optimization problem is a mixed integer nonlinear fractional programming problem. It is difficult to solve integer matrix variables. The variables to be solved include the binary user association matrix X(t) and the base station sleep vector Y(t), which are difficult to solve.

[0090] Therefore, an algorithm based on Dinkelbach and concave-convex optimization theory is used to solve the optimization model, which decomposes the problem P into two sub-problems: inner and outer sub-problems, namely:

[0091] 1) Inner layer problem: Assuming that the base station sleep variable Y(t) is known, an algorithm based on concave-convex optimization theory is used to solve the user association variable X(t);

[0092] 2) Outer layer problem: Use Dinkelbach algorithm to solve the base station sleep variable Y(t).

[0093] By alternately solving the inner and outer layers of the two sub-problems, we eventually converge to the optimal solution of the original problem.

[0094] Specifically, x i,j={0,1} constraint is relaxed to x i,j =[0,1]. Since the optimal solution is at the boundary, the relaxation of this constraint does not affect the solution of the problem.

[0095]

[0096] Then the optimization problem can be expressed as:

[0097]

[0098] The network system achieves maximum carbon emission efficiency * (* indicates the optimal variable), if and only if

[0099]

[0100] in,

[0101] The Dinkelbach algorithm can be used to find the optimal solution to the linear fractional programming problem. Therefore, the problem can be equivalently transformed into:

[0102]

[0103] The Dinkelbach-based iterative algorithm is used to solve the optimization problem (P2) to obtain ξ * , and update ξ in the iteration * , and finally achieve convergence to the optimal carbon efficiency.

[0104] For convenience of representation, let:

[0105] g1(X(t))=f1(X(t))-ξ·f2(X(t)),

[0106] g2(X(t))=ξ·f3(X(t)),

[0107] Since g1(X(t)) is a concave objective function, g2(X(t)) is a concave objective function. Since problem (P2) is a convex programming problem, the Convex-Concave Programming (CCP) method can be used to solve it. Based on this method, g2(X(t)) is a local feasible solution X k (t) is expanded by first order Taylor for iterative approximation, and g2(X(t)) is converted into an affine function g2(X(t):X k (t)), thereby transforming the convex difference problem into a series of convex problems, and finally converging to the global optimal solution by repeatedly solving the local optimal solution.

[0108] g2(X(t)) in X kPerform a first-order Taylor expansion at (t) and obtain:

[0109]

[0110] in, represents the inner product of matrices,

[0111]

[0112] Problem (P2) can be transformed into multiple local solution problems, which can be expressed as follows:

[0113]

[0114] Since the constraints of problem (P3) are all linear constraints, g1(X(t)) is a concave function. is an affine function, so the objective function is a concave function. Therefore, problem (P3) is a convex optimization problem.

[0115] Algorithm input includes: photovoltaic power supply power P pv (t), battery energy B soe (t), maximum number of iterations L max , algorithm accuracy ε=10 -3 , and other necessary parameters. Initialize carbon effect ξ m =0, m=1.

[0116] First, solve the inner layer problem and perform a tentative sleep of the base station to obtain the base station sleep vector Y m (t), initialize ξ l (t) = 0, l = 1, substitute into problem (P3), and use the CCP algorithm to solve the user association matrix X l (t), loop and solve until the optimal user association matrix is found. Then solve the outer layer problem and convert X k (t), Y m Substitute (t) into problem (P2) and solve for the carbon emission efficiency ξ m =CE(X l (t),Y m (t)), after trying all base station sleep modes, the loop is exited and the user association and base station sleep strategy with the highest carbon emission efficiency is obtained.

[0117] That is, by alternately solving the inner and outer layers of the two sub-problems, we can finally converge to the optimal solution of the original problem and obtain the user association and base station sleep strategy with the highest carbon emission efficiency. Figure 2 As shown, this is a flow chart of the iterative optimization algorithm for new energy perception provided in the present invention.

[0118] Step 1: Initialize carbon efficiency m=0, m=1.

[0119] Step 2: Solve the inner layer problem and perform a tentative sleep of the base station to obtain the base station sleep vector Y m (t), initialize ξ l (t) = 0, l = 1, substitute into problem (P3), and use the CCP algorithm to solve the user association matrix X l (t), and solve it cyclically until the optimal user association matrix is found.

[0120] Step 3: Solve the outer problem and convert X k (t), Y m Substitute (t) into problem (P2) and solve for the carbon emission efficiency ξ m =CE(X l (t),Y m (t)), change the base station sleep mode and return to step 2. After trying all base station sleep modes, jump out of the loop.

[0121] Step 4: Obtain the user association and base station sleep strategy with the highest carbon emission efficiency.

[0122] To reduce the difficulty of solving the problem, the model problem is decomposed into two sub-problems: an inner layer and an outer layer. The inner layer is solved using the CCP algorithm, and the outer layer is solved using the Dinkelbach algorithm. By alternately solving the inner and outer layers of the two sub-problems, the optimal solution of the original problem is finally converged.

[0123] Those skilled in the art will appreciate that other existing algorithms may also be used to solve the above model, which will not be described in detail here.

[0124] An embodiment of the present invention provides an electronic device, comprising: a computer-readable storage medium and a processor;

[0125] The computer-readable storage medium is used to store executable instructions;

[0126] The processor is configured to read the executable instructions stored in the computer-readable storage medium and execute the method described in any one of the above embodiments.

[0127] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute the method described in any of the above embodiments.

[0128] An embodiment of the present invention provides a computer program product, including a computer program or instructions, which implements the method described in any of the above embodiments when executed by a processor.

[0129] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A photovoltaic-aware carbon efficiency optimization method based on base station sleep and user association is applied to a hybrid energy supply cellular network. Each base station in the network is preferentially powered by photovoltaics. When photovoltaic energy supply is insufficient, it switches to traditional energy supply. When photovoltaic power is excessive, the excess energy is stored in a battery energy storage system. The base stations include macro base stations and micro base stations. The macro base stations are always in operation, and the micro base stations can selectively sleep. The characteristics are: The method comprises: Under preset constraints, by maximizing the carbon emission efficiency of the hybrid energy supply cellular network, a connection relationship between base stations and users in the hybrid energy supply cellular network and a dormancy strategy of the base stations are obtained; The carbon emission efficiency is the ratio of the total downlink data volume of the hybrid energy cellular network to the total carbon emissions of all base stations; base station B i Carbon emissions E at time slot t c,i (t)E c,i (t) = EF grid ·max{P bs,i (t)-P pv,i (t),0}·Δt+EF pv ·min{P pv,i (t),P bs,i (t)}·Δt; where EF grid EF pv are the carbon emission factors of traditional energy and photovoltaic energy respectively, P pv,i (t), P bs,i (t) are base station B i The photovoltaic function power and the power consumed in time slot t are calculated, and Δt is the time interval for calculating energy consumption. The preset constraints include decision variable value constraints, user-base station connection constraints, base station resource block number upper limit constraints, and user service quality lower limit constraints. The user-base station connection constraints include: users can only connect to base stations in working state, and users can only connect to one base station.

2. The method according to claim 1, wherein Setting an objective function, and maximizing the carbon emission efficiency of the hybrid energy supply cellular network by maximizing the objective function; The objective function is: Where X(t) is the connection matrix between the base station and the user, X(t) = [x i,j (t)],i=0,1,...,N m -1,N m ,N m+1 ,...,N b-1 ,j=1,2,…,N u (t), N m is the number of macro base stations, N b is the total number of base stations, N u (t) is the total number of users at time t, Y(t) is the sleep strategy vector of the base station, Y(t)=[y i (t)],y i (t) is base station B i status, The dimension is 1×N b A row vector whose elements are all 1. The dimension is 1×N b A row vector, the elements of which are all 0, B r is the bandwidth of a single resource block, 1 Nu The dimension is 1×N u The row vector of the row vector, the elements of which are all 1, RB(t)=[RB i,j (t)], RB i,j (t) is user U j Access base station B i The number of resource blocks required, R(t) = [R i,j (t)],R i,j (t) is user U j Access base station B i The information receiving rate at .

3. The method according to claim 2, wherein User service quality is characterized by user online rate; User service quality of the hybrid energy cellular network at time t 4. The method according to claim 2 or 3, wherein: When both the macro base station and the micro base station use frequency division multiple access technology, the upper limit of the number of base station resource blocks is constrained as follows: in, is the total number of resource blocks of the macro base station. The total number of resource blocks of each macro base station is equal. is the total number of resource blocks of the micro base station, and the total number of resource blocks of each micro base station is equal.

5. An electronic device, characterized in that: include: Computer-readable storage media and processor; The computer-readable storage medium is used to store executable instructions; The processor is configured to read the executable instructions stored in the computer-readable storage medium and execute the method according to any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute the method according to any one of claims 1 to 4.

7. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images