Hybrid energy supply oriented heterogeneous cellular network resource allocation method and device

Abstract

The present application relates to the heterogeneous cellular network resource allocation method and device for mixed energy supply, belongs to cellular network resource optimization technical field. The heterogeneous cellular network resource allocation method for mixed energy supply includes, based on the mixed energy supply carbon emission of heterogeneous cellular network and cellular network interaction throughput, network resource allocation model is established;Considering the low carbon difference between different energy types when mixed energy supply, in order to improve the cellular network interaction throughput of heterogeneous cellular network, and the mixed energy supply carbon emission is reduced as optimization target, single target fitness function is established;Based on single target fitness function and network resource allocation model, the optimal network resource allocation scheme is obtained. The present application can reduce carbon emission while meeting the optimization network throughput.

Description

Technical Field

[0001] This invention relates to a method and apparatus for resource allocation in heterogeneous cellular networks oriented towards hybrid energy supply, belonging to the field of cellular network resource optimization technology. Background Technology

[0002] Existing cellular network resource allocation schemes typically revolve around issues such as user association, spectrum allocation, power control, and energy consumption optimization. In heterogeneous cellular networks, communication models are generally established between macro base stations, small base stations, and users. Based on factors such as user link quality and base station load, user access relationships and power allocation schemes are optimized to improve system throughput, spectrum efficiency, or energy utilization efficiency. Most existing cellular network resource allocation schemes primarily aim to increase system throughput, spectrum efficiency, or reduce system energy consumption, without considering the impact of hybrid energy-powered cellular networks on carbon emissions or low-carbon performance indicators. Therefore, their resource allocation results not only affect system throughput but also the grid's energy extraction and low-carbon operation effectiveness. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and apparatus for resource allocation in heterogeneous cellular networks with hybrid energy supply. It takes into account the low-carbon differences between different energy types when using hybrid energy supply, and reduces carbon emissions while optimizing network throughput. This solves the problem that current cellular network resource allocation schemes focus on optimizing energy consumption and throughput, but neglect the low-carbon differences between different energy types, resulting in excessive carbon emissions.

[0004] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution:

[0005] This invention provides a method for resource allocation in heterogeneous cellular networks oriented towards hybrid energy supply, comprising:

[0006] A network resource allocation model is established based on the carbon emissions of hybrid energy supply and the cellular network interaction throughput of heterogeneous cellular networks.

[0007] To improve the cellular network interaction throughput of heterogeneous cellular networks and reduce carbon emissions from hybrid energy supply as optimization objectives, a single-objective fitness function is established.

[0008] Based on the single-objective fitness function and the network resource allocation model, the optimal network resource allocation scheme is obtained; the optimal network resource allocation scheme includes the base station-user association scheme and the transmit power allocation scheme for each user.

[0009] Furthermore, the network resource allocation model includes: a hybrid energy supply carbon emission model, a cellular network interaction throughput model, and constraints;

[0010] The carbon emission model for hybrid energy supply is expressed as follows:

[0011] ;

[0012] in, This represents a set of base stations in a heterogeneous cellular network. The set of base stations includes one macro base station and multiple micro base stations. The macro base station is powered by traditional grid energy, while the micro base stations are powered by both traditional grid energy and local clean energy. Indicates carbon emissions from hybrid energy supply; Indicates base station Traditional power grid energy consumption Indicates the carbon emission factor of traditional power grid energy; Indicates base station Local clean energy consumption, Indicates the local clean energy carbon emission factor;

[0013] The cellular network interaction throughput model is expressed as follows:

[0014] ;

[0015] in, This represents a set of users in a heterogeneous cellular network. This represents the total throughput of cellular network interactions; Indicates user Data rate;

[0016] The constraints include power control constraints and signal-to-interference-plus-noise ratio constraints;

[0017] The power control constraint is expressed as:

[0018] ;

[0019] , ;

[0020] in, Indicates base station To users The power distribution Indicates base station To users The lower bound of the allocated power, Indicates base station To users The upper bound of the allocated power; Indicates base station With users The relationship between users Associated with base station hour, ,otherwise, ; Indicates base station Maximum transmit power;

[0021] The signal-to-interference-plus-noise ratio constraint is expressed as:

[0022] ;

[0023] in, Indicates user Signal-to-interference-to-noise ratio, This indicates the minimum quality of service threshold for users.

[0024] Furthermore, the single-objective fitness function, which aims to improve the cellular network interaction throughput of heterogeneous cellular networks and reduce carbon emissions from hybrid energy supply, includes:

[0025] The carbon emission index is constructed using the following formula:

[0026] ;

[0027] in, Indicators representing carbon emissions. This indicates the carbon emissions from mixed energy supply. Indicates a reference value for carbon emissions;

[0028] The interaction throughput metric is constructed using the following formula:

[0029] ;

[0030] in, Indicates the interaction throughput metric. This represents the total throughput of cellular network interactions. This represents a reference value for throughput.

[0031] A single-objective fitness function is established based on carbon emission indicators and interaction throughput indicators;

[0032] The single-objective fitness function is expressed as:

[0033] ;

[0034] in, Represented by fitness value Minimum is the objective; This represents the optimization adjustment coefficient, used to adjust the optimization focus between the total throughput of cellular network interactions and carbon emissions.

[0035] Furthermore, the process of obtaining the optimal network resource allocation scheme based on the single-objective fitness function and the network resource allocation model includes:

[0036] For the relationship between base stations and users and the power allocated by base stations to users in the network resource allocation model, a candidate location solution vector of hybrid discrete-continuous individual coding is constructed, which is expressed as:

[0037] ;

[0038] in, Represents an individual The set of candidate position solution vectors, Represents an individual The random candidate position solution vector, Represents an individual The random mirror candidate position solution vector, Represents an individual The chaotic candidate position solution vector Represents an individual The chaotic mirror candidate position solution vector;

[0039] ;

[0040] in, Represents an individual Users to be scheduled in the random candidate location solution vector The selection results of the two candidate micro base stations , Indicates the total number of users to be scheduled. This indicates the user to be scheduled. Select the first candidate micro base station, if This indicates the user to be scheduled. Select the second candidate micro base station; Represents an individual Base station to user in random candidate location solution vector The power distribution , This represents the total number of users in a heterogeneous cellular network.

[0041] ;

[0042] in, Represents an individual Users to be scheduled in the chaotic candidate position solution vector The selection results of candidate micro base stations Represents an individual In the chaotic candidate location solution vector, the base station to the user The allocated power;

[0043] right , Perform mirroring operations separately to obtain and ;

[0044] Establish a priority evaluation rule for the feasibility of candidate location solution vectors;

[0045] An improved Harris Eagle optimization algorithm is used to perform a population iterative search on candidate position solution vectors. The historical best individual position solution vector and the global best individual position solution vector are selected by a single-objective fitness function and a feasibility priority evaluation rule for candidate position solution vectors.

[0046] If the population iterative search satisfies the preset maximum number of population iterations or the termination condition, then the global optimal individual position solution vector corresponding to the current population iteration round is output, and the optimal network resource allocation scheme is obtained.

[0047] Furthermore, the feasibility priority evaluation rule for the candidate location solution vector includes:

[0048] The formula for calculating individual constraint violation is:

[0049] ;

[0050] in, express Individual constraint violation of any candidate position solution vector; This represents the function that takes the maximum value.

[0051] like If the candidate position solution vector is feasible, then the solution vector is feasible. If the candidate position solution vector is not feasible, then the solution vector is not feasible.

[0052] For any two candidate position solution vectors, select the feasible candidate position solution vector;

[0053] If any two candidate position solution vectors are feasible, then the feasible candidate position solution vector with the smaller target fitness function is selected.

[0054] If neither of the two candidate position solution vectors is feasible, then the candidate position solution vector with the smaller individual constraint violation is selected.

[0055] Furthermore, the improved Harris Eagle optimization algorithm is used to perform a population-based iterative search on candidate position solution vectors. The historically optimal individual position solution vector and the globally optimal individual position solution vector are selected through a single-objective fitness function and a feasibility-priority evaluation rule for candidate position solution vectors. This includes:

[0056] During the iterative search process of the population, the prey escape energy is calculated using the following formula: ;

[0057] in, Indicates the energy of prey escaping. Indicates a random initial value. Indicates the iteration round, Indicates the maximum number of iterations for the preset population;

[0058] Based on the prey escape energy, the individual's position in the population is updated to obtain the individual's initial position solution vector;

[0059] By adopting the feasibility-first evaluation rule of candidate position solution vectors, the initial position solution vectors are modified to obtain the historical best individual position solution vector and the global best individual position solution vector that satisfy the single-objective fitness function.

[0060] Furthermore, the step of updating the individual's position in the population based on prey escape energy to obtain the individual's initial position solution vector includes:

[0061] like Then it enters the global search phase and generates Uniform random numbers within the interval ;

[0062] like Then, the individual position is updated by referring to the candidate position solution vector randomly selected from the population, as expressed as: ;

[0063] in, Represents an individual The initial position solution vector after position update in the current iteration round. Represents an individual The solution vector at the position at the start of the current iteration round. This means randomly selecting individuals from the population. The candidate position solution vector; , express Uniformly distributed random numbers within the interval are used to generate the search step size;

[0064] like Then, the individual position is updated by referring to the mean of the global optimal individual position solution vector and the position solution vector of the individual in the current iteration, as expressed as:

[0065] ;

[0066] in, This represents the global optimal individual position solution vector in the current iteration round. The mean of the positional solution vectors representing the positions of individuals in the population during the current iteration. , express Uniformly distributed random numbers within the interval are used to generate the search step size; This represents the upper bound of the elements in the solution vector of an individual's position. This represents the lower bound of the elements in the solution vector for an individual's position;

[0067] like Then it enters the local search phase and generates Uniform random numbers within the interval ;

[0068] like and In this case, soft capture is used to update the individual's position, as shown below:

[0069] ;

[0070] in, express Uniformly distributed random numbers within the interval;

[0071] like and Then, hard capture is used to update the individual's position, represented as:

[0072] ;

[0073] like Then the first candidate position solution vector is generated. Second candidate position solution vector ;

[0074] ;

[0075] ;

[0076] in, This represents the standard normal random number generation function;

[0077] The feasibility-first evaluation rule for candidate location solution vectors is adopted. , and The positional solution vectors that meet the requirements are selected from the options and used as... .

[0078] Furthermore, the step of adopting a feasibility-priority evaluation rule for candidate location solution vectors to revise the initial location solution vectors and obtain the historically optimal individual location solution vector and the globally optimal individual location solution vector that satisfy the single-objective fitness function includes:

[0079] If the globally optimal individual position solution vector is not feasible in the current iteration, then the initial position solution vector is modified as follows: Make corrections:

[0080] ;

[0081] ;

[0082] in, , express Users awaiting scheduling The selection results of the two candidate micro base stations express China base station to users The allocated power; Indicate each The corrected result Represents an individual The positional solution vector at the start of the current iteration. The power allocation from the base station to the user This indicates the upper bound of the power allocated by the base station to the user. This indicates the lower bound of the power allocated by the base station to the user; Indicate each The corrected result Represents an individual The historical best individual position solution vector. express Zhongyu Elements corresponding to dimensions This represents the global optimal individual position solution vector in the current iteration round. express Zhongyu Elements corresponding to dimensions; This means generating a random number between 0 and 1. This represents the rounding function;

[0083] If the global optimal individual position solution vector is feasible in the current iteration, then the initial position solution vector is processed as follows: Make corrections:

[0084] ;

[0085] ;

[0086] in, express Zhongyu Elements corresponding to dimensions express Zhongyu Elements corresponding to dimensions.

[0087] Furthermore, the step of adopting a feasibility-priority evaluation rule for candidate location solution vectors to revise the initial location solution vectors and obtain the historically optimal individual location solution vector and the globally optimal individual location solution vector that satisfy the single-objective fitness function also includes:

[0088] If the number of consecutive times the global optimal individual position solution vector is not updated during the population iterative search process meets the preset stagnation threshold, then select a portion of individuals to form an individual perturbation set, and add perturbation to the individual perturbation set;

[0089] If the globally optimal individual position solution vector is feasible in the current iteration, then the initial position solution vector in the individual perturbation set is perturbed and updated in the following way:

[0090] ;

[0091] in, Represents an individual The initial position solution vector after perturbation update This represents the global optimal individual position solution vector in the current iteration round. This represents the upper bound of the elements in the solution vector of an individual's position. This represents the lower bound of the elements in the solution vector for an individual's position. This represents the standard normal random number generation function;

[0092] If the globally optimal individual position solution vector is not feasible in the current iteration, then the initial position solution vector in the individual perturbation set is perturbed and updated in the following way:

[0093] ;

[0094] in, Represents an individual The historical optimal individual position solution vector.

[0095] Another aspect of the present invention provides a heterogeneous cellular network resource allocation device for hybrid energy supply, comprising:

[0096] The network resource allocation model construction module is used to establish a network resource allocation model based on the carbon emissions of hybrid energy supply and the interaction throughput of cellular networks in heterogeneous cellular networks.

[0097] The single-objective fitness function construction module is used to establish a single-objective fitness function with the optimization objectives of improving the cellular network interaction throughput of heterogeneous cellular networks and reducing the carbon emissions of hybrid energy supply.

[0098] The model solving module is used to obtain the optimal network resource allocation scheme based on the single-objective fitness function and the network resource allocation model; the optimal network resource allocation scheme includes the base station-user association scheme and the transmit power allocation scheme for each user.

[0099] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0100] 1. This invention establishes a network resource allocation model based on the carbon emissions of hybrid energy supply in heterogeneous cellular networks and the cellular network interaction throughput. The optimization objective is to improve the cellular network interaction throughput of heterogeneous cellular networks and reduce the carbon emissions of hybrid energy supply. A single-objective fitness function is established, considering the low-carbon differences between different energy types when using hybrid energy supply. This reduces carbon emissions while simultaneously optimizing network throughput. Based on the single-objective fitness function and the network resource allocation model, the optimal network resource allocation scheme is obtained, solving the problem that current cellular network resource allocation schemes focus on optimizing energy consumption and throughput while neglecting the low-carbon differences between different energy types, leading to excessive carbon emissions.

[0101] 2. This invention uses an improved Harris Eagle optimization algorithm to solve the network resource allocation model. It enhances the initial position of individuals in the population by generating a set of candidate position solution vectors, thereby introducing low-power and low-energy-consumption candidate solutions in the initialization stage, resulting in a higher quality initial population.

[0102] 3. This invention also introduces a constraint-guided two-stage search mechanism and a stagnation detection perturbation mechanism. Through the two-stage search of infeasible search and feasible region fine search, the algorithm can preferentially enter the feasible region under strong constraints, avoiding the situation where the algorithm takes too long to search for infeasible solutions. Through stagnation detection perturbation, the selected individual performs a local search around the global optimal position in the feasible region, and performs a larger-scale search around the individual's historical optimal position in the infeasible region, reducing the risk of the algorithm getting stuck in local optima or stagnation in the later stages of the search. Attached Figure Description

[0103] Figure 1 This is a flowchart of a heterogeneous cellular network resource allocation method for hybrid energy supply provided in an embodiment of the present invention;

[0104] Figure 2 This is a schematic diagram of a heterogeneous cellular network provided in an embodiment of the present invention;

[0105] Figure 3 This is a flowchart illustrating the process of solving the optimal network resource allocation scheme provided in an embodiment of the present invention;

[0106] Figure 4 This is a comparison chart of the average convergence curves of different algorithms provided in the embodiments of the present invention;

[0107] Figure 5 This is a comparison chart of the average total throughput of the system under different algorithms provided in the embodiments of the present invention;

[0108] Figure 6 This is a comparison chart of the average grid power supply of the system under different algorithms provided in the embodiments of the present invention;

[0109] Figure 7 This is a comparison chart of system carbon emission indicators under different algorithms provided in the embodiments of the present invention;

[0110] Figure 8 This is a schematic diagram of the structure of a heterogeneous cellular network resource allocation device for hybrid energy supply provided in an embodiment of the present invention; Among them, 1 represents macro base stations, 2 represents micro base stations, 3 represents traditional grid energy, 4 represents local clean energy, and 5 represents users. Detailed Implementation

[0111] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0112] Example 1

[0113] like Figure 1 As shown, this embodiment provides a resource allocation method for heterogeneous cellular networks oriented towards hybrid energy supply, including:

[0114] Step 1: Based on the carbon emissions of hybrid energy supply in heterogeneous cellular networks and the throughput of cellular network interactions, establish a network resource allocation model; specifically:

[0115] like Figure 2 As shown, in this embodiment, the heterogeneous cellular network includes one macro base station 1 and... Micro base stations 2; macro base stations 1 are responsible for providing wide-area coverage and basic access guarantee. Micro base stations 2 are densely deployed within the coverage area of ​​macro base stations 1 to improve system capacity and spatial reuse efficiency. Macro base stations 1 are powered by traditional grid energy 3, and micro base stations 2 are powered by traditional grid energy 3 and local clean energy 4. Local clean energy 4 can come from renewable energy sources such as solar energy and wind energy, and its available energy varies in each discrete time slot.

[0116] The set of base stations is then represented as , Where, number 0 represents a macro base station, and number... Indicates micro base station, Indicates the total number of micro base stations;

[0117] The set of users 5 within the coverage area of ​​the base station is represented as follows: : , This represents the total number of users.

[0118] Network resource allocation models include: a hybrid energy supply carbon emission model, a cellular network interaction throughput model, and constraints.

[0119] The carbon emission model for hybrid energy supply is expressed as follows:

[0120] ;

[0121] in, Indicates carbon emissions from hybrid energy supply; Indicates base station Traditional power grid energy consumption Indicates the carbon emission factor of traditional power grid energy; Indicates base station Local clean energy consumption, Indicates the local clean energy carbon emission factor;

[0122] ;

[0123] in, Indicates base station Total transmission energy consumption within a time slot Indicates base station Clean energy consumption available within a time slot This represents the function that takes the maximum value.

[0124] ;

[0125] in, Indicates base station With users The relationship between users Associated with base station hour, ,otherwise, ; Indicates base station To users The allocated power;

[0126] ;

[0127] in, This represents a function that takes the minimum value.

[0128] The cellular network interaction throughput model is expressed as:

[0129] ;

[0130] in, This represents the total throughput of cellular network interactions; Indicates user Data rate;

[0131] ;

[0132] in, This indicates the bandwidth occupied by the user on the subcarrier; Indicates user Signal-to-interference-plus-noise ratio;

[0133] ;

[0134] in, Indicates base station To users Channel gain, Indicates other base stations In users Occupy the interference power on the subcarrier Indicates user The power of additive white Gaussian noise;

[0135] Constraints include power control constraints and signal-to-interference-plus-noise ratio (SINR) constraints;

[0136] Power control constraints are expressed as:

[0137] ;

[0138] , ;

[0139] in, Indicates base station To users The lower bound of the allocated power, Indicates base station To users The upper bound of the allocated power; Indicates base station Maximum transmit power;

[0140] The signal-to-interference-plus-noise ratio (SINR) constraint is expressed as: ;

[0141] in, This indicates the minimum quality of service threshold for users.

[0142] Step 2: Establish a single-objective fitness function with the optimization objectives of improving the cellular network interaction throughput of heterogeneous cellular networks and reducing carbon emissions from hybrid energy supply; specifically:

[0143] The carbon emission index is constructed using the following formula: ;

[0144] in, Indicators representing carbon emissions. Indicates a reference value for carbon emissions;

[0145] The interaction throughput metric is constructed using the following formula: ;

[0146] in, Indicates the interaction throughput metric. This represents a reference value for throughput.

[0147] A single-objective fitness function is established based on carbon emission indicators and interaction throughput indicators;

[0148] The single-objective fitness function is expressed as: ;

[0149] in, Represents a single-objective fitness function; This indicates the optimization adjustment coefficient. This is used to adjust the optimization focus between the total throughput of cellular network interactions and carbon emissions;

[0150] In this embodiment, The value is 0.5.

[0151] like Figure 3 As shown, step 3 involves solving the network resource allocation model based on the single-objective fitness function to obtain the optimal network resource allocation scheme; specifically:

[0152] For the relationship between base stations and users and the power allocated by base stations to users in the network resource allocation model, a candidate location solution vector of hybrid discrete-continuous individual coding is constructed, which is expressed as:

[0153] ;

[0154] in, Represents an individual The set of candidate position solution vectors, Represents an individual The random candidate position solution vector, Represents an individual The random mirror candidate position solution vector, Represents an individual The chaotic candidate position solution vector Represents an individual The chaotic mirror candidate position solution vector;

[0155] ;

[0156] in, Represents an individual Users to be scheduled in the random candidate location solution vector The selection results of the two candidate micro base stations , Indicates the total number of users to be scheduled. This indicates the user to be scheduled. Select the first candidate micro base station, if This indicates the user to be scheduled. Select the second candidate micro base station; Represents an individual Base station to user in random candidate location solution vector The power distribution , This represents the total number of users in a heterogeneous cellular network.

[0157] right Individuals generated using chaotic mapping Chaotic candidate position solution vector ;

[0158] Generated using chaotic mapping sequence values ​​within the interval and , Represents an individual Users to be scheduled The sequence values ​​used in the chaotic mapping Represents an individual user The sequence values ​​used in the chaotic mapping;

[0159] ;

[0160] ;

[0161] in, Represents an individual Base station to user in random candidate location solution vector The lower bound of the allocated power, Represents an individual Base station to user in random candidate location solution vector The upper limit of the power allocation, Indicates the power bias adjustment parameter;

[0162] ;

[0163] in, Represents an individual Users to be scheduled in the chaotic candidate position solution vector The selection results of candidate micro base stations Represents an individual In the chaotic candidate location solution vector, the base station to the user The allocated power;

[0164] right , Perform mirroring operations separately to obtain and ;

[0165] Establish a priority evaluation rule for the feasibility of candidate location solution vectors, including:

[0166] The formula for calculating individual constraint violation is:

[0167] ;

[0168] in, express Individual constraint violation of any candidate position solution vector;

[0169] like If the candidate position solution vector is feasible, then the solution vector is feasible. If the candidate position solution vector is not feasible, then the solution vector is not feasible.

[0170] For any two candidate position solution vectors, select the feasible candidate position solution vector;

[0171] If any two candidate position solution vectors are feasible, then the feasible candidate position solution vector with the smaller target fitness function is selected.

[0172] If neither of the two candidate position solution vectors is feasible, then the candidate position solution vector with the smaller individual constraint violation is selected.

[0173] An improved Harris Eagle optimization algorithm is used to perform a population-based iterative search on candidate location solution vectors. Historically optimal and globally optimal individual location solution vectors are selected through a single-objective fitness function and a feasibility-priority evaluation rule for candidate location solution vectors. This includes:

[0174] During the iterative search process of the population, the prey escape energy is calculated using the following formula: ;

[0175] in, Indicates the energy of prey escaping. Indicates a random initial value. Indicates the iteration round, Indicates the maximum number of iterations for the preset population;

[0176] Based on prey escape energy, the individual's position in the population is updated to obtain the individual's initial position solution vector; specifically:

[0177] like Then it enters the global search phase and generates Uniform random numbers within the interval ;

[0178] like Then, the individual position is updated by referring to the candidate position solution vector randomly selected from the population, as expressed as: ;

[0179] in, Represents an individual The initial position solution vector after position update in the current iteration round. Represents an individual The solution vector at the position at the start of the current iteration round. This means randomly selecting individuals from the population. The candidate position solution vector; , express Uniformly distributed random numbers within the interval are used to generate the search step size;

[0180] like Then, the individual position is updated by referring to the mean of the global optimal individual position solution vector and the position solution vector of the individual in the current iteration, as expressed as:

[0181] ;

[0182] in, This represents the global optimal individual position solution vector in the current iteration round. The mean of the positional solution vectors representing the positions of individuals in the population during the current iteration. , express Uniformly distributed random numbers within the interval are used to generate the search step size; This represents the upper bound of the elements in the solution vector of an individual's position. This represents the lower bound of the elements in the solution vector for an individual's position;

[0183] like Then it enters the local search phase and generates Uniform random numbers within the interval ;

[0184] like and In this case, soft capture is used to update the individual's position, as shown below:

[0185] ;

[0186] in, express Uniformly distributed random numbers within the interval;

[0187] like and Then, hard capture is used to update the individual's position, represented as:

[0188] ;

[0189] like Then the first candidate position solution vector is generated. Second candidate position solution vector ;

[0190] ;

[0191] ;

[0192] in, This represents the standard normal random number generation function;

[0193] The feasibility-first evaluation rule for candidate location solution vectors is adopted. , and The positional solution vectors that meet the requirements are selected from the options and used as... .

[0194] The feasibility-first evaluation rule for candidate location solution vectors is adopted to revise the initial location solution vectors, resulting in the historically optimal individual location solution vector and the globally optimal individual location solution vector that satisfy the single-objective fitness function; specifically:

[0195] This embodiment introduces a constraint-guided two-stage search mechanism:

[0196] If the globally optimal individual position solution vector in the current iteration is infeasible, then the infeasibility search phase begins, and the initial position solution vector is processed as follows: Make corrections:

[0197] ;

[0198] ;

[0199] in, , express Users awaiting scheduling The selection results of the two candidate micro base stations express China base station to users The allocated power; Indicate each The corrected result Represents an individual The positional solution vector at the start of the current iteration. The power allocation from the base station to the user This indicates the upper bound of the power allocated by the base station to the user. This indicates the lower bound of the power allocated by the base station to the user; Indicate each The corrected result Represents an individual The historical best individual position solution vector. express Zhongyu Elements corresponding to dimensions express Zhongyu Elements corresponding to dimensions; This means generating a random number between 0 and 1. This represents the rounding function;

[0200] If the globally optimal individual position solution vector is feasible in the current iteration, then the process enters the refined search phase of the feasible region, and the initial position solution vector is processed as follows: Make corrections:

[0201] ;

[0202] ;

[0203] in, express Zhongyu Elements corresponding to dimensions express Zhongyu Elements corresponding to dimensions.

[0204] This embodiment also introduces a stagnation detection perturbation mechanism:

[0205] Set a stagnation counter , The update rule is expressed as:

[0206] ;

[0207] If the global optimal individual position solution vector has not been updated for a certain number of consecutive times during the population iterative search process... If the preset threshold for the number of pauses is met, select a subset of individuals to form an individual perturbation set, and add perturbation to the individual perturbation set;

[0208] If the globally optimal individual position solution vector is feasible in the current iteration, then the initial position solution vector in the individual perturbation set is perturbed and updated in the following way:

[0209] ;

[0210] in, Represents an individual The initial position solution vector after perturbation update;

[0211] If the globally optimal individual position solution vector is not feasible in the current iteration, then the initial position solution vector in the individual perturbation set is perturbed and updated in the following way:

[0212] ;

[0213] If the population iterative search satisfies the preset maximum number of population iterations or the termination condition, then the global optimal individual position solution vector corresponding to the current population iteration round is output to obtain the optimal network resource allocation scheme.

[0214] The optimal network resource allocation scheme output in this embodiment includes the base station-user association scheme and the transmit power allocation scheme for each user;

[0215] Based on the optimal network resource allocation scheme, we can obtain the corresponding total system throughput, grid power supply for each base station, clean energy utilization, carbon emission index, interaction throughput index, constraint satisfaction, and final fitness function value.

[0216] This embodiment uses the following parameters for simulation experiments:

[0217] The simulation system adopts a heterogeneous cellular network downlink communication scenario, with one macro base station and nine micro base stations, totaling 59 users. The maximum transmit power of the macro base station is set to 30dBm, the maximum transmit power of the micro base stations is set to 26dBm, and the user receive sensitivity threshold is set to -94dBm; the carrier frequency is 2000MHz. The system spectrum resources are based on RBs as the basic unit, with each RB having a bandwidth of 180kHz, for a total of 110 RBs, resulting in a total system bandwidth of approximately 20MHz; the minimum SINR threshold for users is set to 6. The macro base station is entirely powered by the traditional power grid, while the micro base stations prioritize the use of local clean energy, with any shortfall supplemented by the power grid.

[0218] The available clean energy consumption for each base station is set as follows:

[0219] ;

[0220] in, The first element corresponds to the macro base station, with a value of 0, and the remaining 9 elements correspond to the 9 micro base stations;

[0221] The comparison algorithms employed Particle Swarm Optimization (PSO) and the standard Harris-Hawk optimization algorithm (HHO); the population size was set to 30, and the maximum number of iterations was set to 800. To reduce the influence of randomness, each of the three algorithms underwent 20 independent experiments.

[0222] like Figure 4 As shown in the figure, the vertical axis represents constraint-aware fitness, which is used to uniformly describe the comprehensive performance of the algorithm throughout the search process: when the current best individual satisfies the constraints, the constraint-aware fitness is equal to the fitness function value; when the current best individual is infeasible, a penalty-type evaluation value is constructed based on the fitness function and the total constraint violation. According to this definition, even if the algorithm does not obtain a feasible solution in the early stage, the convergence curve can continuously reflect the entire process of its evolution from infeasible search to feasible region optimization. Therefore, the lower the curve value, the better the comprehensive performance of the algorithm in terms of both feasibility and objective optimization. From the average convergence curve, the evolutionary characteristics of the three algorithms in the search process are significantly different. The method proposed in this embodiment drops rapidly in the early stage of iteration, and enters a stable plateau region after about 80 to 100 generations. Thereafter, it only changes by a very small margin until 800 generations, indicating that it can transition from the early search to the later stable optimization relatively early and has high convergence efficiency.

[0223] like Figure 5 , Figure 6 and Figure 7 As shown, the average total throughput of PSO and HHO is quite similar. The method proposed in this embodiment is slightly lower than the former two in this indicator, but the overall change is not significant. According to the simulation results, the grid energy efficiency of PSO, HHO and the method proposed in this embodiment are 8.001 Mbit / s / W, 5.363 Mbit / s / W and 9.816 Mbit / s / W, respectively. This indicates that the method proposed in this embodiment performs better in terms of data transmission capability per unit grid power supply and has higher resource utilization efficiency.

[0224] Further comparison of the method proposed in this embodiment with the standard HHO reveals that the method proposed in this embodiment reduces the average grid power supply by 47.06% and carbon emissions by 46.28%, indicating that the improved algorithm can more effectively reduce grid consumption and carbon costs while maintaining the basic communication performance of the system.

[0225] Example 2

[0226] like Figure 8 As shown, this embodiment provides a heterogeneous cellular network resource allocation device for hybrid energy supply, including:

[0227] The network resource allocation model construction module is used to establish a network resource allocation model based on the carbon emissions of hybrid energy supply and the interaction throughput of cellular networks in heterogeneous cellular networks.

[0228] The single-objective fitness function construction module is used to establish a single-objective fitness function with the optimization objectives of improving the cellular network interaction throughput of heterogeneous cellular networks and reducing the carbon emissions of hybrid energy supply.

[0229] The model solving module is used to obtain the optimal network resource allocation scheme based on the single-objective fitness function and the network resource allocation model. The optimal network resource allocation scheme includes the base station and user association scheme and the transmit power allocation scheme for each user.

[0230] The specific functions of each module described above are explained in the relevant content of the method in Embodiment 1, and will not be repeated here.

[0231] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0232] This invention is described with reference to flowchart illustrations of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each step in the flowchart can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, generate instructions for implementing the process. Figure 1 One or more processes or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0233] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 The function specified in one or more processes.

[0234] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 Steps of a specified function in one or more processes.

[0235] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A resource allocation method for heterogeneous cellular networks oriented towards hybrid energy supply, characterized in that, include: A network resource allocation model is established based on the carbon emissions of hybrid energy supply and the cellular network interaction throughput of heterogeneous cellular networks. To improve the cellular network interaction throughput of heterogeneous cellular networks and reduce carbon emissions from hybrid energy supply as optimization objectives, a single-objective fitness function is established. Based on the single-objective fitness function and the network resource allocation model, the optimal network resource allocation scheme is obtained; the optimal network resource allocation scheme includes the base station-user association scheme and the transmit power allocation scheme for each user.

2. The resource allocation method for heterogeneous cellular networks oriented towards hybrid energy supply according to claim 1, characterized in that, The network resource allocation model includes: a hybrid energy supply carbon emission model, a cellular network interaction throughput model, and constraints. The carbon emission model for hybrid energy supply is expressed as follows: ； in, This represents a set of base stations in a heterogeneous cellular network. The set of base stations includes one macro base station and multiple micro base stations. The macro base station is powered by traditional grid energy, while the micro base stations are powered by both traditional grid energy and local clean energy. Indicates carbon emissions from hybrid energy supply; Indicates base station Traditional power grid energy consumption Indicates the carbon emission factor of traditional power grid energy; Indicates base station Local clean energy consumption, Indicates the local clean energy carbon emission factor; The cellular network interaction throughput model is expressed as follows: ； in, This represents a set of users in a heterogeneous cellular network. This represents the total throughput of cellular network interactions; Indicates user Data rate; The constraints include power control constraints and signal-to-interference-plus-noise ratio constraints; The power control constraint is expressed as: ； ， ； in, Indicates base station To users The power distribution Indicates base station To users The lower bound of the allocated power, Indicates base station To users The upper bound of the allocated power; Indicates base station With users The relationship between users Associated with base station hour, ,otherwise, ; Indicates base station Maximum transmit power; The signal-to-interference-plus-noise ratio constraint is expressed as: ； in, Indicates user Signal-to-interference-to-noise ratio, This indicates the minimum quality of service threshold for users.

3. The resource allocation method for heterogeneous cellular networks oriented towards hybrid energy supply according to claim 1, characterized in that, The optimization objective is to improve the cellular network interaction throughput of heterogeneous cellular networks and reduce carbon emissions from hybrid energy supply. A single-objective fitness function is established, including: The carbon emission index is constructed using the following formula: ； in, Indicators representing carbon emissions. This indicates the carbon emissions from mixed energy supply. Indicates a reference value for carbon emissions; The interaction throughput metric is constructed using the following formula: ； in, Indicates the interaction throughput metric. This represents the total throughput of cellular network interactions. This represents a reference value for throughput. A single-objective fitness function is established based on carbon emission indicators and interaction throughput indicators; The single-objective fitness function is expressed as: ； in, Represented by fitness value Minimum is the objective; This represents the optimization adjustment coefficient, used to adjust the optimization focus between the total throughput of cellular network interactions and carbon emissions.

4. The resource allocation method for heterogeneous cellular networks oriented towards hybrid energy supply according to claim 2, characterized in that, The optimal network resource allocation scheme, based on a single-objective fitness function and a network resource allocation model, is obtained as follows: For the relationship between base stations and users and the power allocated by base stations to users in the network resource allocation model, a candidate location solution vector of hybrid discrete-continuous individual coding is constructed, which is expressed as: ； in, Represents an individual The set of candidate position solution vectors, Represents an individual The random candidate position solution vector, Represents an individual The random mirror candidate position solution vector, Represents an individual The chaotic candidate position solution vector Represents an individual The chaotic mirror candidate position solution vector; ； in, Represents an individual Users to be scheduled in the random candidate location solution vector The selection results of the two candidate micro base stations , Indicates the total number of users to be scheduled. This indicates the user to be scheduled. Select the first candidate micro base station, if This indicates the user to be scheduled. Select the second candidate micro base station; Represents an individual Base station to user in random candidate location solution vector The power distribution , This represents the total number of users in a heterogeneous cellular network. ； in, Represents an individual Users to be scheduled in the chaotic candidate position solution vector The selection results of candidate micro base stations Represents an individual In the chaotic candidate location solution vector, the base station to the user The allocated power; right , Perform mirroring operations separately to obtain and ; Establish a priority evaluation rule for the feasibility of candidate location solution vectors; An improved Harris Eagle optimization algorithm is used to perform a population iterative search on candidate position solution vectors. The historical best individual position solution vector and the global best individual position solution vector are selected by a single-objective fitness function and a feasibility priority evaluation rule for candidate position solution vectors. If the population iterative search satisfies the preset maximum number of population iterations or the termination condition, then the global optimal individual position solution vector corresponding to the current population iteration round is output, and the optimal network resource allocation scheme is obtained.

5. The resource allocation method for heterogeneous cellular networks oriented towards hybrid energy supply according to claim 4, characterized in that, The feasibility priority evaluation rule for candidate location solution vectors includes: The formula for calculating individual constraint violation is: ； in, express Individual constraint violation of any candidate position solution vector; This represents the function that takes the maximum value. like If the candidate position solution vector is feasible, then the solution vector is feasible. If the candidate position solution vector is not feasible, then the solution vector is not feasible. For any two candidate position solution vectors, select the feasible candidate position solution vector; If any two candidate position solution vectors are feasible, then the feasible candidate position solution vector with the smaller target fitness function is selected. If neither of the two candidate position solution vectors is feasible, then the candidate position solution vector with the smaller individual constraint violation is selected.

6. The resource allocation method for heterogeneous cellular networks oriented towards hybrid energy supply according to claim 4, characterized in that, The improved Harris Eagle optimization algorithm is used to perform a population-based iterative search on candidate position solution vectors. The algorithm selects historically optimal and globally optimal individual position solution vectors through a single-objective fitness function and a feasibility-priority evaluation rule for candidate position solution vectors. This includes: During the iterative search process of the population, the prey escape energy is calculated using the following formula: ; in, Indicates the energy of prey escaping. Indicates a random initial value. Indicates the iteration round, Indicates the maximum number of iterations for the preset population; Based on the prey escape energy, the individual's position in the population is updated to obtain the individual's initial position solution vector; By adopting the feasibility-first evaluation rule of candidate position solution vectors, the initial position solution vectors are modified to obtain the historical best individual position solution vector and the global best individual position solution vector that satisfy the single-objective fitness function.

7. The resource allocation method for heterogeneous cellular networks oriented towards hybrid energy supply according to claim 6, characterized in that, The process of updating an individual's position within the population based on prey escape energy to obtain the individual's initial position vector includes: like Then it enters the global search phase and generates Uniform random numbers within the interval ; like Then, the individual position is updated by referring to the candidate position solution vector randomly selected from the population, as expressed as: ; in, Represents an individual The initial position solution vector after position update in the current iteration round. Represents an individual The solution vector at the position at the start of the current iteration round. This means randomly selecting individuals from the population. The candidate position solution vector; , express Uniformly distributed random numbers within the interval are used to generate the search step size; like Then, the individual position is updated by referring to the mean of the global optimal individual position solution vector and the position solution vector of the individual in the current iteration, as expressed as: ； in, This represents the global optimal individual position solution vector in the current iteration round. The mean of the positional solution vectors representing the positions of individuals in the population during the current iteration. , express Uniformly distributed random numbers within the interval are used to generate the search step size; This represents the upper bound of the elements in the solution vector of an individual's position. This represents the lower bound of the elements in the solution vector for an individual's position; like Then it enters the local search phase and generates Uniform random numbers within the interval ; like and In this case, soft capture is used to update the individual's position, as shown below: ； in, express Uniformly distributed random numbers within the interval; like and Then, hard capture is used to update the individual's position, represented as: ； like Then the first candidate position solution vector is generated. Second candidate position solution vector ; ； ； in, This represents the standard normal random number generation function; The feasibility-first evaluation rule for candidate location solution vectors is adopted. , and The positional solution vectors that meet the requirements are selected from the options and used as... .

8. The method for resource allocation in heterogeneous cellular networks oriented towards hybrid energy supply according to claim 6, characterized in that, The step of adopting a feasibility-priority evaluation rule for candidate location solution vectors to revise the initial location solution vectors and obtain the historically optimal individual location solution vector and the globally optimal individual location solution vector that satisfy the single-objective fitness function includes: If the globally optimal individual position solution vector is not feasible in the current iteration, then the initial position solution vector is modified as follows: Make corrections: ； ； in, , express Users awaiting scheduling The selection results of the two candidate micro base stations express China base station to users The allocated power; Indicate each The corrected result Represents an individual The positional solution vector at the start of the current iteration. The power allocation from the base station to the user This indicates the upper bound of the power allocated by the base station to the user. This indicates the lower bound of the power allocated by the base station to the user; Indicate each The corrected result Represents an individual The historical best individual position solution vector. express Zhongyu Elements corresponding to dimensions This represents the global optimal individual position solution vector in the current iteration round. express Zhongyu Elements corresponding to dimensions; This means generating a random number between 0 and 1. This represents the rounding function; If the global optimal individual position solution vector is feasible in the current iteration, then the initial position solution vector is processed as follows: Make corrections: ； ； in, express Zhongyu Elements corresponding to dimensions express Zhongyu Elements corresponding to dimensions.

9. The resource allocation method for heterogeneous cellular networks oriented towards hybrid energy supply according to claim 6, characterized in that, The step of adopting a feasibility-priority evaluation rule for candidate location solution vectors to revise the initial location solution vectors and obtain the historically optimal individual location solution vectors and the globally optimal individual location solution vectors that satisfy the single-objective fitness function also includes: If the number of consecutive times the global optimal individual position solution vector is not updated during the population iterative search process meets the preset stagnation threshold, then select a portion of individuals to form an individual perturbation set, and add perturbation to the individual perturbation set; If the globally optimal individual position solution vector is feasible in the current iteration, then the initial position solution vector in the individual perturbation set is perturbed and updated in the following way: ； in, Represents an individual The initial position solution vector after perturbation update This represents the global optimal individual position solution vector in the current iteration round. This represents the upper bound of the elements in the solution vector of an individual's position. This represents the lower bound of the elements in the solution vector for an individual's position. This represents the standard normal random number generation function; If the globally optimal individual position solution vector is not feasible in the current iteration, then the initial position solution vector in the individual perturbation set is perturbed and updated in the following way: ； in, Represents an individual The historical optimal individual position solution vector.

10. A resource allocation device for heterogeneous cellular networks oriented towards hybrid energy supply, characterized in that, include: The network resource allocation model construction module is used to establish a network resource allocation model based on the carbon emissions of hybrid energy supply and the interaction throughput of cellular networks in heterogeneous cellular networks. The single-objective fitness function construction module is used to establish a single-objective fitness function with the optimization objectives of improving the cellular network interaction throughput of heterogeneous cellular networks and reducing the carbon emissions of hybrid energy supply. The model solving module is used to obtain the optimal network resource allocation scheme based on the single-objective fitness function and the network resource allocation model; the optimal network resource allocation scheme includes the base station-user association scheme and the transmit power allocation scheme for each user.

Images