Double-layer heterogeneous network spectrum resource allocation method and device and electronic equipment

By introducing discrete walrus optimization mechanism and user information transmission rate weight coefficient in a double-layer heterogeneous network, the problem that the spectrum resource allocation method cannot meet the minimum rate requirements of all users and maximizes the special user rate, achieving more efficient spectrum resources Utilization and shorter calculation time.

CN119052807APending Publication Date: 2024-11-29CHINA INST OF RADIO PROPAGATION
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411033351.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-30
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The existing two-layer heterogeneous network spectrum resource allocation method cannot maximize the sum of information transmission rates of some special users while meeting the minimum information transmission rate requirements of all users, and does not consider the user information transmission rate weight coefficient and discrete walrus optimization mechanism. .

Method used

The spectrum resource allocation method of the two-layer heterogeneous network based on discrete walrus optimization mechanism and user information transmission rate weight coefficient is adopted. By establishing a spectrum resource allocation model, the information transmission rate weight coefficient is set, and the walrus group is updated using predation, migration and evasion strategies. Speed ​​and location, optimize spectrum resource allocation to meet minimum rate requirements and maximize the rate of special users.

Benefits of technology

On the premise of meeting the minimum information transmission rate requirements of all users, the sum of information transmission rates of some special users is maximized, the utilization rate of spectrum resources is improved, the complexity of algorithms is reduced, and the optimal spectrum resource allocation is achieved in a short time. plan.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119052807A_ABST
    Figure CN119052807A_ABST
Patent Text Reader

Abstract

The invention discloses a double-layer heterogeneous network spectrum resource allocation method and device and electronic equipment, and the method comprises the following steps: 1, building a double-layer heterogeneous network spectrum resource allocation model; 2, initializing a discrete sea elephant group, and determining the global optimal speed; 3, updating the speed of each sea elephant according to a predation strategy, a migration strategy or an escape strategy; and 5, judging whether iteration is terminated or not, and outputting an optimal position as an optimal spectrum resource allocation scheme, wherein the adaptive degree of each sea elephant is calculated, and the speed of each sea elephant is determined according to a greedy selection strategy. According to the double-layer heterogeneous network spectrum resource allocation method provided by the invention, the sum of information transmission rates of some special users can be maximized on the premise of meeting the minimum information transmission rate requirements of all users by designing the user information transmission rate weight coefficient.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of wireless communications, and in particular relates to a double-layer heterogeneous network spectrum resource allocation method, device and electronic equipment in the field based on a discrete walrus optimization mechanism and taking into account a user information transmission rate weight coefficient. Background Art

[0002] Spectrum resources are regarded as the core resources for promoting the development of the mobile communication industry and the basis for ensuring the quality of user services. However, with the continuous growth of the number of mobile communication users, business types and network service requirements, the shortage of spectrum resources has increasingly become one of the bottlenecks restricting the development of the wireless communication field. The two-layer heterogeneous network combines the traditional macro base station network with the micro base station or home base station network to adapt to different scenarios and needs. The home base station network provides low-power and small coverage services, while the macro base station network provides high-power and large coverage services. In heterogeneous networks, how to coordinate and optimize the allocation and use of spectrum resources between two different network layers becomes a key issue. Traditional mobile communication systems have the problem of low spectrum resource utilization, and the introduction of two-layer heterogeneous networks is conducive to improving the spectrum resource utilization of the system, thereby improving the throughput and communication performance of the system. Through a reasonable spectrum resource allocation strategy, coordination and optimization between different network layers can be achieved, and the same-layer interference and cross-layer interference can be reduced, thereby achieving higher network coverage and user experience quality. Therefore, the reasonable allocation of spectrum resources in two-layer heterogeneous networks has become one of the research hotspots in the field of wireless communications.

[0003] Jin Donghui et al. proposed a resource allocation algorithm based on reinforcement learning in the article “Resource allocation strategy based on multi-objective optimization in dense heterogeneous networks” published in Telecommunications Technology (2023, 63(04), 466-474) to improve the energy efficiency and spectrum efficiency of the system. Although the algorithm takes into account the cross-layer interference between macro base station users and home base station users and comprehensively considers the service quality of users, the resource blocks occupied by each macro base station user are fixed, which limits the overall performance of the system to a certain extent and cannot be extended to the case where there is same-layer interference between macro base station users. In addition, the algorithm has high complexity and long training time. Zhang Haihui et al. introduced a nonlinear gravitational attenuation factor to improve the original gravitational search algorithm in order to improve the spectrum efficiency and thus improve the overall network efficiency in the article “Cognitive wireless network spectrum allocation based on improved gravitational search algorithm” published in Avionics Technology (2021, 52(02), 51-60) to improve the original gravitational search algorithm, and used the individuals with the highest fitness found in the iterative process to guide the group to find the optimal solution faster. However, this spectrum allocation scheme cannot guarantee that the signal-to-interference-noise ratio at each user is greater than the minimum required value. The computational complexity of the algorithm is high, and the global optimization capability needs to be further improved, and the spectrum utilization rate is not maximized.

[0004] By consulting the literature, it is found that the existing two-layer heterogeneous network spectrum resource allocation method does not take into account the special situation that some special users have high requirements for information transmission rate, and cannot be extended to the situation where the sum of the information transmission rates of some special users is maximized while meeting the minimum information transmission rate requirements of all users. By searching and comparing the existing literature, no two-layer heterogeneous network spectrum resource allocation method that takes into account the user information transmission rate weight coefficient has been found, and no two-layer heterogeneous network spectrum resource allocation method based on the discrete walrus optimization mechanism has been found. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a two-layer heterogeneous network spectrum resource allocation method, device and electronic device based on a discrete walrus optimization mechanism and a user information transmission rate weight coefficient in view of the shortcomings and deficiencies of the current two-layer heterogeneous network spectrum resource allocation method that cannot maximize the sum of the information transmission rates of some special users while meeting the minimum information transmission rate requirements of all users.

[0006] The present invention adopts the following technical solution:

[0007] A method for allocating spectrum resources in a double-layer heterogeneous network, the improvement of which is that it comprises the following steps:

[0008] Step 1: Establish a two-layer heterogeneous network spectrum resource allocation model:

[0009] Assume a radius of R M There are N randomly distributed in the macro cell M There are B macro base station users and B home base stations, and the coverage radius of each home base station is R F And contains N F Home base station users;

[0010] Divide the total system bandwidth evenly into N W The number of the channels allocated to each macro base station user and home base station user constitutes a decision variable. Each dimension of the decision variable takes a value from 1 to N. W Integer between ;

[0011] Assume that users in a home base station will not occupy the same sub-channel. For the nth macro base station user, assume that it occupies the nth sub-channel. sub-channels, and their signal-to-interference-noise ratio is:

[0012]

[0013] In the above formula, Indicates that the macro base station is allocated in The power of the subchannel, h n represents the channel gain from the macro base station to the nth macro base station user, Kn represents the path loss from the macro base station to the nth macro base station user, Λ represents the user occupying the nth The set of labels of the home base stations to which the home users of the sub-channels belong, Indicates that the jth home base station is allocated in The power of the sub-channels, represents the channel gain from the jth home base station to the nth macro base station user, represents the path loss from the jth home base station to the nth macro base station user, δ 2 represents the noise power;

[0014] The information transmission rate of the nth macro base station user is:

[0015]

[0016] In the above formula, W represents the total system bandwidth;

[0017] For the mth user under the bth home base station, assume that it occupies the θth subchannel, b=1,2,...,B, m=1,2,...,N F , its signal-to-interference-noise ratio is:

[0018]

[0019] In the above formula, represents the power allocated to the θth subchannel by the bth home base station, represents the channel gain from the bth home base station to its mth user, represents the path loss from the bth home base station to its mth user, Δ represents the set of home base station labels to which the home user occupying the θth subchannel belongs, represents the power allocated by the kth home base station to the θth subchannel, represents the channel gain from the kth home base station to the mth user under the bth home base station, represents the path loss from the kth home base station to the mth user under the bth home base station, p θ represents the power allocated by the macro base station to the θth subchannel, represents the channel gain from the macro base station to the mth user under the bth home base station, represents the path loss from the macro base station to the mth user under the bth home base station;

[0020] The information transmission rate of the mth user under the bth home base station is:

[0021]

[0022] Set the information transmission rate weight coefficient for each macro base station user and home base station user: It is required that the sum of the information transmission rates of some special users be maximized when all users meet the minimum information transmission rate requirement. If the special users include the nth macro base station user, then its weight coefficient w n =1, otherwise let w n =0; if the special users include the mth user under the bth home base station, then let its weight coefficient Otherwise,

[0023] The optimization goal is:

[0024]

[0025] The constraints are:

[0026] C n ≥C min , n=1,2,...,N M , b=1,2,...,B, m=1,2,...,N F

[0027] In the above formula, C min The minimum information transmission rate required by the user;

[0028] Step 2: Initialize the discrete walrus group and determine the global optimal speed:

[0029] Assume that the size of the discrete walrus group is A, and the total number of macro base station users and home base station users is the dimension of the search space, expressed as Z = N M +B×N F ;

[0030] Assume that when the discrete walrus herd evolves to the tth generation, the speed of the ath walrus is:

[0031]

[0032] In the above formula, a=1,2,...,A,z=1,2,...,Z,each dimension of the initial velocity of each walrus follows a uniform random distribution between 0 and 1;

[0033] The position of the a-th walrus in the t-th generation is expressed as:

[0034]

[0035] In the above formula,

[0036] The mapping rule between the z-dimensional velocity and position of the a-th walrus is:

[0037]

[0038] In the above formula, ceil(·) represents the upward rounding function;

[0039] The position of each walrus corresponds to a spectrum resource allocation scheme, which determines the subchannel allocated to each user. The fitness of the a-th walrus in the t-th generation is:

[0040]

[0041] In the above formula, ε is the penalty factor;

[0042] The speed of the walrus with the largest fitness in the discrete walrus group up to the tth generation is recorded as the global optimal speed:

[0043]

[0044] Step 3, update the speed of each walrus according to the predation strategy, migration strategy or escape strategy:

[0045] When the discrete walrus group iterates to the tth generation, a random number that follows the standard Gaussian distribution is generated for the ath walrus and a random number uniformly distributed between 0 and 1 make:

[0046]

[0047] In the above formula, β1 is a constant, t max is the maximum number of iterations;

[0048] For the ath walrus, if and Then execute the predation strategy, if and Then execute the migration strategy, if Then execute the escape strategy;

[0049] When the ath walrus updates its moving step using a predation strategy or an escape strategy, it first generates a random number between 1 and Z. The set Ω of different integers, where is a random number uniformly distributed between 0 and 1;

[0050] When using the predation strategy update, the update formula for the z-th dimension moving step length of the a-th walrus is:

[0051]

[0052] In the above formula, exp(·) is the natural exponential function, is a random variable that follows a standard normal distribution, is the z-th velocity of the τ-th walrus randomly selected from the discrete walrus group;

[0053] When using the escape strategy to update, the update formula for the z-th dimension moving step of the a-th walrus is:

[0054]

[0055] In the above formula, represents a random variable that follows a standard normal distribution, represents a random variable that is uniformly distributed between 0 and 1;

[0056] When the ath walrus updates its moving step length using a predation strategy or an escape strategy, the update formula for the z-dimensional velocity of the ath walrus is:

[0057]

[0058] When the migration strategy is used for updating, the update formula for the z-th dimension moving step length of the a-th walrus is:

[0059]

[0060] In the above formula, β2 is a constant. is a random variable that follows a standard normal distribution, is the z-th velocity of the ζ-th walrus randomly selected from the discrete walrus group;

[0061] When the ath walrus updates its moving step length using the migration strategy, the update formula for the z-dimensional velocity of the ath walrus is:

[0062] and

[0063] Step 4, calculate the fitness of each walrus and determine the speed of each walrus based on the greedy selection strategy:

[0064] After updating the speed of each walrus, the updated position of each walrus is calculated according to the mapping rule between speed and position shown in step 2, and the updated fitness of each walrus is calculated;

[0065] The speed of the ath walrus is determined according to the greedy selection strategy, which is specifically expressed as:

[0066]

[0067] The speed of the walrus with the maximum fitness after the update is recorded as the global optimal speed until the t+1th generation:

[0068]

[0069] Step 5: determine whether to terminate the iteration and output the optimal position as the optimal spectrum resource allocation solution:

[0070] If the number of iterations does not reach the maximum number of iterations set, then t < t max , then let t = t + 1 and return to step 3 to continue the iteration; otherwise, terminate the iteration, obtain the global optimal position of the discrete walrus group according to the mapping rule between speed and position, and output the global optimal position as the optimal spectrum resource allocation solution.

[0071] A double-layer heterogeneous network spectrum resource allocation device, the improvement of which is that it includes:

[0072] Modeling module: used to establish a double-layer heterogeneous network spectrum resource allocation model;

[0073] Initialization module: used to initialize discrete walrus groups and determine the global optimal speed;

[0074] Update module: used to update the speed of each walrus according to the hunting strategy, migration strategy or escape strategy;

[0075] Calculation module: used to calculate the fitness of each walrus and determine the speed of each walrus based on the greedy selection strategy;

[0076] Output module: used to determine whether to terminate the iteration and output the optimal position as the optimal spectrum resource allocation solution.

[0077] An electronic device, the improvement of which lies in: comprising a memory and a processor, wherein the memory is used to store programs; the processor is coupled to the memory and is used to execute the programs stored in the memory to implement the steps in the above-mentioned dual-layer heterogeneous network spectrum resource allocation method.

[0078] The beneficial effects of the present invention are:

[0079] The double-layer heterogeneous network spectrum resource allocation method proposed in the present invention can maximize the sum of the information transmission rates of some special users while meeting the minimum information transmission rate requirements of all users by designing the user information transmission rate weight coefficient. By adjusting the user information transmission rate weight coefficient, the spectrum resource allocation method can be applied to different scenarios, expanding its scope of application.

[0080] In response to the spectrum allocation problem of two-layer heterogeneous networks, the present invention designs a discrete walrus optimization mechanism to effectively solve the problem. Taking into account the situation that some special users have high requirements for information transmission rate, under the premise of ensuring that the information transmission rate of all users is greater than the minimum required value, the weight coefficient is adjusted to maximize the sum of the information transmission rates of some special users, so that the optimal spectrum resource allocation plan can be found in a relatively short time.

[0081] The discrete walrus optimization mechanism designed in the present invention solves the discrete integer optimization problem and breaks through the application limitation that the original walrus optimization algorithm cannot be used for discrete optimization problems. By designing new speed and position update formulas, the algorithm's calculation complexity is reduced while the global optimization capability of the algorithm is improved, providing a new idea for using swarm intelligence optimization algorithms to solve other integer optimization problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 It is a schematic flow diagram of the method of the present invention;

[0083] Figure 2 It is a curve diagram showing the variation of system throughput with the number of iterations of the two-layer heterogeneous network spectrum allocation method obtained by applying the discrete walrus optimization mechanism, the original walrus optimization algorithm, the particle swarm optimization algorithm and the transient search optimization algorithm;

[0084] Figure 3 It is a graph showing the sum of macro base station user information transmission rates versus the number of iterations of the double-layer heterogeneous network spectrum allocation method obtained by applying the discrete walrus optimization mechanism, the original walrus optimization algorithm, the particle swarm algorithm, and the transient search optimization algorithm. DETAILED DESCRIPTION

[0085] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0086] Embodiment 1, this embodiment discloses a double-layer heterogeneous network spectrum resource allocation method, such as Figure 1 As shown, the following steps are included:

[0087] Step 1: Establish a two-layer heterogeneous network spectrum resource allocation model:

[0088] Assume a radius of R M There are N randomly distributed in the macro cell M There are B macro base station users and B home base stations, and the coverage radius of each home base station is R F And contains N F Home base station users;

[0089] Divide the total system bandwidth evenly into N W The number of the channels allocated to each macro base station user and home base station user constitutes a decision variable. Each dimension of the decision variable takes a value from 1 to N. W Integer between ;

[0090] Assume that users in a home base station will not occupy the same sub-channel. For the nth macro base station user, assume that it occupies the nth sub-channel. sub-channels, and their signal-to-interference-noise ratio is:

[0091]

[0092] In the above formula, Indicates that the macro base station is allocated in The power of the subchannel, h n represents the channel gain from the macro base station to the nth macro base station user, K n represents the path loss from the macro base station to the nth macro base station user, Λ represents the user occupying the nth The set of labels of the home base stations to which the home users of the sub-channels belong, Indicates that the jth home base station is allocated in The power of the sub-channels, represents the channel gain from the jth home base station to the nth macro base station user, represents the path loss from the jth home base station to the nth macro base station user, δ 2 represents the noise power;

[0093] The information transmission rate of the nth macro base station user is:

[0094]

[0095] In the above formula, W represents the total system bandwidth;

[0096] For the mth user under the bth home base station, assume that it occupies the θth subchannel, b=1,2,...,B, m=1,2,...,N F , its signal-to-interference-noise ratio is:

[0097]

[0098] In the above formula, represents the power allocated to the θth subchannel by the bth home base station, represents the channel gain from the bth home base station to its mth user, represents the path loss from the bth home base station to its mth user, Δ represents the set of home base station labels to which the home user occupying the θth subchannel belongs, represents the power allocated by the kth home base station to the θth subchannel, represents the channel gain from the kth home base station to the mth user under the bth home base station, represents the path loss from the kth home base station to the mth user under the bth home base station, p θ represents the power allocated by the macro base station to the θth subchannel, represents the channel gain from the macro base station to the mth user under the bth home base station, represents the path loss from the macro base station to the mth user under the bth home base station;

[0099] The information transmission rate of the mth user under the bth home base station is:

[0100]

[0101] Set the information transmission rate weight coefficient for each macro base station user and home base station user: It is required that the sum of the information transmission rates of some special users be maximized when all users meet the minimum information transmission rate requirement. If the special users include the nth macro base station user, then its weight coefficient w n =1, otherwise let w n =0; if the special users include the mth user under the bth home base station, then let its weight coefficient Otherwise,

[0102] The optimization goal is:

[0103]

[0104] The constraints are:

[0105] C n ≥C min , n=1,2,...,N M , b=1,2,...,B, m=1,2,...,N F

[0106] In the above formula, C min The minimum information transmission rate required by the user;

[0107] For any n∈{1,2...,N M}, b∈{1,2,...,B}, m∈{1,2,...,N F}, let w n =1, The optimization problem then becomes to maximize the system throughput when all users meet the minimum information transmission rate requirement. M}, let w n =1; for any b∈{1,2,...,B}, m∈{1,2,...,N F},set up The optimization problem is then transformed into maximizing the sum of the information transmission rates of all macro base station users when all users meet the minimum information transmission rate requirement. It can be seen that by adjusting the user information transmission rate weight coefficient, the designed spectrum resource allocation method can be applied to different application scenarios, reflecting its universality.

[0108] Step 2: Initialize the discrete walrus group and determine the global optimal speed:

[0109] Assume that the size of the discrete walrus group is A, and the total number of macro base station users and home base station users is the dimension of the search space, expressed as Z = N M +B×N F ;

[0110] Assume that when the discrete walrus herd evolves to the tth generation, the speed of the ath walrus is:

[0111]

[0112] In the above formula, a=1,2,...,A,z=1,2,...,Z,each dimension of the initial velocity of each walrus follows a uniform random distribution between 0 and 1;

[0113] The position of the a-th walrus in the t-th generation is expressed as:

[0114]

[0115] In the above formula,

[0116] The mapping rule between the z-dimensional velocity and position of the a-th walrus is:

[0117]

[0118] In the above formula, ceil(·) represents the upward rounding function;

[0119] The position of each walrus corresponds to a spectrum resource allocation scheme, which determines the subchannel allocated to each user. The fitness of the a-th walrus in the t-th generation is:

[0120]

[0121] In the above formula, ε is the penalty factor;

[0122] The speed of the walrus with the largest fitness in the discrete walrus group up to the tth generation is recorded as the global optimal speed:

[0123]

[0124] Step 3, update the speed of each walrus according to the predation strategy, migration strategy or escape strategy:

[0125] When the discrete walrus group iterates to the tth generation, a random number that follows the standard Gaussian distribution is generated for the ath walrus and a random number uniformly distributed between 0 and 1 make:

[0126]

[0127] In the above formula, β1 is a constant, t max is the maximum number of iterations;

[0128] For the ath walrus, if and Then execute the predation strategy, if and Then execute the migration strategy, if Then execute the escape strategy;

[0129] When the ath walrus updates its moving step using a predation strategy or an escape strategy, it first generates a random number between 1 and Z. The set Ω of different integers, where is a random number uniformly distributed between 0 and 1;

[0130] When using the predation strategy update, the update formula for the z-th dimension moving step length of the a-th walrus is:

[0131]

[0132] In the above formula, exp(·) is the natural exponential function, is a random variable that follows a standard normal distribution, is the z-th velocity of the τ-th walrus randomly selected from the discrete walrus group;

[0133] When using the escape strategy to update, the update formula for the z-th dimension moving step of the a-th walrus is:

[0134]

[0135] In the above formula, represents a random variable that follows a standard normal distribution, represents a random variable that is uniformly distributed between 0 and 1;

[0136] When the ath walrus updates its moving step length using a predation strategy or an escape strategy, the update formula for the z-dimensional velocity of the ath walrus is:

[0137] and abs(·) means taking the absolute value;

[0138] When the migration strategy is used for updating, the update formula for the z-th dimension moving step length of the a-th walrus is:

[0139]

[0140] In the above formula, β2 is a constant. is a random variable that follows a standard normal distribution, is the z-th velocity of the ζ-th walrus randomly selected from the discrete walrus group;

[0141] When the ath walrus updates its moving step length using the migration strategy, the update formula for the z-dimensional velocity of the ath walrus is:

[0142] and

[0143] Step 4, calculate the fitness of each walrus and determine the speed of each walrus based on the greedy selection strategy:

[0144] After updating the speed of each walrus, the updated position of each walrus is calculated according to the mapping rule between speed and position shown in step 2, and the updated fitness of each walrus is calculated;

[0145] The speed of the ath walrus is determined according to the greedy selection strategy, which is specifically expressed as:

[0146]

[0147] The speed of the walrus with the maximum fitness after the update is recorded as the global optimal speed until the t+1th generation:

[0148]

[0149] Step 5: determine whether to terminate the iteration and output the optimal position as the optimal spectrum resource allocation solution:

[0150] If the number of iterations does not reach the maximum number of iterations set, then t < t max , then let t = t + 1 and return to step 3 to continue the iteration; otherwise, terminate the iteration, obtain the global optimal position of the discrete walrus group according to the mapping rule between speed and position, and output the global optimal position as the optimal spectrum resource allocation solution.

[0151] Next, the beneficial effects of the present invention are further illustrated by simulation experiments:

[0152] For a two-layer heterogeneous network, let the number of subchannels be N W =40, the coordinates of the macro base station are (0,0)m, and its coverage radius is R M =400m, N is randomly distributed within its coverage area M = 40 macro base station users and B = 5 home base stations, where each home base station covers a radius R F = 60m, N is randomly distributed within the coverage area of ​​each home base station F = 6 home base station users. The total transmission power of the macro base station and the home base station is 6W and 0.6W respectively, and the power allocated to each subchannel by the macro base station and the home base station is the same, and the noise power δ 2 =0.5mW. The system total bandwidth, path loss and channel gain parameter settings refer to "Macro-and femtocell interference mitigation in OFDMA wireless systems" published by Ning G et al. in "2012IEEE Global Communications Conference" (2012, pp: 5068-5073). The parameter settings of the double-layer heterogeneous network spectrum resource allocation method of the discrete walrus optimization mechanism are as follows: the population size of the discrete walrus group A = 50, the penalty factor ε = 0.01, β1 = 8, β2 = 0.6, C min =0.1Mbit / s. In order to demonstrate the superiority of the designed discrete walrus optimization mechanism, it is compared with the original walrus optimization algorithm, particle swarm optimization algorithm and transient search optimization algorithm. Since the original walrus optimization algorithm, particle swarm optimization algorithm and transient search optimization algorithm cannot solve the discrete optimization problem, the position of each individual in the iterative process of the three algorithms is rounded up as shown in step 2. The population size and maximum number of iterations of all algorithms are set to the same value A = 50 and t max=300, the simulation results are the average of 50 simulation experiments. For other parameter settings of the original walrus optimization algorithm, please refer to "A new bio-inspired metaheuristic algorithm for solving optimization problems based on walruses behavior" published by Trojovsky P et al. in "Scientific Reports" (2023, 13 (1): 8775), for other parameter settings of the particle swarm algorithm, please refer to "A new optimizer using particle swarm theory" published by Kennedy J et al. in "MHS'95. proceedings of the sixth international symposium on micro machine and human science", and for other parameter settings of the transient search optimization algorithm, please refer to "Transient search optimization: a new meta-heuristic optimization algorithm" published by Mohammed H et al. in "Applied Intelligence" (2020, 50: 3926-3941). Figure 2 , 3 In the figure, “DWaOA” represents a two-layer heterogeneous network spectrum resource allocation method based on discrete walrus optimization mechanism, “WaOA” represents a two-layer heterogeneous network spectrum resource allocation method based on original walrus optimization algorithm, “PSO” represents a two-layer heterogeneous network spectrum resource allocation method based on particle swarm optimization algorithm, and “TSO” represents a two-layer heterogeneous network spectrum resource allocation method based on transient search optimization algorithm.

[0153] Figure 2 In order to set the information transmission rate weight coefficient of all users to 1, that is, when the optimization target is system throughput, the discrete walrus optimization mechanism, the original walrus optimization algorithm, the particle swarm optimization algorithm, and the transient search optimization algorithm are used to obtain the curve of system throughput changing with the number of iterations. Figure 2 It can be seen that the original walrus optimization algorithm, particle swarm optimization algorithm and transient search optimization algorithm are prone to local convergence when solving the spectrum resource allocation problem of two-layer heterogeneous networks. The discrete walrus optimization mechanism makes full use of the global optimal speed found in the iterative process of the walrus group to guide the walrus group to update its speed and position, so it can achieve higher convergence accuracy than the original walrus optimization algorithm.

[0154] Figure 3When the information transmission rate weight coefficient of all macro base station users is set to 1 and the information transmission rate weight coefficient of all home base station users is set to 0, that is, when the optimization goal is to maximize the sum of the information transmission rates of all macro base station users, the curve of the sum of the information transmission rates of macro base station users obtained by applying the discrete walrus optimization mechanism, the original walrus optimization algorithm, the particle swarm optimization algorithm, and the transient search optimization algorithm varies with the number of iterations. Figure 3 It can be seen that the sum of the macro base station user information transmission rates obtained by applying the discrete walrus optimization mechanism is greater than the sum of the macro base station user information transmission rates obtained by applying the original walrus optimization algorithm, the particle swarm algorithm, and the transient search optimization algorithm, which proves the effectiveness of the discrete walrus optimization mechanism.

[0155] This embodiment also discloses a dual-layer heterogeneous network spectrum resource allocation device, including:

[0156] Modeling module: used to establish a double-layer heterogeneous network spectrum resource allocation model;

[0157] Initialization module: used to initialize discrete walrus groups and determine the global optimal speed;

[0158] Update module: used to update the speed of each walrus according to the hunting strategy, migration strategy or escape strategy;

[0159] Calculation module: used to calculate the fitness of each walrus and determine the speed of each walrus based on the greedy selection strategy;

[0160] Output module: used to determine whether to terminate the iteration and output the optimal position as the optimal spectrum resource allocation solution.

[0161] This embodiment also discloses an electronic device, including a memory and a processor, wherein the memory is used to store programs; the processor is coupled to the memory and is used to execute the programs stored in the memory to implement the steps in the above-mentioned dual-layer heterogeneous network spectrum resource allocation method.

Claims

1. A method for allocating spectrum resources in a dual-layer heterogeneous network, characterized in that: The steps include: Step 1: Establish a two-layer heterogeneous network spectrum resource allocation model: Assume a radius of R M There are N randomly distributed in the macro cell M There are B macro base station users and B home base stations, and the coverage radius of each home base station is R F And contains N F Home base station users; Divide the total system bandwidth evenly into N W The number of the channels allocated to each macro base station user and home base station user constitutes a decision variable. Each dimension of the decision variable takes a value from 1 to N. W Integer between ; Assume that users in a home base station will not occupy the same sub-channel. For the nth macro base station user, assume that it occupies the nth sub-channel. sub-channels, and their signal-to-interference-noise ratio is: In the above formula, Indicates that the macro base station is allocated in The power of the subchannel, h n represents the channel gain from the macro base station to the nth macro base station user, K n represents the path loss from the macro base station to the nth macro base station user, Λ represents the user occupying the nth The set of labels of the home base stations to which the home users of the sub-channels belong, Indicates that the jth home base station is allocated in The power of the sub-channels, represents the channel gain from the jth home base station to the nth macro base station user, represents the path loss from the jth home base station to the nth macro base station user, δ 2 represents the noise power; The information transmission rate of the nth macro base station user is: In the above formula, W represents the total system bandwidth; For the mth user under the bth home base station, assume that it occupies the subchannels, b=1,2,...,B, m=1,2,...,N F , its signal-to-interference-noise ratio is: In the above formula, Indicates that the bth home base station is allocated in The power of the sub-channels, represents the channel gain from the bth home base station to its mth user, represents the path loss from the bth home base station to its mth user, Δ represents the occupancy of the The set of labels of the home base stations to which the home users of the sub-channels belong, Indicates that the kth home base station is allocated in The power of the sub-channels, represents the channel gain from the kth home base station to the mth user under the bth home base station, represents the path loss from the kth home base station to the mth user under the bth home base station, Indicates that the macro base station is allocated in The power of the sub-channels, represents the channel gain from the macro base station to the mth user under the bth home base station, represents the path loss from the macro base station to the mth user under the bth home base station; The information transmission rate of the mth user under the bth home base station is: Set the information transmission rate weight coefficient for each macro base station user and home base station user: It is required that the sum of the information transmission rates of some special users be maximized when all users meet the minimum information transmission rate requirement. If the special users include the nth macro base station user, then its weight coefficient w n =1, otherwise let w n =0; if the special users include the mth user under the bth home base station, then let its weight coefficient Otherwise, The optimization goal is: The constraints are: C n ≥C min , n=1,2,...,N M ,b=1,2,...,B,m=1,2,...,N F In the above formula, C min The minimum information transmission rate required by the user; Step 2: Initialize the discrete walrus group and determine the global optimal speed: Assume that the size of the discrete walrus group is A, and the total number of macro base station users and home base station users is the dimension of the search space, expressed as Z = N M +B×N F ; Assume that when the discrete walrus herd evolves to the tth generation, the speed of the ath walrus is: In the above formula, a=1,2,...,A,z=1,2,...,Z,each dimension of the initial velocity of each walrus follows a uniform random distribution between 0 and 1; The position of the ath walrus in the tth generation is expressed as: In the above formula, The mapping rule between the z-dimensional velocity and position of the a-th walrus is: In the above formula, ceil(·) represents the upward rounding function; The position of each walrus corresponds to a spectrum resource allocation scheme, which determines the subchannel allocated to each user. The fitness of the a-th walrus in the t-th generation is: In the above formula, ε is the penalty factor; The speed of the walrus with the largest fitness in the discrete walrus group up to the tth generation is recorded as the global optimal speed: Step 3, update the speed of each walrus according to the predation strategy, migration strategy or escape strategy: When the discrete walrus group iterates to the tth generation, a random number that follows the standard Gaussian distribution is generated for the ath walrus and a random number uniformly distributed between 0 and 1 make: In the above formula, β1 is a constant, t max is the maximum number of iterations; For the ath walrus, if and Then execute the predation strategy, if and Then execute the migration strategy, if Then execute the escape strategy; When the ath walrus updates its moving step using a predation strategy or an escape strategy, it first generates a random number between 1 and Z. The set Ω of different integers, where is a random number uniformly distributed between 0 and 1; When using the predation strategy update, the update formula for the z-th dimension moving step length of the a-th walrus is: In the above formula, exp(·) is the natural exponential function, is a random variable that follows a standard normal distribution, is the z-th velocity of the τ-th walrus randomly selected from the discrete walrus group; When using the escape strategy to update, the update formula for the z-th dimension moving step of the a-th walrus is: In the above formula, represents a random variable that follows a standard normal distribution, represents a random variable that is uniformly distributed between 0 and 1; When the ath walrus updates its moving step length using a predation strategy or an escape strategy, the update formula for the z-dimensional velocity of the ath walrus is: and abs(·) means taking the absolute value; When the migration strategy is used for updating, the update formula for the z-th dimension moving step length of the a-th walrus is: In the above formula, β2 is a constant. is a random variable that follows a standard normal distribution, is the z-th velocity of the ζ-th walrus randomly selected from the discrete walrus group; When the ath walrus updates its moving step length using the migration strategy, the update formula for the z-dimensional velocity of the ath walrus is: and Step 4, calculate the fitness of each walrus and determine the speed of each walrus based on the greedy selection strategy: After updating the speed of each walrus, the updated position of each walrus is calculated according to the mapping rule between speed and position shown in step 2, and the updated fitness of each walrus is calculated; The speed of the ath walrus is determined according to the greedy selection strategy, which is specifically expressed as: The speed of the walrus with the maximum fitness after the update is recorded as the global optimal speed until the t+1th generation: Step 5: determine whether to terminate the iteration and output the optimal position as the optimal spectrum resource allocation solution: If the number of iterations does not reach the maximum number of iterations set, then t < t max , then let t = t + 1, and return to step 3 to continue the iteration; Otherwise, the iteration is terminated, and the global optimal position of the discrete walrus group is obtained according to the mapping rule between speed and position, and the global optimal position is output as the optimal spectrum resource allocation solution.

2. A dual-layer heterogeneous network spectrum resource allocation device, characterized in that: include: Modeling module: used to establish a double-layer heterogeneous network spectrum resource allocation model; Initialization module: used to initialize discrete walrus groups and determine the global optimal speed; Update module: used to update the speed of each walrus according to the hunting strategy, migration strategy or escape strategy; Calculation module: used to calculate the fitness of each walrus and determine the speed of each walrus based on the greedy selection strategy; Output module: used to determine whether to terminate the iteration and output the optimal position as the optimal spectrum resource allocation solution.

3. An electronic device, characterized in that: It includes a memory and a processor, wherein the memory is used to store programs; the processor is coupled to the memory and is used to execute the programs stored in the memory to implement the steps in a dual-layer heterogeneous network spectrum resource allocation method as described in claim 1 above.