CN110753365B - Interference coordination method for heterogeneous cellular networks - Google Patents

Abstract

The invention discloses a heterogeneous cellular network interference coordination method, comprising: in a simulated real downlink network link scenario, aiming at minimizing the total transmission power consumption of the base station, taking the user QoS requirements, the maximum transmission power of the macro base station, The maximum transmit power of micro base station, base station access limit and channel multiplexing limit are used as constraints to determine the optimization problem; solve the optimization problem; decode the results obtained from the solution to obtain the corresponding user access, channel allocation and power control strategies. The method of the present invention establishes the collaborative optimization of user access strategy, channel allocation and power control by adopting a network scenario with high reduction degree and reliability, and considering the user access strategy, channel allocation and power control strategy cooperatively based on the scenario. A heuristic algorithm with lower complexity is proposed to solve the problem, which can effectively ensure the differentiated QoS requirements of users and the constraints of the maximum transmit power of the base station, and minimize the total power consumption of the system.

Description

Interference coordination method for heterogeneous cellular networks

technical field

The invention belongs to the field of mobile communication, and in particular relates to a resource scheduling optimization technology in a heterogeneous cellular network.

Background technique

With the increasing demands of mobile users and the diversification of applications, the amount of mobile data is approaching the upper limit of the capacity of the existing cellular mobile network at an exponential rate, which leads to the overload of the cellular network and the deterioration of service quality. Therefore, it is necessary to further improve the cellular mobile network The capacity of the network to meet the ever-increasing demands of users. One solution is to use heterogeneous layered coverage and increase base station deployment, which in turn derives a dense heterogeneous network. Heterogeneous networking can effectively improve the communication capacity of cells.

Although dense heterogeneous networking improves the capacity of mobile communication networks to a certain extent, the dense deployment of base stations brings serious intra-layer interference, and the spectrum sharing mechanism used in heterogeneous networks also causes serious cross-layer interference, which leads to heterogeneous The spectral efficiency of the network is low. In addition, multi-dimensional resource allocation in heterogeneous networks is also a huge challenge due to the characteristics of local hotspots in service loads. If all base stations work at the maximum transmit power, it will not only cause energy waste and reduce energy efficiency, but also cause excessive interference between base stations, reduce channel quality, and affect user experience. By optimizing the association between users and base stations, channels, and power control, under the premise of ensuring user Quality of Service (QoS), reducing interference between base stations and reducing power consumption are the main issues discussed in this field.

At present, the research scope of heterogeneous cellular networks is very wide, including related technologies of cellular network communication, application scenarios, and possible problems in practical applications (such as access strategy, resource management, power control), etc. The paper "Li Jianhui. Research on service offloading technology for dense heterogeneous networks [D]. Beijing University of Posts and Telecommunications, 2017" proposes a delayed service offloading scheme based on effective capacity, which maximizes the energy of the system under the limited conditions of user service quality. Efficiency; the literature "Wang Lu. Research on User Unloading Strategy for Heterogeneous Cellular Networks [D]. University of Science and Technology of China, 2017" comprehensively considers parameters such as user affiliation and the proportion of almost blank subframes to construct a joint utility optimization problem, and adopts Gauss-Seidel method to solve.

Existing research shows that the introduction of an effective resource scheduling strategy in a dense heterogeneous cellular network can improve the spectrum utilization of the system and reduce the power consumption of the base station. It solves some practical problems faced by dense heterogeneous networks; however, in actual communication scenarios, it is necessary to effectively guarantee the differentiated QoS requirements of different users, and there are differences in performance such as the maximum transmit power of different base stations. None of the above-mentioned prior art can take into account well.

SUMMARY OF THE INVENTION

Aiming at the above problems existing in the prior art, the present invention proposes a heterogeneous cellular network interference coordination method, which specifically includes the following steps:

Step S1: In the simulated real downlink network scenario, with the goal of minimizing the total transmit power consumption of the base station, the user QoS requirements, the maximum transmit power of the macro base station, the maximum transmit power of the micro base station, the base station access limit and the channel complexity are respectively taken as the goal. Determine the optimization problem with constraints as constraints;

Step S2: Solve the optimization problem determined in step S1;

Step S3: Decode the result obtained in Step S2 to obtain a corresponding user access, channel allocation and power control strategy.

Further, the real link scenario simulated in step S1 includes: M mobile users, N base stations (including L macro base stations and N-L micro base stations), N base stations are evenly deployed to form a cell, and M mobile users are randomly distributed. In a cell; all base stations are deployed on the same frequency, and the frequency band is divided into I orthogonal channels, each mobile user can connect to only one base station, and obtain services from one or more channels, and other base stations transmit at the same base station. The signal will interfere with the user, and there is no interference between different channels.

Further, the step S1 determines that the optimization problem is a mixed integer nonlinear optimization problem.

Further, the step S2 is specifically solved by using a hybrid heuristic search algorithm based on a genetic algorithm and a particle swarm algorithm.

Further, the specific steps of solving are as follows:

S21: Use the same format to encode the gene sequence in the genetic algorithm and the position information in the particle swarm algorithm, and initialize the genetic algorithm;

S22: Obtain a set of feasible solutions by iteratively using a finite number of genetic algorithms;

S23: Initialize the particle swarm optimization algorithm, and input the results obtained by the genetic algorithm into the particle swarm optimization algorithm to solve;

S24: Re-input the solution obtained by the particle swarm optimization algorithm into the genetic algorithm until the result converges or the upper limit of the number of iterations is reached.

Further, the chromosome sequence in the genetic algorithm used in step S21 and the position information in the particle swarm optimization algorithm are the 0-1 integer matrix corresponding to the user access strategy and the channel allocation strategy, and the real number related to the power control of the base station. The fitness expression of matrix, individual and particle is the total power consumption of the optimization target base station of the problem.

Further, in a large cycle, for each iteration of the genetic algorithm, the parent population is subjected to the crossover operation and mutation operation to generate the child population, and the optimal strategy and optimal fitness are recorded, and then the selection operation is performed to enter the next iteration. , after a finite number of iterations, the genetic algorithm is forced to stop and the current population is output.

Further, every time the particle swarm optimization algorithm is started, the velocities of all particles are reset to 0, and the genetic information of the current population is obtained by inputting the genetic algorithm as the position information. After one iteration, the particle velocities are calculated and the iteration is performed.

Beneficial effects of the present invention: The heterogeneous cellular network interference coordination method of the present invention establishes a network scenario with a relatively high degree of restoration and reliability, and based on the scenario, the user access strategy, channel allocation and power control strategy are considered collaboratively. User access strategy, channel allocation and power control co-optimization scheme, and a heuristic algorithm with lower complexity is proposed to solve the problem. Aiming at the downlink transmission scenario in the heterogeneous cellular network, the method of the present invention can minimize the total power consumption of the system under the condition of effectively ensuring the differentiated QoS requirements of users and the constraint of the maximum transmit power of the base station.

Description of drawings

FIG. 1 is a schematic diagram of a system scenario according to an embodiment of the present invention.

FIG. 2 is a schematic flowchart of a method according to an embodiment of the present invention.

Detailed ways

In order to facilitate those skilled in the art to understand the technical content of the present invention, the content of the present invention will be further explained below with reference to the accompanying drawings.

FIG. 1 shows a network scenario constructed by an embodiment of the present invention, considering the downlink and the scenario of co-frequency deployment of base stations. Among them, MBS represents the macro base station, SBS represents the micro base station, MU represents the mobile user, A _N×M and _BM×I respectively represent the association between the user and the base station and the channel occupied by the user; the dotted line represents the interference signal, and the solid line is useful. Signal.

In this embodiment, a real downlink communication scenario is simulated, specifically: a heterogeneous cellular network downlink system composed of N base stations and M users, including L macro base stations and N-L micro base stations, and M mobile users randomly distributed in the area.

为基站集合，

为用户集合，

表示第n个基站，n＜L时表示宏基站，反之则表示微基站，所有基站同频部署，存在同层干扰和跨层干扰；

表示第m个用户设备，假设每个用户只能关联一个基站，令a_n，m∈{0，1}为用户与基站的关联变量，a_n,m＝1表示用户m接入了基站n，反之a_n,m＝0。set of orders

is the set of base stations,

collection of users,

Indicates the nth base station. When n<L, it means a macro base station. Otherwise, it means a micro base station. All base stations are deployed on the same frequency, and there is co-layer interference and cross-layer interference;

Represents the mth user equipment, assuming that each user can only be associated with one base station, let an _,m ∈{0,1} be the association variable between the user and the base station, an _n,m =1 indicates that user m accesses base station n , otherwise an _,m =0.

为基站资源块的集合，即每个基站共有I个子信道。b_m,i表示用户m对第i个信道的占用情况，b_m,i＝1表示用户m占用了第i个子信道，并且会受到其他基站在这个信道上发射信号的干扰。令P_S表示微基站最大发射功率，P_M表示宏基站的最大发射功率。另外网络中存在集中控制器，可以实时感知基站和用户状态，并做出相应决策。Multiple users accessing the same base station will share the system resources of the base station, such as obtaining services by time division multiplexing or frequency division multiplexing.

is a set of base station resource blocks, that is, each base station has a total of I subchannels. b _m,i indicates the occupancy of the i-th channel by user m, and b _m,i =1 indicates that the user m occupies the i-th sub-channel and will be interfered by signals transmitted by other base stations on this channel. Let PS _denote the maximum transmit power of the micro base station, and _{PM denote} the maximum transmit power of the macro base station. In addition, there is a centralized controller in the network, which can sense the status of base stations and users in real time, and make corresponding decisions.

For each scheduling period of the network, it is assumed that the controller calculates user access, channel allocation and power control strategies in real time according to the current user location information and channel state information. The corresponding policy obeys the following constraints:

User access strategy: each mobile user is only allowed to establish a connection with one base station, and the base station serves no upper limit (but cannot exceed the number of sub-channels that can be allocated).

Channel allocation strategy: All base stations are deployed on the same frequency, and the frequency domain is divided into multiple orthogonal sub-channels. The base station can provide users with one or more sub-channels and ensure that each user occupies at least one sub-channel.

Power control strategy: The base station can arbitrarily allocate power to each sub-channel, and only needs to satisfy the constraint that the total power does not exceed the maximum transmit power of the base station.

Based on the network scenario, a heterogeneous cellular network interference coordination method of the present invention specifically includes the following steps:

S1. In the simulated real downlink network scenario, aiming at minimizing the total transmit power consumption of the base station, the user QoS requirements, the maximum transmit power of the macro base station, the maximum transmit power of the micro base station, the base station access limit and the channel reuse limit are respectively Constraints to determine the optimization problem. Specifically, the objective function is to minimize the total power consumption of the system, and the QoS requirement _vm of the mobile user, the maximum transmit power p _n1,i of the macro base station, the maximum transmit power p _n2,i of the micro base station, and the user access indicator a are respectively taken as the objective function. The values of _n,m and the channel assignment indicator b _m,i are constraints, and the first optimization problem is obtained.

The optimization problem expression is as follows:

P1:

s.t.

C1: a _n,m ∈ {0,1}

C2:

C3: b _m,i ∈ {0,1}

C4:

C5:

C6:

C7:

对应了宏基站n₁和微基站n₂在信道i上的发射功率，两约束分别保证了宏基站与微基站的总发射功率不会超过各自的发射功率上限P_M/P_S；约束条件C7保证了用户m得到的吞吐量C_m满足其QoS需求v_m(最小数据速率)。最大发射功率P_M＝50w，P_S＝20w，用户差异化QoS需求v_m随机分布于[10⁵,3×10⁵]bit/s。Among them, P1 represents the objective function of the optimization problem; in the constraint condition C1, an _,m represents the associated variable between the user and the base station, an _n,m =1 means that the user m accesses the base station n, an _n,m =0 means that The user does not establish a connection with the station establishment; Constraint C2 ensures that the user accesses at least one base station at most; In Constraint C3, b _m,i represents the occupancy of the i-th subchannel by user m, and b _m,i =1 represents the user m occupies the i-th subchannel, and will be interfered by signals transmitted by other base stations on this channel. b _m,i = 0 means that the user is not associated with channel m and channel i; constraint C4 ensures that for a specific base station, it Each sub-channel can only be occupied by one mobile user (users accessing different base stations can reuse the same channel); in the constraints C5 and C6,

Corresponding to the transmit power of macro base station n ₁ and micro base station n ₂ on channel i, the two constraints respectively ensure that the total transmit power of macro base station and micro base station will not exceed their respective transmit power upper limit _{P M} /PS ; Constraint C7 It is guaranteed that the throughput C _m obtained by user m satisfies its QoS requirement _vm (minimum data rate). The maximum transmit power P _M =50w, P _S =20w, and the user's differentiated QoS requirement _vm is randomly distributed in [10 ⁵ , 3×10 ⁵ ]bit/s.

The QoS of a mobile user is the data rate obtained by the user from the associated base station, and the calculation formula is:

Among them, C _n,m,i is the data rate provided by base station n for user m on the ith subchannel, and the expression is as follows:

where B is the bandwidth of each orthogonal sub-channel, N ₀ is the power spectral density of the background noise, and H _n,m is the channel gain model:

H _n,m =PL _n,m × _{H R} ×HS

_{HR is Rayleigh fading due to multipath effect, and H S is} random shadow fading. Specifically, PL _n,m is expressed as:

Among them, d _n,m represents the distance between the user and the base station.

I _n,m,i is the interference received by user m on the i-th subchannel of base station n, and the expression is as follows:

S2. Use genetic algorithm and particle swarm algorithm to jointly solve the above mixed integer nonlinear programming problem,

The problem determined in step S1 is an NP-hard problem, which is difficult to solve directly, so a hybrid heuristic search algorithm with low complexity is used to solve the problem. In the large loop of the algorithm, the genetic algorithm and the particle swarm algorithm are repeatedly run to achieve a good global search and speed up the convergence. The specific process is shown in Figure 2, and the specific steps are as follows:

S21. Use the same format to encode the gene sequence in the genetic algorithm and the position information and fitness in the particle swarm algorithm, and initialize the algorithm; the specific implementation is as follows

为0-1整数型矩阵，

为实数矩阵，分别对应用户与基站的关联情况(用户接入第几个基站)、信道分配策略(用户占用接入基站的哪几个信道)、功率分配策略(基站在每个子信道上分配的功率)。For genetic algorithms, each individual in the population represents a set of strategies, and the chromosomes of the individual

is a 0-1 integer matrix,

It is a real number matrix, which corresponds to the association between the user and the base station (the number of base stations the user accesses), the channel allocation strategy (which channels the user occupies to access the base station), and the power allocation strategy (the number of channels allocated by the base station on each subchannel). power).

The fitness of an individual represents the performance of a set of strategies. In this problem, the total power consumption of the optimization target is used to measure the performance of the strategy:

In the particle swarm optimization algorithm, each particle represents a set of strategies, and has three attributes: position, speed and fitness. The position and fitness of the particle correspond to the chromosome and fitness of the individual genetic algorithm, respectively, and represent the current strategy and performance index; the speed of the particle is calculated by the algorithm and is the direction of strategy change that is expected to obtain better fitness.

Initialize the number of populations and particle swarm particles NP, the number of loop iterations T, the number of iterations in the genetic algorithm stage T ₁ and the maximum number of iterations in the particle swarm optimization stage T ₂ , and set the algorithm parameters: the number of offspring populations in the genetic algorithm NP ₁ , the crossover The probability P _c , the mutation probability P _m , the acceleration ω in the particle swarm optimization algorithm, the current collective centroid weight c ₁ , and the historical best centroid weight c ₂ .

S22, obtain a set of feasible solutions by using the finite-order genetic algorithm;

The operation steps of each genetic algorithm iteration are crossover operation, mutation operation and selection operation.

Crossover operation: For each generation of the population, randomly select two parents, exchange a part of the chromosome segments with the probability of P _c , and check whether the obtained new strategy satisfies the constraints, and if the conditions are met, a child individual is obtained.

Mutation operation: randomly select an individual, perform mutation operation on it with probability P _m , and extract a chromosome segment to randomly change in the feasible region.

The rules of mutation operation (ie mutation operator) are:

代表抽取到的片段为整数型变量，

代表抽取到的片段为实数型变量，γ₁、γ₂均为[0,1]区间上的随机数(根据

规模大小，γ₁、γ₂可能为随机数列或矩阵)，round计算符表示取与括号内数值最接近的整数，x_max、x_min、y_max、y_min分别表示抽取染色体片段每个分量可以取到的最大值或最小值。in,

Represents the extracted segment as an integer variable,

Represents that the extracted segments are real-number variables, and γ ₁ and γ ₂ are random numbers in the [0,1] interval (according to

Scale, γ ₁ , γ ₂ may be random number columns or matrices), the round operator means to take the integer closest to the value in the brackets, x _max , x _min , y _max , y _min respectively indicate that each component of the extracted chromosome segment can be The maximum or minimum value obtained.

Repeat the above steps to generate NP ₁ offspring individuals, calculate the fitness of each individual, and use the fitness as the weight, and use the roulette algorithm to select NP individuals as the next generation population. In particular, if the individual with the best fitness in the current individual is eliminated, the individual is forced to replace the individual with the lowest fitness in the next generation of the population to ensure that the algorithm proceeds in a better direction.

After iterating T ₁ times, enter the next stage.

S23, initialize the particle swarm algorithm, and input the result obtained in S22 into the particle swarm optimization algorithm to solve;

In particle swarm optimization, the position of each particle represents a plan, and the speed determines the speed of the iterative evolution of the plan. In order to maintain the consistency of the algorithm, the expression of the position is the same as the expression of the chromosome in the genetic algorithm, and the speed is defined as two iterations for each particle. The amount of change in position, that is:

表示当前粒子的位置。pbest _r,d is the position of the current best individual, gbest _r,d is the position where the historical best fitness appears,

Indicates the current particle position.

表示S23要求每次粒子群算法启动前将速度重置为0，输入遗传算法的解并进行一次迭代之后，重新加入速度分量进行优化。In the above formula,

Indicates that S23 requires that the velocity be reset to 0 before each particle swarm optimization algorithm is started, and after inputting the solution of the genetic algorithm and performing one iteration, the velocity component is re-added for optimization.

表示粒子的惯性，即算法会尝试保持上一次策略迭代的变化方向，以期获得更好的结果，

表示粒子会向当前集体质心靠拢，

表示粒子会向历史最佳的位置靠拢。in,

represents the inertia of the particle, i.e. the algorithm will try to maintain the direction of change from the last policy iteration in the hope of getting better results,

means that the particles will move closer to the current collective center of mass,

Indicates that the particle will move closer to the historically optimal position.

The positions and velocities of all particles are updated in each iteration, and the iterative expressions of integer variables and real variables are as follows:

Among them, x represents an integer variable, and y represents a continuous variable.

Calculate the fitness of all particles, and record the position of the particle with the highest fitness as the current collective centroid. If the fitness is better than the best in history, record the current position as the best in history, and the current fitness as the best in history Spend.

After the particle swarm optimization iterates T ₂ times, it enters the next stage.

S24, re-input the solution obtained in S23 into the genetic algorithm, until the result converges or reaches the upper limit T of the number of iterations;

It should be noted that the genetic algorithm has good global search performance, but the convergence speed is slow. The particle swarm optimization algorithm converges faster, but it is easy to prematurely converge to a local optimum. Coarse-grained searches are performed using genetic algorithms, and the results are fed into particle swarm optimization to accelerate algorithm convergence. When the outer loop of the algorithm reaches the upper limit or the fitness of the best individual for several generations is no longer reduced, the algorithm ends, and the genetic information and fitness of the best individual are output.

S3. Decode the finally obtained historical best collective centroid to obtain a corresponding user access, channel allocation and power control strategy, and the historical best fitness is the minimum total power consumption of the system.

Those of ordinary skill in the art will appreciate that the embodiments described herein are intended to assist readers in understanding the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. Various modifications and variations of the present invention are possible for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the scope of the claims of the present invention.

Claims (4)

1. A heterogeneous cellular network interference coordination method specifically comprises the following steps:

step S1: under a simulated real downlink network link scene, determining an optimization problem by respectively taking a user QoS requirement, a macro base station maximum transmitting power, a micro base station maximum transmitting power, a base station access limit and a channel multiplexing limit as constraint conditions with the aim of minimizing the total transmitting power consumption of a base station;

the real link scenario simulated in step S1 includes: the mobile communication system comprises M mobile users and N base stations, wherein the N base stations comprise L macro base stations and N-L micro base stations, the N base stations are uniformly deployed to form a cell, and the M mobile users are randomly distributed in the cell; all base stations are deployed at the same frequency, the frequency band is divided into I orthogonal channels, each mobile user can be connected with only one base station and obtains service from one or more channels, signals transmitted by other base stations in the same channel can generate interference on the user, and interference does not exist between different channels;

the step S1 determines that the optimization problem is specifically a mixed integer nonlinear optimization problem;

the mixed integer nonlinear optimization problem specifically comprises the following steps: minimizing the total power consumption of the system as an objective function, respectively based on the QoS requirement v of the mobile user_mMacro base station maximum transmit power

Maximum transmitting power of micro base station

User access indicator a_n,mAnd a channel allocation indicator b_m,iObtaining a first optimization problem by taking the value of (a) as a constraint condition;

the optimization problem expression is as follows:

s.t.

C1:a_n,m∈{0,1}

C3:b_m,i∈{0,1}

wherein P1 represents the objective function of the optimization problem;

in constraint C1, a_n,mRepresenting the association variable of the user with the base station, a_n,m1 indicates that user m accesses base station n, a_n,mIf 0, it means that the user has not established connection with the base station;

constraint C2 ensures that a user has at least and at most one access to a base station;

in constraint C3, b_m,iRepresenting the occupation of the ith sub-channel by user m, b_m,i1 indicates that user m occupies the ith sub-channel and is interfered by signals transmitted by other base stations on the channel, b_m,i0 means that user m is not associated with channel i;

constraint C4 ensures that each sub-channel of a particular base station can only be occupied by one mobile subscriber, and that subscribers accessing different base stations reuse the same channel;

in the constraints C5 and C6,

corresponds to a macro base station n₁And a micro base station n₂On the channel i, the two constraints respectively ensure that the total transmission power of the macro base station and the micro base station does not exceed the respective upper limit P of the transmission power_M/P_S；

Constraint C7 ensures throughput C for user m_mMinimum data rate v to meet its QoS requirements_mMaximum transmission power P_M＝50w，P_SUser differentiated QoS requirements v 20w_mIs randomly distributed in [10 ]⁵,3×10⁵]bit/s；

The QoS of a mobile user, i.e. the data rate the user gets from the associated base station, is calculated as:

wherein, C_n,m,iFor the data rate provided by base station n for user m on the ith subchannel, the expression is as follows:

where B is the bandwidth of each orthogonal sub-channel, N₀Power spectral density, H, of background noise_n,mFor the channel gain model:

H_n,m＝PL_n,m×H_R×H_S

H_Ris Rayleigh fading, H, due to multipath effects_SFor random shadow fading, in particular PL_n,mThe expression is as follows:

wherein d is_n,mRepresents the distance between the user and the base station;

I_n,m,ifor the interference experienced by user m on the ith subchannel of base station n, the expression is as follows:

step S2: solving the optimization problem determined in the step S1;

specifically, a hybrid heuristic search algorithm based on a genetic algorithm and a particle swarm algorithm is used for solving, and the solving specifically comprises the following steps:

s21: using the same format to encode gene sequences in the genetic algorithm and position information in the particle swarm algorithm, and initializing the genetic algorithm;

s22: iteratively solving a group of feasible solutions by utilizing a finite genetic algorithm;

s23: initializing a particle swarm algorithm, and inputting a result obtained by the genetic algorithm into the particle swarm optimization algorithm for solving;

s24: inputting the solution obtained by the particle swarm optimization algorithm into the genetic algorithm again until the result is converged or the upper limit of the iteration times is reached;

step S3: and decoding the result obtained in the step S2 to obtain the corresponding user access, channel allocation and power control strategy.

2. The method for coordinating interference in heterogeneous cellular networks according to claim 1, wherein the chromosome sequences in the genetic algorithm and the location information in the particle swarm optimization algorithm used in step S21 are 0-1 integer matrices corresponding to the user access policy and the channel allocation policy, and real matrices related to power control of the base station, and the fitness expression of the individuals and the particles is the total power consumption of the optimization target base station of the problem.

3. The method according to claim 1, wherein in a major loop, after performing crossover operation and mutation operation on a parent population to generate a child population for each iteration of the genetic algorithm, recording an optimal strategy and optimal fitness, performing selection operation to enter the next iteration, after a limited number of iterations, forcibly stopping the genetic algorithm, and outputting the current population.

4. The method according to claim 1, wherein every time the particle swarm optimization algorithm is started, all particle speeds are reset to 0, a genetic algorithm is input to obtain gene information of a current population as position information, and after one iteration, the particle speeds are calculated and the iteration is executed.

Images