Method and apparatus for realizing optimization of base station antenna parameters

Abstract

The present invention discloses a method and apparatus for realizing optimization of base station antenna parameters. The method includes: determining the number of cells to be optimized and the number of base station antenna weight parameters; determining one or more service scenarios that meet the current service scenario as the solution space of the particle swarm, and performing time period cutting according to the start and end time points of the service scenario to obtain the number of cut time periods; determining the dimension of the position vector, and taking the optimization direction of the antenna weight as the velocity vector of a single particle; setting the number of particles in the particle swarm; initializing each particle; using the particle swarm algorithm to iteratively update the position vector and velocity vector of each particle until the iteration end condition is reached, then pulling all particles back to the solution space to determine the global historical optimal solution; determining each antenna weight according to the global historical optimal value. By using the present invention, the requirements for large-scale and multi-parameter optimization of the communication network can be met, and the efficiency and timeliness of network optimization can be improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a method and device for optimizing base station antenna parameters. Background Art

[0002] At present, with the large-scale application of 5G (fifth-generation mobile communication technology), high-speed, low-latency, and wide-coverage communication networks have become an urgent need and the goal pursued by the communications industry. However, with the advent of 5G, network optimization has become a thorny problem. In order to achieve goals such as large connections, low latency, and high capacity, it is necessary to optimize and adjust the base station parameters reasonably. Unlike in the past, due to its technical characteristics, 5G often requires more base stations to achieve the same coverage, and more base stations mean that the corresponding base station parameters have risen sharply. Faced with a large number of parameter settings, the traditional method of manual adjustment based on expert experience has become increasingly difficult. Only the mutual cooperation between a large number of base station clusters can achieve high-quality network coverage, and the order of magnitude of parameters of a large number of base stations has far exceeded the scope that can be accommodated by the traditional expert experience adjustment method. Moreover, in cities, users' spatiotemporal activities lead to tidal, fluctuating, and periodic changes in communication demand. Traditional methods cannot perform quantitative analysis when facing tidal optimization, and cannot combine users to produce targeted solutions, which invisibly increases the cost of network coverage. Summary of the invention

[0003] The present invention provides a method and device for realizing base station antenna parameter optimization, so that the optimization of communication network base station parameters can meet the requirements of large-scale, multi-parameter and tidal, and improve the optimization efficiency.

[0004] To this end, the present invention provides the following technical solutions:

[0005] In one aspect, the present invention provides a method for optimizing base station antenna parameters, the method comprising:

[0006] Determine the number m of cells to be optimized and the number k of base station antenna weight parameters;

[0007] Determine one or more business solutions that meet the current business scenario as the solution space of the particle swarm, and cut the time period according to the start and end time points of the business solution to obtain the number of cut time periods d;

[0008] Determine the dimension of the position vector according to the number m of cells to be optimized, the number k of base station antenna weight parameters, and the number d of time periods, and use the optimization direction of the antenna weight as the velocity vector of a single particle;

[0009] Set the number of particles N in the particle swarm;

[0010] For each particle, initialize the position vector using random numbers and ensure that the positions of all particles are within the solution space, and initialize the velocity vector using a zero vector; use the particle swarm algorithm to iteratively update the position vector and velocity vector of each particle until the iteration end condition is reached, then pull all particles back into the solution space to determine the global historical optimal solution;

[0011] Determine the antenna weights according to the global historical optimal value.

[0012] Optionally, the antenna weight parameters include: azimuth angle, downward tilt angle, horizontal beamwidth, vertical beamwidth.

[0013] Optionally, the service plan includes any one or more of the following fields: global cell identification code, service plan start time, duration, azimuth angle, downward tilt angle, horizontal beamwidth, vertical beamwidth, service plan score for service metrics.

[0014] Optionally, the service metrics include any one or more of the following: coverage target, capacity target, quality target, data service capability target of the current service scenario.

[0015] Optionally, setting the number N of particles in the particle swarm includes:

[0016] Set multiple candidate particle swarm sizes;

[0017] Based on each candidate particle swarm size respectively, use the particle swarm algorithm to iteratively update the position vector and velocity vector of each particle until the set stop condition is reached;

[0018] Determine the best candidate particle swarm size among them as the number N of particles in the particle swarm according to the iteration results corresponding to each candidate particle swarm size.

[0019] Optionally, the method further includes:

[0020] Determine the objective function value based on the updated position vector of each particle;

[0021] Determine the score of each particle according to the solution space and the objective function value;

[0022] Determine the historical optimal value of each current particle and the global historical optimal value according to the score of each particle.

[0023] Optionally, determining the objective function based on the updated position vector of each particle includes:

[0024] Based on the updated position vector of each particle, simulate the reference signal received power Rsrp value and / or signal-to-interference-plus-noise ratio Sinr value of each grid in the target area;

[0025] Calculate the objective function value according to the Rsrp value and / or the Sinr value.

[0026] Optionally, the pulling all particles back to the solution space includes:

[0027] For a particle that is not currently in the solution space, calculate the distances between the particle and each point in the solution space;

[0028] Update the position vector of the particle to the position vector of the point in the solution space that is closest to the particle.

[0029] Optionally, the iteration end condition is: reaching the set maximum number of iterations, or the global historical optimal value reaching the set threshold.

[0030] On the other hand, the present invention also provides a device for optimizing base station antenna parameters, the device includes:

[0031] An optimization cell determination module, configured to determine the number m of cells to be optimized and the number k of base station antenna weight parameters;

[0032] A solution space determination module, configured to determine one or more service scenarios that meet the current service scenario as the solution space of the particle swarm, and perform time period slicing according to the start and end time points of the service scenario to obtain the number d of sliced time periods;

[0033] An optimization parameter determination module, configured to determine the dimension of the position vector according to the number m of cells to be optimized, the number k of base station antenna weight parameters, and the number d of time periods, and use the optimization direction of the antenna weights as the velocity vector of a single particle;

[0034] A particle scale determination module, configured to set the number N of particles in the particle swarm;

[0035] An initialization module, configured to, for each particle, initialize the position vector using a random number, and place the positions of the particles in the solution space, and initialize the velocity vector using a zero vector;

[0036] An iterative processing module, configured to iteratively update the position vector and the velocity vector of each particle using the particle swarm algorithm until the iteration end condition is reached, pull all particles back to the solution space, and determine the global historical optimal solution;

[0037] An optimization parameter determination module, configured to determine each antenna weight according to the global historical optimal value.

[0038] Optionally, the particle scale determination module includes:

[0039] A candidate particle swarm scale setting unit, configured to set multiple candidate particle swarm scales;

[0040] An iterative optimization processing unit for iteratively updating the position vector and velocity vector of each particle based on each candidate particle swarm size using a particle swarm algorithm until a set stop condition is reached.

[0041] A selection unit for determining the best candidate particle swarm size among the iterative results corresponding to each candidate particle swarm size as the number N of particles in the particle swarm.

[0042] Optionally, the iterative processing module is further configured to determine an objective function value based on the updated position vector of each particle, determine the score of each particle according to the solution space and the objective function value, and determine the historical optimal value and the global historical optimal value of each current particle according to the score of each particle.

[0043] Optionally, the apparatus further includes:

[0044] A simulation module for simulating the reference signal received power Rsrp value and / or the signal-to-interference-plus-noise ratio Sinr value of each grid in the target area based on the updated position vector of each particle, and calculating the objective function value according to the Rsrp value and / or the Sinr value.

[0045] The iterative processing module obtains the objective function value by calling the simulation module.

[0046] The method and apparatus for optimizing base station antenna parameters provided by the present invention determine the number m of cells to be optimized and the number k of base station antenna weight parameters, determine one or more service scenarios that meet the current service scenario as the solution space of the particle swarm, and incorporate the time dimension. According to the number m of cells to be optimized, the number k of base station antenna weight parameters, and the number d of time periods, the dimension of the position vector is determined. The optimization direction of the antenna weight is used as the velocity vector of a single particle, and the particle swarm algorithm is used to automatically calculate and optimize the parameters of the cells to be optimized. Compared with the traditional manual on-site method, using the solution of the present invention, the optimization result can be automatically sent to the parameter platform in the form of a work order for automatic adjustment, without manual on-site adjustment, saving labor costs while improving the adjustment efficiency.

[0047] Furthermore, when performing parameter adjustment, the solution of the present invention uses an artificial intelligence algorithm to realize the automatic calculation and analysis of base station parameters. During this process, no manual intervention or participation is required, and it does not depend on the network optimization-related knowledge and work experience of operators, enabling anyone to quickly start operating and achieve professional-level parameter optimization and adjustment, greatly reducing the operator threshold and labor costs.

[0048] Furthermore, the artificial intelligence algorithm used in the solution of the present invention can autonomously learn and optimize the solution based on historical data. Compared with the expert experience algorithm, it has better universality and flexibility, is applicable to various different types of business scenarios, has a certain degree of self - adaptability and environmental perception, and can be adapted to different business scenarios.

[0049] Furthermore, in view of the characteristic that the pedestrian flow shows tidal changes over time, the solution of the present invention innovatively introduces the time dimension, which can enable the optimization of base station parameters to better adapt to the characteristics of the tidal changes in pedestrian flow. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a schematic diagram of the velocity and position of particles during the iterative process of the existing particle swarm algorithm;

[0051] Figure 2 is a flowchart of a method for implementing the optimization of base station antenna parameters according to the present invention;

[0052] Figure 3 is an example of multiple service solutions used in an embodiment of the present invention;

[0053] Figure 4 is for Figure 3 a schematic diagram of segmenting the time period for multiple service solutions used in the example;

[0054] Figure 5 is a schematic diagram of the initialization of the position vector in an embodiment of the present invention;

[0055] Figure 6 is a schematic diagram of the initialization of the velocity vector in an embodiment of the present invention;

[0056] Figure 7 is a schematic diagram of the relationship between the particle swarm size and the communication network optimization effect in an embodiment of the present invention;

[0057] Figure 8 is a schematic diagram of the relationship between the particle swarm size and the computational resource consumption in an embodiment of the present invention;

[0058] Figure 9 is a flowchart of using the particle swarm algorithm to iteratively update the position vector and velocity vector of each particle in an embodiment of the present invention;

[0059] Figure 10 is a schematic diagram of pulling particles back into the solution space in an embodiment of the present invention;

[0060] Figure 11 is an example of the historical optimal value of a single particle and the global historical optimal value in an embodiment of the present invention;

[0061] Figure 12The present invention is a schematic diagram of the structure of a device for optimizing base station antenna parameters. DETAILED DESCRIPTION

[0062] In traditional network optimization, users often give feedback about poor network quality and then make adjustments manually. This leads to untimely and inaccurate adjustments and high labor costs. Especially after 5G commercialization, the number of base stations has increased dramatically, and the corresponding labor costs have also increased dramatically. With the development of technology, base station parameters can now be adjusted remotely through parameter platforms, but manual adjustments still have the disadvantage of being ineffective and unable to coordinate multiple base stations. In some cases, multiple base stations may interfere with each other, and it is often impossible to quickly find the optimal solution. In addition, when faced with large-scale, multi-parameter optimization, manual calculation and adjustment alone takes a lot of time. At the same time, good optimization relies too much on the experience and network optimization knowledge of experts. People without a lot of professional experience and network optimization knowledge cannot make accurate, fast and good parameter adjustments.

[0063] The existing method of extracting expert experience into an algorithm, although it eliminates the step of manual calculation, is slow to update, overly dependent on expert experience, not universal, and cannot accurately predict regional traffic using expert experience, and lacks prediction of changes in human flow, and cannot quantify the impact of changes in the population on changes in communication needs, and therefore cannot adapt to the optimization needs of tidal changes in human flow. Therefore, in the face of tidal changes, in order to achieve wider coverage, more base station cells are often required, which invisibly increases the cost of base station construction and maintenance costs.

[0064] In response to the above problems, the embodiments of the present invention provide a method and device for optimizing base station antenna parameters. The method and device are combined with a particle swarm algorithm based on the expert experience algorithm, and use an artificial intelligence algorithm to perform autonomous learning and optimization based on historical data. Compared with the expert experience algorithm, the method and device are more universal and flexible, can be applied to a variety of different types of business scenarios, and have strong adaptability and environmental perception.

[0065] In order to enable persons skilled in the art to better understand the solutions of the embodiments of the present invention, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and implementation modes.

[0066] The solution of the present invention uses a particle swarm optimization algorithm to optimize the base station parameters. The following first briefly describes the particle swarm optimization algorithm (PSO).

[0067] The particle swarm algorithm is a random search algorithm based on group collaboration developed by simulating the foraging behavior of bird flocks. It mainly involves the following points:

[0068] 1) Initialization parameters:

[0069] Set the particle swarm size N, the dimension d of the position and velocity vectors. N represents the total number of particles. At each moment, each particle is in the solution space, that is, at each moment, the value of the objective function can be calculated for each particle. d represents the dimension of the solution space and is related to the objective function.

[0070] 2) Initialize the particles:

[0071] For each particle, initialize the position vector Xi and the velocity vector Vi using random numbers to disperse the particle positions as much as possible in the solution space.

[0072] 3) Iteration process:

[0073] For the t-th iteration, calculate the objective function value of each particle. For the i-th particle, record the position with the best objective function value in the history of this particle as Pi(t), and record the position with the best objective function value of all particles in the history as G(t).

[0074] To minimize y = x1 2 + x2 2 as an example, for particle i, after calculating t y values, take the position vector (x1, x2) corresponding to the minimum y as Pi(t). For all particles, after calculating N×t y values, take the position vector (x1, x2) corresponding to the minimum y as G(t).

[0075] Refer to Figure 1 as shown, update the velocity and position for the next iteration. Vi(t) is the velocity of the particle at the start of the t-th iteration, and Xi(t) is the position of the particle at the start of the t-th iteration, then:

[0076] Vi(t + 1) = wVi(t)+c1×r1×(Pi(t)-Xi(t)) + c2× r2× (G(t) - Xi(t));

[0077] Xi(t + 1) = Xi(t) + Vi(t + 1);

[0078] Among them, c1 and c2 are acceleration constants used to adjust the maximum learning step size; r1 and r2 are two random functions with values in the range [0,1], used to increase the search randomness; w is the inertia weight, a non-negative number, used to adjust the search range of the solution space.

[0079] 4) Judge convergence:

[0080] After reaching the maximum number of iterations or when it is judged that the solution reaches a certain threshold, output the global optimal historical optimal value as the optimal solution.

[0081] The method for optimizing a communication network provided by an embodiment of the present invention combines a particle swarm algorithm and performs autonomous learning and optimization based on historical data of base station parameters, effectively improving the efficiency of base station parameter optimization.

[0082] As Figure 2 shown, it is a flowchart of the method for optimizing base station antenna parameters according to the present invention, including the following steps:

[0083] Step 201, determine the number m of cells to be optimized and the number k of base station antenna weight parameters.

[0084] In an embodiment of the present invention, the antenna weight parameters can be set as needed. For example, they can include but are not limited to the following: azimuth angle, downward tilt angle, horizontal beamwidth, vertical beamwidth.

[0085] Step 202, determine one or more service scenarios that meet the current service scenario as the solution space of the particle swarm, and perform time period cutting according to the start and end time points of the service scenario to obtain the number d of cut time periods.

[0086] The purpose of determining the solution space is that during the process of optimizing antenna parameters using the particle swarm algorithm, it can meet the requirements of other service scenarios, and the final result output will limit the result within the solution space, that is, the service scenario, to avoid the situation where the objective function value is high but other measurement indicators are too poor.

[0087] In an embodiment of the present invention, the service scenario can include but is not limited to any one or more of the following fields: Cell Global Identity (CGI), service scenario start time, duration, azimuth angle, downward tilt angle, horizontal beamwidth, vertical beamwidth, score of the service scenario for service indicators. The CGI is used to identify the area covered by a cell (base station / a sector cell). In addition, the score of the service scenario for service indicators can be the score of one or more indicators. For example, the service indicators can include but are not limited to any one or more of the following: coverage target, capacity target, quality target, data service ability target of the current service scenario. Correspondingly, there is a score for each service indicator. It should be noted that when determining the solution space, normalization processing needs to be performed on each score so that the score is distributed between [0, 1].

[0088] It should be noted that the service scenarios and quantities of a cell within a day can be inconsistent, that is, different time periods can have different service scenarios, and the number of service scenarios used in different time periods can also be different. In view of this situation, in order to improve the algorithm performance and reduce the calculation amount, the service scenarios can be sliced at the maximum granularity according to time periods, specifically, slicing can be performed at the start and end time points of all service scenarios. As Figure 3 shown, it is a schematic diagram of all service scenarios used.Figure 4 is corresponding to Figure 3 a schematic diagram for segmenting time periods for multiple service scenarios used in the example. Finally, the number d of segmented time periods is obtained. Figure 4 In the shown example, the number d of segmented time periods is 6.

[0089] The number d of time periods is the number of time periods segmented according to the service scenarios at different time periods within a day. That is to say, taking one day as a unit, one day is segmented into d time periods.

[0090] Step 203: Determine the dimension of the position vector according to the number m of cells to be optimized, the number k of base station antenna weight parameters, and the number d of time periods, and use the optimization direction of the antenna weights as the velocity vector of a single particle.

[0091] Specifically, set the dimension D of the position vector to m×k×d. Where m is the number of cells to be optimized, k is the number of base station antenna weight parameters, and d is the number of time periods segmented according to the start and end time points of each service scenario.

[0092] In the embodiment of the present invention, the position vector of a single particle is represented by Xi, and the position vector of each particle represents all parameters of all cells. For example, if 50 cells need to be optimized at one time, the base station weight parameter k = 4, and the number d of time periods = 6, then the dimension of the position vector is dimensions.

[0093] The dimension of the velocity vector Vi is the same as that of the position vector Xi. The direction of the velocity vector Vi represents the optimization direction of the current particle, and the magnitude of the velocity vector Vi represents the optimization amount of the current particle.

[0094] Step 204: Set the number N of particles in the particle swarm.

[0095] The number N of particles in the particle swarm represents the particle swarm size. The position vector of each particle represents a solution, and the particle swarm size N represents that there are N solutions. The particle swarm algorithm will select a global optimal value from these N solutions as the final result.

[0096] Step 205: For each particle, initialize the position vector with a random number, and make the positions of all particles be in the solution space, and initialize the velocity vector with a zero vector.

[0097] For each particle, its position vector Xi(0) can be initialized with a random number, and its velocity vector Vi(0) can be initialized with a zero vector, where i = 1, 2,..., N. To ensure that the positions of all particles are within the solution space, the range of the random number can be limited.

[0098] For exampleFigure 5 and Figure 6 As shown respectively in Figure 6 and , schematic diagrams of position vector initialization and velocity vector initialization are shown respectively.

[0099] Step 206: Use the particle swarm algorithm to iteratively update the position vector and velocity vector of each particle. After reaching the iteration end condition, pull all the particles back to the solution space to determine the global historical optimal solution.

[0100] It should be noted that during the iteration process, particles are allowed to appear outside the solution space. However, after the iteration ends, if there are particles not in the solution space, the particles not in the solution space need to be pulled back to the solution space respectively. After all the particles are pulled back to the solution space, the global historical optimal solution is determined again.

[0101] The update calculation of the position vector and velocity vector during the iteration process, the objective function used, and the specific implementation method of how to pull the particles not in the solution space back to the solution space will be described in detail later.

[0102] Step 207: Determine each antenna weight according to the global historical optimal value.

[0103] The iteration end condition is: reaching the set maximum number of iterations, or the global historical optimal value reaching the set threshold.

[0104] It should be noted that the setting of the particle swarm size N will affect the consumption degree of computer resources and the optimization effect of the algorithm. Specifically, when N is too large, it is easier to find the global optimal value, that is, the best parameter setting value, but at the same time, it will increase the consumption of computing resources; when N is too small, it may fall into the local optimal value and fail to find the actual global optimal value, which will affect the optimization effect to a certain extent. Figure 7 and Figure 8 As shown respectively in Figure 7 and Figure 8 , the relationships between the particle swarm size and the communication network optimization effect, as well as the relationship between the particle swarm size and the consumption of computing resources are shown respectively. Considering these situations, in practical applications, an appropriate N value can be set in combination with the actual situation.

[0105] In a non - restrictive embodiment, when setting the particle swarm size N in the above - mentioned step 204, it can be set according to empirical values.

[0106] Further, in order to balance the relationship between the optimization effect and the consumption of computing resources, in another non-limiting embodiment, multiple candidate particle swarm sizes can be set first; based on each candidate particle swarm size, the particle swarm algorithm is used to iteratively update the position vector and velocity vector of each particle until the set stopping condition is reached; the best candidate particle swarm size is determined according to the iterative results corresponding to each candidate particle swarm size as the number N of particles in the particle swarm. For example, a relatively conservative value is set first, and then the value of N is incremented by a certain step (such as 3-5) to obtain multiple candidate N values. Iterative optimization is performed respectively based on multiple different N values but the same other parameters until the set stopping condition is reached (the stopping condition can specifically be iterating a certain number of times, or iterating for a certain period of time, or the optimization effect does not change significantly, etc.). According to the iterative results corresponding to the number N of particles in the multiple different particle swarms, the best number N of particles in the particle swarm is determined. Specifically, the best N value that balances the optimization effect and the consumption of computing resources can be selected by comprehensively considering the optimization effect and the consumption of computing resources in each case of the N value.

[0107] It should be noted that in the solution of the present invention, the Rsrp (Reference Signal Receiving Power) and / or Sinr (Signal to Interference plus Noise Ratio) mentioned later can be used as the index parameters for evaluating the optimization effect, and the optimization effect represents the sum of the objective function values corresponding to the optimized cells. The optimization effect does not change significantly means that the change in the optimization effect does not exceed the set value, such as not exceeding 5%.

[0108] The following combines Figure 9 to illustrate the above iterative update process in detail. The iterative process of this embodiment is illustrated by taking the set maximum number of iterations as an example.

[0109] Refer to Figure 9 , Figure 9 which is a flowchart for iteratively updating the position vector and velocity vector of each particle using the particle swarm algorithm in an embodiment of the present invention, including the following steps:

[0110] Step 901, initialize the iterative parameters: particle size N, acceleration constants c1 and c2, inertia weight w, maximum number of iterations T, random functions r1 and r2, and iteration round t = 0.

[0111] Among them, the acceleration constant c1 is used to learn its own experience, and the acceleration constant c2 is used to learn the group experience. When the value of c is too small, the convergence speed will be prolonged. When the value of c is too large, the speed will be too fast and the optimal value will be missed. In specific applications, the acceleration constants c1 and c2 can be determined according to experiments. For example, c1 = c2 = 2 can be set.

[0112] Among them, the random function r1 is used to introduce random fluctuations for learning its own experience, and r2 is used to introduce random fluctuations for learning the group experience. The purpose of setting the random function is to introduce non-linear capabilities into the algorithm to prevent the algorithm from falling into local optimal values. The values of r1 and r2 should not be too large or too small. When they are too large, it is difficult to converge due to excessive randomness. When they are too small, the algorithm will fall into local optimal values due to insufficient non-linear capabilities and thus cannot achieve the optimal effect. Therefore, in the embodiments of the present invention, r1 and r2 can be set to random(0,1), and better results can be obtained.

[0113] Among them, the inertia weight w affects the convergence speed of the algorithm. When the inertia weight w is too small, the convergence speed is too slow and it takes a long time or multiple iterations. When the inertia weight w is too large, the optimal value may be missed and the algorithm is also difficult to converge. Therefore, in the embodiments of the present invention, the value of the inertia weight w can be set in the following manner: the value of w is set to linearly decrease with the increase of the number of iterations. Specifically, a larger value is set at the beginning to accelerate the convergence speed of the algorithm, and this value can be appropriately adjusted smaller with the increase of the number of iterations to achieve the optimal effect. For example, a linearly decreasing manner from 0.9 to 0.1 can be selected.

[0114] Among them, the maximum number of iterations T affects the final optimization effect. When the value of T is too small, the number of algorithm iterations is insufficient and it ends before finding the optimal value, resulting in poor optimization effects. When the value of T is too large, it takes too much time, but the optimization effect does not increase significantly. Therefore, it is particularly important to select an appropriate maximum number of iterations, and a selection method similar to the particle swarm size N value can be used.

[0115] Step 902, input a service plan that meets the service scenario as the solution space of the particle swarm algorithm.

[0116] Step 903, determine the dimension D of the position vector.

[0117] The dimension D of the position vector is m×k×d, where m is the number m of cells to be optimized, k is the number of base station antenna weight parameters, and d is the number of time periods obtained by cutting the service plan according to different time periods within a day.

[0118] Step 904, initialize the particle position vector X and the velocity vector V according to the solution space, so that the particle positions are within the solution space.

[0119] It should be noted that the particle position X refers to the position vectors of all particles in the particle swarm. Similarly, the velocity vector V refers to the velocity vectors of all particles.

[0120] Then iterative calculations are started, that is, steps 905 to 914 below.

[0121] First, in step 905, the objective function value yi(t) is determined based on the position vector Xi(t) of each particle.

[0122] Specifically, in step 906, the simulation engine can be used to perform simulation calculations on the input data (i.e., the position vectors of each particle in the current round), and the objective function value yi(t) is calculated according to the simulation results.

[0123] Specifically, the weights of all cells (i.e., the position vectors of each particle in the current round) are input into the simulation software, and the values and / or values of all cells in the target area are obtained through simulation.

[0124] It should be noted that when inputting the position vector of a particle into the simulation engine, it is necessary to determine whether the particle is in the solution space. If it is not in the solution space, its position vector is directly assigned 0. The purpose is to avoid using a service plan that is not in the solution space as the historical optimal solution for other particles to learn, preventing the appearance of an unacceptable service plan. If the particle position is in the solution space, the position vector of the particle is input into the simulation engine as a parameter, and the simulation engine obtains the values and / or values of all cells in the target area, that is, the indicators for measuring the signal strength and / or signal-to-noise ratio of all cells. Then, the objective function value yi(t) is calculated according to the simulation results.

[0125] The target area can be defined as the smallest circumscribed rectangle of the non-empty value area in the simulation engine.

[0126] For example, when considering both the values and values of each cell in the target area, the objective function value yi(t) can be calculated according to the following formula:

[0127] ;

[0128] Where:

[0129] / (ScoreMax - ScoreMin);

[0130] ;

[0131] Where d is the number of time periods for cutting. For the current time period t of the influence time, taking the time period in Figure 4 as an example, the corresponding time of the first time period is 5 hours, and the corresponding time of the second time period is 4 hours, and so on. is the score of the current input position for the service indicator, which is an influence factor introduced when calculating the objective function; n is the number of input scores. In the target area, n non-adjusted cells are input as the basic coverage of the adjustment area, and according to the antenna type of the input basic coverage cells, empirical scores are assigned respectively as the basic scores when calculating the objective function. During the current influence time ScoreMin is the theoretically occurring socrePso minimum value according to the formula, and ScoreMax is the theoretically occurring socrePso maximum value according to the formula.

[0132] Among them, and are weight coefficients, and specific values can be taken according to different service scenarios. For example, if great importance is attached to while not paying attention to in the service scenario, then can be increased, can be decreased.

[0133] avgRsrp ;

[0134] ;

[0135] Among them, represents the allowed minimum value, which is generally set to -140;

[0136] represents the allowed minimum value, which is generally set to -126.

[0137] Among them, represents the j th grid of the simulation, , a represents the number of grid cells in the target area of the simulation, avgRsrp represents the Rsrp of the target area, avgSinr represents the Sinr of the target area.

[0138] Among them, is in the unit of dBm, is in the unit of dB.

[0139] It should be noted that in practical applications, the objective function may have other forms, and the embodiments of the present invention do not limit this.

[0140] In step 907, determine the score of each particle according to the objective function value.

[0141] Specifically, in one embodiment, the objective function value corresponding to each particle can be used as the score of the particle. Of course, in practical applications, the score calculation method is not limited to this.

[0142] The score of the particle can be used to represent the optimization effect.

[0143] In step 908, determine the historical optimal value Pi(t) of each particle and the historical optimal values of all particles, that is, the global historical optimal value G(t).

[0144] Determine the historical optimal value Pi(t) of each current particle and the global historical optimal value G(t) of all particles according to the score of each particle i. Specifically, after obtaining the score of each particle, the score can be compared with the historical optimal solution of the particle, and the larger value is assigned to the historical optimal solution of the particle; similarly, the global historical optimal solution can be obtained in a similar way. At this time, the score of the particle represents the quality of the current position, that is, the parameter. The higher the score, the better the parameter.

[0145] It should be noted that the score of the particle not in the solution space can be directly assigned a value of 0. The purpose is to prevent the particle not in the solution space from being learned by the particle swarm, resulting in an unsatisfactory final optimization effect. At the same time, it should be noted that an overly sparse solution space will result in too many particles with a score of 0. At this time, it is necessary to increase the particle swarm size to accelerate the convergence speed. Of course, this situation will increase the consumption of computing resources.

[0146] In step 909, determine whether the maximum number of iterations is reached, that is, judge whether t is greater than T. If so, execute step 914; otherwise, execute step 910.

[0147] In step 910, for each particle i, update its position vector Xi(t) and velocity vector Vi(t).

[0148] Specifically, the update can be performed according to the following formula:

[0149] ;

[0150] Xi(t + 1) = Xi(t) + Vi(t + 1);

[0151] In step 911, judge whether the number of iterations t is less than the maximum number of iterations T; if so, execute step 912; otherwise, execute step 913.

[0152] In step 912, increment the iteration count t by 1, then return to step 905 to perform the next iteration optimization process.

[0153] In step 913, calculate the distance between the particle position and the solution space, and pull the particle back into the solution space. Then return to step 906 to re - call the simulation engine for simulation calculation and update the objective function value yi(t).

[0154] Specifically, for each particle not in the solution space, the distance of each particle to all service solutions in the solution space can be calculated respectively. For example, use the distance formula in multi - dimensional space to calculate the distance of the current particle to each point in the solution space, and update the position vector of the current particle to the position vector of the nearest point in the solution space.

[0155] The distance formula is:

[0156]

[0157] Where n represents the vector dimension, i represents the iteration of n, x is the value of the i - th dimension of the particle position vector, and y is the value of the i - th dimension of the solution space position vector.

[0158] As Figure 10 shown, it is a schematic diagram of pulling the particle back into the solution space in an embodiment of the present invention. Among them, each arrow - marked line segment represents an iteration process. During the iteration process, particle A sometimes appears outside the solution space. After the iteration process ends, the corresponding position of particle A is , by calculating the distance between and each point in the solution space, according to the calculation result, it can be known that the position of point B in the solution space is the nearest, so update the position of particle A to the position of point B.

[0159] After all particles are pulled back into the solution space, it is also necessary to input all particles into the simulation engine to recalculate the objective function value and determine the global historical optimal solution.

[0160] In step 914, output the global historical optimal value G(t).

[0161] As Figure 11 shown, it is an example of the historical optimal value of a single particle and the global historical optimal value in an embodiment of the present invention.

[0162] The method for optimizing base station antenna parameters provided by the present invention determines the number m of cells to be optimized and the number k of base station antenna weight parameters, determines one or more service scenarios that meet the current service scenario as the solution space of the particle swarm, incorporates the time dimension, determines the dimension of the position vector according to the number m of cells to be optimized, the number k of base station antenna weight parameters, and the number d of time periods, takes the optimization direction of the antenna weight as the velocity vector of a single particle, and uses the particle swarm algorithm to automatically calculate and optimize the parameters of the cells to be optimized. Compared with the traditional method of manually adjusting parameters on site, the method for optimizing base station antenna parameters provided by the present invention can realize the automatic adjustment of network parameters, greatly save manpower, and improve the timeliness of adjustment at the same time.

[0163] Correspondingly, the present invention also provides a device for optimizing base station antenna parameters, as Figure 12 shown, a schematic structural diagram of the device.

[0164] The device 120 for optimizing base station antenna parameters provided in this embodiment includes the following modules:

[0165] The cell to be optimized determination module 121 is used to determine the number m of cells to be optimized and the number k of base station antenna weight parameters;

[0166] The solution space determination module 122 is used to determine one or more service scenarios that meet the current service scenario as the solution space of the particle swarm, and perform time period cutting according to the start and end time points of the service scenario to obtain the number d of cut time periods;

[0167] The optimization parameter determination module 123 is used to determine the dimension of the position vector according to the number m of cells to be optimized, the number k of base station antenna weight parameters, and the number d of time periods, and take the optimization direction of the antenna weight as the velocity vector of a single particle;

[0168] The particle scale determination module 124 is used to set the number N of particles in the particle swarm;

[0169] The initialization module 125 is used to initialize the position vector for each particle using random numbers, and make the positions of the particles in the solution space, and initialize the velocity vector using a zero vector;

[0170] The iterative processing module 126 is used to iteratively update the position vector and velocity vector of each particle using the particle swarm algorithm until the iteration end condition is reached, pull all the particles back to the solution space, and determine the global historical optimal solution;

[0171] The optimization parameter determination module 127 is used to determine each antenna weight according to the global historical optimal value.

[0172] It should be noted that in a specific embodiment, the particle swarm size N can be set according to empirical values. Considering that the setting of the particle swarm size N will affect the consumption degree of computer resources and the optimization effect of the algorithm. Specifically, when N is too large, it is easier to find the global optimal value, that is, the best parameter setting value, but at the same time, it will increase the consumption of computing resources; when N is too small, it may fall into a local optimal value and fail to find the actual global optimal value, which will affect the optimization effect to a certain extent. Therefore, in another specific embodiment, the particle size determination module 902 can first set multiple different N values, perform optimization iterations based on the same parameters and different N values, and select the best N value according to the iteration results. Correspondingly, in this embodiment, the particle size determination module 124 may include the following units:

[0173] A candidate particle swarm size setting unit, configured to set multiple candidate particle swarm sizes;

[0174] An iterative optimization processing unit, configured to respectively iterate and update the position vector and velocity vector of each particle by using the particle swarm algorithm based on each candidate particle swarm size until the set stop condition is reached;

[0175] A selection unit, configured to determine the best candidate particle swarm size among them as the number N of particles in the particle swarm according to the iteration results corresponding to each candidate particle swarm size.

[0176] In the embodiment of the present invention, the determination of the global historical optimal value can comprehensively consider the Rsrp value and / or Sinr value in the target area. Therefore, an objective function related to the Rsrp value and / or Sinr value is set, and the score of each particle is determined according to the objective function value, and then the global historical optimal value is obtained. Correspondingly, the iterative processing module 126 is further configured to determine the objective function value based on the updated position vector of each particle, determine the score of each particle according to the solution space and the objective function value, and determine the historical optimal value and global historical optimal value of each current particle according to the score of each particle.

[0177] For the convenience of calculation, a simulation module (not shown) can be set, which is configured to simulate and obtain the reference signal receiving power Rsrp value and / or signal-to-interference-plus-noise ratio Sinr value of each grid in the target area based on the updated position vector of each particle, and calculate the objective function according to the Rsrp value and / or the Sinr value.

[0178] Correspondingly, the iterative processing module 126 can obtain the value of the objective function by calling the simulation module. Specifically, the iterative processing module 126 inputs the updated position vector of each particle into the simulation module, and the simulation module outputs the calculated objective function value to the iterative processing module 126 after calculation.

[0179] The setting and calculation method of the objective function can be referred to the description in the method embodiment of the present invention above, and will not be elaborated here.

[0180] For the specific implementation manners of the modules in the device for realizing the optimization of base station antenna parameters of the present invention, reference can be made to the description in the method embodiment of the present invention above, and will not be elaborated here.

[0181] The device for realizing the optimization of base station antenna parameters provided by the present invention determines the number m of cells to be optimized and the number k of base station antenna weight parameters, determines one or more service scenarios that meet the current service scenario as the solution space of the particle swarm, integrates the time dimension, determines the dimension of the position vector according to the number m of cells to be optimized, the number k of base station antenna weight parameters, and the number d of time periods, uses the optimization direction of the antenna weights as the velocity vector of a single particle, and uses the particle swarm algorithm to automatically calculate and optimize the parameters of the cells to be optimized. Using the solution of the present invention, the optimization adjustment of the base station antenna parameters can be realized efficiently and accurately, without manual on-site adjustment, saving manpower, improving timeliness, and reducing the limitation of weather on the adjustment; moreover, there is no need for the operator to have complete network optimization-related knowledge and work experience, reducing the threshold of users, and a continuous and stable optimization effect can be achieved.

[0182] The solution of the present invention uses an artificial intelligence algorithm, has the characteristics of environmental perception ability, wide application range, rich adaptable service types, etc., and is more universal than the expert experience algorithm, and can be adapted to different service scenarios.

[0183] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0184] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the differences between each embodiment and other embodiments are emphasized. Moreover, the system embodiments described above are only illustrative. The modules and units described as separate components may or may not be physically separated, that is, they may be located on one network unit, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative work.

[0185] The above has introduced the embodiments of the present invention in detail. In this text, specific implementation manners are used to elaborate on the present invention. The descriptions of the above embodiments are only used to help understand the method and device of the present invention. They are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention. The content of this specification should not be construed as a limitation to the present invention. Therefore, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for optimizing base station antenna parameters, characterized in that The method includes: Determining the number m of cells to be optimized and the number k of base station antenna weight parameters; Determining one or more service scenarios that meet the current service scenario as the solution space of the particle swarm, and performing time period cutting according to the start and end time points of the service scenario to obtain the number d of cut time periods; Determining the dimension of the position vector according to the number m of cells to be optimized, the number k of base station antenna weight parameters, and the number d of time periods, and taking the optimization direction of the antenna weight as the velocity vector of a single particle; Setting the number N of particles in the particle swarm; For each particle, initializing the position vector with random numbers and making the positions of the particles within the solution space, and initializing the velocity vector with a zero vector; using the particle swarm algorithm to iteratively update the position vector and velocity vector of each particle until the iteration end condition is reached, then pulling all particles back to the solution space to determine the global historical optimal solution; Determining each antenna weight according to the global historical optimal value; Wherein, the service scenario includes any one or more of the following fields: global cell identification code, service scenario start time, duration, azimuth angle, tilt angle, horizontal beamwidth, vertical beamwidth, score of the service scenario for service metrics.

2. The method according to claim 1, characterized in that The antenna weight parameters include: azimuth angle, tilt angle, horizontal beamwidth, vertical beamwidth.

3. The method according to claim 2, wherein The service metrics include any one or more of the following: coverage target, capacity target, quality target, data service capability target of the current service scenario.

4. The method according to claim 1, characterized in that The setting the number N of particles in the particle swarm includes: Setting multiple candidate particle swarm sizes; Based on each candidate particle swarm size respectively, using the particle swarm algorithm to iteratively update the position vector and velocity vector of each particle until the set stop condition is reached; Determining the best candidate particle swarm size among them as the number N of particles in the particle swarm according to the iteration results corresponding to each candidate particle swarm size.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: Determining the objective function value based on the updated position vector of each particle; Determining the score of each particle according to the solution space and the objective function value; Determining the historical optimal value of each current particle and the global historical optimal value according to the score of each particle.

6. The method according to claim 5, wherein The determining the objective function based on the updated position vector of each particle includes: Based on the updated position vector of each particle, simulating to obtain the reference signal received power Rsrp value and / or signal-to-interference-plus-noise ratio Sinr value of each grid in the target area; Calculating the objective function value according to the Rsrp value and / or the Sinr value.

7. The method according to claim 6, wherein The pulling all particles back to the solution space includes: For the particles that are not currently in the solution space, calculating the distances between the particles and the points in the solution space; Updating the position vector of the particle to the position vector of the point in the solution space that is closest to the particle.

8. The method according to claim 6, wherein The iteration end condition is: reaching the set maximum number of iterations, or the global historical optimal value reaching the set threshold.

9. An apparatus for optimizing base station antenna parameters, characterized in that, The device includes: A cell to be optimized determination module, configured to determine the number m of cells to be optimized and the number k of base station antenna weight parameters; A solution space determination module, configured to determine one or more service scenarios that meet the current service scenario as the solution space of the particle swarm, and perform time period cutting according to the start and end time points of the service scenario to obtain the number of cut time periods d; An optimization parameter determination module, configured to determine the dimension of the position vector according to the number m of cells to be optimized, the number k of base station antenna weight parameters, and the number d of time periods, and use the optimization direction of the antenna weights as the velocity vector of a single particle; A particle scale determination module, configured to set the number N of particles in the particle swarm; An initialization module, configured to, for each particle, initialize the position vector using random numbers, and ensure that the positions of the particles are within the solution space, and initialize the velocity vector using a zero vector; An iterative processing module, configured to iteratively update the position vector and velocity vector of each particle using the particle swarm algorithm until the iterative end condition is reached, pull all particles back to the solution space, and determine the global historical optimal solution; An optimization parameter determination module, configured to determine each antenna weight according to the global historical optimal value; Wherein, the service scenario includes any one or more of the following fields: global cell identification code, service scenario start time, duration, azimuth angle, tilt angle, horizontal beamwidth, vertical beamwidth, and the score of the service scenario for service metrics.

10. The device according to claim 9, wherein The particle scale determination module includes: A candidate particle swarm scale setting unit, configured to set multiple candidate particle swarm scales; An iterative optimization processing unit, configured to respectively perform iterative update on the position vector and velocity vector of each particle using the particle swarm algorithm based on each candidate particle swarm scale until a set stop condition is reached; A selection unit, configured to determine the best candidate particle swarm scale among them as the number N of particles in the particle swarm according to the iterative results corresponding to each candidate particle swarm scale.

11. The apparatus according to claim 9 or 10, wherein The iterative processing module is further configured to determine the objective function value based on the updated position vector of each particle, determine the score of each particle according to the solution space and the objective function value, and determine the historical optimal value and the global historical optimal value of each particle currently according to the score of each particle.

12. The device according to claim 11, characterized in that, The apparatus further includes: A simulation module, configured to simulate the reference signal received power Rsrp value and / or the signal-to-interference-plus-noise ratio Sinr value of each grid in the target area based on the updated position vector of each particle, and calculate the objective function value according to the Rsrp value and / or the Sinr value; The iterative processing module obtains the objective function value by calling the simulation module.

Images