A Frequency Planning Method for Wireless Communication Networks Based on Genetic Algorithms

Abstract

This invention discloses a frequency planning method for wireless communication networks based on a genetic algorithm, relating to the field of frequency planning technology for wireless communication networks. The method includes: collecting network traffic, user distribution, base station location, antenna orientation, and historical interference data; preprocessing the collected data; integrating the preprocessed data and calculating interference values ​​to identify cell pairs with interference relationships; assessing the current network interference situation to obtain interference assessment results; using the interference assessment results as the basis for formulating a dynamic frequency reuse strategy, and adjusting the frequency reuse mode; constructing an interference prediction model using a random forest algorithm, outputting the predicted cell pairs and interference levels; adjusting frequency allocation based on the predicted cell pairs and interference levels; using the frequency allocation adjustment results as the initial population for genetic algorithm optimization, evaluating the fitness of each individual, and iteratively updating based on the fitness evaluation results.

Description

Technical Field

[0001] This invention relates to the field of frequency planning technology for wireless communication networks, and in particular to a frequency planning method for wireless communication networks based on a genetic algorithm. Background Technology

[0002] Frequency planning in wireless communication networks is crucial for ensuring efficient spectrum utilization and minimizing interference. As mobile communication technology has evolved from 2G and 3G to 4G and is gradually moving towards 5G and higher standards, the scarcity of frequency resources and the growth in user demand have made optimizing frequency allocation a complex and critical task. Traditional frequency planning methods mainly rely on static rules and fixed patterns. These methods performed well in early networks, but they are inadequate in the face of increasingly complex network environments. For example, in early cellular systems, frequency reuse patterns were usually fixed, meaning each cell used a predefined set of frequencies, while adjacent cells used different frequency combinations to avoid interference. However, this static approach cannot adapt to dynamically changing traffic distribution and user behavior, resulting in low spectrum utilization and potentially causing unnecessary interference.

[0003] In recent years, with the advancement of machine learning and artificial intelligence technologies, researchers have begun to explore more intelligent and adaptive frequency planning strategies. Among these, methods based on genetic algorithms (GA) have attracted widespread attention due to their global search capabilities and excellent handling of nonlinear problems. By simulating the survival-of-the-fittest principle in the natural selection process, GA can find optimal or near-optimal solutions in a large-scale solution space, making it suitable for solving complex combinatorial optimization problems. Furthermore, combining it with other intelligent algorithms such as random forests (RF) can further improve prediction accuracy and decision quality. Specifically, random forests can effectively handle high-dimensional data and provide powerful classification and regression capabilities, which is particularly important for building accurate interference prediction models. Nevertheless, existing technologies still have several shortcomings, especially in terms of real-time performance and flexibility. Most existing solutions lack effective utilization of historical interference data and fail to fully consider potential interference situations in the future, resulting in untimely and inaccurate frequency adjustments. In addition, traditional methods often ignore the influence of changes in user distribution and factors such as base station location and antenna orientation, which limits their effectiveness in practical applications. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a frequency planning method for wireless communication networks based on genetic algorithms to solve the problem that most existing solutions lack effective use of historical interference data and fail to fully consider potential interference in the future, resulting in untimely and inaccurate frequency adjustments.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a frequency planning method for wireless communication networks based on a genetic algorithm, comprising,

[0008] Collect network traffic, user distribution, base station location, antenna orientation and historical interference data, and preprocess the collected data;

[0009] By integrating the preprocessed data and calculating the interference value, cell pairs with interference relationships are identified, the interference situation in the current network is assessed, and the interference assessment result is obtained.

[0010] Based on the interference assessment results, a dynamic frequency reuse strategy is formulated, and the frequency reuse mode is adjusted.

[0011] An interference prediction model is constructed using the random forest algorithm. Historical interference data, base station location, and antenna orientation are input into the interference prediction model to predict the interference situation in the future period and output the predicted cell pairs and interference level.

[0012] The frequency allocation is adjusted based on the predicted cell pairs and interference levels to obtain the frequency allocation adjustment results;

[0013] The frequency allocation adjustment results are used as the initial population for genetic algorithm optimization. Each individual represents a frequency planning scheme. The fitness of each individual is evaluated, and the results are iteratively updated based on the fitness evaluation results.

[0014] As a preferred embodiment of the wireless communication network frequency planning method based on genetic algorithm described in this invention, the steps of collecting network traffic, user distribution, base station location, antenna orientation, and historical interference data, and preprocessing the collected data, are as follows:

[0015] Network traffic is collected through the Network Management System (NMS), user distribution is obtained using location information from GPS positioning services, base station locations are extracted from the operator's infrastructure database, antenna directions are extracted from base station configuration files, and historical interference data is obtained by compiling past fault reports and maintenance records.

[0016] The standard deviation method is used to identify and remove obviously erroneous and unreasonable data points. For partially missing data, interpolation is used to fill in the missing data appropriately.

[0017] Z-score normalization is used to convert data of different dimensions into the same dimension.

[0018] As a preferred embodiment of the wireless communication network frequency planning method based on genetic algorithm described in this invention, the steps of integrating preprocessed data, calculating interference values, identifying cell pairs with interference relationships, evaluating the interference situation in the current network, and obtaining interference evaluation results are as follows:

[0019] The preprocessed data is integrated into an interference matrix, and the interference value I is calculated based on the interference matrix. mk ;

[0020] Set I1, when I mk When I > I1, it is assumed that there is an interference relationship between cell m and cell k;

[0021] Traverse the interference matrix and find all expressions that satisfy I mk Cell pairs with values ​​greater than I1 are marked as interference pairs.

[0022] The number of interference pairs and the average interference value are used as evaluation indicators to assess the interference situation in the current network.

[0023] As a preferred embodiment of the frequency planning method for wireless communication networks based on genetic algorithms described in this invention, the specific steps for formulating a dynamic frequency reuse strategy based on interference evaluation results are as follows:

[0024] Based on the network interference assessment value, the adaptive dynamic frequency reuse strategy is selected as the optimal solution.

[0025] Calculate the adaptive reuse distance between adjacent base stations based on the interference intensity and distribution;

[0026] Based on the calculated adaptive reuse distance, frequencies are reallocated to each cell;

[0027] A frequency pool is obtained by combining all available frequencies;

[0028] Based on the interference assessment results, the cells are sorted from largest to smallest interference value, with priority given to cells with severe interference.

[0029] Traverse the sorted list of cells, select the frequency with the least interference to allocate to the current cell, and remove the allocated frequency from the frequency pool;

[0030] For each cell with an allocated frequency, check whether its neighboring cells meet the adaptive reuse distance requirement. If not, reselect a frequency for allocation until the reuse distance requirement is met.

[0031] As a preferred embodiment of the frequency planning method for wireless communication networks based on genetic algorithms described in this invention, the specific steps for adjusting the frequency reuse mode are as follows:

[0032] The frequency reuse mode is dynamically adjusted based on real-time interference assessment results.

[0033] Continuously monitor network interference, obtain real-time interference data, periodically calculate interference assessment results, and recalculate the adaptive reuse distance between base stations on the same frequency based on the new interference assessment results. If the new adaptive reuse distance does not match the current allocation, the frequency is reallocated.

[0034] As a preferred embodiment of the wireless communication network frequency planning method based on genetic algorithm described in this invention, the following steps are employed: A random forest algorithm is used to construct an interference prediction model. Historical interference data, base station location, and antenna direction are input into the interference prediction model to predict interference conditions over a future period. The predicted cell pairs and interference levels are then output.

[0035] The feature vector X is obtained by integrating historical interference data, base station location data, and antenna orientation data.

[0036] The random forest algorithm is used to train the disturbance prediction model. The random forest consists of T trees, and each tree randomly selects u features for splitting during training.

[0037] The preprocessed data is input into a trained random forest model to predict the disturbance situation in the future.

[0038] As a preferred embodiment of the frequency planning method for wireless communication networks based on genetic algorithms described in this invention, the specific steps for adjusting frequency allocation according to the predicted interfering cell pairs and interference levels to obtain the frequency allocation adjustment result are as follows:

[0039] A priority function is constructed based on the interference value, distance, and gain between adjacent cells to evaluate the priority F(m,k) of frequency allocation adjustment between cells m and k;

[0040] Set an adjustment threshold F1. When F(m,k) is greater than the adjustment threshold F1, it indicates that the frequency allocation between cell m and cell k needs to be adjusted.

[0041] As a preferred embodiment of the frequency planning method for wireless communication networks based on genetic algorithms described in this invention, the steps of using the frequency allocation adjustment results as the initial population for genetic algorithm optimization, evaluating the fitness of each individual, and iteratively updating based on the fitness evaluation results are as follows:

[0042] Based on the frequency allocation adjustment results, the initial population for the genetic algorithm is constructed, denoted as S = [s1, s2, ..., s]. R ], where R represents the total number of cells in the network, s R This represents the BCCH frequency of the Rth cell, where each individual cell represents a frequency planning scheme.

[0043] The fitness assessment results are obtained by evaluating the quality of each individual based on the initial population.

[0044] Based on the results of fitness evaluation, the selection, crossover, and mutation operations of the genetic algorithm are performed;

[0045] After completing the selection, crossover, and mutation operations of the genetic algorithm, a new generation of individuals is obtained. Then, the new generation of individuals is evaluated using the latest fitness function A(S), and the old individuals with low fitness are replaced to form a new population. This process is repeated iteratively until the maximum number of iterations is reached or the fitness no longer improves significantly.

[0046] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the frequency planning method for wireless communication networks based on genetic algorithms as described in the first aspect of the present invention.

[0047] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the frequency planning method for wireless communication networks based on genetic algorithms as described in the first aspect of the present invention.

[0048] The beneficial effects of this invention are as follows: By meticulously processing data, accurately assessing interference, formulating dynamic frequency reuse strategies, constructing interference prediction models, flexibly adjusting frequency allocation, and globally optimizing using genetic algorithms, this invention comprehensively improves the effectiveness of frequency planning for wireless communication networks. Each of these steps plays a unique role and together constitutes an intelligent and adaptive frequency planning system, which significantly improves spectrum utilization, reduces interference risks, and enhances network performance and quality of service. This method is particularly suitable for modern complex wireless communication networks, such as 4G LTE, 5G, and future 6G networks, demonstrating its broad application prospects in the field of wireless communication technology. Attached Figure Description

[0049] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart of the frequency planning method for wireless communication networks based on genetic algorithms in Example 1.

[0051] Figure 2 This is a flowchart of the frequency allocation adjustment results in Example 1. Detailed Implementation

[0052] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0053] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0054] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0055] Example 1, referring to Figure 1 and Figure 2 This is the first embodiment of the present invention, which provides a frequency planning method for wireless communication networks based on a genetic algorithm, including the following steps:

[0056] Step 1: Collect network traffic, user distribution, base station location, antenna orientation, and historical interference data, and preprocess the collected data;

[0057] Network traffic is collected through the Network Management System (NMS), user distribution is obtained using location information from GPS positioning services, base station locations are extracted from the operator's infrastructure database, antenna orientation, including the azimuth and downtilt angle of each antenna, is extracted from the base station configuration file, and historical interference data is obtained by sorting through past fault reports and maintenance records.

[0058] The standard deviation method is used to identify and remove obviously erroneous and unreasonable data points, such as traffic peaks that exceed the normal range or unreasonable user location coordinates. For partially missing data, interpolation is used to fill in the missing data.

[0059] Z-score normalization is used to convert data of different dimensions to the same dimension. The quality of the normalized data is then assessed using the following formula:

[0060]

[0061] Where Q represents the data quality assessment value, and n represents the number of data types. Index variables representing data, Indicates the first The cleanliness level of this data ranges from 0 to 1, with a higher value indicating a higher degree of cleanliness. Indicates the first The normalization degree of the data ranges from 0 to 1. The larger the value of 1, the higher the normalization degree. α represents the weight of the cleanliness and β represents the weight of the normalization.

[0062] The data quality assessment value Q ranges from 0 to 1, with a higher value indicating higher data quality.

[0063] Step 2: Integrate the preprocessed data and calculate the interference value to identify cell pairs with interference relationships, assess the interference situation in the current network, and obtain the interference assessment result.

[0064] The preprocessed data are integrated into an interference matrix, and the interference value is calculated based on the interference matrix. The expression is as follows:

[0065]

[0066] Where m and k represent the cell index variables, I mk P represents the interference value of cell m to cell k. m G represents the transmit power of cell m. mk d represents the antenna gain between cell m and cell k, d0 represents the target distance, and d mk Let Z0 represent the distance between cell m and cell k, γ represent the path loss exponent, θ represent the antenna azimuth difference, σ represent the standard deviation of the antenna azimuth difference, Z0 represent the noise power, and exp represent the natural exponential function.

[0067] I mk The range of is (0, +∞), and the larger the value, the more severe the interference.

[0068] Set I1, when I mk When I > I1, it is assumed that there is an interference relationship between cell m and cell k;

[0069] Traverse the interference matrix and find all expressions that satisfy I mk Cell pairs with values ​​greater than I1 are marked as interference pairs.

[0070] The interference situation in the current network is evaluated using the number of interference pairs and the average interference value as evaluation indicators. The expression is:

[0071]

[0072] Where E represents the network interference assessment value, and M represents the number of interference pairs;

[0073] The range of the network interference assessment value E is (0, +∞), and the larger the value, the more severe the interference in the network.

[0074] Step 3: Based on the interference assessment results, formulate a dynamic frequency reuse strategy and adjust the frequency reuse mode.

[0075] Based on network interference assessment values, an adaptive dynamic frequency reuse strategy is selected as the optimal solution. This strategy can flexibly adjust the frequency reuse mode according to real-time interference assessment results to maximize spectrum utilization and minimize interference.

[0076] Reasons for selection: It can dynamically adjust according to real-time interference conditions, adapt to network changes, is not limited to fixed multiplexing modes, and can be optimized according to interference intensity and distribution. By adjusting the multiplexing distance and frequency allocation, it can effectively improve spectrum utilization and reduce interference.

[0077] Based on the interference intensity and distribution, the adaptive reuse distance between adjacent base stations is calculated using the following expression:

[0078]

[0079] Among them, D mk I1 represents the adaptive reuse distance between cell m and cell k, I1 represents the preset interference value, and ∈ represents the distance adjustment coefficient.

[0080] D mk The range of is (0, +∞). The larger the value, the greater the minimum reuse distance required and the more stringent the interference control.

[0081] Based on the calculated adaptive reuse distance, frequencies are reallocated to each cell, following these principles: avoid using frequencies with significant interference in adjacent cells simultaneously, prioritize allocating frequencies with less interference to cells with severe interference, and ensure the continuity and stability of frequency allocation.

[0082] A frequency pool is obtained by combining all available frequencies;

[0083] Based on the interference assessment results, the cells are sorted from largest to smallest interference value, with priority given to cells with severe interference.

[0084] Traverse the sorted list of cells, select the frequency with the least interference to allocate to the current cell, remove the allocated frequency from the frequency pool, update the frequency pool, and ensure that neighboring cells do not use frequencies with high interference.

[0085] For each cell with an allocated frequency, check whether its neighboring cells meet the adaptive reuse distance requirement. If not, reselect a frequency for allocation until the reuse distance requirement is met. Specifically:

[0086] Calculate the actual distance between each frequency-allocated cell and all neighboring cells, and compare the calculated actual distance with the adaptive reuse distance previously calculated based on the interference assessment results;

[0087] If the actual distance is greater than or equal to the adaptive reuse distance, the reuse distance requirement is considered to be met.

[0088] If the actual distance is less than the adaptive reuse distance, it is considered that the reuse distance requirement is not met. When the reuse distance requirement is not met, the frequency is re-selected and allocated for the current cell or its neighboring cells. When re-selecting the frequency, the frequency with less interference is given priority until all cell pairs using the same frequency meet the adaptive reuse distance requirement.

[0089] Genetic algorithms are used to optimize frequency allocation, reducing interference and improving spectrum utilization.

[0090] Dynamic frequency reuse strategy refers to flexibly adjusting the frequency reuse mode based on interference assessment results, rather than using a fixed reuse mode;

[0091] Based on real-time interference assessment results, the frequency reuse mode is dynamically adjusted. The adjustment cycle can be set according to the actual network conditions, such as once per hour or once per day.

[0092] Continuously monitor network interference, obtain real-time interference data, periodically calculate interference assessment results, and recalculate the adaptive reuse distance between base stations on the same frequency based on the new interference assessment results. If the new adaptive reuse distance does not match the current allocation, the frequency is reallocated.

[0093] Adjusting the frequency reuse mode refers to adjusting the minimum distance between base stations on the same or adjacent frequencies according to the interference intensity and distribution, in order to reduce interference, reallocate frequencies to each cell, and avoid the simultaneous use of frequencies with greater interference in adjacent cells.

[0094] Frequency reuse refers to the technique of reusing the same frequency in different geographical areas to improve spectrum efficiency.

[0095] Step 4: Construct an interference prediction model using the random forest algorithm. Input historical interference data, base station location, and antenna direction into the interference prediction model to predict the interference situation in the future period and output the predicted cell pairs and interference level.

[0096] The feature vector X is obtained by integrating historical interference data, base station location data, and antenna orientation data, and its expression is:

[0097] X = [X1, X2, X3];

[0098] Where X1 represents historical interference data, X2 represents base station location, and X3 represents antenna direction data;

[0099] The random forest algorithm is used to train the disturbance prediction model. The random forest consists of T trees, and each tree randomly selects u features for splitting during training.

[0100] The preprocessed data is input into the trained random forest model to predict the disturbance situation in the future period. The expression is:

[0101]

[0102] Y = [Y1, Y2];

[0103] Where Y represents the prediction result, Y1 represents the predicted cell pairs with interference relationship, Y2 represents the probability distribution of the predicted interference value, and argm axm,k represents finding the cell pair (m,k) with the highest probability. Y represents the probability that the t-th tree predicts interference between cell m and cell k. 2t The degree of interference in the prediction of the t-th tree is represented, and T represents the total number of trees in the random forest;

[0104] The value range of Y1 is [0, 1], representing the probability of interference, and the value range of Y2 is [0, 1], representing different degrees of interference.

[0105] Step 5: Adjust the frequency allocation based on the predicted interfering cell pairs and the degree of interference to obtain the frequency allocation adjustment result;

[0106] A priority function is constructed based on the interference values, distance, and gain between adjacent cells to evaluate the priority F(m,k) of frequency allocation adjustment between cells m and k. The expression is as follows:

[0107]

[0108] Where δ1 represents the end time of the specified time interval, and δ0 represents the start time of the specified time interval. dδ represents the interference value between cell m and cell k during time δ, where dδ represents the time derivative element of the integral variable, and D mk G represents the adaptive reuse distance between cell m and cell k, λ represents the positive weighting coefficient, and G... mk This represents the antenna gain between cell m and cell k;

[0109] The range of F(m,k) is [-∞, +∞]. A larger positive value indicates a higher adjustment priority, and a negative value indicates the opposite.

[0110] Set an adjustment threshold F1. When F(m,k) is greater than the adjustment threshold F1, it indicates that the frequency allocation between cell m and cell k needs to be adjusted.

[0111] When F(m,k) is greater than the adjustment threshold F1, for each cell pair to be adjusted, check whether a new unused frequency that can meet the adaptive reuse distance requirement can be found from the frequency pool.

[0112] When a suitable frequency is found, it is allocated directly, and the frequency pool is updated; when no suitable frequency is found, frequencies from other nearby cells are reallocated to free up available frequencies.

[0113] Step 6: Use the frequency allocation adjustment results as the initial population for genetic algorithm optimization, evaluate the fitness of each individual, and iteratively update based on the fitness evaluation results;

[0114] Based on the frequency allocation adjustment results, the initial population for the genetic algorithm is constructed, denoted as S = [s1, s2, ..., s]. R ], where R represents the total number of cells in the network, s R This represents the BCCH frequency of the Rth cell, where each individual cell represents a frequency planning scheme.

[0115] The fitness assessment is obtained by evaluating the quality of each individual based on the initial population. This fitness assessment measures an individual's performance in reducing interference and improving spectrum utilization. The expression is:

[0116]

[0117] Where A(S) is the fitness function, which comprehensively evaluates the performance of a given frequency planning scheme S in reducing interference and improving spectrum utilization, U i D represents the number of users covered by the i-th cell. i This represents the adaptive reuse distance between the i-th cell in S and its neighboring cells. A larger value indicates a more stable frequency allocation. η represents a positive weighting coefficient.

[0118] The range of A(S) is [-∞, +∞]. The larger the positive value, the higher the fitness of the individual, that is, the better the frequency planning scheme. In this way, we can effectively select individuals with better performance.

[0119] Based on the results of fitness evaluation, the selection, crossover, and mutation operations of the genetic algorithm are performed;

[0120] Selection: Several individuals are randomly selected from the current population according to their fitness ratio as parents. Individuals with higher fitness have a greater probability of being selected, thus ensuring the transmission of excellent traits.

[0121] Crossover: Perform multi-point crossover on selected parent individuals to generate new offspring. The selection of crossover points is based on the geographical distribution between cells to maintain the spatial continuity of frequency allocation. For example, a two-point crossover strategy can be adopted, taking into account the frequency combination of multiple cells at the same time, to ensure that the new individuals inherit the advantages of the parent.

[0122] Mutation: Low-probability mutation operations are performed on newly generated offspring to randomly change the BCCH frequency points of certain cells in order to explore more possible frequency allocation schemes. The probability and magnitude of mutation can be dynamically adjusted according to historical optimization results to ensure the diversity of the search space.

[0123] After completing the selection, crossover, and mutation operations of the genetic algorithm, a new generation of individuals is obtained. Then, the new generation of individuals is evaluated using the latest fitness function A(S), and the old individuals with low fitness are replaced to form a new population. This process is repeated iteratively until the maximum number of iterations is reached or the fitness no longer improves significantly.

[0124] This embodiment also provides a computer device applicable to the frequency planning method for wireless communication networks based on genetic algorithms, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the frequency planning method for wireless communication networks based on genetic algorithms as proposed in the above embodiment.

[0125] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0126] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the frequency planning method for wireless communication networks based on genetic algorithms as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0127] In summary, this invention comprehensively improves the effectiveness of frequency planning in wireless communication networks through: meticulous data processing, accurate interference assessment, formulation of dynamic frequency reuse strategies, construction of interference prediction models, flexible adjustment of frequency allocation, and global optimization using genetic algorithms. Each of these steps plays a unique role and together constitutes an intelligent and adaptive frequency planning system, significantly improving spectrum utilization, reducing interference risks, and enhancing network performance and quality of service. This method is particularly suitable for modern complex wireless communication networks, such as 4G LTE, 5G, and future 6G networks, demonstrating its broad application prospects in the field of wireless communication technology.

[0128] In summary, the wireless communication network frequency planning method based on genetic algorithms proposed in this invention demonstrates excellent performance in several key indicators, including data processing accuracy, interference control effectiveness, frequency allocation flexibility, and user satisfaction. Particularly noteworthy is that this invention, by introducing intelligent algorithms and technical means, successfully addresses the shortcomings of existing technologies, such as the inability of static frequency reuse modes to adapt to dynamically changing network environments and the lack of predictive management of future interference trends. This method not only improves spectrum utilization but also significantly reduces interference risks, enhances network performance and service quality, and possesses significant innovative and practical value.

[0129] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A frequency planning method for wireless communication networks based on genetic algorithms, characterized in that: include: Collect network traffic, user distribution, base station location, antenna orientation and historical interference data, and preprocess the collected data; The preprocessed data is integrated into an interference matrix, and the interference value is calculated based on the interference matrix. Set interference threshold ,when > At that time, it was believed that the community and community There are interference relationships between them. Traverse the interference matrix and find all those that satisfy the condition. > The cell pairs are identified and marked as interference pairs. The number of interference pairs and the average interference value are used as evaluation indicators to assess the interference situation in the current network. The preprocessed data are integrated into an interference matrix, and the interference value is calculated based on the interference matrix. The expression is as follows: ； in, and The index variable representing the cell. Indicates the community For the community The interference value, Indicates the community The transmission power, Indicates the community Arrive at the community Antenna gain between Indicates the distance to the target. Indicates the community Arrive at the community The distance between them This represents the path loss index. Indicates the antenna directional angle difference. This represents the standard deviation of the antenna directional angle difference. Indicates noise power. Represents the natural exponential function; Based on the interference assessment results, a dynamic frequency reuse strategy is formulated, and the frequency reuse mode is adjusted. The specific steps are as follows: Based on the network interference assessment value, the adaptive dynamic frequency reuse strategy is selected as the optimal solution. Calculate the adaptive multiplexing distance between adjacent base stations based on the interference intensity and distribution. ; Based on the calculated adaptive reuse distance, frequencies are reallocated to each cell; A frequency pool is obtained by combining all available frequencies; Based on the interference assessment results, the cells are sorted from largest to smallest interference value, with priority given to cells with severe interference. Traverse the sorted list of cells, select the frequency with the least interference to allocate to the current cell, and remove the allocated frequency from the frequency pool; For each cell with an allocated frequency, check whether its neighboring cells meet the adaptive reuse distance requirement. If not, reselect a frequency for allocation until the reuse distance requirement is met. The frequency reuse mode is dynamically adjusted based on real-time interference assessment results. Continuously monitor network interference, obtain real-time interference data, periodically calculate interference assessment results, and recalculate the adaptive reuse distance between base stations on the same frequency based on the new interference assessment results. If the new adaptive reuse distance does not match the current allocation, the frequency is reallocated. The feature vector X is obtained by integrating historical interference data, base station location data, and antenna orientation data. The random forest algorithm is used to train the disturbance prediction model. The random forest consists of T trees, and each tree randomly selects u features for splitting during training. The preprocessed data is input into the trained random forest model to predict the disturbance situation in the future period. The frequency allocation is adjusted based on the predicted cell pairs and interference levels to obtain the frequency allocation adjustment results; The frequency allocation adjustment results are used as the initial population for genetic algorithm optimization. The fitness of each individual is evaluated, and the results are iteratively updated based on the fitness evaluation results.

2. The frequency planning method for wireless communication networks based on genetic algorithms as described in claim 1, characterized in that: The process involves collecting network traffic, user distribution, base station location, antenna orientation, and historical interference data, and then preprocessing the collected data. The specific steps are as follows: Network traffic is collected through the Network Management System (NMS), user distribution is obtained using location information from GPS positioning services, base station locations are extracted from the operator's infrastructure database, antenna directions are extracted from base station configuration files, and historical interference data is obtained by compiling past fault reports and maintenance records. The standard deviation method is used to identify and remove obviously erroneous and unreasonable data points. For partially missing data, interpolation is used to fill in the missing data appropriately. Z-score normalization is used to convert data of different dimensions into the same dimension.

3. The frequency planning method for wireless communication networks based on genetic algorithms as described in claim 1, characterized in that: The frequency allocation is adjusted based on the predicted interfering cell pairs and the degree of interference to obtain the frequency allocation adjustment result. The specific steps are as follows: A priority function is constructed based on the interference values, distance, and gain between neighboring cells to evaluate the cells. and Priority of frequency allocation adjustment ; Set the adjustment threshold F1, when When it is greater than the adjustment threshold F1, it indicates that the cell and community The frequency allocation between them needs to be adjusted.

4. The frequency planning method for wireless communication networks based on genetic algorithms as described in claim 3, characterized in that: The frequency allocation adjustment results are used as the initial population for genetic algorithm optimization. Each individual represents a frequency planning scheme. The fitness of each individual is evaluated, and iterative updates are performed based on the fitness evaluation results. The specific steps are as follows: Based on the frequency allocation adjustment results, the initial population for the genetic algorithm is constructed, denoted as: ,in, This represents the total number of cells in the network. Indicates the first The BCCH frequency of each cell; The fitness assessment results are obtained by evaluating the quality of each individual based on the initial population. Based on the results of fitness evaluation, the selection, crossover, and mutation operations of the genetic algorithm are performed; After completing the selection, crossover, and mutation operations of the genetic algorithm, a new generation of individuals is obtained. Then, the latest fitness function is used. The new generation of individuals is evaluated and replaced with older individuals with low fitness to form a new population. This process is repeated until the maximum number of iterations is reached or fitness no longer increases significantly.

Images