Mobility load balancing processing method and system based on weight adaptive adjustment

Abstract

The invention relates to the technical field of communication, in particular to a mobility load balancing processing method and system based on weight adaptive adjustment. Comprising the following steps: generating a plurality of communication cells according to a communication structure, and constructing an association transfer model; obtaining load monitoring packets of each communication cell according to a preset time node, and generating a load abnormal value of each communication cell according to all the load monitoring packets; setting a plurality of risk cells according to all the load abnormal values, and generating an adjustment sub-strategy of each risk cell according to a preset load adjustment model; the method comprises the following steps: analyzing historical execution records of each communication cell, setting a plurality of load adjustment models, setting initial weight parameters of each load adjustment scene to ensure that the weight is adjusted in the direction of improving the overall communication quality, and setting an iteration sub-model of each load adjustment scene to improve the overall communication quality. The system oscillation caused by the weight parameter optimization process is avoided, and the weight self-adaption efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a mobility load balancing processing method and system based on weight adaptive adjustment. Background Technology

[0002] Mobile communication systems often experience load imbalances between base stations due to uneven user distribution and dynamic changes in traffic volume, resulting in congestion in some cells while neighboring cells have idle resources. This imbalance reduces network resource utilization and affects the communication quality of the mobile communication system.

[0003] To address these issues, mobility load balancing technology has been introduced. Its core principle is to guide users to migrate from high-load cells to low-load cells by adjusting inter-cell handover parameters, thereby achieving load balancing. Traditional mobility load balancing methods often rely on fixed weights or empirical values ​​to set handover parameters, lacking accurate perception of real-time network status and user dynamic characteristics. This can easily lead to increased handover failure rates or over-compensation, impacting the communication efficiency of mobile communication systems. Summary of the Invention

[0004] The purpose of this application is to solve the above-mentioned technical problems by providing a mobility load balancing processing method and system based on weight adaptive adjustment, which aims to ensure the communication quality and load balancing of mobile communication systems.

[0005] In some embodiments of this application, by analyzing the historical execution records of each communication cell, multiple load adjustment models are set, and by setting initial weight parameters for each load adjustment scenario, the weights are ensured to be adjusted in a direction that improves the overall communication quality. Simultaneously, by setting iterative sub-models for each load adjustment scenario, system oscillations caused by the optimization process of the weight parameters are avoided, thus improving the efficiency of weight adaptation. This ensures the communication quality and load balancing of the mobile communication system.

[0006] In some embodiments of this application, a mobility load balancing processing method based on weight adaptive adjustment is provided, including: Multiple communication cells are generated based on the communication structure, and an associated transfer model is constructed. The load monitoring packets of each communication cell are obtained according to the preset time nodes, and the load anomaly values ​​of each communication cell are generated based on all the load monitoring packets. Multiple risk cells are set based on all abnormal load values, and adjustment sub-strategies for each risk cell are generated based on a preset load adjustment model. This involves generating multiple communication cells, including: Establish a sequence of communication cells A, A=(a1, a2, ..., a3) i …a n ), where a iLet i be the i-th communication cell; n is the number of communication cells.

[0007] In some embodiments of this application, a preset load adjustment model is included, including: Multiple operational characteristic indicators are generated based on historical execution records; Establish multiple load adjustment scenarios based on all operational characteristic indicators; Establish a load adjustment scenario sequence B, B=(b1,b2…b i …b m ), where b i Let m be the i-th load adjustment scenario; m is the number of load adjustment scenarios. b is set sequentially according to the load adjustment scenario sequence B. i Target load adjustment scenario; Generate the initial weighting strategy and iterative sub-model for the target load adjustment scenario; Generate a regulation sub-model for the target load regulation scenario based on the initial weighting strategy and the iterative sub-model; The adjustment sub-models for each load adjustment scenario are generated sequentially; A load regulation model is generated based on all the regulation sub-models.

[0008] In some embodiments of this application, multiple risk cells are set based on all abnormal load values, including: Based on the communication cell sequence A, a is set sequentially. i For the target communication cell; Generate the load anomaly value f of the target communication cell at the current monitoring time node based on the load monitoring packet; The load anomaly values ​​for each communication cell at the current monitoring time point are generated sequentially; Establish a sequence of load anomalies F, F=(f1,f2…f i …f n ), where f i The load anomaly value of the i-th communication cell at the current monitoring time point; Preset load anomaly threshold F1; If f i >F1, set the i-th communication cell as a risk cell at the current monitoring time point.

[0009] In some embodiments of this application, generating the load anomaly value f of the target communication cell at the current monitoring time node includes: f=g*[ Y1(i)*(s i -s' i )]; g=U1*[ Y2(i)*(s i -s1i )]; Where g is the load compensation coefficient; θ1 is the number of load indicators; s i s' is the reference value of the i-th load indicator of the target communication cell at the current monitoring time point; i U1 is the standard threshold for the i-th load metric; U1 is the preset first conversion coefficient; s 1i Y1(i) is the first-level reference value of the i-th load index of the target communication cell generated based on the correlation transfer model; Y1(i) is the selection coefficient; if (s i -s' i If )≥0,Y1(i)=1; if (s i -s' i If ) < 0, Y1(i) = 0; Y2(i) is the selection coefficient; if (s i -s 1i If )≥0,Y2(i)=1; if (s i -s 1i If ) < 0, Y2(i) = 0.

[0010] In some embodiments of this application, adjustment sub-strategies for each risk cell are generated according to a preset load adjustment model, including: Establish a sequence A1 of risky communities at the current monitoring time point; A1=(a 11 a 12 …a 1i …a 1n1 ), where a 1i Let n1 be the i-th risky cell at the current monitoring time point; n1 is the number of risky cells at the current monitoring time point; and n1 <n; Based on the risk zone sequence A1, a is set sequentially. 1i Targeted high-risk communities; Obtain the operational data packets of the target high-risk cell; The primary adjustment model for the target risk cell is set based on the running data packet; Based on the correlation transfer model, establish the correlation sequence A2 of the target risk area, A2=(a 21 a 22 …a 2i …a 2n2 ), where a 2i Let be the i-th associated cell of the target risk cell; n2 is the number of associated cells of the target risk cell; Based on the primary regulation model and the associated cell sequence A2, a regulation sub-strategy is generated for the target risk cell; The adjustment sub-strategies for each risk zone are generated sequentially.

[0011] In some embodiments of this application, a first-level adjustment model for the target risk cell is defined, including: b is set sequentially according to the load adjustment scenario sequence B. i For the comparison and adjustment scenario; Generate a matching value d between the target risk cell and the scenario to be compared and adjusted based on the running data packet; d=U2*[ (v i -v' i )]; Wherein, U2 is the preset second conversion coefficient; The number of operational characteristic indicators; v i To generate a reference value for the i-th runtime characteristic indicator based on the runtime data packet; v' i This is the reference value for the i-th operational characteristic indicator in the scenario to be compared and adjusted; The matching values ​​for the target risk cell and each load adjustment scenario are generated sequentially; The adjustment sub-model corresponding to the load adjustment scenario with the maximum value among all matching values ​​is set as the first-level adjustment model of the target risk cell.

[0012] In some embodiments of this application, the adjustment sub-strategy for generating the target risk cell includes: Generate a first-level weighting strategy and a first-level iterative model based on the first-level adjustment model; Multiple transition time nodes are generated based on the first-level iterative model, and a sequence of transition time nodes T, T = (t1, t2, ..., t3) is established. i …t r ), where t i Let r be the i-th transfer time node; r is the number of transfer time nodes; Based on the first-level weighting strategy, the transfer evaluation value of each associated cell in the associated cell sequence A2 at the transfer time node t1 is generated; The transfer sub-strategy is set based on all transfer evaluation values ​​at transfer time node t1; Obtain the feedback data packet at the transfer time node t2 and generate the optimized profit value h1; The correction instructions for the first-level weight strategy are generated based on the optimized return value and the first-level iterative model. A secondary weighting strategy is generated based on the correction results; The transition sub-strategy for transition time node t2 is generated based on the second-level weight strategy; Obtain the feedback data packet at the transfer time node t3 and generate the optimized profit value h2; Determine whether to generate a stop instruction based on the optimized profit value h1 and the optimized profit value h2; If h1 ≥ h2, generate a stop instruction; If h1 < h2, generate a transfer sub-strategy for the transfer time node t3, and perform iterative loops until a stop instruction is generated or the transfer time ends; Generate an adjustment sub-strategy for the target risk cell according to the stop instruction.

[0013] In some embodiments of the present application, the transfer sub-strategy for setting the transfer time node t1 includes: Set a 2i as the target associated cell according to the associated cell sequence A2 in sequence; Generate a transfer evaluation value k for the target associated cell at the transfer time node t1; k = U3 * β i * s 2i ; Among them, U3 is a preset third conversion coefficient; θ1 is the number of load indicators; β i is the weight coefficient for setting the i-th load indicator based on the first-level weight strategy; s 2i is the reference value of the i-th load indicator of the target associated cell at the transfer time node t1; Generate the transfer evaluation values of each associated cell at the transfer time node t1 in sequence; Set the associated cell corresponding to the maximum value among all the transfer evaluation values as the target transfer cell at the transfer time node t1; Set the target transfer amount according to the first-level iteration model; Generate a transfer sub-strategy for the transfer time node t1 according to the target transfer amount and the target transfer cell.

[0014] In some embodiments of the present application, a mobility load balancing processing system based on weight adaptive adjustment is provided, including: A central control unit, configured to generate multiple communication cells according to a communication structure and construct an associated transfer model; A monitoring unit, configured to obtain load monitoring packets of each communication cell according to a preset time node; The central control unit includes: A first processing module, configured to generate load outliers of each communication cell according to all the load monitoring packets; A second processing module, configured to set multiple risk cells according to all the load outliers; <9000221>A third processing module, configured to generate an adjustment sub-strategy for each risk cell according to a preset load adjustment model. <00 / / 00223>In some embodiments of the present application, the central control unit further includes: A fourth processing module, configured to establish a communication cell sequence A, A = (a1, a2…a i …a n ), where ai Let i be the i-th communication cell; n is the number of communication cells; The fifth processing module is used to generate multiple runtime characteristic indicators based on historical execution records; Establish multiple load adjustment scenarios based on all operational characteristic indicators; Establish a load adjustment scenario sequence B, B=(b1,b2…b i …b m ), where b i Let m be the i-th load adjustment scenario; m is the number of load adjustment scenarios. b is set sequentially according to the load adjustment scenario sequence B. i Target load adjustment scenario; Generate the initial weighting strategy and iterative sub-model for the target load adjustment scenario; Generate a regulation sub-model for the target load regulation scenario based on the initial weighting strategy and the iterative sub-model; The adjustment sub-models for each load adjustment scenario are generated sequentially; A load regulation model is generated based on all the regulation sub-models.

[0016] Compared with existing technologies, the mobility load balancing method and system based on weight adaptive adjustment in this application have the following advantages: By analyzing historical execution records of various communication cells, multiple load adjustment models are established. Initial weight parameters for each load adjustment scenario are set to ensure that weights are adjusted in a way that improves overall communication quality. Furthermore, by setting iterative sub-models for each load adjustment scenario, system oscillations caused by weight parameter optimization are avoided, thus improving the efficiency of weight adaptive adjustment. This ensures both communication quality and load balancing in the mobile communication system. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the mobility load balancing method based on weight adaptive adjustment in a preferred embodiment of this application. Detailed Implementation

[0018] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0019] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0020] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0021] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0022] like Figure 1 As shown, the preferred embodiment of the mobility load balancing method based on weight adaptive adjustment in this application includes: S101: Generate multiple communication cells based on the communication structure and construct an associated transfer model; S102: Obtain the load monitoring packets of each communication cell according to the preset time node, and generate the load anomaly value of each communication cell based on all load monitoring packets; S103: Set multiple risk cells based on all abnormal load values, and generate adjustment sub-strategies for each risk cell based on the preset load adjustment model. This involves generating multiple communication cells, including: Establish a sequence of communication cells A, A=(a1, a2, ..., a3) i …a n ), where a i Let i be the i-th communication cell; n is the number of communication cells.

[0023] Specifically, the communication structure refers to all the communication base stations of the local mobile communication system. Multiple communication cells are established based on all the mobile communication cells in all the communication base stations. In this case, all users in a single mobile communication cell are in the same communication cell, and a single communication cell corresponds to one mobile cell.

[0024] Specifically, it is determined in turn whether communication users can be transferred between each communication cell. If a certain communication cell is a neighbor cell in the current communication cell's neighbor cell list, then the communication cell is an associated cell of the current communication cell, and the association relationship between the two is generated. Based on all the association relationships, an association transfer model is constructed.

[0025] Specifically, the preset load adjustment model includes: Multiple operational characteristic indicators are generated based on historical execution records; Establish multiple load adjustment scenarios based on all operational characteristic indicators; Establish a load adjustment scenario sequence B, B=(b1,b2…b i …b m ), where b i Let m be the i-th load adjustment scenario; m is the number of load adjustment scenarios. b is set sequentially according to the load adjustment scenario sequence B. i Target load adjustment scenario; Generate the initial weighting strategy and iterative sub-model for the target load adjustment scenario; Generate a regulation sub-model for the target load regulation scenario based on the initial weighting strategy and the iterative sub-model; The adjustment sub-models for each load adjustment scenario are generated sequentially; A load regulation model is generated based on all the regulation sub-models.

[0026] Specifically, historical execution records refer to the time points when MLB was triggered in the past, the various load index parameters and channel quality data of the source cell that triggered it, the target cell selection record, the number of successfully transferred users, and the evaluation parameters of the effect after equalization.

[0027] Specifically, load metrics include physical resource block utilization, number of connected users, and service throughput.

[0028] Specifically, the operational characteristic indicators include, but are not limited to, the difference between the real-time value and the standard threshold of each load indicator, and parameters such as channel quality. By quantifying each operational characteristic indicator, multiple value ranges for each operational characteristic indicator are generated. Based on the random combination of all value ranges, various load adjustment scenarios are established.

[0029] Specifically, the initial weight parameters of each load index are generated by selecting and recording the target transfer cell corresponding to the target load adjustment scenario, and the adjustment amount of the target user transfer amount and weight parameters in the target load adjustment scenario is set according to the number of successfully transferred users, and the corresponding weight adjustment period is set.

[0030] Specifically, an initial weight strategy for the target load adjustment scenario is generated based on the initial weight parameters, and an iterative sub-model is constructed based on the target user transfer volume and the adjustment amount of the weight parameters, and the weight adjustment cycle.

[0031] Specifically, the adjustment amount of the weight parameter is the increase or decrease value of the weight of each load indicator in a single iteration. The larger the target user transfer volume, the larger the corresponding load indicator adjustment amount, thereby reducing the weight optimization cost while ensuring the efficiency of load balancing adjustment.

[0032] Specifically, the target user transfer volume is set as the average number of users successfully transferred in previous attempts.

[0033] In a preferred embodiment of this application, multiple risk cells are set based on all abnormal load values, including: Based on the communication cell sequence A, a is set sequentially. i For the target communication cell; Generate the load anomaly value f of the target communication cell at the current monitoring time node based on the load monitoring packet; The load anomaly values ​​for each communication cell at the current monitoring time point are generated sequentially; Establish a sequence of load anomalies F, F=(f1,f2…f i …f n ), where f i The load anomaly value of the i-th communication cell at the current monitoring time point; Preset load anomaly threshold F1; If f i >F1, set the i-th communication cell as a risk cell at the current monitoring time point.

[0034] Specifically, the load anomaly threshold can be set based on historical parameters. When the load anomaly value of a communication cell exceeds the preset load anomaly threshold, it indicates that the current communication cell is under high load and the load needs to be transferred in a timely manner to avoid affecting the overall communication quality.

[0035] Specifically, the load anomaly value f of the target communication cell at the current monitoring time point is generated, including: f=g*[ Y1(i)*(s i -s' i )]; g=U1*[ Y2(i)*(s i -s 1i )]; Where g is the load compensation coefficient; θ1 is the number of load indicators; s i s' is the reference value of the i-th load indicator of the target communication cell at the current monitoring time point; i U1 is the standard threshold for the i-th load metric; U1 is the preset first conversion coefficient; s 1i Y1(i) is the first-level reference value of the i-th load index of the target communication cell generated based on the correlation transfer model; Y1(i) is the selection coefficient; if (s i -s' i If )≥0,Y1(i)=1; if (s i -s' i If ) < 0, Y1(i) = 0; Y2(i) is the selection coefficient; if (s i -s 1i If )≥0,Y2(i)=1; if (s i -s 1i If ) < 0, Y2(i) = 0.

[0036] Specifically, by presetting a first conversion coefficient, the load compensation coefficient g is made to be within a preset value range, and [ Y2(i)*(s i -s 1i The larger the value of ], the larger the corresponding load compensation coefficient g value. The mapping relationship between the two can be set according to historical parameters, and the load compensation coefficient g value is always greater than 1.

[0037] Specifically, the primary reference value s 1i It is the average of the reference values ​​of the load index of all associated cells of the target communication cell at the current monitoring time node at the i-th load index.

[0038] Specifically, the standard threshold for each load indicator can be set based on historical parameters. When the reference value of a single load indicator is greater than the preset standard threshold, it indicates that the load indicator of the target communication cell is in an overloaded or high-load state.

[0039] Specifically, by quantifying each load indicator, all load indicators are made to be within the same value range. The larger the reference value of each load indicator, the greater the possibility that the load indicator is in an overload or high load state.

[0040] Specifically, the larger the load anomaly value, the greater the likelihood that the target communication cell is currently under high load at the current monitoring time point.

[0041] In a preferred embodiment of this application, adjustment sub-strategies for each risk cell are generated based on a preset load adjustment model, including: Establish a sequence A1 of risky communities at the current monitoring time point; A1=(a 11 a 12 …a 1i …a 1n1 ), where a 1i Let n1 be the i-th risky cell at the current monitoring time point; n1 is the number of risky cells at the current monitoring time point; and n1 <n; Based on the risk zone sequence A1, a is set sequentially. 1i Targeted high-risk communities; Obtain the operational data packets of the target high-risk cell; The primary adjustment model for the target risk cell is set based on the running data packet; Based on the correlation transfer model, establish the correlation sequence A2 of the target risk area, A2=(a 21 a 22 …a 2i …a 2n2 ), where a 2i Let be the i-th associated cell of the target risk cell; n2 is the number of associated cells of the target risk cell; Based on the primary regulation model and the associated cell sequence A2, a regulation sub-strategy is generated for the target risk cell; The adjustment sub-strategies for each risk zone are generated sequentially.

[0042] Specifically, the first-level adjustment model for the target risk area includes: b is set sequentially according to the load adjustment scenario sequence B. i For the comparison and adjustment scenario; Generate a matching value d between the target risk cell and the scenario to be compared and adjusted based on the running data packet; d=U2*[ (v i -v' i )]; Wherein, U2 is the preset second conversion coefficient; The number of operational characteristic indicators; v i To generate a reference value for the i-th runtime characteristic indicator based on the runtime data packet; v' i This is the reference value for the i-th operational characteristic indicator in the scenario to be compared and adjusted; The matching values ​​for the target risk cell and each load adjustment scenario are generated sequentially; The adjustment sub-model corresponding to the load adjustment scenario with the maximum value among all matching values ​​is set as the first-level adjustment model of the target risk cell.

[0043] Specifically, the larger the similarity value is, the more similar the real-time communication state of the target risk cell is to the communication state corresponding to the to-be-compared adjustment scenario, and the higher the load adjustment efficiency of the adjustment sub-model in the to-be-compared adjustment scenario for the target risk cell.

[0044] Specifically, by presetting a second conversion coefficient, the fit value d is within a preset value range, and (v i -v' i )] The larger the value is, the greater the difference between the operating state of the target risk cell and the operating state corresponding to the to-be-compared adjustment scenario, and the smaller the corresponding fit value d.

[0045] In the preferred embodiment of the embodiment of the present application, generating an adjustment sub-strategy for the target risk cell includes: Generating a first-level weight strategy and a first-level iterative model according to the first-level adjustment model; Generating multiple transfer time nodes according to the first-level iterative model, and establishing a transfer time node sequence T, T=(t1, t2…t i …t r ), where t i is the i-th transfer time node; r is the number of transfer time nodes; Generating transfer evaluation values of each associated cell in each associated cell sequence A2 at the transfer time node t1 according to the first-level weight strategy; Setting a transfer sub-strategy for the transfer time node t1 according to all transfer evaluation values; Obtaining a feedback data packet at the transfer time node t2, and generating an optimized benefit value h1; Generating a correction instruction for the first-level weight strategy according to the optimized benefit value and the first-level iterative model; Generating a second-level weight strategy according to the correction result; Generating a transfer sub-strategy for the transfer time node t2 according to the second-level weight strategy; Obtaining a feedback data packet at the transfer time node t3, and generating an optimized benefit value h2; Judging whether to generate a stop instruction according to the optimized benefit value h1 and the optimized benefit value h2; If h1≥h2, generating a stop instruction; If h1<h2, generating a transfer sub-strategy for the transfer time node t3, and iterating in a loop until a stop instruction is generated or the transfer time ends; Generating an adjustment sub-strategy for the target risk cell according to the stop instruction.

[0046] Specifically, setting the initial weight strategy in the first-level adjustment model as the first-level weight strategy, and setting the iterative sub-model in the first-level adjustment model as the first-level iterative model.

[0047] Specifically, based on current communication requirements, multiple performance indicators are generated, including but not limited to communication signal quality, load balancing rate, and ping-pong handover count. By analyzing each performance indicator, corresponding optimization benefit values ​​are generated. If the current optimization benefit value is greater than the optimization benefit value of the previous transfer time node, it indicates that the optimal weight has not been reached and iteration needs to continue. If the current optimization benefit value is less than or equal to the optimization benefit value of the previous transfer time node, it indicates that the optimal weight was reached at the previous transfer time node. The target transfer cell of the target risk cell is selected according to the optimal weight parameter, thereby generating the corresponding adjustment sub-strategy.

[0048] Specifically, the transition sub-strategy for transition time node t1 includes: Based on the associated cell sequence A2, set a sequentially. 2i Associate the target cell; Generate the transfer evaluation value k of the target associated cell at transfer time node t1; k=U3*[ β i *s 2i ]; Where U3 is the preset third conversion coefficient; θ1 is the number of load indicators; β i The weighting coefficient of the i-th load metric is set based on the first-level weighting strategy; s 2i The reference value of the i-th load index of the target associated cell at the transfer time node t1; The transfer evaluation values ​​of each associated cell at the transfer time node t1 are generated sequentially; The associated cell corresponding to the maximum value among all transfer evaluation values ​​is set as the target transfer cell at transfer time node t1; The target transfer amount is set according to the first-level iterative model; A transfer sub-strategy is generated based on the target transfer amount and the target transfer cell, with a transfer time node t1.

[0049] Specifically, by presetting a third fixed coefficient, the transfer evaluation value is kept within a preset value range, and [ β i *s 2i The larger the value of ], the smaller the corresponding transfer evaluation value. The mapping relationship between the two can be set according to historical parameters.

[0050] Specifically, the higher the transfer evaluation value, the lower the overall load of the target associated cell, the higher the reception success rate of transferred users in the target risk cell, and the more beneficial it is to ensure the overall load balance of the system by transferring some communication users of the target risk cell to the target associated cell.

[0051] It is understandable that in the above embodiments, by analyzing the historical execution records of each communication cell, multiple load adjustment models are set, and initial weight parameters for each load adjustment scenario are set to ensure that the weights are adjusted in a direction that improves the overall communication quality. Simultaneously, by setting iterative sub-models for each load adjustment scenario, system oscillations caused by the optimization process of the weight parameters are avoided, thus improving the efficiency of weight adaptation. This ensures the communication quality and load balancing of the mobile communication system.

[0052] In another preferred embodiment of the mobility load balancing processing method based on weight adaptive adjustment in any of the above preferred embodiments, this preferred embodiment provides a mobility load balancing processing system based on weight adaptive adjustment, including: The central control unit is used to generate multiple communication cells based on the communication structure and to construct an associated transfer model; The monitoring unit is used to acquire load monitoring packets for each communication cell according to preset time nodes; The central control unit includes: The first processing module is used to generate load anomaly values ​​for each communication cell based on all load monitoring packets. The second processing module is used to set multiple risk cells based on all abnormal load values; The third processing module is used to generate adjustment sub-strategies for each risk cell based on a preset load adjustment model.

[0053] In a preferred embodiment of this application, the central control unit further includes: The fourth processing module is used to establish the communication cell sequence A, A=(a1, a2…a…). i …a n ), where a i Let i be the i-th communication cell; n is the number of communication cells; The fifth processing module is used to generate multiple runtime characteristic indicators based on historical execution records; Establish multiple load adjustment scenarios based on all operational characteristic indicators; Establish a load adjustment scenario sequence B, B=(b1,b2…b i …b m ), where b i Let m be the i-th load adjustment scenario; m is the number of load adjustment scenarios. b is set sequentially according to the load adjustment scenario sequence B. i Target load adjustment scenario; Generate the initial weighting strategy and iterative sub-model for the target load adjustment scenario; Generate a regulation sub-model for the target load regulation scenario based on the initial weighting strategy and the iterative sub-model; The adjustment sub-models for each load adjustment scenario are generated sequentially; A load regulation model is generated based on all the regulation sub-models.

[0054] Based on the first concept of this application, by analyzing the historical execution records of each communication cell, multiple load adjustment models are established. By setting initial weight parameters for each load adjustment scenario, the weights are ensured to be adjusted in a direction that improves overall communication quality. Simultaneously, by setting iterative sub-models for each load adjustment scenario, system oscillations caused by the optimization process of the weight parameters are avoided, thus improving the efficiency of weight adaptiveness. This ensures the communication quality and load balancing of the mobile communication system.

[0055] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application.

Claims

1. A mobility load balancing process based on weight adaptive adjustment, characterized in that, The method comprises the following steps: generating a plurality of communication cells according to a communication structure and constructing an association transfer model; obtaining a load monitoring package of each communication cell according to a preset time node, and generating a load abnormal value of each communication cell according to all the load monitoring packages; setting a plurality of risk cells according to all the load abnormal values, and generating an adjustment sub-strategy of each risk cell according to a preset load adjustment model; wherein generating a plurality of communication cells comprises: A sequence of communication cells A, A=(a1, a2…a i …a n ), is established, where a i is the i-th communication cell; n is the number of communication cells.

2. The mobility load balancing processing method based on weight adaptive adjustment according to claim 1, wherein the preset load adjustment model comprises: generating a plurality of running characteristic indexes according to historical execution records; establishing a plurality of load adjustment scenes according to all the running characteristic indexes; Establish a load adjustment scenario sequence B, B=(b1,b2…b i …b m ), where b i Let m be the i-th load adjustment scenario; m is the number of load adjustment scenarios. According to the load adjustment scenario sequence B, b is set in turn i Target load adjustment scenario; generating an initial weight strategy and an iteration sub-model of a target load adjustment scene; generating an adjustment sub-model of the target load adjustment scene according to the initial weight strategy and the iteration sub-model; generating adjustment sub-models of each load adjustment scene in sequence; generating a load adjustment model according to all the adjustment sub-models.

3. The mobility load balancing processing method based on weight adaptive adjustment according to claim 2, wherein setting a plurality of risk cells according to all the load abnormal values comprises: According to the communication cell number sequence A, a is set in turn i Target communication cell; generating a load abnormal value f of a target communication cell at a current monitoring time node according to a load monitoring package; generating load abnormal values of each communication cell at the current monitoring time node in sequence; A load abnormal value sequence F, F=(f1, f2…f i …f n ), is established, wherein f i is a load abnormal value of the ith communication cell at a current monitoring time node. presetting a load abnormal value threshold F1; If f i >F1, set the i-th communication cell as a risky cell at the current monitoring time node.

4. The mobility load balancing process based on weight adaptive adjustment according to claim 3, wherein, generating the load abnormal value f of the target communication cell at the current monitoring time node comprises: f = g * [i Y1(i) * (s i -s' i )]; g = U1* [Y2(i) * (s Y2(i)*(s i -s 1i )]; Wherein, g is a load compensation coefficient; θ1 is a load index number; s i is a reference value of the i th load index of the target communication cell at a current monitoring time node; s i is a standard threshold value of the i th load index; U1 is a preset first conversion coefficient; s 1i is a first-level reference value of the i th load index of the target communication cell generated based on an association transfer model; Y1(i) is a selection coefficient; if (s i -s' i )≥0, Y1(i)=1; if (s i -s' i )<0, Y1(i)=0; Y2(i) is a selection coefficient; if (s i -s 1i )≥0, Y2(i)=1; if (s i -s 1i )<0, Y2(i)=0.

5. The mobility load balancing process based on weight adaptive adjustment according to claim 3, wherein, generating an adjustment sub-strategy of each risk cell according to a preset load adjustment model comprises: establishing a risk cell number sequence A1 of the current monitoring time node; A1=(a 11 , a 12 …a 1i , a 1n1 ), wherein a 1i is the i-th risk cell of the current monitoring time node; n1 is the number of risk cells of the current monitoring time node; and n1n. According to the risk cell number sequence A1, a is set in turn 1i Target risk cell; obtaining a running data package of a target risk cell; setting a first-level adjustment model of the target risk cell according to the running data package; According to the association transfer model, the association cell series A2 of the target risk cell is established, A2=(a 21 , a 22 …a 2i …a 2n2 ), wherein a 2i is the ith association cell of the target risk cell; n2 is the number of association cells of the target risk cell; generating an adjustment sub-strategy of the target risk cell according to the first-level adjustment model and an associated cell number sequence A2; generating adjustment sub-strategies of each risk cell in sequence.

6. The mobility load balancing process based on weight adaptive adjustment according to claim 5, wherein, setting the first-level adjustment model of the target risk cell comprises: According to the load adjustment scene sequence B, b is set in turn i For the contrast adjustment scene to be compared generating a fitting value d of the target risk cell and a to-be-compared adjustment scene according to a running data package; d = U2* [ (v - v') / (v + v') ] + 1 (v i -v' i )]; U2 is a preset second conversion coefficient; is a number of running characteristic indexes; i is a reference value of the i-th running characteristic index generated according to the running data packet;v i is a reference value of the i-th running characteristic index in the to-be-compared adjustment scene; generating fitting values of the target risk cell and each load adjustment scene in sequence; setting an adjustment sub-model of a load adjustment scene corresponding to a maximum value in all the fitting evaluation values as the first-level adjustment model of the target risk cell.

7. The mobility load balancing process based on weight adaptive adjustment according to claim 5, wherein, generating the adjustment sub-strategy of the target risk cell comprises: generating a first-level weight strategy and a first-level iteration model according to the first-level adjustment model; According to the first-order iteration model, a plurality of transfer time nodes are generated, and a transfer time node sequence T is established, T=(t1, t2…t i …t r ), wherein t i is the i-th transfer time node; r is the number of transfer time nodes; generating a transfer evaluation value of each associated cell in each associated cell number sequence A2 at a transfer time node t1 according to the first-level weight strategy; setting a transfer sub-strategy of the transfer time node t1 according to all the transfer evaluation values; obtaining a feedback data package of a transfer time node t2, and generating an optimization benefit value h1; generating a correction instruction of the first-level weight strategy according to the optimization benefit value and the first-level iteration model; generating a second-level weight strategy according to a correction result; generating a transfer sub-strategy of the transfer time node t2 according to the second-level weight strategy; obtaining a feedback data package of a transfer time node t3, and generating an optimization benefit value h2; determining whether to generate a stop instruction according to the optimization benefit value h1 and the optimization benefit value h2; if h1≥h2, generating the stop instruction. If h1 < h2, a transfer sub-strategy of a transfer time node t3 is generated, and the iteration is circular until a stop instruction or a transfer time is ended; An adjustment sub-strategy of a target risk cell is generated according to the stop instruction.

8. The mobility load balancing process based on weight adaptive adjustment according to claim 7, wherein, The transfer sub-strategy of the transfer time node t1 is set, including: According to the number of associated cells, the sequence A2 is set in order a 2i Target associated cell; A transfer evaluation value k of the target associated cell at the transfer time node t1 is generated; k=U3*[ β i *s 2i ]; Wherein, U3 is a preset third conversion coefficient; θ1 is a load index number; β i is a weight coefficient of the i th load index based on a first weight strategy; s 2i is a reference value of the i th load index of the target associated cell at the transfer time node t1. The transfer evaluation values of all associated cells at the transfer time node t1 are generated in sequence; The associated cell corresponding to the maximum value in all transfer evaluation values is set as the target transfer cell of the transfer time node t1; The target transfer amount is set according to the first iteration model; The transfer sub-strategy of the transfer time node t1 is generated according to the target transfer amount and the target transfer cell.

9. A mobility load balancing processing system based on adaptive adjustment of weights, employing the mobility load balancing processing method based on adaptive adjustment of weights according to any one of claims 1-8, characterized in that, Including: The central control unit is used for generating a plurality of communication cells according to a communication structure, and constructing an associated transfer model; The monitoring unit is used for obtaining load monitoring packages of each communication cell according to a preset time node; The central control unit includes: The first processing module is used for generating load abnormal values of each communication cell according to all load monitoring packages; The second processing module is used for setting a plurality of risk cells according to all load abnormal values; The third processing module is used for generating an adjustment sub-strategy of each risk cell according to a preset load adjustment model.

10. The mobility load balancing process system based on weight adaptive adjustment as claimed in claim 9, wherein, The central control unit further includes: The fourth processing module is configured to establish a communication cell sequence A, A=(a1, a2...an), wherein ai represents the i-th communication cell; n represents the number of communication cells. i …a n ), wherein ai represents the i-th communication cell; n represents the number of communication cells. i ai represents the i-th communication cell; n represents the number of communication cells. The fifth processing module is used for generating a plurality of running characteristic indexes according to historical execution records; A plurality of load adjustment scenes are established according to all running characteristic indexes; Establish a load adjustment scenario sequence B, B=(b1,b2…b i …b m ), where b i Let m be the i-th load adjustment scenario; m is the number of load adjustment scenarios. According to the load adjustment scenario sequence B, b is set in turn i Target load adjustment scenario; An initial weight strategy and an iteration sub-model of a target load adjustment scene are generated; An adjustment sub-model of the target load adjustment scene is generated according to the initial weight strategy and the iteration sub-model; The adjustment sub-models of all load adjustment scenes are generated in sequence; A load adjustment model is generated according to all adjustment sub-models.

Images