TWI778772B - System and method for optimizing network - Google Patents

Abstract

A system and a method for optimizing a network are provided. The method includes: storing a plurality of algorithms; receiving network information of the network; determining a optimization target according to the network information; selecting an algorithm from the plurality of algorithms according to the optimization target; generating at least one network configuration according to the network information and the algorithm; and access the network and configuring the network according to the at least one network configuration.

Description

System and method for optimizing a network

The present invention relates to a system and method for optimizing a network.

The communication network will reduce parameters such as the connection rate or average rate of users due to the use requirements of multiple users or multiple applications, thereby greatly reducing the service quality. However, the existing end-to-end (E2E) test service management system can only propose a common solution for different equipment (for example, equipment of different brands) for a single optimization goal, but cannot provide a solution for multiple fields. Optimization goals come up with solutions. In addition, the conventional quality of service (quality of service, QoS) guarantee mechanism can ensure that certain application requirements have a fixed network bandwidth available for use. However, when there are too many user equipments with the same allocation and retention priority (ARP), the existing technology may still fail to guarantee the service quality of each user equipment.

The present invention provides a system and method for optimizing the network, which can configure the network for multiple optimization objectives.

A system for optimizing a network of the present invention includes a processor, a storage medium and a transceiver. The storage medium stores multiple modules. The processor is coupled to the storage medium and the transceiver, and accesses and executes a plurality of modules, wherein the plurality of modules include an algorithm pool, a data collection module, a data analysis and processing module, an optimization direction or a target selection module, Network and application service joint optimization module and network and application service parameter setting module. The algorithm pool stores multiple algorithms. The data collection module receives network information of the network through the transceiver. The data analysis and processing module determines the optimization target according to the network information. The optimization direction or goal selection module selects an algorithm from a plurality of algorithms according to the optimization goal. The network and application service joint optimization module generates at least one network configuration according to the network information and the algorithm. The network and application service parameter setting module accesses the network through the transceiver, and configures the network according to at least one network configuration.

In an embodiment of the present invention, the above-mentioned network and application service joint optimization module is configured to perform: calculating at least one estimated service quality and at least one application demand value according to network information based on an algorithm; based on an algorithm and generating an objective optimization function according to at least one estimated service quality and at least one application demand value; and generating at least one network configuration according to the objective optimization function.

In an embodiment of the present invention, the above-mentioned network and application service joint optimization module is further configured to perform: calculating a plurality of target optimization function values respectively corresponding to a plurality of network configurations according to the target optimization function; and in response to At least one network configuration is selected from the plurality of network configurations corresponding to the largest target optimization function value among the plurality of target optimization function values.

In an embodiment of the present invention, the above-mentioned network and application service joint optimization module is further configured to perform: calculating a plurality of target optimization function values corresponding to a plurality of network configurations according to the target optimization function, wherein a plurality of The target optimization function value includes at least one target optimization function value corresponding to at least one network configuration; and is selected from a plurality of network configurations in response to the at least one target optimization function value being greater than the target optimization function value corresponding to the initial network configuration At least one network configuration.

In an embodiment of the present invention, the above-mentioned network information corresponds to a plurality of user equipments, wherein the plurality of user equipments include at least one user equipment, wherein the network and application service joint optimization module is further configured to execute: according to the network The calculation of the road information corresponds to a plurality of estimated service qualities of a plurality of user equipments, wherein the plurality of estimated service qualities includes at least one estimated service quality corresponding to at least one user equipment; the calculation according to the network information corresponds to at least one estimated service quality. at least one application demand value of the estimated service quality; at least one combination of at least one user equipment and at least one application demand is generated in response to the at least one estimated service quality being greater than or equal to the service quality threshold and the at least one application demand value being greater than the application demand threshold, Wherein at least one application requirement corresponds to at least one application requirement value; and an objective optimization function is generated according to at least one combination.

In an embodiment of the present invention, the aforementioned network and application service joint optimization module determines the service quality threshold and the application demand threshold according to an algorithm, and calculates a plurality of estimated service qualities according to the initial network configuration.

In an embodiment of the present invention, the objective optimization function value of the above-mentioned objective optimization function is equal to the number of at least one combination that satisfies the service quality threshold and the application requirement threshold, wherein at least one combination includes combinations.

In an embodiment of the present invention, the objective optimization function value of the above-mentioned objective optimization function is equal to the sum of application demand values satisfying at least one combination of the service quality threshold and the application demand threshold.

In an embodiment of the present invention, the above-mentioned at least one combination includes a combination in the first combination set and a combination in the second combination set, wherein the objective optimization function value of the objective optimization function is equal to the first element of the first combination set. A linear combination of the number and the second number of elements of the second combination set.

In an embodiment of the present invention, the above-mentioned modules further include an optimization program startup module. The optimization program activation module instructs the data collection module to receive network information in response to receiving the command through the transceiver to perform an optimization process of the network.

In an embodiment of the present invention, the above-mentioned modules further include an optimization program startup module. The optimizer activation module periodically instructs the data collection module to receive network information to perform network optimization.

In an embodiment of the present invention, the above-mentioned network information includes at least one of the following: radio access network data, radio access network user identification data, multivariate data, component default configuration data, network Signal data, core network data, transmission network parameters, external input data and configuration data for optimizing system operation.

In an embodiment of the present invention, the above-mentioned radio access network data includes at least one of the following: general/advanced configuration management, general/advanced performance management, fault management, user terminal reporting, and minimum drive test .

In an embodiment of the present invention, the above-mentioned radio access network user identification data includes at least one of the following: radio access network user equipment location data, service type, traffic demand and moving path.

In an embodiment of the present invention, the above-mentioned multivariate data includes at least one of the following: application service quality, object shape and quantity identification, field map, object positioning data, and other multivariate information.

In an embodiment of the present invention, the above-mentioned device default configuration data includes at least one of the following: radio access network configuration parameters, core network parameters, antenna directions, mobile network terminal parameters, and application services parameter.

In an embodiment of the present invention, the above-mentioned network signal data includes at least one of the following: radio frequency signal scanning and drive test mobile phone.

In an embodiment of the present invention, the above-mentioned core network data includes at least one of the following: user plane function data, service quality management data, and core network signaling data.

In an embodiment of the present invention, the above-mentioned transmission network parameters include at least one of the following: a packet queue status parameter, a packet processing mechanism parameter, and a packet processing performance parameter.

In an embodiment of the present invention, the above-mentioned external input data includes at least one of the following: radio access network intent, application service intent, and field restriction.

In an embodiment of the present invention, the above-mentioned optimization system operation configuration data includes at least one of the following: a system configuration file, an optimization program trigger condition configuration file, and an optimization target configuration file.

In an embodiment of the present invention, the above-mentioned data analysis and processing module calculates parameters according to network information, and determines an optimization target according to the parameters, wherein the parameters include at least one of the following: network data, including at least one of the following One: Connectivity, Retention, Availability, Mobility, Integrity, Utilization, RF Signal Pattern, Configuration Management, Core Network Data, and Transmission Network Data; Multiple Information, including at least one of the following: Service quality, service demand, schedule information, user trajectory, and traffic; and device stability data, including at least one of the following: device stability and device alarms.

A method for optimizing a network of the present invention includes: storing a plurality of algorithms; receiving network information of the network; determining an optimization target according to the network information; selecting an algorithm from a plurality of algorithms according to the optimization target; generating at least one network configuration from the road information and the algorithm; and accessing the network and configuring the network according to the at least one network configuration.

Based on the above, the present invention can set optimization goals in combination with relevant data such as application requirements and service quality. The invention can improve the target optimization function value, allocate network resources accurately according to the application requirements, and automatically complete the optimization work, so as to achieve a more detailed and accurate optimization effect that is closer to the user's needs, so as to ensure the user's understanding of the network or application. The quality of experience (QoE) of the service.

FIG. 1 is a schematic diagram of a system 100 for optimizing a network according to an embodiment of the present invention. System 100 may include processor 110 , storage medium 120 , and transceiver 130 .

The processor 110 is, for example, a central processing unit (CPU), or other programmable general-purpose or special-purpose micro control unit (micro control unit, MCU), microprocessor (microprocessor), digital signal processing digital signal processor (DSP), programmable controller, application specific integrated circuit (ASIC), graphics processor (graphics processing unit, GPU), image signal processor (image signal processor, ISP) ), image processing unit (IPU), arithmetic logic unit (ALU), complex programmable logic device (CPLD), field programmable gate array (field programmable gate array) , FPGA) or other similar elements or a combination of the above. The processor 110 may be coupled to the storage medium 120 and the transceiver 130 , and access and execute a plurality of modules and various application programs stored in the storage medium 120 .

The storage medium 120 is, for example, any type of fixed or removable random access memory (random access memory, RAM), read-only memory (ROM), and flash memory (flash memory). , a hard disk drive (HDD), a solid state drive (SSD), or similar components or a combination of the above components for storing a plurality of modules or various application programs executable by the processor 110 . In this embodiment, the storage medium 120 can store an optimization program startup module 121 , a data collection module 122 , a data analysis and processing module 123 , an optimization direction or target selection module 124 , and a joint optimization module for network and application services. The functions of the group 125 , the network and application service parameter setting module 126 , the algorithm pool 127 , and the database 128 and other modules will be described later.

Data can be exchanged between the various modules through the database 128, and the database 128 can store the outputs of the various modules. The database 128 may adopt one or more database technologies or memory storage technologies, such as associative database, non-associative database, memory database or memory temporary storage, etc., the invention is not limited thereto.

The transceiver 130 transmits and receives signals in a wireless or wired manner. Transceiver 130 may also perform operations such as low noise amplification, impedance matching, frequency mixing, up or down frequency conversion, filtering, amplification, and the like. The system 100 can access the network to be optimized through the transceiver 130 .

The algorithm pool 127 can store multiple algorithms, such as a multi-domain joint optimization algorithm, a fault handling algorithm, or a basic parameter setting algorithm, but the invention is not limited thereto. The algorithm pool 127 may also store information related to the algorithm, such as application demand functions, conversion functions used to generate the objective optimization function, initial network configuration, service quality thresholds or application demand thresholds.

The optimization program activation module 121 can activate the system 100 to perform the optimization process of the network. In one embodiment, the optimization program startup module 121 can receive an instruction from the user equipment through the transceiver 130, and instruct the data collection module 122 to receive the network information of the network to be optimized according to the instruction to perform the operation of the network. Optimize the process. In other words, the user can send an instruction by operating the user equipment to start the optimization process of the network. In one embodiment, the optimization program activation module 121 may periodically instruct the data collection module 122 to receive network information of the network to be optimized to perform the optimization process of the network. In other words, the optimization program enabling module 121 can automatically schedule and execute the optimization process of the network.

The data collection module 122 can access the network to be optimized through the transceiver 130 to obtain network information of the network. Network information may include radio access network data, radio access network user identification data, multivariate data, component default configuration data, network signal data, core network data, transmission network parameters, external input data or Optimize system operation configuration data, etc., but the present invention is not limited to this. The data collection module 122 can store the obtained data in the storage medium 120 . The data collection module 122 may further perform data preprocessing on the acquired data, wherein the data preprocessing may include data extraction, data pre-analysis, data association, data labeling, data cleaning, or data type conversion. The data collection module 122 may obtain initial data from one or more data sources, and then perform preprocessing on the initial data. Then, the data collection module 122 can store the preprocessed data in the storage medium 120 for use by other modules.

The radio access network data may include general/advanced configuration management, general/advanced performance management, fault management, user terminal reporting or minimum drive test (minimization of drive test, MDT) data, but the invention is not limited thereto . Advanced data (such as advanced configuration management or advanced performance management) usually has a large amount of data, so it is easy to consume hardware resources or network transmission resources, but advanced data can provide more detailed and real-time data, including Short-term operation status file data (for example: 1 minute PM file or interface), short-term operation status streaming report data (eg: streaming PM) or application service short-term operation status report, etc.

The radio access network user identification data may include radio access network user equipment positioning data, service types, traffic requirements or moving paths, etc., but the invention is not limited thereto.

The multivariate data may include data such as application service quality, object shape and quantity identification, field map, object positioning data, or other multivariate information, but the present invention is not limited thereto.

The component default configuration data may include radio access network configuration parameters, core network parameters, antenna directions, mobile network (for example: 4G/5G network) terminal parameters or application service parameters and other data, but the present invention does not limited to this.

The network signal data may include data such as radio frequency signal scanning or mobile phone driving test, but the present invention is not limited thereto.

The core network data may include user plane function data, service quality management data, or core network signaling data, etc., but the present invention is not limited thereto.

The transmission network parameters may include data such as packet queue status parameters, packet processing mechanism parameters, or packet processing performance parameters, but the invention is not limited thereto.

The external input data may include data such as radio access network intentions, application service intentions, or field restrictions, but the invention is not limited thereto.

The optimized system operation configuration data may include data such as a system configuration file, an optimization program trigger condition configuration file, or an optimization target configuration file, but the invention is not limited thereto.

The data analysis and processing module 123 can determine the optimization target according to the network information. Specifically, the data analysis and processing module 123 can analyze the network information to calculate the parameters, and input the parameters to the machine learning model to determine the optimization target according to the output of the machine learning model. Parameters can include network data, multivariate information, or device stability data, etc. On the other hand, after the parameters are calculated, the data analysis and processing module 123 can also select the optimization parameters as the optimization target.

The optimization objective can be related to multiple parameters. For example, optimization goals may include quality of service and number of users. The system 100 can improve the network according to the optimization objective, thereby improving the service quality of the network and increasing the number of user equipments served by the network. In one embodiment, the data analysis and processing module 123 can calculate an objective function value according to the network information of the network. If the objective function value is smaller than the objective threshold, the data analysis and processing module 123 may determine that the system 100 performs an optimization process for the network, thereby generating an optimization objective. For example, the objective function value may be the number of users that the network can serve. The data analysis and processing module 123 can calculate an objective function value of 10 according to the network information. That is, the maximum number of user equipments that the network can serve may be 10. Assuming that the target threshold is 20, the data analysis and processing module 123 may generate an optimization target for increasing the number of users that can be served in response to the target function value being less than the target threshold.

Optimization goals can be related to network load balancing, network interference mitigation, network mobility optimization, network coverage adjustment, network configuration conflict handling, network neighbor cell optimization, network antenna tracking users, and network transmission reliability optimization, network packet storage and queuing mechanism optimization, network multi-input multi-output (MIMO) mechanism optimization, network beamforming mechanism optimization, network resource allocation hierarchical processing, on/off Network advanced data collection, network equipment fault handling, application service demand resource classification processing, enabling/disabling application service advanced data collection or application service equipment fault handling and other items, but the present invention is not limited to this.

Network data can include accessibility, retention, availability, mobility, integrity, utilization, RF fingerprint, group configuration management (configuration management, CM), core network data or transmission network data and other data, but the present invention is not limited to this.

The multiple pieces of information may include data such as service quality, service demand, schedule information, user trajectory or traffic, but the invention is not limited thereto.

The device stability data may include data such as device stability or device alarm, but the present invention is not limited thereto.

The optimization direction or goal selection module 124 may select a specific algorithm from a plurality of algorithms in the algorithm pool 127 according to the optimization goal. In one embodiment, the optimization direction or target selection module 124 may select the algorithm according to a look-up table. For example, the algorithm pool 127 may store the mapping relationship between optimization objectives and algorithms. The optimization direction or target selection module 124 may select the algorithm associated with the optimization target from the lookup table according to the mapping relationship. In one embodiment, the optimization direction or target selection module 124 may input the optimization target into the machine learning model to determine the algorithm according to the output of the machine learning model. The algorithm is, for example, a multi-domain joint optimization algorithm, a fault handling algorithm, or a basic parameter setting algorithm, among which the multi-domain joint optimization algorithm may consider the optimization objective during operation.

After determining the algorithm, the network and application service joint optimization module 125 can generate at least one network configuration according to the network information and the algorithm. The network and application service joint optimization module 125 can generate at least one network configuration according to a rule-based method. Specifically, the network and application service joint optimization module 125 can calculate at least one estimated service quality and at least one application demand value according to the network information. Next, the network and application service joint optimization module 125 can generate an objective optimization function according to the at least one estimated service quality and at least one application requirement value based on the algorithm, and generate at least one network configuration according to the objective optimization function value.

為與用戶設備n相關的第m個應用需求的索引，並且

為對應於演算法的初始網路配置。在一實施例中，網路及應用服務聯合優化模組125可根據機器學習模型計算預估服務品質

。

…(1) It is assumed that the network information includes related information of a plurality of user equipments, wherein the plurality of user equipments include at least one user equipment. The network and application service joint optimization module 125 can calculate a plurality of estimated service qualities corresponding to a plurality of user equipments according to network information based on an algorithm, wherein the plurality of estimated service qualities can include corresponding to at least one user equipment at least one estimated quality of service. Equation (1) represents an estimated quality of service, where n is the index of the user equipment,

is the index of the mth application requirement related to user equipment n, and

is the initial network configuration corresponding to the algorithm. In one embodiment, the network and application service joint optimization module 125 can calculate the estimated service quality according to the machine learning model

.

…(1)

。

…(2) Next, the network and application service joint optimization module 125 may calculate at least one application demand value corresponding to at least one estimated service quality according to the network information based on the algorithm. Equation (2) represents the application demand value corresponding to Equation (1). The network and application service joint optimization module 125 can select the application requirement function R corresponding to the algorithm from the algorithm pool 127 . In one embodiment, the network and application service joint optimization module 125 can calculate the application demand value according to the machine learning model

.

…(2)

為用戶設備n與應用需求

的組合，

為對應於用戶設備n與應用需求

的服務品質閾值，並且X為應用需求閾值。網路及應用服務聯合優化模組125可從演算法池127中選出與演算法相對應的服務品質閾值

以及應用需求閾值X。

…(3) The network and application service joint optimization module 125 can select a set of combinations that meet specific conditions from multiple combinations of user equipment and application requirements, as shown in equation (3), where S is a set of combinations that meet specific conditions, NM is a collection of all combinations of user equipment and application requirements,

For user equipment n and application requirements

The combination,

To correspond to user equipment n and application requirements

and X is the application demand threshold. The network and application service joint optimization module 125 can select the service quality threshold corresponding to the algorithm from the algorithm pool 127

and the application requirement threshold X.

…(3)

，如方程式（4）所示，其中

為對應於網路配置

的目標優化函數值，並且

為轉換函數。網路及應用服務聯合優化模組125可從演算法池127中選出與演算法相對應的轉換函數

。網路及應用服務聯合優化模組125可通過調整轉換函數

來使系統100產生的網路配置能兼具公平性與效能的考量。

…(4) After obtaining the set S, the network and application service joint optimization module 125 can generate an objective optimization function according to the set S

, as shown in equation (4), where

to correspond to the network configuration

the objective optimization function value of , and

is the conversion function. The network and application service joint optimization module 125 can select the conversion function corresponding to the algorithm from the algorithm pool 127

. The network and application service joint optimization module 125 can adjust the conversion function by

This enables the network configuration generated by the system 100 to have both fairness and performance considerations.

…(4)

。如此，則目標優化函數

的目標優化函數值

可等於滿足服務品質閾值

以及應用需求閾值X的至少一組合的數量（即：集合S中的元素的數量），如方程式（5）所示。

…(5) In one embodiment, the conversion function

. So, the objective optimization function

The objective optimization function value of

may be equal to meeting the quality of service threshold

and the number of at least one combination of the application requirement threshold X (ie: the number of elements in the set S), as shown in equation (5).

…(5)

。如此，則目標優化函數

的目標優化函數值

可等於滿足服務品質閾值

以及應用需求閾值X的至少一組合的應用需求值總和，如方程式（6）所示。

…(6) In one embodiment, the conversion function

. So, the objective optimization function

The objective optimization function value of

may be equal to meeting the quality of service threshold

and the sum of application demand values for at least one combination of application demand thresholds X, as shown in equation (6).

…(6)

如方程式（7）所示。如此，則目標優化函數

的目標優化函數值

可等於組合集合A的元素個數

與組合集合B的元素個數

的線性組合，如方程式（8）所示。網路及應用服務聯合優化模組125可通過將方程式（7）中的常數a和常數b調整為相同而使系統100產生的網路配置可公平地兼顧組合集合A和組合集合B中的組合之網路效能。

…(7)

…(8) In one embodiment, the combination of user equipment and application requirements may include combinations in combination set A and combinations in combination set B, and the conversion function

As shown in equation (7). So, the objective optimization function

The objective optimization function value of

Can be equal to the number of elements in the combined set A

and the number of elements in the combined set B

A linear combination of , as shown in Equation (8). The network and application service joint optimization module 125 can adjust the constant a and the constant b in equation (7) to be the same, so that the network configuration generated by the system 100 can fairly take into account the combinations in the combination set A and the combination set B network performance.

…(7)

…(8)

、

、

、…、

等k個網路配置作為目標優化函數

的自變數（argument）以產生

、

、

、…、

等k個目標優化函數值。網路及應用服務聯合優化模組125可根據

、

、

、…、

等k個目標優化函數值以從

、

、

、…、

等k個網路配置中選出至少一網路配置。 The network and application service joint optimization module 125 can obtain a plurality of objective optimization function values corresponding to a plurality of network configurations according to the above method. For example, the joint network and application services optimization module 125 may

,

,

, …,

Wait for k network configurations as the objective optimization function

the arguments to produce

,

,

, …,

Wait for k objective optimization function values. The network and application service joint optimization module 125 can be based on

,

,

, …,

etc. k objective optimization function values to get from

,

,

, …,

At least one network configuration is selected from among the k network configurations.

、

、

、…、

等k個網路配置中選出對應於最大目標優化函數值

的網路配置

，如方程式（9）所示。

…(9) In one embodiment, the network and application services joint optimization module 125 can be

,

,

, …,

Select the optimal function value corresponding to the maximum objective among the k network configurations

network configuration

, as shown in Equation (9).

…(9)

、

、

、…、

等k個網路配置中選出對應於目標優化函數值

的網路配置

，其中目標優化函數值

大於對應於初始網路配置

的目標優化函數值

，如方程式（10）所示。

…(10) In one embodiment, the network and application services joint optimization module 125 can be

,

,

, …,

Select the value corresponding to the objective optimization function from the k network configurations

network configuration

, where the objective optimization function value

greater than that corresponds to the initial network configuration

The objective optimization function value of

, as shown in Equation (10).

…(10)

後，網路及應用服務參數設定模組126可通過收發器130存取網路，並且根據網路配置

來配置網路。舉例來說，網路及應用服務參數設定模組126可通過收發器130通訊連接至諸如網路服務元件、應用服務元件或應用程式介面（application programming interface，API）端點等網路中的元件，並且根據網路配置

來配置所述元件。 Select the network configuration

After that, the network and application service parameter setting module 126 can access the network through the transceiver 130, and according to the network configuration

to configure the network. For example, the web and application service parameter setting module 126 may communicate through the transceiver 130 to elements in the network such as web service elements, application service elements, or application programming interface (API) endpoints. , and according to the network configuration

to configure the element.

Each module in the storage medium 120 may belong to the same or different components (eg, hardware components, software components or virtual components). Taking an open radio access network (O-RAN) architecture as an example, FIG. 2 is a schematic diagram of an O-RAN architecture according to an embodiment of the present invention, wherein End-to-End is end-to-end (E2E), OAM is operation management and maintenance, SMO is service management orchestrator (SMO), Non-RT RIC is non-real time radio access network intelligent controller, Near -RT RIC is near-real time radio access network controller, O-CU is O-RAN central unit, O-DU is open network decentralized unit (O-RAN distributed unit), O-RU is O-RAN radio unit, O-Cloud is O-RAN cloud and MEC is multi-access edge computing ( multi-access edge computing).

The optimizer startup module 121 may be part of the End-to-End management and control module, SMO, Non-RT RIC, or Near-RT RIC and other components.

The data collection module 122 may be part of an End-to-End management and control module, SMO, Non-RT RIC, or Near-RT RIC, among other components.

The data analysis and processing module 123 may be part of the End-to-End management and control module, SMO, Non-RT RIC or Near-RT RIC and other components.

The optimization direction or target selection module 124 may be part of an End-to-End management and control module, SMO, Non-RT RIC or Near-RT RIC, among other components.

The network and application service joint optimization module 125 may be part of an End-to-End management and control module, SMO, Non-RT RIC or Near-RT RIC, among other components.

The network and application service parameter setting module 126 may be part of the End-to-End management and control module, SMO, Non-RT RIC or Near-RT RIC and other components.

The database 128 may be part of an End-to-End Management and Control Module, SMO, Non-RT RIC or Near-RT RIC, among other components.

The aforementioned sources of network information for optimizing the network can be, for example, O-RAN OAM, 5G RAN network element management system, 4G RAN network element management system, MEC network element management system, application service parameters and guidance controllers, transmission Network management system, core network element management system or other external information.

The aforementioned network elements such as network service elements, application service elements or API endpoints may include O-RAN OAM, Non-RT RIC, 4G RAN network element management system, 5G RAN network element management system, MEC network element management system, Transport Network Management System, Core Network Element Management System, Application Service Parameter and Directive Controller, End-to-End Management and Control Module, A1 or O1 interface endpoint on Near-RT RIC, or on O - E2 or O1 interface endpoint on CU or O-DU.

3 is a flowchart illustrating a method for optimizing a network under an O-RAN architecture according to an embodiment of the present invention. In this embodiment, it is assumed that the optimization program activation module 121, the data collection module 122, the data analysis and processing module 123, the optimization direction or target selection module 124, the network and application service joint optimization module 125, and the network And the application service parameter setting module 126 belongs to the network element 31, and the algorithm pool 127 belongs to the rApp of the network element 31 or the xApp of the network element 32 distributedly or centrally, wherein the network element 31 can be Non- RT RIC, and network element 32 may be a Near-RT RIC. Multi-domain data sources 33 may include O-RAN OAM, 5G RAN network element management system, 4G RAN network element management system, MEC network element management system, application service parameter and guidance controller, transmission network management system, core network network Meta management system or other external sources. Network and application service element management and control modules or API endpoints 34 may include O-RAN OAM, Non-RT RIC, 4G RAN network element management system, 5G RAN network element management system, MEC network element management system, transport network Road Management System, Core Network Element Management System, Application Service Parameter and Guidance Controller, End-to-End Management and Control Module, A1 or O1 interface endpoint on Near-RT RIC, or on O-CU or E2 or O1 interface endpoint on O-DU.

In step S301 , the optimization program activation module 121 deployed in the network element 31 can periodically trigger or trigger the execution of the optimization process of the network according to an event.

In step S302 , the data collection module 122 attached to the network element 31 can obtain network information from the multi-domain data source 33 , wherein the network information may include radio access network-related PM data (eg, user rate, user quantity, base station air medium resource utilization, network signal quality or user report data), signaling data (such as data related to QoS traffic notifications) or application service data (such as service quality, service utilization, user trajectory, Sub-regional shape identification or quantity identification).

In step S303 , the data collection module 122 may store the obtained network information in the database 128 .

In step S304, the data analysis and processing module 123 attached to the network element 31 can analyze the network information obtained in step S302 to generate an optimization target. For example, the data analysis and processing module 123 can predict network demand traffic of each user and service, predict user network signal quality, or analyze each user and application service combination according to the machine learning model.

In step S305 , the data analysis and processing module 123 may store the generated optimization target in the database 128 .

In step S306 , the optimization direction or target selection module 124 assigned to the network element 31 can determine whether optimization needs to be performed according to the information such as the optimization target generated in step S304 , and select a corresponding algorithm from the algorithm pool 127 . The optimization direction or target selection module 124 can take application service requirements into consideration, cooperate with application services and network services, and make comprehensive adjustments to parameters or guidelines. For example, when a cell coverage area or a certain area will experience network congestion and cannot meet QoS requirements, mark the corresponding optimization direction or goal, such as load balancing, interference suppression, coverage adjustment, configuration conflict processing, RAN resource classification processing , Application service classification processing, start network advanced data collection or start application service advanced data collection.

The algorithm pool 127 may be composed of one or more rApps deployed on the network element 31 or one or more xApps deployed on the network element 32 . An rApp or xApp can support algorithmic operations individually or in combination. The data required for the operation of the xApp can be obtained from the database 128, provided by the Near-RT RIC platform, or obtained through the O-RAN A1-EI (A1-enrichment information) standard interface. The data required for rApp operation can be obtained from the Non-RT RIC platform or the SMO platform, or through an external database.

In step S307 , the optimization direction or target selection module 124 may store the selected algorithm in the database 128 .

In step S308, the network and application service joint optimization module 125 deployed in the network element 31 can execute the selected algorithm to generate at least one network configuration according to the rule-based method. If the multi-domain joint optimization algorithm is executed, the network and application service joint optimization module 125 may consider performing optimization for the target optimization function. The optimization result may instruct the system 100 to adjust the network parameters, adjust the application service parameters, issue a policy to the network device, or issue a policy to the application service.

In one embodiment, the network and application service joint optimization module 125 that executes the algorithm can cooperate with the rApp deployed in the network element 31 to obtain calculation data and perform the calculation to generate results. The results may include suggestions for network parameter adjustment, suggestions for adjustment of application service parameters, instructions for network devices, or instructions for application services.

In one embodiment, the network and application service joint optimization module 125 that executes the algorithm can cooperate with the xApp deployed in the network element 32 to obtain calculation data and perform the calculation to generate results. The results may include suggestions for network parameter adjustment, suggestions for adjustment of application service parameters, instructions for network devices, or instructions for application services.

In one embodiment, the network and application service joint optimization module 125 that executes the algorithm can cooperate with the rApp deployed in the network element 31 and the xApp deployed in the network element 32 to obtain calculation data and execute the calculation. post results. The results may include suggestions for network parameter adjustment, suggestions for adjustment of application service parameters, instructions for network devices, or instructions for application services.

In step S309, the network and application service parameter setting module 126 deployed in the network element 31 can configure the network and application service element management and control module or the API endpoint 34 according to the obtained network configuration. For example, the network and application service parameter setting module 126 can call the Non-RT RIC to perform the O-RAN A1 standard interface network guide setting (eg, direct the Near-RT RIC to change the user air interface resource usage priority), call O-RAN OAM executes O-RAN O1 standard interface settings (eg: network mobility parameters or base station transmit power settings), and calls Near-RT RIC to execute O-RAN E2 standard interface parameter settings or guidance (eg: antenna beam direction) Adjustment or user scheduling priority related guidelines), call application service parameters and instruct the controller to execute application service standard/custom parameter settings or issue service guidelines (for example: adjust specific application service demand bandwidth or enable application service to save bandwidth transport mechanism).

FIG. 4 is a schematic diagram illustrating an application scenario of the system 100 according to an embodiment of the present invention. The coverage area 411 of the base station 410 may include user equipment 41 with application requirements A and B, user equipment 42 with application requirements B, and user equipment 43 with application requirements C. The coverage area 421 of the base station 420 may include user equipment 44 with application requirement D, user equipment 45 with application requirement A, and user equipment 46 with application requirement B.

It is assumed that the purpose of optimizing the network is to improve the network quality of user equipment with high application requirements and good QoS. The system 100 can find a set S of combinations of user equipments and application requirements that meet the conditions according to the aforementioned equation (3). If the set S only includes the combination corresponding to the user equipment 41 and the combination corresponding to the user equipment 44 , the system 100 can configure the network of the base station 410 and the base station 420 to improve the network usage of the user equipment 41 or the user equipment 44 .

The system 100 can increase the resource scheduling priority of the user equipment 41 or the user equipment 44 . For example, the system 100 may instruct the application service parameter and direction controller to increase the network bandwidth corresponding to the combination of user equipment 41 or user equipment 44 .

In one embodiment, the system 100 can increase the network bandwidth of the user equipment 41 or 44 by reducing the network bandwidth of the user equipment 42 , 43 , 45 or 46 .

In one embodiment, the system 100 can improve the resource scheduling priority of the user equipment 41 or the user equipment 44 by adjusting the beam of the base station 410 or the base station 420 . For example, the system 100 may adjust the omnidirectional beam of the base station 410 to be a directional beam directed at the user equipment 41 .

FIG. 5 is a schematic diagram illustrating another application scenario of the system 100 according to an embodiment of the present invention. The coverage area 511 of the base station 510 may include the UAV (or inspection robot) 51 , the UAV 52 and the UAV 53 . The coverage area 521 of the base station 520 may include the drone 54 and the drone 55 .

相關。 This embodiment mainly aims to improve the network quality of the UAV flying in a specific area, so as to monitor the specific area perfectly. The system 100 can calculate the application demand value of the UAV according to the position of the UAV. In other words, the location information of the UAV can be used as the independent variable of the application demand function R as shown in Equation (2). The system 100 may set the application demand threshold X, as shown in equation (3), to 0, so that all drones in FIG. 5 belong to the set S. In other words, all UAVs can be optimized with the objective function value

related.

來計算目標優化函數值

。系統100可根據目標優化函數值

產生網路配置，並可根據網路配置來設定各個基地台或用戶設備的參數，藉以優化網路。 In one embodiment, the system 100 may be based on a transfer function as used in Equation (6)

to calculate the objective optimization function value

. The system 100 can optimize the function value according to the objective

The network configuration is generated, and the parameters of each base station or user equipment can be set according to the network configuration, so as to optimize the network.

來計算目標優化函數值

。方程式（7）中的組合集合A可對應於具有高應用需求值的無人機，並且組合集合B可對應於具有低應用需求值的無人機。系統100可設定方程式（7）中的常數a以及常數b，以使常數a大於常數b，藉以調整無人機對目標優化函數值

的權重。如此，在計算目標優化函數值

時，具有高應用需求之無人機的影響將會高於具有低應用需求之無人機的影響。由系統100根據目標優化函數值

所產生的網路配置將會著重於改善具有高應用需求之無人機的網路使用狀況。 In one embodiment, the more important the location of the drone is, the higher the application demand value of the drone is. Relatively speaking, if the location of the drone is less important, the application demand value of the drone is lower. The system 100 can be based on a transfer function as shown in equation (7)

to calculate the objective optimization function value

. Combination set A in equation (7) may correspond to drones with high application demand values, and combination set B may correspond to drones with low application demand values. The system 100 can set the constant a and the constant b in Equation (7), so that the constant a is greater than the constant b, so as to adjust the value of the UAV to the target optimization function

the weight of. In this way, when calculating the objective optimization function value

, the impact of UAVs with high application requirements will be higher than the impact of UAVs with low application requirements. The function value is optimized by the system 100 according to the objective

The resulting network configuration will focus on improving the network usage of drones with high application requirements.

FIG. 6 is a flowchart illustrating a method for optimizing a network according to an embodiment of the present invention, wherein the method can be implemented by the system 100 shown in FIG. 1 . In step S601, a plurality of algorithms are stored. In step S602, network information of the network is received. In step S603, the optimization target is determined according to the network information. In step S604, an algorithm is selected from a plurality of algorithms according to the optimization objective. In step S605, at least one network configuration is generated according to the network information and the algorithm. In step S606, the network is accessed, and the network is configured according to at least one network configuration.

To sum up, the present invention has the following features and effects: the present invention can jointly consider multiple data and network data, and can jointly set parameters or establish guidelines for network and application services. The present invention can achieve a more detailed optimization method by combining data application and optimization in multiple fields. The system of the present invention can automatically complete the optimization-related work; the present invention can consider application requirements and target optimization functions, and allocate network resources accurately according to application requirements, so as to comprehensively consider fairness and operation efficiency, and optimize the effect closer to user requirements. The present invention can be applied to O-RAN architecture, and can be used in special network fields with high controllability of application service content, such as smart factories, smart medical care, smart health, smart logistics, smart monitoring, and high-definition video streaming It can play practical benefits in private network scenarios such as live broadcast, augmented reality services, virtual reality services or drones, and greatly improve the quality of user experience.

100: System

110: Processor

120: Storage Media

121: Optimizer startup module

122: Data collection module

123: Data analysis and processing module

124:Optimize the direction or target selection module

125: Network and application service joint optimization module

126: Network and application service parameter setting module

127: Algorithm Pool

128:Database

130: Transceiver

31, 32: Network Components

33: Multidisciplinary Sources

34: Network and application service element management and control modules or API endpoints

410, 420, 510, 520: base station

411, 421, 511, 521: Coverage

41, 42, 43, 44, 45, 46, 51, 52, 53, 54, 55: User Equipment

S601, S602, S603, S604, S605, S606: Steps

FIG. 1 is a schematic diagram illustrating a system for optimizing a network according to an embodiment of the present invention. FIG. 2 is a schematic diagram illustrating an O-RAN architecture according to an embodiment of the present invention. FIG. 3 is a flowchart illustrating a method for optimizing a network under an O-RAN architecture according to an embodiment of the present invention. FIG. 4 is a schematic diagram illustrating an application context of the system according to an embodiment of the present invention. FIG. 5 is a schematic diagram illustrating another application scenario of the system according to an embodiment of the present invention. FIG. 6 is a flowchart illustrating a method for optimizing a network according to an embodiment of the present invention.

S601, S602, S603, S604, S605, S606: Steps

Claims (21)

A system for optimizing a network, comprising: a transceiver; a storage medium storing a plurality of modules; and a processor coupled to the storage medium and the transceiver, and accessing and executing the plurality of modules, wherein The multiple modules include: an algorithm pool, storing multiple algorithms; a data collection module, receiving network information of the network through the transceiver; a data analysis and processing module, according to the network The information determines the optimization goal; the optimization direction or goal selection module selects an algorithm from the plurality of algorithms according to the optimization goal; the network and application service joint optimization module, based on the algorithm and according to the Network information calculates at least one estimated service quality and at least one application demand value, generates an objective optimization function according to the at least one estimated service quality and the at least one application demand value based on the algorithm, and optimizes according to the objective The function computes a plurality of objective optimization function values corresponding to a plurality of network configurations, respectively, and calculates a plurality of objective optimization function values from the multiple network configurations in response to the at least one network configuration corresponding to the largest objective optimization function value among the plurality of objective optimization function values. The at least one network configuration is selected from the network configurations; and a network and application service parameter setting module accesses the network through the transceiver, and configures the network according to the at least one network configuration network. The system of claim 1, wherein the network and application services joint optimization module is further configured to perform: The at least one network configuration is selected from the plurality of network configurations in response to the at least one target optimization function value being greater than the target optimization function value corresponding to the initial network configuration. The system of claim 1, wherein the network information corresponds to a plurality of user equipments, wherein the plurality of user equipments include at least one user equipment, wherein the network and application services joint optimization module is further configured to perform: calculating, according to the network information, a plurality of estimated quality of service corresponding to the plurality of user equipments, wherein the plurality of estimated service qualities include the at least one quality of service corresponding to the at least one user equipment estimating service quality; calculating the at least one application demand value corresponding to the at least one estimated service quality according to the network information; in response to the at least one estimated service quality being greater than or equal to a service quality threshold and the at least one application requirement value is greater than an application requirement threshold value to generate at least one combination of the at least one user equipment and at least one application requirement, wherein the at least one application requirement corresponds to the at least one application requirement value; and according to the at least one application requirement value Combining produces the objective optimization function. The system of claim 3, wherein the network and application service joint optimization module determines the service quality threshold and the application demand threshold according to the algorithm, and calculates the plurality of thresholds according to an initial network configuration Estimated service quality. The system of claim 3, wherein an objective optimization function value of the objective optimization function is equal to a number of the at least one combination that satisfies the quality of service threshold and the application demand threshold, wherein the at least one combination includes all said combination. The system of claim 3, wherein an objective optimization function value of the objective optimization function is equal to a sum of application demand values satisfying the at least one combination of the service quality threshold and the application demand threshold. The system of claim 3, wherein the at least one combination includes a combination in a first set of combinations and a combination in a second set of combinations, wherein the objective optimization function value of the objective optimization function is equal to the first combination set The linear combination of the first element number of and the second element number of the second combination set. The system of claim 1, wherein the plurality of modules further comprises: an optimizer startup module, in response to receiving an instruction through the transceiver, instructing the data collection module to receive the network information for execution The optimization process of the network. The system of claim 1, wherein the modules further comprise: an optimization program startup module, which periodically instructs the data collection module to receive the network information to perform an optimization process of the network . The system of claim 1, wherein the network information includes at least one of the following: radio access network data, radio access network user identification data, multivariate data, component default configuration data, network Signal data, core network data, transmission network parameters, external input data and configuration data for optimizing system operation. The system of claim 10, wherein the radio access network data includes at least one of: general/advanced configuration management, general/advanced performance management, fault management, user terminal reporting, and minimum path Measurement. The system of claim 10, wherein the radio access network user identification data includes at least one of the following: radio access network user equipment location data, service type, traffic demand, and travel path. The system of claim 10, wherein the multivariate data includes at least one of the following: application service quality, object shape and quantity identification, field map, object location data, and other multivariate information. The system of claim 10, wherein the element default configuration data includes at least one of the following: radio access network configuration parameters, core network parameters, antenna directions, mobile network terminal parameters, and applications service parameters. The system of claim 10, wherein the network signal data includes at least one of the following: a radio frequency signal scan and a drive test mobile phone. The system of claim 10, wherein the core network data includes at least one of the following: user plane function data, quality of service management data, and core network signaling data. The system of claim 10, wherein the transport network parameters include at least one of the following: a packet queue status parameter, a packet processing mechanism parameter, and a packet processing performance parameter number. The system of claim 10, wherein the external input data includes at least one of the following: radio access network intent, application service intent, and field restrictions. The system of claim 10, wherein the optimization system operation configuration data includes at least one of the following: a system configuration file, an optimization program trigger condition configuration file, and an optimization target configuration file. The system of claim 1, wherein the data analysis and processing module calculates parameters according to the network information, and determines the optimization goal according to the parameters, wherein the parameters include at least one of the following: Network data, including at least one of the following: Connectivity, Retention, Availability, Mobility, Integrity, Utilization, RF Signal Patterns, Configuration Management, Core Network Data, and Transmission Network Data; Multiple Information , including at least one of the following: service quality, service demand, schedule information, user trajectory, and traffic; and device stability data, including at least one of the following: device stability and device alarms. A method for optimizing a network, comprising: storing a plurality of algorithms; receiving network information of the network; determining an optimization target according to the network information; selecting from the plurality of algorithms according to the optimization target algorithm; At least one estimated service quality and at least one application demand value are calculated according to the network information based on the algorithm, and generated according to the at least one estimated service quality and the at least one application demand value based on the algorithm an objective optimization function, calculating a plurality of objective optimization function values corresponding to a plurality of network configurations according to the objective optimization function, and in response to the at least one network configuration corresponding to the maximum value of the plurality of objective optimization function values selecting the at least one network configuration from the plurality of network configurations by optimizing a function value with an objective; and accessing the network and configuring the network according to the at least one network configuration.

Images