Wireless communication network operation optimization management system based on multimode communication

Abstract

The invention discloses a wireless communication network operation optimization management system based on multimode communication, which belongs to the technical field of wireless communication network management, and adopts a three-layer closed-loop optimization architecture of data acquisition, analysis decision and dynamic execution to generate a structured demand vector index by user service type and service environment information coding. Based on a multi-attribute decision theory engine, a demand vector index and a network performance index are dynamically matched, an optimal communication mode combination is output to construct a multi-mode composite network environment, cross-mode resource allocation is executed, meanwhile, the global load state is monitored, and cross-mode load balancing control is implemented. According to the method, end-to-end intelligent closed-loop optimization starting from service requirements is realized, the overall resource utilization rate, the user experience quality and the operation and maintenance automation level of the multimode heterogeneous network are effectively improved, and the technical problems of service and resource static matching, difficulty in cross-mode collaboration and easiness in local overload in traditional network management are solved.

Description

Technical Field

[0001] This invention relates to the field of wireless communication network management technology, and more specifically, to a wireless communication network operation optimization and management system based on multimode communication. Background Technology

[0002] Wireless communication network operation and management refers to a series of technologies and activities that continuously monitor, control, optimize, and maintain wireless communication networks to ensure that the network can operate stably, efficiently, securely, and economically under limited wireless resources (such as spectrum and power) and meet users' quality of service requirements.

[0003] Multimode communication means that wireless communication networks support multiple communication modes, such as 4G, 5G, Wi-Fi, and satellite communication, which greatly expands the "objects of management" from managing a single-standard network to managing a heterogeneous and integrated network ecosystem, and the complexity of management increases exponentially.

[0004] Against this backdrop, efficiently managing and optimizing the operation of multimode wireless communication networks has become crucial for improving user experience and network operational efficiency. However, traditional wireless network operation and management solutions suffer from the following technical limitations when facing these complex scenarios:

[0005] 1. Static matching of business needs and network resources, lacking refinement and personalization: Traditional network access selection is mostly based on simple network-side signal quality indicators (such as Received Signal Strength Indicator (RSSI)) or static service type labels. This approach cannot deeply perceive and understand the specific service type currently being run by the user and the service environment it depends on. Static matching of business needs and network resources lacks refinement and personalization, and cannot deeply perceive and understand the specific service type currently being run by the user and the context environment it depends on.

[0006] Second, cross-mode resource coordination is difficult, resulting in an "island" effect: the management of key network resources such as spectrum, power, and routing is usually carried out independently within each communication mode, lacking a unified view and collaborative optimization mechanism across modes;

[0007] 3. Load balancing capabilities are limited to a single mode, which can easily lead to local overload: Conventional load balancing technologies are mainly designed for a single network standard and balance the load through scheduling among users. However, when a large number of users or high-bandwidth services are concentrated in a specific network mode, the system lacks the ability to intelligently guide some service traffic to other relatively idle network modes.

[0008] Therefore, in response to the actual technical deficiencies, a wireless communication network operation optimization and management system based on multi-mode communication is proposed. Summary of the Invention

[0009] The purpose of this invention is to address practical technical deficiencies. It provides a wireless communication network operation optimization and management system based on multi-mode communication. This system is an end-to-end closed-loop management framework that is driven by business needs, performs cross-mode network combination and resource coordination, and intelligent global load control. It achieves a leap from decentralized and isolated to unified coordination, from static configuration to dynamic closed loop, and from single-mode management to cross-mode integration at the system architecture level.

[0010] The objective of this invention can be achieved through the following technical solution: a wireless communication network operation optimization and management system based on multi-mode communication, including a multi-dimensional network monitoring module, a communication mode decision module, a cross-mode resource dynamic allocation module, and a global load balancing control module;

[0011] The multi-dimensional network monitoring module is used to collect network performance indicators of various network modes in real time, and simultaneously obtain information on user service types and business environment during service operation, and uniformly send it to the communication mode decision module.

[0012] The communication mode decision module receives user service type and service environment information, performs multi-dimensional feature fusion encoding on the user service type and service environment information to generate structured demand vector indicators, and dynamically matches and analyzes the demand vector indicators with the network performance indicators of each network mode based on the multi-attribute decision theory engine, outputs the optimal communication mode combination, constructs a multi-mode composite network environment based on the communication mode combination, and sends it to the cross-mode resource dynamic allocation module and the global load balancing control module.

[0013] The cross-mode resource dynamic allocation module performs cross-mode resource dynamic allocation operations based on the multi-mode composite network environment; the global load balancing control module is used to monitor the global load status of each communication mode in the multi-mode composite network environment, and implement cross-mode load balancing control based on the load status to avoid overload of a single network mode.

[0014] Furthermore, the multi-dimensional network monitoring module includes a network performance indicator acquisition subsystem and a service identification and environmental awareness subsystem.

[0015] The network performance metrics acquisition subsystem is used to collect network performance metrics from multiple points such as terminal modems, base station gateway systems, and core network probes, including signal strength, bandwidth utilization, latency, jitter, and packet loss rate. The service identification and environment awareness subsystem is used to identify user service types and obtain service environment information during service operation.

[0016] Furthermore, a lightweight convolutional neural network is used to analyze traffic characteristics such as packet size, packet arrival interval, and service flow duration. Based on the traffic characteristics, the user service type is identified. Specifically, the feature parameters of the traffic characteristics are input into a pre-trained and optimized deep learning classification model, and the user service type classification is output, including at least one of ultra-low latency communication, enhanced mobile broadband, massive machine communication, video streaming, and large file transfer.

[0017] Business environment information includes various data such as the user device's mobile speed, geographical location information, remaining battery power, user priority level, and historical network quality data.

[0018] Furthermore, the specific process of generating structured demand vector indicators includes:

[0019] The communication mode decision module pre-stores a service type-basic service template library. When it receives a user service type sent by the multi-dimensional network monitoring module, it calls the basic service quality requirement set corresponding to the specific user service type from the service type-basic service template library. The requirement vector in the basic service quality requirement set includes latency requirements, bandwidth requirements, jitter requirements, packet loss rate requirements, stability weight, cost weight, and energy consumption weight.

[0020] The acquired business environment information is used as a dynamic adjustment factor in the set of basic service quality requirements to correct the demand vector in real time, resulting in a corrected and quantified demand vector. All quantified demand vectors are then fused and encoded according to preset dimensions to output a unified structured demand vector index.

[0021] Furthermore, the process of dynamically matching and analyzing the demand vector indicators with the network performance indicators of each network mode includes: the multi-attribute decision theory engine adopts a pre-trained and optimized multi-mode selection decision model, takes the demand vector indicators and network performance indicators as state inputs, takes the communication mode selection action as output, and outputs a combination of multiple network modes as the optimal communication mode combination.

[0022] Furthermore, the cross-mode resource dynamic allocation operation specifically includes dynamic allocation of spectrum resources, transmit power control, and data transmission route optimization.

[0023] Furthermore, the process of implementing cross-modal load balancing control based on load status includes:

[0024] In a multi-mode composite network environment, through multi-source data collection, normalized resource calculation and business weighted fusion, normalized load status indicators across network modes are generated. The changes in load status indicators are continuously monitored using time series, and standard threshold ranges are dynamically defined. The current load status indicator trend is compared with the standard threshold range, and each network mode is labeled with a load level, including normal network, light-load network, heavy-load network and overload network.

[0025] For network modes marked as heavily loaded or overloaded, the service flows connected to that network mode are obtained, and their "portability" is determined. If the portability index is met, the network mode is marked as a network to be migrated. Among all network modes marked as normal or lightly loaded, an optimal network mode is selected for the network to be migrated based on the priority of user equipment service needs and the real-time load of the network mode.

[0026] The present invention also proposes a method for optimizing and managing the operation of wireless communication networks based on multi-mode communication, comprising the following steps:

[0027] S1: During the operation of the wireless communication network, network performance indicators of multiple available communication network modes are collected in real time.

[0028] S2: Synchronously identify the current user's service type and obtain the business environment information during service operation, and generate demand vector indicators based on the user's service type and business environment information;

[0029] S3: Input the generated demand vector index and the network performance index into the multi-mode selection decision model, and output the optimal communication mode combination through fusion analysis;

[0030] S4: Construct a multi-mode composite network environment based on the optimal communication mode combination, and perform cross-mode resource dynamic allocation in the multi-mode composite network environment;

[0031] S5: Monitors the global load status of each communication mode in a multi-mode composite network environment, and implements cross-mode load balancing control based on the load status to avoid overload of a single network mode.

[0032] Compared with the prior art, the advantages of this invention are:

[0033] 1. This invention collects network performance indicators, user service types, and service environment information in real time. It generates structured demand vector indicators by fusing and encoding user service types and service environment information. Based on a multi-attribute decision theory engine, it dynamically matches the demand vector indicators with network performance indicators and outputs the optimal communication mode combination to construct a multi-mode composite network environment. Then, it performs cross-mode spectrum resource allocation, power control, and routing optimization. At the same time, it monitors the global load status and implements cross-mode load balancing control, realizing end-to-end intelligent closed-loop optimization starting from service needs. This effectively improves the overall resource utilization, user experience quality, and operation and maintenance automation level of multi-mode heterogeneous networks, thereby solving the technical problems of static matching of services and resources, difficulty in cross-mode collaboration, and easy local overload in traditional network management.

[0034] 2. Construct a multi-mode composite network environment by combining real-time optimal communication modes. The construction of the multi-mode composite network environment and the rapid mode switching mechanism ensure that the failure or performance degradation of a single network node will no longer cause service interruption. The system can quickly migrate services to a healthy network, significantly improving the availability of services across the entire network.

[0035] 3. In a multi-mode composite network environment, through multi-source data collection, normalized resource calculation and service weighted fusion, normalized load status indicators across network modes are generated. By continuously monitoring the changes in load status indicators using time series, and based on trend prediction, users / services are proactively guided and migrated between heterogeneous networks, fundamentally avoiding overload of a single network and improving the overall capacity and resource utilization of the entire multi-mode network system. Attached Figure Description

[0036] Figure 1 This is a system principle block diagram of the present invention;

[0037] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0039] Example 1: This invention discloses a wireless communication network operation optimization and management system based on multi-mode communication. Please refer to [link / reference]. Figures 1-2 It includes a multi-dimensional network monitoring module, a communication mode decision module, a cross-mode resource dynamic allocation module, and a global load balancing control module.

[0040] The multi-dimensional network monitoring module is used to collect network performance indicators of various network modes in real time, and simultaneously obtain information on user service types and business environment during service operation, and uniformly send it to the communication mode decision module.

[0041] The multi-dimensional network monitoring module includes a network performance indicator acquisition subsystem and a service identification and environment perception subsystem. The network performance indicator acquisition subsystem is used to collect network performance indicators from multiple points such as terminal modems, base station gateway systems, and core network probes. Network performance indicators include signal strength, bandwidth utilization, latency, jitter, and packet loss rate. The service identification and environment perception subsystem is used to identify user service types and obtain service environment information during service operation.

[0042] A lightweight convolutional neural network (1D-CNN) is used to analyze traffic characteristics such as packet size, packet arrival interval, and service flow duration. Based on the traffic characteristics, the user service type is identified. Specifically, the feature parameters of the traffic characteristics are input into a pre-trained and optimized deep learning classification model, and the user service type classification is output. The user service type includes at least one of ultra-low latency communication, enhanced mobile broadband, massive machine communication, video streaming, and large file transfer.

[0043] Business environment information includes various data such as the user device's mobile speed, geographical location information, remaining battery power, user priority level, and historical network quality data.

[0044] The communication mode decision module receives user service type and service environment information, performs multi-dimensional feature fusion encoding on the user service type and service environment information to generate structured demand vector indicators, and dynamically matches and analyzes the demand vector indicators with the network performance indicators of each network mode based on the multi-attribute decision theory engine, outputs the optimal communication mode combination, constructs a multi-mode composite network environment based on the communication mode combination, and sends it to the cross-mode resource dynamic allocation module and the global load balancing control module.

[0045] The specific process of fusing user business types and business environment information through multi-dimensional features to generate structured demand vector indicators includes:

[0046] The communication mode decision module pre-stores a service type-basic service template library. When it receives a user service type (such as "remote control", "video streaming", or "large file download") sent by the multi-dimensional network monitoring module, it retrieves the corresponding basic service quality requirement set for the specific user service type from the service type-basic service template library. The requirement vector in the basic service quality requirement set includes latency requirements, bandwidth requirements, jitter requirements, packet loss rate requirements, stability weight, cost weight, and energy consumption weight. This basic service quality requirement set defines the core quantitative indicator benchmark values ​​for this type of user service. For example, remote control service corresponds to the benchmark of "extremely low latency (less than 10ms)" and "ultra-high reliability (greater than 99.999%)"; video streaming service corresponds to the benchmark of "high bandwidth (greater than 5Mbps)" and "medium latency (less than 200ms)".

[0047] The acquired business environment information is used as a dynamic adjustment factor in the set of basic service quality requirements to correct the demand vector in real time, resulting in a corrected and quantified demand vector. The correction and adjustment rules include mobility adjustment, resource and cost adjustment, and spatiotemporal and network status adjustment. All quantified demand vectors are fused and encoded according to preset dimensions to output a unified structured demand vector index.

[0048] Among them, mobility adjustment: if the user is in a high-speed mobile state, the absolute requirements for latency will be appropriately relaxed, but at the same time the weight of the indicators of "network switching stability" and "connection retention" will be significantly increased; resource and cost adjustment: the weight and threshold of the "energy efficiency" indicator will be dynamically adjusted according to the terminal's battery level (prioritizing energy saving when the battery is low); spatiotemporal and network status adjustment: combining historical congestion patterns (such as identifying the current time when the office area is in midday Wi-Fi congestion) and the current location (such as being at the edge of multi-mode coverage), the indicators such as "network reliability" and "minimum guaranteed bandwidth" will be conservatively modified, or the "multi-mode redundancy" requirement will be increased;

[0049] Here, "pre-defined dimensions" refers to a set of core evaluation indicators that are pre-defined during the system design phase to comprehensively characterize the network service quality requirements of a user service. Based on communication theory and industry standards, these indicators are pre-determined and their core function is to provide a unified and structured framework for "fusion and coding." The modified and quantified requirements (e.g., latency ≤ 20ms, bandwidth ≥ 10Mbps, high cost weight, low energy efficiency weight, etc.) are filled into a standardized data structure (i.e., a vector) according to this pre-defined and fixed dimension order and format. The final output is a "unified and structured requirement vector indicator," for example, requirement vector indicator = [latency target value, bandwidth target value, packet loss rate target value, stability weight, cost weight, energy efficiency weight].

[0050] Generating a structured demand vector index by combining user service type and service environment information is a process of multi-dimensional feature fusion and quantitative mapping. This process aims to transform qualitative and discrete service and environmental information into a set of quantitative, continuous, machine-decision-aware numerical vectors. This structured demand vector index fully represents "what quality of service the network needs to provide for this specific service type under the current specific service environment".

[0051] The process of dynamically matching and analyzing demand vector indicators with network performance indicators of various network models includes:

[0052] The multi-attribute decision theory engine adopts a pre-trained and optimized multi-mode selection decision model. It takes the demand vector index and network performance index as state inputs and the communication mode selection action as output. The model is trained and optimized through a reward mechanism aimed at maximizing the overall utility of the network, and outputs a combination of multiple network modes as the optimal communication mode combination.

[0053] By constructing a multi-mode composite network environment using real-time optimal communication mode combinations, the construction of the multi-mode composite network environment and the rapid mode switching mechanism ensure that the failure or performance degradation of a single network node no longer leads to service interruption. The system can quickly migrate services to a healthy network, significantly improving the availability of services across the entire network.

[0054] Example 2: The cross-mode resource dynamic allocation module performs cross-mode resource dynamic allocation operations based on a multi-mode composite network environment. The operations include dynamic allocation of spectrum resources, transmit power control, and data transmission route optimization.

[0055] For dynamic spectrum resource allocation: the service demand vector and interference status index of each communication mode in a multi-mode composite network environment are obtained. The service demand vector is generated by fusing and encoding user service type and service environment information. The interference status index can be obtained through multiple channels such as network-side measurement, terminal-side reporting, spectrum sensing and historical database, and fused into an "interference map". With the service demand vector as the optimization target and the real-time sensed "interference map" as the core constraint, the optimal spectrum resource allocation scheme is dynamically solved through a built-in optimization solver. This is a dynamic optimization process of "demand-driven and interference-constrained".

[0056] More specifically, regarding network-side measurements: each base station / access point continuously measures the signal strength and characteristics received from other network systems (such as adjacent 5G base stations, co-frequency Wi-Fi APs, and radar systems) to generate external interference reports; regarding terminal-side reporting: user terminals, acting as distributed sensors, can report the signal strength and quality information of all networks they scan that are currently not connected but can be detected, helping the network side to create a more detailed local interference map; regarding spectrum sensing: when the network or terminal has cognitive radio capabilities, it can directly detect spectrum holes (unoccupied spectrum) and interference temperature in the target frequency band; regarding historical databases: combined with geographical location information, it calls up historically accumulated typical interference pattern data for the area (e.g., in an office building, the 5G frequency band of operator A and the Wi-Fi frequency band of enterprises in the building have serious mutual interference during a specific period).

[0057] For transmit power control operations: obtain the channel quality, network load and service reliability requirements of each communication mode, input them into a pre-trained and optimized power control decision model, and use optimization algorithms to dynamically adjust the transmit power level of the base station or user terminal to achieve joint optimization of coverage, capacity and energy consumption;

[0058] Channel quality is periodically reported by user terminals, including reference signal received power, signal-to-noise ratio / signal-to-interference-plus-noise ratio, channel rank indication, and precoding matrix indication. This quantifies the robustness and potential transmission capacity of the current wireless link. Good channel quality means that reliable communication can be achieved with lower power. Network load information is obtained from resource status counters within the base station / access point, including physical resource block utilization, number of active users, packet queue length, and average scheduling wait time. The purpose is to quantify the service pressure and resource scarcity of the current cell or access point. High load usually means increased interference between users, requiring more refined power control. Service reliability requirements come from the relevant dimensions of the "demand vector indicators" output by the communication mode decision module for the current user or service flow. The core is the quantitative constraints on reliability (such as target block error rate, target packet loss rate) and latency (such as maximum allowable latency). The purpose is to define the service target baseline for power control.

[0059] For data transmission routing optimization: In a multi-mode composite network environment, a multi-path network transmission mechanism is established to monitor each available transmission network path in real time. Data traffic is dynamically allocated based on the path performance indicators and service flow characteristics of each path, and seamless switching and path redundancy are supported. The path performance indicators are obtained through end-to-end active probing or passive measurement, and the service flow characteristics are directly derived from the identification of service demand vectors.

[0060] Through the combined effects of spectrum allocation, power control, and data transmission routing optimization, the system ensures that the selected "multi-mode composite network environment" can carry user services in the most efficient and reliable way, ultimately realizing the core value of improving user experience and network efficiency.

[0061] The global load balancing control module is used to monitor the global load status of each communication mode in a multi-mode composite network environment, and implement cross-mode load balancing control based on the load status to avoid overload of a single network mode.

[0062] The process of implementing cross-mode load balancing control based on load status includes:

[0063] In a multi-mode composite network environment, normalized load status indicators are generated across network modes through multi-source data collection, normalized resource calculation, and service weighted fusion. Time series data is used to continuously monitor changes in load status indicators and dynamically define standard threshold ranges. These standard threshold ranges are not fixed values ​​but are dynamically adjusted based on network capacity, historical patterns, and current service mix types. The current load status indicator trend is compared with the standard threshold range, and each network mode is labeled with a load level, including normal network, lightly loaded network, heavily loaded network, and overloaded network. Based on the current trend, the load changes of each network in the next few minutes are predicted, enabling the system to identify networks that are "currently overloaded" and "about to be overloaded."

[0064] For network modes marked as heavily loaded or overloaded, the service flows connected to these network modes are obtained, and their "portability" is determined. If the portability criteria are met, the network mode is marked as a network to be migrated. Among all network modes marked as normal or lightly loaded, an optimal network mode target is selected for the network to be migrated based on a collaborative decision made according to the priority of user equipment service needs and the real-time load of the network mode.

[0065] The criteria for determining "portability" include: service requirement compatibility (whether the service's requirement vector, such as low latency, can be met by other candidate networks), terminal capability (whether the user terminal supports switching to the candidate network), migration cost (estimating the impact of possible short-term interruptions and signaling overhead on user experience), and service priority (combining user level and service criticality). When all of the above criteria are met, the network mode marked as a heavy-load network or an overloaded network is deemed to meet the portability index.

[0066] The criteria for selecting an optimal network mode for a network to be migrated include: capacity margin (the target network has sufficient resources to handle the service), performance matching degree (the performance of the target network, such as predicted latency, can meet the demand vector indicators of the service), migration efficiency (historical performance such as handover latency and success rate from the source network to the target network), and global optimization (the migrated service will not immediately cause the target network to become a new hotspot). When a normal network or a lightly loaded network meets all of the above criteria, it is taken as the optimal network mode target for the network to be migrated.

[0067] Global load balancing is an intelligent and automated traffic scheduling system that proactively guides and migrates users / services between heterogeneous networks by monitoring the entire network status in real time. This fundamentally avoids overload on a single network and improves the overall capacity and resource utilization of the entire multi-mode network system.

[0068] As can be seen from Embodiments 1 and 2, this invention also proposes a method for optimizing and managing the operation of wireless communication networks based on multi-mode communication. Please refer to [link / reference]. Figure 2 It includes the following steps:

[0069] S1: During the operation of the wireless communication network, network performance indicators of multiple available communication network modes are collected in real time, including signal strength, bandwidth utilization, latency, jitter, and packet loss rate.

[0070] S2: Synchronously identify the current user's service type and obtain the business environment information during service operation, and generate demand vector indicators based on the user's service type and business environment information;

[0071] S3: Input the generated demand vector index and all network performance indexes into the multi-mode selection decision model, and output the optimal combination of communication modes through fusion analysis;

[0072] S4: Construct a multi-mode composite network environment based on the optimal combination of communication modes, and perform cross-mode dynamic resource allocation in the multi-mode composite network environment;

[0073] S5: Monitors the global load status of each communication mode in a multi-mode composite network environment, and implements cross-mode load balancing control based on the load status to avoid overload of a single network mode.

[0074] In summary, the architecture employs a three-layer closed-loop optimization approach: "data acquisition - analysis and decision-making - dynamic execution." This approach integrates business requirements, network status, and resource status in real time, supporting collaborative management in heterogeneous network environments (such as 4G / 5G / Wi-Fi / satellite / IoT private networks). Specifically:

[0075] By fusing and encoding user service type and service environment information to generate structured demand vector indicators, and based on a multi-attribute decision theory engine, the demand vector indicators are dynamically matched with network performance indicators to output the optimal communication mode combination to construct a multi-mode composite network environment. This enables cross-mode spectrum resource allocation, power control, and routing optimization. Simultaneously, it monitors the global load status and implements cross-mode load balancing control, achieving end-to-end intelligent closed-loop optimization starting from service requirements. This effectively improves the overall resource utilization, user experience quality, and operation and maintenance automation level of multi-mode heterogeneous networks, thus solving the technical problems of static matching of services and resources, difficulty in cross-mode collaboration, and susceptibility to local overload in traditional network management.

[0076] The above description is merely a preferred embodiment of the present invention; however, the scope of protection of the present invention is not limited thereto; any equivalent substitutions or modifications made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and improved concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A wireless communication network operation optimization and management system based on multimode communication, characterized in that: It includes a multi-dimensional network monitoring module, a communication mode decision module, a cross-mode resource dynamic allocation module, and a global load balancing control module; The multi-dimensional network monitoring module is used to collect network performance indicators of various network modes in real time, and simultaneously obtain information on user service types and business environment during service operation, and uniformly send it to the communication mode decision module. The communication mode decision module receives user service type and service environment information, performs multi-dimensional feature fusion encoding on the user service type and service environment information to generate structured demand vector indicators, and dynamically matches and analyzes the demand vector indicators with the network performance indicators of each network mode based on the multi-attribute decision theory engine, outputs the optimal communication mode combination, constructs a multi-mode composite network environment based on the communication mode combination, and sends it to the cross-mode resource dynamic allocation module and the global load balancing control module. The cross-mode resource dynamic allocation module performs cross-mode resource dynamic allocation operations based on the multi-mode composite network environment; the global load balancing control module is used to monitor the global load status of each communication mode in the multi-mode composite network environment, and implement cross-mode load balancing control based on the load status to avoid overload of a single network mode.

2. The wireless communication network operation optimization management system based on multi-mode communication according to claim 1, characterized in that: The multi-dimensional network monitoring module includes a network performance indicator acquisition subsystem and a service identification and environmental perception subsystem. The network performance metrics acquisition subsystem is used to collect network performance metrics from multiple points such as terminal modems, base station gateway systems, and core network probes, including signal strength, bandwidth utilization, latency, jitter, and packet loss rate. The service identification and environment awareness subsystem is used to identify user service types and obtain service environment information during service operation.

3. The wireless communication network operation optimization management system based on multimode communication according to claim 2, characterized in that: A lightweight convolutional neural network is used to analyze traffic characteristics such as packet size, packet arrival interval, and service flow duration. Based on the traffic characteristics, the user service type is identified. Specifically, the feature parameters of the traffic characteristics are input into a pre-trained and optimized deep learning classification model, and the user service type classification is output, including at least one of ultra-low latency communication, enhanced mobile broadband, massive machine communication, video streaming, and large file transfer. Business environment information includes various data such as the user device's mobile speed, geographical location information, remaining battery power, user priority level, and historical network quality data.

4. The wireless communication network operation optimization management system based on multimode communication according to claim 3, characterized in that: The specific process of generating structured demand vector indicators includes: The communication mode decision module pre-stores a service type-basic service template library. When it receives a user service type sent by the multi-dimensional network monitoring module, it calls the basic service quality requirement set corresponding to the specific user service type from the service type-basic service template library. The requirement vector in the basic service quality requirement set includes latency requirements, bandwidth requirements, jitter requirements, packet loss rate requirements, stability weight, cost weight, and energy consumption weight. By using business environment information as a dynamic adjustment factor in the set of basic service quality requirements, the demand vector is corrected in real time to obtain a corrected and quantified demand vector. This vector is then fused and encoded according to preset dimensions to output a unified and structured demand vector index.

5. The wireless communication network operation optimization management system based on multimode communication according to claim 4, characterized in that: The process of dynamically matching and analyzing the demand vector index with the network performance index of each network mode includes: the multi-attribute decision theory engine adopts a pre-trained and optimized multi-mode selection decision model, takes the demand vector index and network performance index as state input, takes the communication mode selection action as output, and outputs a combination of multiple network modes as the optimal communication mode combination.

6. The wireless communication network operation optimization management system based on multimode communication according to claim 5, characterized in that: The cross-mode resource dynamic allocation operation specifically includes dynamic allocation of spectrum resources, transmit power control, and data transmission route optimization.

7. The wireless communication network operation optimization management system based on multimode communication according to claim 6, characterized in that: The process of implementing cross-mode load balancing control based on load status includes: In a multi-mode composite network environment, through multi-source data collection, normalized resource calculation and business weighted fusion, normalized load status indicators across network modes are generated. The changes in load status indicators are continuously monitored using time series, and standard threshold ranges are dynamically defined. The current load status indicator trend is compared with the standard threshold range, and each network mode is labeled with a load level, including normal network, light-load network, heavy-load network and overload network. For network modes marked as heavily loaded or overloaded, obtain the service flows connected to that network mode and determine its "portability". If the portability index is met, mark the network mode as a network to be migrated. Among all network modes marked as normal or lightly loaded, make a collaborative decision based on the service demand priority of user equipment and the real-time load of the network mode to select the optimal network mode for the network to be migrated.

8. A method for optimizing and managing the operation of a wireless communication network based on multi-mode communication, comprising the wireless communication network operation optimization and management system based on multi-mode communication as described in any one of claims 1-7, characterized in that, Includes the following steps: S1: During the operation of the wireless communication network, network performance indicators of multiple available communication network modes are collected in real time. S2: Synchronously identify the current user's business type and obtain the business environment information during business operation, and generate demand vector indicators based on the user's business type and business environment information; S3: Input the generated demand vector index and the network performance index into the multi-mode selection decision model, and output the optimal communication mode combination through fusion analysis; S4: Construct a multi-mode composite network environment based on the optimal communication mode combination, and perform cross-mode resource dynamic allocation in the multi-mode composite network environment; S5: Monitors the global load status of each communication mode in a multi-mode composite network environment, and implements cross-mode load balancing control based on the load status to avoid overload of a single network mode.

Images