Resource management method and system for multi-network environment

Abstract

The invention provides a resource management method and system for a multi-network environment, and the method comprises the steps: carrying out the initialization after intelligent terminal equipment is started; wherein the intelligent terminal equipment comprises a 5G module, an FTTR module and an AI processing module; first network data of the 5G module, second network data of the FTTR module and user service demand data are collected and sent to the AI processing module; obtaining an optimal transmission path, bandwidth allocation information and time slot scheduling information output by the AI processing module according to the first network data, the second network data and the user service demand data; and if the AI processing module outputs abnormal state information according to the first network data, the second network data and the user service demand data, selecting a preset solution strategy according to the abnormal state information to quickly recover the network function. According to the invention, an intelligent resource scheduling scheme in a multi-network environment is provided, and the fault recovery speed is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of network resource management, and in particular to a resource management method and system for a multi-network environment. Background Art

[0002] With the large-scale deployment of 5G networks, leveraging core technologies such as millimeter wave frequency bands, Massive MIMO, and network slicing, a next-generation communications system has been established, featuring three key features: enhanced mobile broadband (eMBB), ultra-reliable and low-latency (uRLLC), and massive machine-type communications (mMTC). This significantly enhances users' immersive service experience. Furthermore, FTTR (Fiber to the Room) technology, through a network of hybrid optical cables and distributed intelligent ONUs, extends 10Gbps-class all-optical networks to every functional area of ​​a home, enabling Wi-Fi 7 hotspot coverage with microsecond-level transmission latency. This completely resolves the persistent issues of traditional home networks, such as multi-wall attenuation and cross-layer roaming interruptions.

[0003] Based on this, the hybrid access mode of 5G network and FTTR technology, that is, the deep integration of wireless and fiber optic technologies, can build an end-to-end network system with wide coverage, high speed and low latency to meet the diverse business needs in home, enterprise and industrial scenarios.

[0004] However, existing 5G and FTTR technologies still face some issues or technical deficiencies in their converged applications. First, at the resource coordination level, the resource scheduling mechanisms for wireless channels and optical fiber media are disconnected, resulting in spectrum resource utilization efficiency of less than 60%, and a lack of coordinated optimization between Massive MIMO beamforming and PON dynamic bandwidth allocation strategies. Second, regarding fault recovery, the intelligent diagnostic capabilities of cross-domain transmission paths are weak. Correlating optical link outages with radio access network congestion takes over 15 minutes to locate, severely impacting the SLA achievement rate for high-value services. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a resource management method and system for a multi-network environment, which solves the problems of incomplete resource scheduling design and slow fault recovery speed in the hybrid access mode of 5G network and FTTR technology in the existing technology.

[0006] According to an embodiment of the present invention, a first aspect provides a resource management method for a multi-network environment, comprising:

[0007] After the smart terminal device is started, it is initialized; wherein the smart terminal device includes a 5G module, an FTTR module and an AI processing module;

[0008] Collecting the first network data of the 5G module, the second network data of the FTTR module, and user service demand data and sending them to the AI ​​processing module;

[0009] Obtaining the optimal transmission path, bandwidth allocation information, and time slot scheduling information output by the AI ​​processing module based on the first network data, the second network data, and the user service demand data;

[0010] Among them, if the AI ​​processing module outputs abnormal status information based on the first network data, the second network data and the user service demand data, a preset solution strategy is selected according to the abnormal status information to quickly restore the network function.

[0011] Optionally, the AI ​​processing module outputs an optimal transmission path according to the first network data, the second network data, and the user service demand data, including:

[0012] Obtain the business type from the user's business demand data;

[0013] According to the service type, the first network data and the second network data, the optimal transmission path weight based on the 5G module and the FTTR module is calculated through DQN algorithm reinforcement learning, and the optimal transmission path weight is used to indicate the optimal transmission path.

[0014] Optionally, the AI ​​processing module outputs bandwidth allocation information according to the first network data, the second network data, and the user service demand data, including:

[0015] Obtain the business type from the user's business demand data;

[0016] According to the service type, the first network data and the second network data, the bandwidth ratio of each service type in the 5G module and the FTTR module is dynamically adjusted through an improved weighted fair queue algorithm.

[0017] Optionally, the AI ​​processing module outputs time slot scheduling information according to the first network data, the second network data, and the user service demand data, including:

[0018] Obtain the business type from the user's business demand data;

[0019] If the current service type is a preset priority service, the preset priority service is dynamically inserted into the allocated time slot of the 5G module, and resource compensation is performed for the interrupted service.

[0020] Optionally, before the AI ​​processing module outputs the optimal transmission path, bandwidth allocation information, and time slot scheduling information based on the first network data, the second network data, and the user service demand data, the AI ​​processing module includes:

[0021] A feature vector is constructed according to the first network data, the second network data and the user service demand data.

[0022] Optionally, the abnormal state information includes at least abnormal optical fiber state and abnormal signal transmission;

[0023] Selecting preset resolution strategies based on fiber status abnormality information to quickly restore network functions, including: when a fiber break or excessive attenuation is detected in the FTTR module, real-time evaluation of available bandwidth in the 5G module is performed, switching preset priority services to the 5G module, and temporarily storing or downgrading remaining services;

[0024] Select a preset solution strategy based on the abnormal status information of signal transmission abnormality to quickly restore network functions, including: when it is detected that the received signal strength of the 5G module is lower than the preset strength value, switch to the FTTR module, and start the beam search algorithm to re-lock the available beam.

[0025] Optionally, after obtaining the optimal transmission path, bandwidth allocation information, and time slot scheduling information output by the AI ​​processing module based on the first network data, the second network data, and the user service demand data, the method further includes:

[0026] According to the traffic load conditions of the 5G module and the FTTR module, the working modes of the 5G module and the FTTR module are adjusted, and the FTTR dominant mode is enabled within a preset period of time.

[0027] The second aspect provides a resource management system for a multi-network environment, including a controller and an intelligent terminal device, wherein the intelligent terminal device includes a 5G module, an FTTR module and an AI processing module, and the controller implements the resource management method for a multi-network environment as described above.

[0028] Optionally, seamless switching is achieved between the 5G module and the FTTR module through hard switching or soft switching.

[0029] Optionally, the 5G module dynamically adjusts the antenna array phase through beamforming;

[0030] The FTTR module enables forward error correction to automatically detect and correct errors in transmission, and monitors and adjusts the transmitted and received optical powers based on dynamic optical power.

[0031] Compared to existing technologies, this invention offers the following advantages: It deeply integrates 5G modules, FTTR optical network access modules, and AI processing modules. The AI ​​processing module analyzes first and second network data and user service demand data in real time, enabling dynamic switching of multiple network interfaces, collaborative resource allocation, and intelligent signal optimization. Furthermore, the AI ​​processing module pre-defines solution strategies, effectively presetting them to quickly restore network functionality and improve fault recovery speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 Schematic diagram of the implementation process of the resource management method for a multi-network environment according to an embodiment of the present invention;

[0033] Figure 2 A schematic diagram of the structure of a resource management system for a multi-network environment according to an embodiment of the present invention;

[0034] Figure 3 This is a schematic diagram of the actual structure of a resource management system for a multi-network environment according to an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0036] like Figure 1 As shown, the embodiment of the present invention proposes a resource management method for a multi-network environment, including but not limited to the following steps:

[0037] S101. After the smart terminal device is started, it is initialized; wherein the smart terminal device includes a 5G module, an FTTR module, and an AI processing module;

[0038] S102: Collect the first network data of the 5G module, the second network data of the FTTR module, and user service demand data and send them to the AI ​​processing module;

[0039] S103: Obtaining the optimal transmission path, bandwidth allocation information, and time slot scheduling information output by the AI ​​processing module based on the first network data, the second network data, and the user service demand data;

[0040] S104. If the AI ​​processing module outputs abnormal status information based on the first network data, the second network data, and the user service demand data, a preset solution strategy is selected based on the abnormal status information to quickly restore network functions.

[0041] It should be noted that smart terminal devices are network devices that deeply integrate 5G modules, FTTR optical network access modules, and AI processing modules, and are applied in a multi-network environment. Therefore, the above steps S101 to S104 are resource management methods proposed for multi-network environment applications based on smart terminal devices.

[0042] In the above step S101, the initialization is used to enable the 5G module, FTTR module and AI processing module to complete hardware status verification, protocol stack loading and network connection. In a specific application, the above step S101 will activate the 5G module, FTTR module and AI processing module in sequence to complete the hardware status verification. Then the 5G module loads the 3GPP R15 protocol stack, the FTTR module loads the ITU-T G.9807.1 standard protocol, and the AI ​​module loads the pre-trained neural network model LSTM network to complete the protocol stack loading. Finally, the 5G module scans the signal strength of the surrounding base stations and establishes a connection with the external 5G network. The FTTR module detects the OLT registration status and establishes a connection with the FTTR network within the home or enterprise to complete the network connection.

[0043] In the above step S102, the first network data of the 5G module includes but is not limited to channel quality index (CQI), signal strength (RSRP), bit error rate (BER), uplink and downlink bandwidth utilization, latency (RTT), etc.; the second network data of the FTTR module includes but is not limited to optical fiber link optical power (TX / RX optical power), PON port throughput, OLT synchronization status, etc.; user service demand data includes but is not limited to IPTV video stream bit rate, number of IoT device connections, real-time requirements, extraction of QoS requirements, etc.

[0044] In an embodiment of the present invention, the AI ​​processing module optimizes network resource allocation through an AI algorithm, analyzes first network data, second network data, and user service demand data in real time, and realizes dynamic switching of multiple network interfaces, resource collaborative allocation, and intelligent signal optimization.

[0045] Exemplarily, in step S103 above, the AI ​​processing module outputs the optimal transmission path based on the first network data, the second network data, and the user service demand data, including:

[0046] Obtain the business type from the user's business demand data;

[0047] According to the service type, the first network data and the second network data, the optimal transmission path weight based on the 5G module and the FTTR module is calculated through DQN algorithm reinforcement learning, and the optimal transmission path weight is used to indicate the optimal transmission path.

[0048] It should be noted that the first network data is used to provide the characteristics of the 5G module, the second network data is used to provide the characteristics of the FTTR module, and the user service demand data is used to provide global characteristics. The DQN algorithm calculates the Q value vector based on the above characteristics, and then converts it into a specific path weight to evaluate the weights of all candidate paths. Finally, the optimal transmission path is obtained based on the candidate path with the optimal transmission path weight.

[0049] Exemplarily, the calculation formula for the optimal transmission path weight is:

[0050]

[0051] Where α is the bandwidth resource weight coefficient, which represents the current service's dependence on bandwidth resources. The normalized value range is 0 ≤ α ≤ 1. When enhanced mobile broadband (eMBB) services, such as 8K video transmission, are detected, the α value is automatically increased. The LSTM network in the AI ​​processing module predicts bandwidth demand 3 to 6 seconds in the future, enabling proactive adjustments. β is the latency sensitivity weight coefficient, reflecting the service's tolerance for transmission latency. The normalized constraint is α + β = 1. The β value is automatically increased for ultra-reliable low-latency communication (URLLC) services.

[0052] The embodiments of the present invention are illustrated using a home application scenario, such as a smart home 4K video stream multi-path transmission scenario, where a user simultaneously conducts a 4K video conference and downloads an 8K ultra-high-definition video. The service type for 4K video conferencing is URLLC, and the service type for 8K ultra-high-definition video downloads is eMBB. For each service type, the candidate paths include:

[0053] Path 1: 5G wireless direct connection to the core network;

[0054] Path 2: 5G → FTTR fiber → core network;

[0055] Path 3: 5G → FTTR → MEC edge node.

[0056] Through the above steps, for URLLC, the weight of Path 1 > the weight of Path 2 > the weight of Path 3, so Path 1 is the optimal transmission path. For eMBB, the weight of Path 2 > the weight of Path 1 > the weight of Path 3, so Path 2 is the optimal transmission path.

[0057] Exemplarily, in step S103 above, the AI ​​processing module outputs bandwidth allocation information based on the first network data, the second network data, and the user service demand data, including:

[0058] Obtain the business type from the user's business demand data;

[0059] According to the service type, the first network data and the second network data, the bandwidth ratio of each service type in the 5G module and the FTTR module is dynamically adjusted through an improved weighted fair queue algorithm.

[0060] It should be noted that the improved weighted fair queuing algorithm performs hierarchical weight allocation based on the above-mentioned characteristics of the 5G module, the FTTR module, and global characteristics, and uses AI to predict traffic patterns and adjust weights in advance.

[0061] The present invention uses a home application scenario as an example, such as a smart home 4K video streaming multipath transmission scenario, where a user simultaneously conducts a 4K video conference and downloads an 8K UHD video. The 4K video conference service type is URLLC, and the 8K UHD video download service type is eMBB. Through the above steps, video conferencing can be allocated to the latency-sensitive 5G module, while video downloads can be allocated to the high-bandwidth FTTR module. Furthermore, the video download traffic is split according to the bandwidth allocation ratio obtained in the above steps.

[0062] Exemplarily, in step S103 above, the AI ​​processing module outputs time slot scheduling information according to the first network data, the second network data, and the user service demand data, including:

[0063] Obtain the business type from the user's business demand data;

[0064] If the current service type is a preset priority service, the preset priority service is dynamically inserted into the allocated time slot of the 5G module, and resource compensation is performed for the interrupted service.

[0065] It should be noted that in this embodiment of the present invention, different service types have different priorities. The preset priority services are higher priority services, such as URLLC, while the remaining services are lower priority services, such as eMBB. The timeslot scheduling information shown in the above steps is a preemptive scheduling strategy based on the preset priority services.

[0066] In one embodiment, before step S103, the step includes: constructing a feature vector based on the first network data, the second network data, and the user service demand data to provide the above-mentioned 5G module characteristics, FTTR module characteristics, and global characteristics. Exemplarily, the calculation formula of the feature vector is:

[0067] X t =[CQI 5G ,RSRP 5G ,BER 5G ,Power FTTR ,Throughput FTTR,QoS App ]

[0068] Among them, CQI 5G Indicates 5G channel quality indicator, RSRP 5G Indicates 5G signal strength, BER 5G Indicates 5G bit error rate, Power FTTR Indicates FTTR transmit and receive power, Throughout FTTR Indicates FTTR upstream optical port throughput, QoS APP Indicates the quality of service.

[0069] In the embodiment of the present invention, an AI processing module is also used to pre-formulate a solution strategy, that is, a preset solution strategy is used to quickly restore network functions and improve the fault recovery speed.

[0070] The abnormal status information in the above step S104 includes abnormal optical fiber status and abnormal signal transmission.

[0071] Among them, a preset solution strategy is selected according to the abnormal status information of the optical fiber status to quickly restore network functions, including: when the optical fiber breakage or excessive attenuation of the FTTR module is detected, the available bandwidth in the 5G module is evaluated in real time, the preset priority services are switched to the 5G module, and the remaining services are temporarily stored or downgraded.

[0072] In one embodiment, abnormal fiber status can be determined by monitoring with an optical power monitoring unit. When the optical power falls below a threshold of -28dBm, an alarm indicating abnormal fiber status is generated. Alternatively, the FTTR downstream rate can be monitored. When the FTTR downstream rate drops by more than 30%, an alarm indicating abnormal fiber status is generated. Furthermore, it is understood that after the fault is recovered, the system automatically switches back to single-path FTTR transmission.

[0073] Among them, a preset solution strategy is selected according to the abnormal status information of the signal transmission abnormality to quickly restore the network function, including: when it is detected that the receiving signal strength of the 5G module is lower than the preset strength value, it switches to the FTTR module, and starts the beam search algorithm to re-lock the available beam.

[0074] In another embodiment of the present invention, after obtaining the optimal transmission path, bandwidth allocation information, and time slot scheduling information output by the AI ​​processing module based on the first network data, the second network data, and the user service demand data, the method further includes: adjusting the operating modes of the 5G module and the FTTR module based on the traffic load of the 5G module and the FTTR module, and enabling the FTTR dominant mode during a preset period. This allows for dynamic power consumption management to be adjusted based on user network application conditions, thereby achieving energy efficiency optimization.

[0075] For example, according to the traffic load of the 5G module and the FTTR module, the operating mode of the 5G module and the FTTR module is adjusted, including: adjusting the PCIe clock frequency of the 5G module and the WiFi7 module according to the real-time load (such as the number of concurrent connections and CPU utilization). In a specific application, when the traffic load of the 5G module and the FTTR module is idle, if the PCIe clock frequency of the 5G module and the WiFi7 module is reduced to below 200MHz, the power consumption can be reduced by about 20%; at the same time, the DDR operating frequency of the FTTR main chip is adjusted to switch between 1600MHz, 1866MHz, and 2133MHz, or, when the operating efficiency requirement is not high, one core between CPU0 / 1 of the ZX279133 main chip is turned off to further reduce power consumption. This method can reduce power consumption by about 2W again.

[0076] For example, the preset time period is the off-peak period at night. In the above-mentioned FTTR-dominant mode, the transmitter of the XGPON optical module will be turned off, and only the receiving link will be kept on standby, reducing power consumption by about 1W; at the same time, the RF front end of the 5G module will be turned off.

[0077] like Figure 2 As shown, another embodiment of the present invention further provides a resource management system for a multi-network environment, including a controller 10 and an intelligent terminal device 20, wherein the intelligent terminal device 20 includes a 5G module 21, an FTTR module 22 and an AI processing module 23, wherein the controller 10 implements the above-mentioned resource management method for a multi-network environment. In a specific application, the controller will intelligently allocate network resources based on the analysis results of the AI ​​processing module, coordinate the work between the 5G module and the FTTR module, and achieve efficient collaboration between devices. At the same time, the fault self-healing capability introduced by the AI ​​processing module reduces manual intervention, reduces the complexity and cost of network operation and maintenance, enhances the reliability and fault tolerance of the network, and ensures the continuity of network services.

[0078] In the resource management system for a multi-network environment provided by an embodiment of the present invention, the 5G module and the FTTR module have also made improvements in signal enhancement and seamless switching of the uplink network to optimize the link quality. Exemplarily, seamless switching is achieved between the 5G module and the FTTR module through hard switching or soft switching; wherein, hard switching is: a Make-Before-Break strategy is adopted between 5G and FTTR to maintain dual-channel concurrency for 300ms; soft switching is: the same type of network (such as multiple 5G base stations) adopts PDCP layer data replication and forwarding. Exemplarily, the 5G module dynamically adjusts the antenna array phase through Beamforming technology to compensate for multipath fading and improve signal strength and quality; forward error correction (FEC) and dynamic optical power adjustment (DOPA) are enabled on the FTTR side, which can significantly reduce bit errors and improve reception accuracy.

[0079] like Figure 3 As shown, the embodiment of the present invention also provides a better implementation structure of the intelligent terminal device, including:

[0080] The 5G module's hardware utilizes the highly integrated Unisoc V510 chip as its core processor. This chip supports multiple communication modes, including 2G / 3G / 4G / 5G, including the Sub-6 GHz band and mainstream global frequency bands such as n1 / n3 / n8 / n20 / n41 / n77 / n78 / n79. It automatically adapts to 5G NSA and SA dual-mode networking, offering a downlink rate of 2.45 Gbps and an uplink rate of 1.25 Gbps. The 5G module's hardware interface supports a PCI-Express 3.0 controller, enabling high-speed data communication with FTTR modules. The 5G module's antenna array supports four external high-gain omnidirectional antennas. The V510 chip's SPI interface communicates with the TCHV4018L control unit chip for data transmission, while GPIO pins are used to control the module's power on / off and reset.

[0081] The FTTR module's hardware utilizes ZTE Microelectronics' ZX279133 and ZX279201A chipset. The ZX279133 boasts robust XG(S)PON protocol processing capabilities, while the ZX279201A chip is specifically designed to provide downstream OLT PON interfaces for FTTR products. Optimized for the specific characteristics of home FTTR networks, the PON mechanism achieves low-latency transmission, provides real-time channels for management and control, and high-priority services, and reduces latency within the home network. The ZX279133 communicates with the 5G module via a PCIE 3.0 interface and with the Rockchip RK3588 AI processing module via an RGMII interface. The ZX279133 provides wireless network throughput of up to 3.6 Gbps via a PCIE 3.0 X2 external WiFi 7 module. It also offers a phone port, USB 3.0, and GE / 2.5GE Ethernet ports. The ZX279201A chip transmits data with the main chip ZX279133 through the internally integrated UNI Serdes interface, providing users with a GPON OLT optical port that supports 2.5Gbps downstream / 1.25Gbps upstream.

[0082] The hardware component of the AI ​​processing module uses the Rockchip RK3588 chip, which includes a neural network processing unit (NPU) that supports basic AI functions and has a core computing energy efficiency of up to 6TOPS / W, which is used to accelerate artificial intelligence algorithms and machine learning tasks. The RK3588 chip supports audio, video, edge computing, and access to multiple sensors. It uses a multi-core heterogeneous computing architecture, including high-performance CPU, GPU, NPU, and DSP components. It can flexibly allocate computing resources between different tasks, thereby achieving efficient dynamic collaborative control. In addition, the RK3588 chip communicates with the FTTR main chip ZX279133 chip through the internal integrated RGMII interface for data transmission at a rate of up to 1Gbps.

[0083] In summary, the user interface of the intelligent terminal device based on the 5G module, FTTR module, and AI processing module includes GE / 2.5GE Ethernet ports, multiple USB 3.0 ports, voice ports, video input ports, HDMI video output ports, and a wireless WiFi7 port. This provides an interface for users to interact with the converged terminal and supports user configuration and management of the device. Through the user interface, users can configure and manage the converged terminal, such as setting network parameters and viewing device status.

[0084] Based on the implementation structure of the above-mentioned smart terminal device, the example of the present invention also provides a better implementation structure of a resource management system for a multi-network environment, including the above-mentioned smart terminal device, and a controller for coordinating the work between the 5G module, FTTR module and AI processing module.

[0085] The controller hardware utilizes TaiSilicon's TCHV4018L chip, based on the ARM Cortex-M0 core. It supports multiple communication protocols and is suitable for low-cost control units. As a standalone control unit, the TCHV4018L coordinates the operating status of each module, optimizes resource allocation, and ensures efficient and stable data transmission. The TCHV4018L uses a low-speed bus interface and multiple GPIO pins to communicate with the 5G module. This low-speed bus interface transmits data, while the GPIO pins enable module power on / off and reset control. The TCHV4018L uses a low-speed bus and multiple GPIO pins to efficiently exchange data with the FTTR module. This low-speed bus transmits data with the FTTR module, while the GPIO pins enable module control and status monitoring. The TCHV4018L uses a low-speed bus and multiple GPIO pins to exchange data with the AI ​​processing module, using the low-speed bus for data transmission and the GPIO pins for module control and data acquisition.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A resource management method for a multi-network environment, characterized in that: include: After the smart terminal device is started, it is initialized; wherein the smart terminal device includes a 5G module, an FTTR module and an AI processing module; Collecting the first network data of the 5G module, the second network data of the FTTR module, and user service demand data and sending them to the AI ​​processing module; Obtaining the optimal transmission path, bandwidth allocation information, and time slot scheduling information output by the AI ​​processing module based on the first network data, the second network data, and the user service demand data; If the AI ​​processing module outputs abnormal status information based on the first network data, the second network data and the user service demand data, a preset solution strategy is selected according to the abnormal status information to quickly restore network functions.

2. The resource management method for a multi-network environment according to claim 1, wherein: The AI ​​processing module outputs an optimal transmission path according to the first network data, the second network data, and the user service demand data, including: Obtain the business type from the user's business demand data; According to the service type, the first network data and the second network data, the optimal transmission path weight based on the 5G module and the FTTR module is calculated through DQN algorithm reinforcement learning, and the optimal transmission path weight is used to indicate the optimal transmission path.

3. The resource management method for a multi-network environment according to claim 1, wherein: The AI ​​processing module outputs bandwidth allocation information according to the first network data, the second network data, and the user service demand data, including: Obtain the business type from the user's business demand data; According to the service type, the first network data and the second network data, the bandwidth ratio of each service type in the 5G module and the FTTR module is dynamically adjusted through an improved weighted fair queue algorithm.

4. The resource management method for a multi-network environment according to claim 1, wherein: The AI ​​processing module outputs time slot scheduling information according to the first network data, the second network data, and the user service demand data, including: Obtain the business type from the user's business demand data; If the current service type is a preset priority service, the preset priority service is dynamically inserted into the allocated time slot of the 5G module, and resource compensation is performed for the interrupted service.

5. The resource management method for a multi-network environment according to any one of claims 1 to 4, characterized in that: Before the AI ​​processing module outputs the optimal transmission path, bandwidth allocation information, and time slot scheduling information based on the first network data, the second network data, and the user service demand data, the AI ​​processing module includes: A feature vector is constructed according to the first network data, the second network data and the user service demand data.

6. The resource management method for a multi-network environment according to claim 1, wherein: The abnormal state information includes at least optical fiber state abnormality and signal transmission abnormality; Selecting preset resolution strategies based on fiber status abnormality information to quickly restore network functions, including: when a fiber break or excessive attenuation is detected in the FTTR module, real-time evaluation of available bandwidth in the 5G module is performed, switching preset priority services to the 5G module, and temporarily storing or downgrading remaining services; Select a preset solution strategy based on the abnormal status information of signal transmission abnormality to quickly restore network functions, including: when it is detected that the received signal strength of the 5G module is lower than the preset strength value, switch to the FTTR module, and start the beam search algorithm to re-lock the available beam.

7. The resource management method for a multi-network environment according to claim 1, wherein: After obtaining the optimal transmission path, bandwidth allocation information, and time slot scheduling information output by the AI ​​processing module based on the first network data, the second network data, and the user service demand data, the method further includes: According to the traffic load conditions of the 5G module and the FTTR module, the working modes of the 5G module and the FTTR module are adjusted, and the FTTR dominant mode is enabled within a preset period of time.

8. A resource management system for a multi-network environment, characterized in that: It includes a controller and an intelligent terminal device, the intelligent terminal device includes a 5G module, an FTTR module and an AI processing module, and the controller implements the resource management method for a multi-network environment as described in any one of claims 1 to 7.

9. The resource management system for a multi-network environment according to claim 8, wherein: The 5G module and the FTTR module achieve seamless switching through hard switching or soft switching.

10. The resource management system for a multi-network environment according to claim 8 or 9, characterized in that: The 5G module dynamically adjusts the antenna array phase through beamforming; The FTTR module enables forward error correction to automatically detect and correct errors in transmission, and monitors and adjusts the transmitted and received optical powers based on dynamic optical power.

Images