Method and device for multi-network coordination pre-scheduling based on terminal image and service intention

Abstract

This invention provides a multi-network collaborative pre-scheduling method and apparatus based on terminal profiling and service intent, relating to the field of communication technology. The method includes: collecting service data streams; determining the service intent type and the terminal sending the service data stream based on the service data streams; determining whether the terminal is an abnormal terminal based on a terminal profiling database; and when preset event triggering conditions are met, determining the target cell of the terminal by querying a digital twin, and generating a pre-scheduling strategy for the target cell. The preset event triggering conditions include the service intent type being voice service and the terminal being an abnormal terminal; the terminal being an abnormal terminal and located within a cross-vendor boundary area; or a prediction that a high-speed mobile terminal is about to enter a weak coverage area. This invention overcomes the problems of existing technologies, such as lack of terminal hardware compatibility awareness, delayed scheduling response, and insufficient cross-domain collaboration.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a multi-network collaborative pre-scheduling method and apparatus based on terminal profiles and service intentions. Background Technology

[0002] In the context of 5G commercialization and the joint construction and sharing of telecommunications networks, users have stringent requirements for the continuity of services such as voice, data, and high-speed rail. The fragmentation of existing terminal models, the mixed use of equipment from different manufacturers, and the complex coverage of multiple frequency bands have led to prolonged voice fallback time, high dropout rates for some terminals in the boundary areas of different manufacturers, and frequent service interruptions in the high-speed rail scenario.

[0003] Existing scheduling schemes rely on physical layer parameters or QCI, and cannot detect terminal hardware compatibility defects; intent-driven architectures focus on offline simulation, and cannot proactively predict network status in millisecond-level critical scenarios such as voice fallback and high-speed rail movement; moreover, existing technologies handle voice, data, and mobility issues in isolation, making it difficult to achieve cross-domain collaborative protection.

[0004] Therefore, it is urgent to achieve millisecond-level proactive pre-scheduling and cross-domain collaboration in scenarios such as cross-vendor boundaries and high-speed mobile environments to solve the technical challenges of intelligent operation and maintenance and service continuity assurance of 5G networks. Summary of the Invention

[0005] This invention provides a multi-network collaborative pre-scheduling method and device based on terminal profile and service intent, which can significantly improve the handover success rate of abnormal terminals in cross-vendor boundary scenarios, reduce EPSFB voice fallback latency, and achieve continuous private network coverage and closed-loop self-evolution of system knowledge in high-speed rail scenarios.

[0006] Firstly, a multi-network collaborative pre-scheduling method based on terminal profiles and service intent is provided, including: Collect business data streams; Based on the business data stream, determine the business intent type corresponding to the business data stream, as well as the terminal that sent the business data stream; where the business intent type includes voice services. Based on the terminal profile database, determine whether the terminal is an abnormal terminal; among them, the abnormal terminal model is the terminal model whose abnormal call drop rate is greater than the preset call drop rate threshold in the cross-manufacturer boundary area. The cross-manufacturer boundary area is the geographical area where the base station coverage areas provided by different equipment manufacturers overlap. When it is determined that the terminal meets the preset event triggering conditions, the target cell of the terminal is determined by querying the digital twin, and a pre-scheduling strategy for the target cell is generated. The target cell is the cell that provides subsequent service support for the terminal, and the digital twin is a data model used to reflect the real-time status of the network around the terminal. The preset event triggering conditions include that the service intent type is voice service and the terminal is an abnormal terminal.

[0007] In this way, by integrating abnormal terminal identification from the terminal profile database, business intent perception, and real-time status query of the digital twin, when a voice service is identified and the terminal is an abnormal terminal, a pre-scheduling strategy is proactively triggered. This solves the problem that existing technologies cannot detect terminal hardware compatibility defects and significantly improves the business continuity assurance capability of high-risk terminals in voice service scenarios.

[0008] In some possible implementations, the preset event triggering conditions also include the terminal being an abnormal terminal and the terminal currently being located within the cross-vendor boundary area; when the preset event triggering conditions are that the terminal is an abnormal terminal and the terminal is currently located within the cross-vendor boundary area, the pre-scheduling strategy includes extending the terminal's mobility management timer and disabling the cell fast reselection function.

[0009] In this way, to address the risk of handover failures for abnormal terminals in cross-manufacturer boundary areas, the handover tolerance time is increased and ping-pong reselection is avoided by extending the mobility management timer and disabling fast reselection, thus effectively suppressing disconnections caused by terminal hardware compatibility defects.

[0010] In some possible implementations, the preset event triggering conditions also include determining, based on the terminal's real-time location and movement trajectory, that the terminal enters an area where the reference signal reception power is lower than a preset coverage threshold within a preset time window; When the preset event trigger condition is determined based on the terminal's real-time location and movement trajectory, and the terminal enters an area where the reference signal reception power is lower than a preset coverage threshold within a preset time window, the pre-scheduling strategy includes activating the target cell within the preset time window and adjusting the handover trigger threshold of the current serving cell.

[0011] In this way, by activating the target cell in advance and adjusting the handover threshold based on location prediction, proactive pre-scheduling can be achieved in high-speed mobile scenarios, avoiding handover failures or service interruptions caused by delayed measurement and reporting.

[0012] In some possible implementations, when the preset event triggering condition is that the service intent type is voice service and the terminal is an abnormal terminal, the pre-scheduling strategy includes sending inter-system measurement configuration to the terminal and performing a handover from the current serving cell to the target cell when the handover condition is met.

[0013] In this way, for abnormal terminals that do not support VoNR, the EPSFB voice fallback latency is significantly reduced and the call connection rate is improved by using cross-system measurement configuration and handover fallback to replace blind redirection.

[0014] In some possible implementations, the terminal is determined to be an abnormal terminal based on the terminal profile database, including: obtaining the terminal's model information, as well as the total number of handover attempts by the terminal in the cross-manufacturer boundary area and the number of times the connection is dropped due to the wireless link control layer reaching the maximum retransmission count; The abnormal call drop rate of the terminal in the cross-vendor boundary area is calculated. The abnormal call drop rate is the ratio of the number of call drops to the total number of handover attempts. When the abnormal call drop rate is greater than a preset threshold, the terminal is determined to be an abnormal terminal.

[0015] In this way, the risk of abnormal disconnection of terminals is quantified based on historical statistics, so that the judgment of abnormal terminals has an objective and reproducible standard, which facilitates the dynamic updating of the profile database.

[0016] In some possible implementations, the target network coverage cell is activated in advance, and the handover trigger threshold of the current serving cell is adjusted, including: The real-time location information of the terminal is obtained through joint positioning using time lead and angle of arrival. Based on the terminal's real-time location information and historical trajectory, determine the terminal's future location within a preset time window; When the terminal is determined to enter an area where the reference signal received power is lower than a preset coverage threshold based on its future location, a target cell within a preset distance ahead of the area is identified, and an activation command is sent to the target cell; and the handover threshold of the measurement reporting event that triggers intra-frequency handover of the terminal's current serving cell is extended to a preset handover duration.

[0017] In this way, by combining positioning and trajectory prediction, the future location of the terminal can be accurately determined. Before entering a weak coverage area, the target cell can be activated in advance and the handover threshold can be extended, so that the terminal can be prioritized to hand over to the target cell, thus achieving proactive protection of "network following the vehicle".

[0018] In some possible implementations, digital twins are constructed in the following ways: Collect measurement report data from multiple frequency bands, store the signal quality parameters of each frequency point in spatial grid units, and form a multi-frequency band signal quality distribution; Collect interface status data between base stations from different manufacturers, and generate a cross-manufacturer interface status view based on the interface status data; Collect topology data from base stations of different manufacturers and generate network topology relationships based on the topology data; A digital twin is established based on the multi-band signal quality distribution, the status view of interfaces between different manufacturers, and the network topology.

[0019] In this way, the digital twin integrates multi-band signal quality, cross-vendor interface health, and shared topology, providing a real-time, multi-dimensional network status view for pre-scheduling decisions and supporting millisecond-level queries.

[0020] Secondly, a multi-network collaborative pre-scheduling device based on terminal profiles and service intentions is provided, including: The first processing module is used to collect business data streams; The second processing module is used to determine the business intent type corresponding to the business data stream and the terminal that sends the business data stream based on the business data stream; wherein the business intent type includes voice service. The third processing module is used to determine whether a terminal is an abnormal terminal based on the terminal profile database. The abnormal terminal model is the terminal model whose abnormal call drop rate is greater than a preset call drop rate threshold in the cross-manufacturer boundary area. The cross-manufacturer boundary area is a geographical area where the base station coverage areas provided by different equipment manufacturers overlap. The fourth processing module is used to determine the target cell of the terminal by querying the digital twin when the terminal meets the preset event triggering conditions, and generate a pre-scheduling strategy for the target cell. The target cell is the cell that provides subsequent service carrying for the terminal, and the digital twin is a data model used to reflect the real-time status of the network around the terminal. The preset event triggering conditions include that the service intent type is voice service and the terminal is an abnormal terminal.

[0021] In some possible implementations, the fourth processing module is also used for: When the preset event triggering conditions also include the terminal being an abnormal terminal and the terminal currently being located within the cross-vendor boundary area, the pre-scheduling strategy includes extending the terminal's mobility management timer and disabling the cell fast reselection function.

[0022] In some possible implementations, the fourth processing module is also used for: When the preset event triggering conditions also include determining, based on the terminal's real-time location and movement trajectory, that the terminal enters an area where the reference signal reception power is lower than a preset coverage threshold within a preset time window, the pre-scheduling strategy includes activating the target cell within the preset time window and adjusting the handover trigger threshold of the current serving cell.

[0023] In some possible implementations, the second processing module is also used for: When the preset event trigger condition is that the service intent type is voice service and the terminal is an abnormal terminal, the inter-system measurement configuration is sent to the terminal, and the handover from the current serving cell to the target cell is executed when the handover condition is met.

[0024] In some possible implementations, the third processing module is also used for: Obtain the terminal's model information, as well as the total number of handover attempts within the cross-manufacturer boundary area and the number of times the connection drops due to the wireless link control layer reaching the maximum retransmission count; The abnormal call drop rate of the terminal in the cross-vendor boundary area is calculated. The abnormal call drop rate is the ratio of the number of call drops to the total number of handover attempts. When the abnormal call drop rate is greater than a preset threshold, the terminal is determined to be an abnormal terminal.

[0025] In some possible implementations, the fourth processing module is also used to: acquire the real-time location information of the terminal, wherein the real-time location information is obtained by joint positioning using time lead and angle of arrival; Based on the terminal's real-time location information and historical trajectory, determine the terminal's future location within a preset time window; When the terminal is determined to enter an area where the reference signal received power is lower than a preset coverage threshold based on its future location, a target cell within a preset distance ahead of the area is identified, and an activation command is sent to the target cell; and the handover threshold of the measurement reporting event that triggers intra-frequency handover of the terminal's current serving cell is extended to a preset handover duration.

[0026] In some possible implementations, the device also includes a fifth processing module for: acquiring measurement report data of multiple frequency bands, storing signal quality parameters of each frequency point in spatial grid units, and forming a multi-frequency band signal quality distribution; Collect interface status data between base stations from different manufacturers, and generate a cross-manufacturer interface status view based on the interface status data; Collect topology data from base stations of different manufacturers and generate network topology relationships based on the topology data; A digital twin is established based on the multi-band signal quality distribution, the status view of interfaces between different manufacturers, and the network topology.

[0027] Thirdly, an electronic device is provided, comprising: one or more processors; one or more memories; and one or more programs, wherein the one or more programs are stored in the one or more memories, and the one or more programs include instructions that, when executed by the one or more processors, cause the configured device to perform the method as described in the first aspect.

[0028] Fourthly, a computer storage medium is provided, the computer storage medium storing instructions that, when executed by a computer, cause the computer to perform the method described in the first aspect or the second aspect.

[0029] Fifthly, a computer program product is provided, the computer program product storing instructions that, when executed by a computer, cause the computer to perform the method described in the first aspect or the second aspect.

[0030] The beneficial effects of the second to fifth aspects can be referred to the introduction of the beneficial effects of the first aspect above, and will not be repeated here. Attached Figure Description

[0031] Figure 1 This is an overall architecture diagram of a multi-network collaborative pre-scheduling method based on terminal profiling and service intent provided in an embodiment of the present invention; Figure 2This is a schematic diagram of the terminal profile and business intent joint identification and high-priority event triggering process provided in an embodiment of the present invention; Figure 3 This is a flowchart of the EPSFB intelligent fallback process provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of a forward collaborative guarantee mechanism for a high-speed rail scenario provided by an embodiment of the present invention; Figure 5 A flowchart of a multi-network collaborative pre-scheduling method based on terminal profile and service intent provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of a multi-network collaborative pre-scheduling device based on terminal profiling and service intent, which can be used to implement the method of the present invention; Figure 7 This is a schematic diagram of an electronic device provided by the present invention. Detailed Implementation

[0032] The solutions provided by the embodiments of the present invention will now be described with reference to the accompanying drawings. In the embodiments of the present invention, "multiple" refers to two or more objects, and "various kinds" refers to two or more types. Terms such as "first," "second," etc., are only used to distinguish similar objects and are not necessarily used to describe a specific order or number of objects.

[0033] Currently, mobile service robots rely mainly on several existing technologies to perform tasks in complex and dynamic environments, but these methods all have significant limitations.

[0034] With the large-scale commercialization of 5G networks and the collaborative construction and sharing of networks by telecommunications companies, mobile users have placed stringent demands on the continuity and high-quality experience of services such as voice calls, high-speed data, and high-speed rail travel. However, existing network infrastructure suffers from fragmented terminal models, mixed use of equipment from different manufacturers, and complex multi-frequency band coverage, leading to quality bottlenecks such as prolonged voice fallback time, high drop rates on specific terminals, and frequent service interruptions in high-speed rail scenarios. Therefore, network-side resource scheduling and service assurance capabilities have become crucial in determining user experience.

[0035] However, under current technological conditions, intelligent scheduling and service continuity assurance in 5G networks face severe constraints. On the one hand, existing scheduling schemes mainly include uplink pre-scheduling based on Quality of Service Category Identifier (QCI) and discrete control scheduling algorithms. These schemes primarily rely on physical layer parameters such as Channel Quality Indicator (CQI), queue length, or QCI for static or semi-static allocation, lacking a deep understanding of terminal hardware compatibility defects. For example, they cannot identify Radio Link Control (RLC) retransmission issues at cross-vendor boundaries for specific terminal models, leading to high abnormal call drop rates.

[0036] On the other hand, existing intent-driven architectures primarily focus on long-term policy orchestration and offline simulation, resulting in a passive response. In millisecond-critical scenarios such as Evolved Packet System fallback (EPSFB) and high-speed rail mobile applications, they cannot predict network status in advance based on service intent, nor can they proactively reserve resources across domains, leading to delayed voice connection times or service interruptions. Furthermore, existing technologies generally handle voice, data, and mobility issues in isolation, failing to construct a real-time online view that integrates multi-band reference signal received power (RSRP) quality maps, cross-vendor logical interface (Xn) status, and 4G shared network topology. Therefore, in environments with joint network construction and sharing among telecommunications companies and heterogeneous multi-vendor networks, it is difficult to achieve globally optimal service continuity assurance.

[0037] In summary, given the inability to change terminal compatibility defects, delayed scheduling response, and insufficient cross-domain collaboration capabilities, how to deeply integrate terminal hardware profiling and service intent perception to construct a lightweight multi-frequency, multi-network digital twin and achieve millisecond-level proactive pre-scheduling and cross-domain collaboration has become a pressing technical challenge in the field of intelligent operation and maintenance and service continuity assurance of 5G networks.

[0038] To address the aforementioned technical issues, this invention proposes a multi-network collaborative pre-scheduling method based on terminal profiling and service intent. First, by analyzing terminal capability signaling and historical dropout records, a dynamic profiling library is constructed, including terminal models and hardware compatibility defects. Terminals of specific models exhibiting Radio Link Control (RLC) retransmission risks at inter-vendor boundaries are marked as high-risk terminals. Simultaneously, user plane data streams are analyzed in real-time to identify service intent types, including new VoNR voice calls, high-definition video, and general browsing. Second, a lightweight network digital twin is constructed, integrating multi-band RSRP quality maps, inter-vendor logical interface (Xn) status, and 4G shared network topology relationships, enabling millisecond-level online awareness of multi-frequency and multi-network status. When a high-priority event trigger condition is detected, such as a voice call intent, a high-risk terminal entering an inter-vendor boundary area, or a high-speed train about to enter a weak coverage tunnel, the optimal target cell is immediately selected from the digital twin, or the network status is predicted, generating a differentiated pre-scheduling strategy. Specifically, for the EPSFB voice fallback scenario in Evolved Packet System (EPS), target cells are pre-selected based on the 4G frequency quality map, and the handover process is triggered by in-frequency / out-frequency measurement event configuration to replace blind redirection. For high-risk terminals, the mobility management timer is extended and fast reselection is disabled. For high-speed rail scenarios, the foreground private network cell is activated in advance, and the threshold value for the measurement reporting event that triggers in-frequency handover for the current serving cell is increased. Finally, the pre-scheduling strategy is distributed to the base station for execution through a standard interface, realizing proactive resource reservation and cross-domain collaboration based on service intent. The entire method, through the deep integration of terminal profiling, intent awareness, and digital twins, constructs a closed-loop intelligent pre-scheduling system of "perception, decision-making, and execution," significantly improving the service continuity assurance capability and user experience in complex 5G multi-network environments.

[0039] Figure 1 This is a block diagram illustrating the overall architecture of a multi-network collaborative pre-scheduling method based on terminal profiling and service intent, as shown in an embodiment of the present invention. Figure 1 As shown, the system consists of three main parts: the data source layer, the intelligent perception layer, and the decision and execution layer.

[0040] The data source layer is used to collect raw input data. Specifically, the SEQ signaling monitoring platform collects terminal capability indicator (IE) fields and historical disconnection records; the UPF user plane function module provides service flow types and protocol characteristics after DPI deep packet inspection and parsing; and the OMC operation and maintenance center provides real-time KPI indicators, neighbor cell relationships, load, interference, and other network status data for 4G / 5G cells.

[0041] The intelligent perception layer comprises three modules: a terminal profiling engine, a service intent recognizer, and a lightweight network digital twin. The terminal profiling engine, based on data from the SEQ signaling monitoring platform, parses terminal models and correlates them with a historical disconnection database. Terminals of specific models exhibiting a high proportion of RLC retransmission records in inter-vendor boundary areas are marked as high-risk terminals, generating a dynamically updated database of problematic terminal behaviors. The service intent recognizer, based on data output from the UPF user plane function module, identifies application layer protocols and outputs structured service intent tags such as VoNR, high-definition video, and general browsing. The lightweight network digital twin aggregates multi-band RSRP quality maps, Xn interface health status between inter-vendor base stations (gNodeBs), and topology relationships provided by the OMC operation and maintenance center to generate a multi-dimensional network status view.

[0042] The decision-making and execution layer includes an AI-powered collaborative decision-making engine and a strategy generation module. This layer monitors high-priority event triggering conditions in real time. Examples include voice-related business intents, high-risk terminals entering cross-vendor boundary areas, and high-speed trains about to enter weak-coverage tunnels. Based on the event type, it obtains terminal profiles, business intents, and digital twin data from the intelligent perception layer to generate differentiated pre-scheduling strategies. These include handover fallback instructions in EPSFB scenarios, T304 timer extension and fast reselection disabling for high-risk terminals, and pre-activation of the front-end private network cell and A3 threshold offset in high-speed rail scenarios. Finally, the strategies are distributed to the base station gNodeB for execution.

[0043] This architecture is the first to incorporate terminal hardware compatibility defects into the profiling system, and integrates a proactive service intent triggering mechanism with a lightweight multi-frequency multi-network digital twin, achieving millisecond-level proactive pre-scheduling and service continuity assurance in 5G co-construction and sharing and multi-vendor heterogeneous environments.

[0044] The following describes a multi-network collaborative pre-scheduling method based on terminal profiles and service intentions, as illustrated in an embodiment of the present invention. The method specifically includes the following steps: In step S1, a terminal profile database is constructed and high-risk terminals are identified.

[0045] In this embodiment of the invention, the UECapability IE field reported by the terminal is analyzed in real time by accessing the SEQ signaling monitoring platform to extract the terminal model information. Simultaneously, the historical disconnection database is consulted to statistically analyze RLC layer retransmission failure records for this model in cross-vendor boundary areas.

[0046] Specifically, for any terminal model T, let Ntotal be the total number of handover attempts occurring in the cross-vendor boundary area, and Nfail be the number of disconnections due to RLC reaching the maximum retransmission count. Then, the abnormal disconnection rate Rfail = Nfail / Ntotal. When Rfail ≥ 80%, the terminal model is marked as a high-risk terminal and stored in the terminal profile library in the form of a triple, such as terminal model, risk scenario, and abnormal disconnection rate. In addition, the profile library also stores descriptions of known hardware compatibility defects, such as "XXX model mobile phone has a high RLC retransmission failure rate during boundary handover."

[0047] In a preferred embodiment of the present invention, the terminal profile database also supports dynamic updates: when a newly connected terminal generates a new disconnection record in the existing network, the abnormal disconnection rate of the new model terminal is periodically recalculated and the risk label is updated, thereby enabling a rapid response to defects of the new model terminal.

[0048] In step S2, the business intent type is identified in real time.

[0049] In this embodiment of the invention, the UPF user plane function module can also perform DPI (Deep Packet Inspection) on the user plane data stream. The DPI engine parses the application layer protocol header and payload to identify the application layer protocol type of the current service and its corresponding service intent.

[0050] Figure 2 This is a schematic diagram illustrating the joint identification of terminal profile and business intent, and the high-priority event triggering process, as shown in an embodiment of the present invention. Figure 2 As shown, the process consists of three parts: an input module, a processing module, and a fusion module.

[0051] The input module comprises two parallel input sources. The first is the SEQ signaling stream, used for terminal profiling. For example, the specific model can be obtained by parsing the UE Capability IE field reported by the terminal, and then associated with a historical disconnection database. If the RLC retransmission failure rate of a terminal device of that model exceeds 80% in the cross-vendor boundary area, it is marked as a high-risk terminal. The second input source is the UPF user plane stream, used for service intent identification. It can identify the application layer protocol type by performing DPI deep packet inspection on the user plane data. For example, if RTP / RTCP is detected, it is determined to be a VoNR or VoLTE voice service; if HLS or MP4 stream is detected, it is marked as high-definition video; other ordinary data is marked as ordinary browsing or audio speed testing.

[0052] Next, the processing module transforms the obtained terminal profile information and business intent information into structured intermediate results. The terminal profile result includes the terminal's category determination information and corresponding risk level; the category determination information includes high-risk terminals or ordinary terminals. The business intent result includes a business type label generated by the business intent recognizer, which indicates the type of business the terminal is currently performing.

[0053] The fusion module, acting as an information fusion node, combines the terminal risk level with the service type label, and optionally overlays a user value score to generate a three-dimensional vector. This three-dimensional vector includes at least the terminal risk level, the service type label, and the user value score. The fusion module has built-in trigger condition judgment logic: when the service type label is voice service and the terminal is marked as a high-risk terminal, it immediately outputs an indication signal to trigger a high-priority event and sends this signal to the AI ​​collaborative decision engine to initiate the pre-scheduling process.

[0054] Specifically, deep packet inspection is performed on the business data stream to parse the application layer protocol header of the business data stream.

[0055] When the payload type field in the header of the Real-Time Transport Protocol or Real-Time Transport Control Protocol indicates VoIP or Adaptive Multi-Rate Coding, the service intent type is determined to be voice service.

[0056] When the content type field in the protocol header of the Hypertext Transfer Protocol indicates that it is a video streaming media format or contains a streaming media request, the business intent type is determined to be a video service. Otherwise, determine the business intent type as ordinary data business.

[0057] Next, the business intent is output in the form of structured tags, which include a timestamp, terminal identifier, and business intent type field. For example: {"timestamp":"2025-03-20T10:30:00Z","imsi":"46001xxxxxxxxxx","intent":"voice service"}.

[0058] In a preferred embodiment of the present invention, when the service intent cannot be parsed through deep packet inspection, an alternative solution can be adopted: obtain the Single Network Slice Selection Assistance Information (S-NSSAI) identifier of the terminal through the Session Management Function (SMF) or Network Repository Function (NRF) of the core network. If the slice type is enhanced mobile broadband (eMBB) and the 5G QoS Identifier (5QI) value of the QoS Flow is 1, it is inferred to be a voice service; if the 5QI value is 2 or 3, it is inferred to be a video service.

[0059] In step S3, a lightweight multi-frequency multi-network digital twin is constructed.

[0060] In this embodiment of the invention, multi-source data from the OMC operation and maintenance center is aggregated in real time to construct a network digital twin. This twin comprises the following three layers of data views: The first layer is a multi-band RSRP quality map. By aggregating minimized road test MDT data or measurement report MR data from frequency bands such as L900, L1800, and N78, it stores the average reference signal received power (RSRP) and reference signal received quality (RSRQ) for each frequency point in each spatial grid, and marks the measurement timestamp.

[0061] Example data format: {"grid_id":"G12345","freq":"1800MHz","avg_RSRP":-98,"avg_RSRQ":-12,"timestamp":"2025-03-20T10:30:00Z"}.

[0062] The second layer is the health status of the inter-vendor interface. This is achieved by collecting metrics such as the Xn interface connection status, handover success rate, and average handover latency between gNodeBs from different vendors. For each pair of inter-vendor boundary cells, a health score (Health_score) is maintained in real time, with a value range of [0, 100]. This score is calculated based on the ratio of the number of successful handovers to the total number of handovers within the most recent preset time period, weighted by the average signaling interaction latency.

[0063] The third layer is the 4G shared network topology. It stores the list of co-built and shared 4G cells, neighbor cell configurations, frequency priorities, and home operator information, forming a complete network topology map of the target fallback cell.

[0064] In a preferred embodiment of the present invention, the digital twin uses an in-memory database to store the above-mentioned views and consumes the OMC's key performance indicator message queue in real time through a stream processing engine to ensure millisecond-level timeliness of the data. When a query request arrives, the system quickly retrieves the RSRP value of the corresponding grid and the status of surrounding cells in the digital twin based on the terminal's current or predicted location.

[0065] It should be understood that the digital twin constructed in this embodiment of the invention differs from traditional twin systems that rely on physical modeling and offline data-driven approaches. Instead, it employs intelligent mapping technology based on the fusion of multi-source heterogeneous data. Specifically, by collecting and processing massive amounts of network data in real time, including signal quality parameters from measurement reports, key performance indicators output by the operation and maintenance center, and interface status information between base stations from different manufacturers, a virtual mapping space that can reflect the real-time network status surrounding the terminal is dynamically constructed using machine learning algorithms such as regression analysis, cluster analysis, and time series prediction.

[0066] This digital twin possesses the following core artificial intelligence capabilities: First, real-time perception capability, which consumes key performance indicator message queues at millisecond levels through a stream processing engine and automatically identifies signal quality fluctuations and interface health changes using sliding window statistics and anomaly detection algorithms; Second, predictive reasoning capability, which uses time-series prediction algorithms such as Gaussian process regression or Kalman filtering based on historical trajectory data to make short-term predictions of the future location and signal strength of high-speed mobile terminals, providing a basis for proactive pre-scheduling decisions; Third, adaptive update capability, which adopts an incremental learning mechanism, using the feedback from the effect of strategy execution as new samples to dynamically adjust the statistical parameters and model weights within the twin, forming a closed-loop self-evolution.

[0067] In step S4, high-priority events are detected and pre-scheduling is triggered.

[0068] In this embodiment of the invention, the system continuously monitors the following three types of high-priority event triggering conditions. When any one of the conditions is met, the process proceeds to step S5: The first type is triggered by service intent. When the service intent tag is voice service, and the terminal is currently camped on a 5G cell and does not support the new 5G call capabilities, the triggering condition is determined to be met.

[0069] The second type is terminal risk triggering. When the current terminal is identified as a high-risk terminal, and the system determines, through joint positioning using time lead and angle of arrival, that the terminal is currently located within the boundary area of ​​different manufacturers, the triggering condition is met. Here, time lead is used to estimate the distance from the terminal to the base station, and angle of arrival is used to estimate the terminal's azimuth.

[0070] The third category is triggered by high-speed rail scenarios. By sampling and matching the sequence of base stations along the railway line with continuous time lead, if the train's speed is detected to be higher than a preset speed threshold, and based on historical trajectory predictions, it will enter a weak network coverage area or tunnel within a preset time window, then the triggering condition is met. The specific prediction algorithm can employ linear regression or Kalman filtering, taking the location sequence of the most recent second preset time window as input and outputting the predicted location for the next third preset time window.

[0071] In step S5, a differentiated pre-scheduling strategy is generated.

[0072] In this embodiment of the invention, the AI ​​collaborative decision engine reads the corresponding state data from the digital twin based on the event type detected in step S4, and generates a specific pre-scheduling strategy.

[0073] For the Evolved Packet System fallback scenario, the decision engine first accesses the 4G shared network topology in the digital twin to filter out the list of 4G neighboring cells corresponding to the current 5G coverage area; then for each candidate 4G cell, it queries its frequency band, expected signal strength in the RSRP quality map, real-time load, and whether there are alarm messages; finally, it selects the optimal target cell based on the weighted scoring function.

[0074] For example, the scoring function can be expressed as Score = 0.4 × (RSRP normalized value) - 0.3 × (load rate) - 0.2 × (handover delay normalized value) + 0.1 × (priority offset). After selecting the target cell, a pre-scheduling policy is generated: A2 event measurement configuration and A5 event measurement configuration are sent to the terminal. The A5 event is used to report that the target cell's signal quality is better than the serving cell's signal quality and exceeds a preset offset. For example, the A5 event offset is set to +3dB. After the terminal reports a measurement report that meets the A5 event conditions, the handover procedure is executed instead of the redirection procedure.

[0075] In step S6, the pre-scheduling strategy is sent to the base station for execution.

[0076] Specifically, the pre-scheduling strategy is distributed to the corresponding base station gNodeB through a standard interface.

[0077] The policy distribution channel supports the E2 interface of the O-RAN RIC platform, the application programming interface of 3GPP NWDAF, and the vendor's proprietary northbound interface.

[0078] For parameter adjustment strategies, configuration is performed via the NETCONF protocol or the Yang model; for handover strategies, it is implemented by sending RRC reconfiguration messages or handover request messages to the gNodeB. After strategy execution, the execution effect is monitored asynchronously: for EPSFB handover, the total latency from call setup request to handover completion is recorded and compared with historical blind redirection latency; for high-risk terminals, their handover success rate in cross-vendor boundary areas is continuously tracked. When the effect data accumulates to a preset threshold, it is fed back to the terminal profile library and digital twin to optimize risk labels and digital twin parameters, forming a closed-loop self-evolution.

[0079] The above steps describe the implementation process of a complete embodiment of the present invention. All steps are automatically completed by the computer processor, database, and network interface without manual intervention. The present invention, through the deep integration of terminal profiling, service intent awareness, and digital twins, achieves a fundamental shift from "passive response" to "proactive pre-scheduling," significantly improving the service continuity assurance capability in complex 5G multi-network environments.

[0080] In summary, this invention addresses the technical bottleneck of existing technologies that cannot achieve proactive pre-scheduling within milliseconds by constructing a lightweight network digital twin and integrating terminal capability profiling with service intent recognition. Specifically, it achieves the following technical effects: First, it establishes a terminal hardware and software defect database, identifies high-risk terminals including specific models, and dynamically adjusts parameters accordingly to resolve abnormal disconnection issues caused by terminal incompatibility, achieving precise protection based on terminal profiling. Second, before high-risk events such as voice call initiation or high-speed rail movement are triggered, based on the real-time network status of the digital twin, it pre-executes cross-domain resource reservation and handover fallback, minimizing EPSFB latency, ensuring service continuity, and achieving millisecond-level proactive pre-scheduling driven by service intent. Third, without changing terminal hardware or the existing network slicing architecture, it solves the network performance fragmentation problem in multi-vendor heterogeneous environments through terminal, network, and cloud collaborative closed loops, providing ordinary users with a VIP-level communication experience and achieving a globally optimal experience through multi-network collaboration.

[0081] The following is in conjunction with the appendix Figure 3 This invention provides a detailed explanation of the EPSFB intelligent fallback process and high-risk terminal parameter adjustment. Taking a 5G co-construction and sharing network environment as an example, this embodiment illustrates the specific implementation of the multi-network collaborative pre-scheduling method based on terminal profiling and service intent. Specifically, it includes the following steps: In step S301, the 5G user terminal initiates a voice call and receives a call setup request.

[0082] In step S302, it is determined whether the current terminal supports VoNR capability. If it does, a VoNR call is established directly; if it does not, the Evolved Packet System (EPSFB) fallback procedure is triggered.

[0083] In step S303, the terminal profile database is queried to obtain the fallback requirement type and protection priority of the terminal. If the terminal is marked as a high-risk terminal, that is, a terminal model with an abnormal call drop rate greater than a preset threshold in the cross-manufacturer boundary area, a high priority protection level is assigned to prioritize the allocation of high-quality fallback resources.

[0084] In step S304, the network accesses the frequency quality map of the 4G shared network topology in the lightweight network digital twin, and filters 4G target cells that meet predetermined conditions based on the geographical location and coverage of the current 5G serving cells.

[0085] The screening criteria include: a reference signal received power of not less than -105dBm, a cell load of not more than 70%, and no hardware faults or transmission alarms. The best candidate cell among those meeting the criteria is selected as the fallback target cell according to a preset scoring rule.

[0086] In step S305, the measurement configurations for events A2 and A5 are sent to the terminal.

[0087] Among them, the A2 event is used to trigger the terminal to start inter-system measurement; the A5 event is used for the terminal to report a measurement report in which the signal quality of the target cell is better than that of the current serving cell and exceeds the preset offset.

[0088] In step S306, after the terminal measures and reports a measurement report that meets the A5 event conditions, it executes a handover procedure to switch the terminal from the current 5G cell to the optimal 4G target cell selected in step S304, replacing the blind redirection method used in the traditional scheme.

[0089] In step S307, the terminal completes the voice call connection on the target 4G cell and continues the call.

[0090] In this embodiment of the invention, a handover mechanism based on terminal profiling and digital twins is introduced into the EPSFB scenario for the first time. This replaces traditional blind redirection with a network-side-led, measurement-driven approach, fully compatible with the 3GPP R15 standard. Actual testing shows that this process significantly reduces EPSFB latency and substantially improves call completion rate.

[0091] For high-risk terminal trigger scenarios, when a terminal is marked as a high-risk terminal and is currently located in the cross-vendor boundary area, the decision engine generates the following strategy: extend the terminal's T304 timer from the default 100ms to 200ms and disable the cell fast reselection function. This strategy is issued to the base station for execution in the form of parameter configuration.

[0092] For high-speed rail scenarios, the decision engine pre-activates private network cells within a 2-kilometer radius of the target location based on predicted location, and applies a +3dB bias to the A3 event threshold of the currently serving cell, enabling the terminal to prioritize switching to the private network cell. Simultaneously, if a VoNR call is currently in progress, an additional Physical Resource Block (PRB) reservation strategy is generated: 20% of the PRB resources are reserved in the target private network cell, and the terminal's scheduling priority is increased.

[0093] The pre-scheduling strategies in the above three scenarios all obtain real-time network status by querying digital twins and are triggered based on joint identification of terminal profiles and service intentions, realizing the transformation from passive response to proactive pre-scheduling, and significantly improving the service continuity assurance capability in complex 5G multi-network environments.

[0094] Figure 4 This is a schematic diagram illustrating a forward collaborative guarantee mechanism for a high-speed rail scenario, as shown in an embodiment of the present invention. In this embodiment, taking a high-speed rail scenario as an example, the specific application of the multi-network collaborative pre-scheduling method based on terminal profiling and business intent in a high-speed mobile environment is explained. Specifically, the train's operating speed is 350 km / h, and the travel path includes tunnels with weak public network coverage or areas without dedicated network coverage.

[0095] First, the position prediction module uses a combined timing advance and angle of arrival (AER) positioning technique to obtain the train's current position in real time. Specifically, the base station calculates the timing advance by measuring the arrival time of the uplink detection reference signal from the terminal, and then estimates the distance from the terminal to the base station; simultaneously, it determines the terminal's orientation relative to the base station by measuring the AER from the detection reference signal using a multi-antenna array.

[0096] The system continuously records historical train trajectory data, including location sequences and corresponding timestamps, and uses linear regression or Kalman filtering algorithms to predict the train's trajectory within the next 10 seconds. For example, if the predicted trajectory indicates that the train will enter a tunnel with weak public network coverage or an area without dedicated network coverage in 10 seconds, a forward collaborative support process is triggered.

[0097] Next, the decision engine performs pre-scheduling operations. When the predicted path determines that the train is about to leave the private network coverage area, the decision engine immediately executes the following private network cell pre-activation operations and handover threshold dynamic offset operations. Among them, private network cell pre-activation involves identifying private network cells to be activated within 2 kilometers ahead of the train, sending activation commands to these cells in advance, and completing parameter configuration, including establishing neighbor cell relationships, preset handover parameters, and setting frequency priority, to ensure that the target cells are ready when the train arrives, eliminating cell wake-up delay.

[0098] The dynamic offset operation for the handover threshold applies a +3dB positive offset to the A3 event handover threshold of the current serving cell. That is, based on the original A3 event judgment formula (neighbor cell signal quality > serving cell signal quality + offset), the offset value is increased by 3dB. This allows the terminal to trigger measurement reporting to the private network cell before the serving cell signal becomes severely degraded, avoiding handover failures caused by measurement reporting delays or channel abrupt changes.

[0099] Finally, during the decision-making process, the service enhancement module simultaneously monitors whether there are ongoing VoNR voice calls within the train carriages. If a VoNR call is detected, the Physical Resource Block (PRB) resource reservation policy is additionally executed. 20% of the PRB resources are reserved in the target private network cell ahead, and the scheduling priority of the terminal is increased. Specifically, a higher proportional fairness factor is allocated to the terminal through the base station scheduler to ensure that voice service data packets receive priority processing during air interface transmission, reducing packet loss rate and jitter.

[0100] Through the coordinated protection of the above mechanisms, in actual tests, under high-speed mobile conditions of 350km / h, the private network dwell rate can be maintained above 98%, and the average opinion score of voice MOS is stable above 4.0, achieving a proactive protection effect of network following the vehicle. This embodiment of the invention is the first to combine location prediction with network resource pre-activation, replacing the traditional passive handover optimization with an active pre-scheduling method, significantly improving the service continuity protection capability in high-speed mobile scenarios.

[0101] Figure 5 This is a flowchart illustrating a multi-network collaborative pre-scheduling method based on terminal profiling and service intent, as shown in an embodiment of the present invention. Figure 5 As shown, it includes the following steps: 510: Collect business data streams.

[0102] 520: Based on the business data stream, determine the business intent type corresponding to the business data stream, as well as the terminal that sent the business data stream; where the business intent type includes voice services.

[0103] 530: Based on the terminal profile database, determine whether the terminal is an abnormal terminal; among them, the abnormal terminal model is the terminal model whose abnormal call drop rate is greater than the preset call drop rate threshold in the cross-manufacturer boundary area. The cross-manufacturer boundary area is the geographical area where the base station coverage areas provided by different equipment manufacturers overlap.

[0104] 540: When it is determined that the terminal meets the preset event triggering conditions, the target cell of the terminal is determined by querying the digital twin, and a pre-scheduling strategy for the target cell is generated. The target cell is the cell that provides subsequent service carrying for the terminal, and the digital twin is a data model used to reflect the real-time status of the network around the terminal. The preset event triggering conditions include that the service intent type is voice service and the terminal is an abnormal terminal.

[0105] By integrating abnormal terminal identification from the terminal profile database, business intent perception, and real-time status query of the digital twin, when a voice service is identified and the terminal is an abnormal terminal, a pre-scheduling strategy is proactively triggered. This solves the problem that existing technologies cannot detect terminal hardware compatibility defects and significantly improves the business continuity assurance capability of high-risk terminals in voice service scenarios.

[0106] In some embodiments, the preset event triggering condition further includes the terminal being an abnormal terminal and the terminal currently located within the cross-vendor boundary area; when the preset event triggering condition is that the terminal is an abnormal terminal and the terminal is currently located within the cross-vendor boundary area, the pre-scheduling strategy includes extending the terminal's mobility management timer and disabling the cell fast reselection function.

[0107] To address the risk of handover failures in cross-vendor boundary areas for abnormal terminals, the handover tolerance time is increased and ping-pong reselection is avoided by extending the mobility management timer and disabling fast reselection, effectively suppressing disconnections caused by terminal hardware compatibility defects.

[0108] In some embodiments, the preset event triggering conditions further include determining, based on the real-time location and movement trajectory of the terminal, that the terminal enters an area where the reference signal reception power is lower than a preset coverage threshold within a preset time window; When the preset event trigger condition is determined based on the terminal's real-time location and movement trajectory, and the terminal enters an area where the reference signal reception power is lower than a preset coverage threshold within a preset time window, the pre-scheduling strategy includes activating the target cell within the preset time window and adjusting the handover trigger threshold of the current serving cell.

[0109] In this way, by activating the target cell in advance and adjusting the handover threshold based on location prediction, proactive pre-scheduling can be achieved in high-speed mobile scenarios, avoiding handover failures or service interruptions caused by delayed measurement and reporting.

[0110] In some embodiments, when the preset event triggering condition is that the service intent type is voice service and the terminal is an abnormal terminal, the pre-scheduling strategy includes sending inter-system measurement configuration to the terminal and performing a handover from the current serving cell to the target cell when the handover condition is met.

[0111] For abnormal terminals that do not support VoNR, the EPSFB voice fallback latency is significantly reduced and the call connection rate is improved by using cross-system measurement configuration and handover fallback to replace blind redirection.

[0112] In some embodiments, determining whether a terminal is an abnormal terminal based on a terminal profile database includes: obtaining the terminal's model information, the total number of handover attempts by the terminal in the cross-manufacturer boundary area, and the number of times the terminal dropped the connection due to the wireless link control layer reaching the maximum retransmission count. The abnormal call drop rate of the terminal in the cross-vendor boundary area is calculated. The abnormal call drop rate is the ratio of the number of call drops to the total number of handover attempts. When the abnormal call drop rate is greater than a preset threshold, the terminal is determined to be an abnormal terminal.

[0113] By quantifying the risk of abnormal disconnection of terminals based on historical statistics, the determination of abnormal terminals has an objective and reproducible standard, which facilitates the dynamic updating of the profile database.

[0114] In some embodiments, the target network coverage cell is activated in advance, and the handover trigger threshold of the currently serving cell is adjusted, including: The real-time location information of the terminal is obtained through joint positioning using time lead and angle of arrival. Based on the terminal's real-time location information and historical trajectory, determine the terminal's future location within a preset time window; When the terminal is determined to enter an area where the reference signal received power is lower than a preset coverage threshold based on its future location, a target cell within a preset distance ahead of the area is identified, and an activation command is sent to the target cell; and the handover threshold of the measurement reporting event that triggers intra-frequency handover of the terminal's current serving cell is extended to a preset handover duration.

[0115] By combining positioning and trajectory prediction, the future location of the terminal can be accurately determined. Before entering a weak coverage area, the target cell can be activated in advance and the handover threshold can be extended, so that the terminal can be prioritized to hand over to the target cell, thus achieving proactive protection of "network following the vehicle".

[0116] In some embodiments, a digital twin is constructed in the following manner: Collect measurement report data from multiple frequency bands, store the signal quality parameters of each frequency point in spatial grid units, and form a multi-frequency band signal quality distribution; Collect interface status data between base stations from different manufacturers, and generate a cross-manufacturer interface status view based on the interface status data; Collect topology data from base stations of different manufacturers and generate network topology relationships based on the topology data; A digital twin is established based on the multi-band signal quality distribution, the status view of interfaces between different manufacturers, and the network topology.

[0117] The digital twin integrates multi-band signal quality, cross-vendor interface health, and shared topology, providing a real-time, multi-dimensional network status view for pre-scheduling decisions and supporting millisecond-level queries.

[0118] Figure 6 This is a schematic diagram of a multi-network collaborative pre-scheduling device based on terminal profiles and service intentions, as shown in an embodiment of the present invention. Figure 6 As shown, it includes: The first processing module is used to collect business data streams; The second processing module is used to determine the business intent type corresponding to the business data stream and the terminal that sends the business data stream based on the business data stream; wherein the business intent type includes voice service. The third processing module is used to determine whether a terminal is an abnormal terminal based on the terminal profile database. The abnormal terminal model is the terminal model whose abnormal call drop rate is greater than a preset call drop rate threshold in the cross-manufacturer boundary area. The cross-manufacturer boundary area is a geographical area where the base station coverage areas provided by different equipment manufacturers overlap. The fourth processing module is used to determine the target cell of the terminal by querying the digital twin when the terminal meets the preset event triggering conditions, and generate a pre-scheduling strategy for the target cell. The target cell is the cell that provides subsequent service carrying for the terminal, and the digital twin is a data model used to reflect the real-time status of the network around the terminal. The preset event triggering conditions include that the service intent type is voice service and the terminal is an abnormal terminal.

[0119] In some embodiments, the fourth processing module is further configured to: When the preset event triggering conditions also include the terminal being an abnormal terminal and the terminal currently being located within the cross-vendor boundary area, the pre-scheduling strategy includes extending the terminal's mobility management timer and disabling the cell fast reselection function.

[0120] In some embodiments, the fourth processing module is further configured to: When the preset event triggering conditions also include determining, based on the terminal's real-time location and movement trajectory, that the terminal enters an area where the reference signal reception power is lower than a preset coverage threshold within a preset time window, the pre-scheduling strategy includes activating the target cell within the preset time window and adjusting the handover trigger threshold of the current serving cell.

[0121] In some embodiments, the second processing module is further configured to: When the preset event trigger condition is that the service intent type is voice service and the terminal is an abnormal terminal, the inter-system measurement configuration is sent to the terminal, and the handover from the current serving cell to the target cell is executed when the handover condition is met.

[0122] In some embodiments, the third processing module is further configured to: Obtain the terminal's model information, as well as the total number of handover attempts within the cross-manufacturer boundary area and the number of times the connection drops due to the wireless link control layer reaching the maximum retransmission count; The abnormal call drop rate of the terminal in the cross-vendor boundary area is calculated. The abnormal call drop rate is the ratio of the number of call drops to the total number of handover attempts. When the abnormal call drop rate is greater than a preset threshold, the terminal is determined to be an abnormal terminal.

[0123] In some embodiments, the fourth processing module is further configured to: The real-time location information of the terminal is obtained through joint positioning using time lead and angle of arrival. Based on the terminal's real-time location information and historical trajectory, determine the terminal's future location within a preset time window; When the terminal is determined to enter an area where the reference signal received power is lower than a preset coverage threshold based on its future location, a target cell within a preset distance ahead of the area is identified, and an activation command is sent to the target cell; and the handover threshold of the measurement reporting event that triggers intra-frequency handover of the terminal's current serving cell is extended to a preset handover duration.

[0124] In some embodiments, the apparatus further includes a fifth processing module, configured to: Collect measurement report data from multiple frequency bands, store the signal quality parameters of each frequency point in spatial grid units, and form a multi-frequency band signal quality distribution; Collect interface status data between base stations from different manufacturers, and generate a cross-manufacturer interface status view based on the interface status data; Collect topology data from base stations of different manufacturers and generate network topology relationships based on the topology data; A digital twin is established based on the multi-band signal quality distribution, the status view of interfaces between different manufacturers, and the network topology.

[0125] Those skilled in the art will readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, the present invention can be implemented in hardware or a combination of hardware and computer software. Whether a function is implemented in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the present invention.

[0126] It should be noted that, Figure 6 The division of modules / units is illustrative and represents only one logical functional division; in actual implementation, other division methods are possible. For example, two or more functions can be integrated into a single data acquisition module. The integrated modules described above can be implemented either in hardware or as software functional modules.

[0127] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-mentioned multi-network collaborative pre-scheduling methods based on terminal profiles and service intentions. That is, an electronic device according to an embodiment of the present invention may include, but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the multi-network collaborative pre-scheduling method based on terminal profiles and service intentions shown in any embodiment of the present invention by calling the computer program.

[0128] In one alternative embodiment, an electronic device is provided, such as Figure 7 As shown, Figure 7 The illustrated electronic device 700 includes a processor 701 and a memory 703. The processor 701 and the memory 703 are connected, for example, via a bus 702. Optionally, the electronic device 700 may further include a transceiver 704, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 704 is not limited to one type, and the structure of the electronic device 700 does not constitute a limitation on the embodiments of the present invention.

[0129] Among them, electronic devices can also be terminal devices, which can be any device that can install applications, including at least one of smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0130] It should be noted that, Figure 7 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0131] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-mentioned multi-network collaborative pre-scheduling methods based on terminal profiles and service intentions.

[0132] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0133] It should be noted that the terms "first," "second," etc., used in the specification and claims of this invention are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of the invention described herein can be implemented in an order other than that shown or described.

[0134] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this invention can be specifically implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, this invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0135] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A multi-network collaborative pre-scheduling method based on terminal profiles and service intents, characterized in that, include: Collect business data streams; Based on the business data stream, determine the business intent type corresponding to the business data stream, and the terminal that sent the business data stream; The service intent types mentioned above include voice services; Based on the terminal profile database, determine whether the terminal is an abnormal terminal; wherein, the abnormal terminal model is a terminal model whose abnormal call drop rate is greater than a preset call drop rate threshold in the cross-manufacturer boundary area, and the cross-manufacturer boundary area is a geographical area where the base station coverage areas provided by different equipment manufacturers overlap. When it is determined that the terminal meets the preset event triggering conditions, the target cell of the terminal is determined by querying the digital twin, and a pre-scheduling strategy for the target cell is generated. The target cell is the cell that provides subsequent service carrying for the terminal. The digital twin is a data model used to reflect the real-time status of the network around the terminal. The preset event triggering conditions include the service intent type being voice service and the terminal being an abnormal terminal.

2. The method according to claim 1, characterized in that, The preset event triggering conditions also include that the terminal is an abnormal terminal and the terminal is currently located within the boundary area of ​​different manufacturers; The method further includes: When the preset event triggering condition is that the terminal is an abnormal terminal and the terminal is currently located in the cross-vendor boundary area, the pre-scheduling strategy includes extending the terminal's mobility management timer and disabling the cell fast reselection function.

3. The method according to claim 1, characterized in that, The preset event triggering conditions also include determining, based on the real-time location and movement trajectory of the terminal, that the terminal enters an area where the reference signal reception power is lower than a preset coverage threshold within a preset time window; The method further includes: When the preset event triggering condition is determined based on the real-time location and movement trajectory of the terminal, that the terminal enters an area where the reference signal reception power is lower than a preset coverage threshold within a preset time window, the pre-scheduling strategy includes activating the target cell within the preset time window and adjusting the handover trigger threshold of the current serving cell.

4. The method according to claim 1, characterized in that, When the preset event triggering condition is that the service intent type is voice service and the terminal is an abnormal terminal, the pre-scheduling strategy includes: The system sends inter-system measurement configuration to the terminal and performs a handover from the current serving cell to the target cell when the handover conditions are met.

5. The method according to any one of claims 1-3, characterized in that, The step of determining whether the terminal is an abnormal terminal based on the terminal profile database includes: Obtain the model information of the terminal, as well as the total number of handover attempts of the terminal in the cross-manufacturer boundary area and the number of times the connection is dropped due to the wireless link control layer reaching the maximum retransmission count; Calculate the abnormal disconnection rate of the terminal in the cross-manufacturer boundary area, wherein the abnormal disconnection rate is the ratio of the number of disconnections to the total number of handover attempts; When the abnormal disconnection rate exceeds a preset threshold, the terminal is determined to be an abnormal terminal.

6. The method according to claim 3, characterized in that, The step of activating the target network coverage cell in advance and adjusting the handover trigger threshold of the currently serving cell includes: The real-time location information of the terminal is obtained, wherein the real-time location information is obtained by joint positioning using time advance and angle of arrival; Based on the real-time location information and historical trajectory of the terminal, the future location of the terminal within a preset time window is determined; When the terminal is determined to enter an area where the reference signal reception power is lower than a preset coverage threshold based on the future location, a target cell within a preset distance in front of the area is determined, and an activation command is sent to the target cell. In addition, the handover threshold value of the measurement reporting event that triggers intra-frequency handover of the current serving cell of the terminal is extended to a preset handover value.

7. The method according to any one of claims 1-3, characterized in that, The digital twin is constructed in the following manner: Collect measurement report data from multiple frequency bands, store the signal quality parameters of each frequency point in spatial grid units, and form a multi-frequency band signal quality distribution; Collect interface status data between base stations from different manufacturers, and generate a cross-manufacturer interface status view based on the interface status data; Collect topology data from base stations of different manufacturers, and generate network topology relationships based on the topology data; The digital twin is established based on the multi-band signal quality distribution, the cross-vendor interface status view, and the network topology.

8. A multi-network collaborative pre-scheduling device based on terminal profiles and service intent, characterized in that, include: The first processing module is used to collect business data streams; The second processing module is used to determine the service intent type corresponding to the service data stream and the terminal that sends the service data stream based on the service data stream; wherein the service intent type includes voice service. The third processing module is used to determine whether the terminal is an abnormal terminal based on the terminal profile database. The abnormal terminal model is a terminal model whose abnormal call drop rate is greater than a preset call drop rate threshold in the cross-manufacturer boundary area. The cross-manufacturer boundary area is a geographical area where the base station coverage areas provided by different equipment manufacturers overlap. The fourth processing module is used to determine the target cell of the terminal by querying the digital twin when it is determined that the terminal meets the preset event triggering conditions, and generate a pre-scheduling strategy for the target cell. The target cell is the cell that provides subsequent service carrying for the terminal, the digital twin is a data model used to reflect the real-time status of the network around the terminal, and the preset event triggering conditions include the service intent type being voice service and the terminal being an abnormal terminal.

9. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the multi-network collaborative pre-scheduling method based on terminal profile and service intent as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to enable the computer to implement the multi-network collaborative pre-scheduling method based on terminal profile and service intent as described in any one of claims 1 to 7.

Images