Dynamic resource isolation method and device, electronic equipment and storage medium

Abstract

The invention provides a dynamic resource isolation method and device, electronic equipment and a storage medium, and relates to the technical field of communication. A resource allocation strategy containing a private network slice minimum bandwidth reservation proportion and a public network user maximum bandwidth limit proportion can be generated based on the service type identification result and the service flow prediction data; and base station scheduling parameters and core network flow routing can be adjusted according to the strategy so as to realize public and private network user resource isolation through a dynamic bandwidth allocation mechanism and network slice mapping, and meanwhile, network state changes are evaluated regularly and a resource allocation strategy is adjusted based on an evaluation result. The problem that in the prior art, resource allocation cannot adapt to dynamic service requirements can be solved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of communication technology, in particular to a dynamic resource isolation method and device, electronic equipment and storage medium. BACKGROUND

[0002] Coal mine multi-frequency hybrid networking is the core communication infrastructure of intelligent mine, mainly serving key scenarios such as collaborative control of mining equipment and remote operation. It cooperates with 4G wide coverage and 5G high-speed network to build a converged communication system that takes into account cost and performance, covering the whole process from physical layer multi-band resource allocation to core network strategy control. Key links include network slice division, service priority management, and base station load balancing. Traditional solutions mostly use a combination of VLAN logical isolation and static QoS configuration. However, due to the lack of historical stability coefficient and adaptive parameters, the fixed weight allocation strategy not only lacks the ability to respond to dynamic business needs, but also is prone to problems such as insufficient private network bandwidth during sudden data transmission of mining equipment, and loss of control commands due to private network resources being occupied by public network video streams. In addition, there are coordination defects in the physical layer (multi-band interference), logical layer (slice boundary ambiguity), and strategy layer (manual adjustment delay), which cannot meet the millisecond-level business needs of unmanned driving and other services, thereby affecting coal mine production safety and network resource utilization. SUMMARY

[0003] The present disclosure provides a dynamic resource isolation method and device, electronic equipment and storage medium. Its main purpose is to at least partially solve one of the technical problems in the related art.

[0004] According to a first aspect of the present disclosure, a dynamic resource isolation method is provided, comprising: collecting network state information in real time, including base station load, user quantity and service throughput, and detecting key service trigger signals; Based on the service type identification result and the service traffic prediction data, a resource allocation strategy is generated, which includes the minimum bandwidth reservation ratio of private network slices and the maximum bandwidth limitation ratio of public network users; According to the resource allocation strategy, adjust the base station scheduling parameters and the core network traffic routing, realize the resource isolation of public and private network users through dynamic bandwidth allocation mechanism and network slice mapping; Periodically evaluate the network state changes, and adjust the resource allocation strategy based on the evaluation results.

[0005] According to a second aspect of the present disclosure, a dynamic resource isolation device is provided, comprising: The acquisition unit is configured to collect network state information in real time, including base station load, user quantity and service throughput, and detect key service trigger signals; The generating unit is configured to generate a resource allocation strategy based on the service type identification result and the service traffic prediction data, the resource allocation strategy including a minimum bandwidth reservation ratio of a private network slice and a maximum bandwidth limitation ratio of a public network user. The isolating unit is configured to adjust base station scheduling parameters and core network traffic routing according to the resource allocation strategy, and to realize resource isolation of public and private network users through a dynamic bandwidth allocation mechanism and network slice mapping. The adjusting unit is configured to periodically evaluate network state changes and adjust the resource allocation strategy based on the evaluation result.

[0006] According to a third aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of the first aspect.

[0007] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to perform the method of the first aspect.

[0008] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the method of the first aspect.

[0009] The dynamic resource isolation method and device, electronic device, and storage medium provided by the present disclosure can solve the problems of insufficient private network bandwidth, public network users occupying private network resources, and resource allocation failing to adapt to dynamic service demand in the prior art, which are caused by the lack of real-time network state acquisition and key service trigger signal detection, the failure to generate a targeted resource allocation strategy based on service type identification and traffic prediction, the reliance on a static configuration rather than a dynamic mechanism to realize public and private network resource isolation, and the failure to periodically adjust the resource allocation strategy. The present disclosure can guarantee the bandwidth demand of mine key services, avoid contention for public and private network resources, and improve network resource utilization. Furthermore, the present disclosure can dynamically adapt the resource allocation strategy to meet millisecond-level service response demand.

[0010] It should be appreciated that the content described in this section is not intended to identify key or critical features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0011] The accompanying drawings are included to provide a better understanding of the present application, and are not intended to limit the present disclosure. Among them: Figure 1 A flowchart of a dynamic resource isolation method provided by an embodiment of the present disclosure; Figure 2 A structural diagram of a dynamic resource isolation device provided by an embodiment of the present disclosure; Figure 3 A schematic block diagram of an example electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0012] The exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to assist in understanding, and should be considered as merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, in order to make the description clear and concise, the description of well-known functions and structures is omitted in the following description.

[0013] The dynamic resource isolation method and device, electronic device and storage medium of the embodiments of the present disclosure are described below with reference to the accompanying drawings.

[0014] Figure 1 A flowchart of a dynamic resource isolation method provided by an embodiment of the present disclosure.

[0015] As Figure 1 shown, the method comprises the following steps: Step 101, real-time collection of network state information, including base station load, number of users and traffic throughput, and detection of key service trigger signals.

[0016] In the embodiments of the present disclosure, this step is the basic data acquisition link of the dynamic resource isolation method, aiming to master the network running state and key business demand in real time, and to provide accurate basis for the generation of subsequent resource allocation strategies. Specifically, this step continuously collects core state information reflecting the current network running status through the network monitoring module, and the network state information at least covers the base station load for representing the network resource occupation degree, the user quantity for representing the network service object size, and the service throughput for reflecting the service data transmission efficiency. At the same time, the business sensing module detects the trigger signal indicating the start, switching or demand change of the key business in real time, and the key business trigger signal can be derived from the mining equipment cooperative control, remote operation and other business instructions with high requirements for network delay and bandwidth in the mine scene. Through real-time acquisition of the above information, the limitation of relying on static data configuration in the traditional scheme can be broken, and it is ensured that the subsequent resource allocation strategy matches the actual network state and business demand; as an implementation manner, the continuous acquisition of network state information can be realized through the preset acquisition frequency (such as millisecond level or second level), and the key business trigger signal can be detected through the identification of specific business data packet characteristics.

[0017] Through real-time acquisition of network state information and detection of key business trigger signal, dynamic and accurate basic data support can be provided for subsequent resource allocation, avoiding the problem that resource allocation is out of touch with actual network and business demand due to the reliance on static data, and the demand change of key business can be captured in time, laying a foundation for subsequent priority guarantee of key business resources, and then improving the timeliness and accuracy of resource allocation.

[0018] Step 102, generating a resource allocation strategy based on the business type recognition result and the business traffic prediction data, the resource allocation strategy including a minimum bandwidth reservation ratio of a private network slice and a maximum bandwidth limitation ratio of public network users.

[0019] In the embodiments of the present disclosure, this step is the core step of policy generation of the dynamic resource isolation method, aiming to combine the business attribute difference with the future resource demand prediction to transform into accurate resource allocation rules, and provide clear basis for subsequent network parameter adjustment. Specifically, this step first acquires a business type identification result, which is used to distinguish the resource demand priority of different businesses (such as the difference between key control businesses and ordinary data transmission businesses in a mine scene), and acquires business traffic prediction data, which is used to predict the resource consumption scale of each type of business in a future specific period; based on the comprehensive analysis of these two types of information, a resource allocation strategy is generated to adapt to the current and future network state, and the core content of the strategy includes the minimum bandwidth reservation ratio of the private network slice and the maximum bandwidth limitation ratio of the public network user, wherein the minimum bandwidth reservation ratio of the private network slice is used to guarantee the basic bandwidth of the key business not to be occupied, and the maximum bandwidth limitation ratio of the public network user is used to define the upper limit of the resource use of the ordinary business to avoid occupying the private network resources. As an implementation manner, the business type identification can be completed by analyzing the business data characteristics, the business traffic prediction can be derived based on the historical business law, and then the bandwidth ratio parameters conforming to the actual scene are determined.

[0020] By combining the business type identification result and the business traffic prediction data to generate the strategy, the problem that the static strategy is disconnected with the business demand can be avoided; at the same time, by clearly defining the minimum bandwidth reservation of the private network and the maximum bandwidth limitation of the public network, the resource supply for the key business can be directly guaranteed and the public network business from occupying too much resources can be prevented, and the accuracy and forward-looking of the resource allocation can be effectively improved, which lays a key foundation for subsequent implementation of public and private network resource isolation.

[0021] Step 103, according to the resource allocation strategy, adjusting the base station scheduling parameter and the core network traffic routing, realizing the resource isolation of the public and private network users through the dynamic bandwidth allocation mechanism and the network slice mapping.

[0022] In the embodiments of the present disclosure, this step is the core step of the strategy of the dynamic resource isolation method, and is aimed at converting the generated resource allocation strategy into actual network operation configuration, so as to realize effective resource isolation between public network users and private network users. Specifically, this step is based on the resource allocation strategy (including the minimum bandwidth reservation ratio of the private network slice and the maximum bandwidth limitation ratio of the public network user) to adjust the scheduling parameters of the base station on the one hand, so as to adapt to the bandwidth demand difference of different users, and to optimize the traffic routing of the core network on the other hand, so as to guide the public network and private network business data to be transmitted along the respective exclusive paths; in this process, the dynamic bandwidth allocation mechanism is used to adapt to the fluctuation of the business traffic in real time, so as to ensure that the resources are allocated on demand, and at the same time, the network slice mapping technology is used to map the public network and private network business to independent resource slices respectively, so as to divide the resource boundary from the logical or physical layer, and finally to achieve the isolation of the public network users and the private network users in the use of resources, so as to avoid mutual occupation. As an implementation manner, the base station scheduling parameter adjustment can involve the setting of the bandwidth allocation weight, the core network traffic routing optimization can be directed to the routing priority configuration of a specific business, and the network slice mapping can realize the binding of the slice and the physical frequency band.

[0023] By converting the resource allocation strategy into the specific configuration of the base station and the core network, and combining the dynamic bandwidth allocation and the network slice mapping mechanism, the public network and private network resource isolation can be directly implemented, which can not only avoid the occupation of the key resources of the private network by the public network business, but also dynamically allocate the private network resources on demand, effectively solve the problem of incomplete resource isolation and insufficient flexibility under the traditional static configuration, and guarantee the stable operation of the key business and the efficient use of network resources.

[0024] In step 104, the network state change is periodically evaluated, and the resource allocation strategy is adjusted based on the evaluation result.

[0025] In the embodiments of the present disclosure, this step is the strategy optimization step of the dynamic resource isolation method, and is aimed at ensuring that the resource allocation strategy is always adapted to the actual network operation demand by continuously monitoring the network dynamic change, so as to avoid the low efficiency of resource utilization or the insufficient guarantee of key business caused by the solidification of the strategy. Specifically, this step comprehensively evaluates the current running state change of the network according to the preset time period or evaluation mechanism, and the evaluation focuses on whether the network state is still adapted to the existing resource allocation strategy (such as the fluctuation of the base station load, the deviation between the actual change of the business traffic and the prediction data, etc.); after obtaining the evaluation result, if it is determined that the network state change has affected the adaptability of the existing strategy, the resource allocation strategy is adjusted based on the evaluation result, for example, the minimum bandwidth reservation ratio of the private network slice or the maximum bandwidth limitation ratio of the public network user is optimized, so as to maintain the matching degree of the strategy and the actual demand of the network. As an implementation manner, a fixed evaluation time interval can be set, the network key performance indicators are collected to complete the state evaluation, and the strategy parameters are adjusted according to the deviation degree evaluated.

[0026] By regularly evaluating the network state and dynamically adjusting the resource allocation strategy, the limitations of the traditional static strategy of "once set, long-term unchanged" can be broken, and the dynamic changes of the network state can be effectively responded to. At the same time, the accurate adjustment based on the evaluation results can ensure that the resource allocation strategy always meets the actual demand, further consolidates the resource isolation effect of the public and private network, guarantees the continuous and stable operation of the key business, and improves the long-term utilization efficiency of network resources.

[0027] The dynamic resource isolation method provided by the present disclosure can collect network state information in real time and detect key business trigger signals, and can generate a resource allocation strategy containing a private network slice minimum bandwidth reservation ratio and a public network user maximum bandwidth limitation ratio based on business type identification results and business traffic prediction data. It can also adjust base station scheduling parameters and core network traffic routing based on this strategy to achieve public and private network user resource isolation through dynamic bandwidth allocation mechanism and network slice mapping. At the same time, it regularly evaluates the changes in the network state and adjusts the resource allocation strategy based on the evaluation results. Therefore, it can solve the problems of insufficient private network bandwidth, public network users occupying private network resources, and resource allocation unable to adapt to dynamic business demand in the prior art due to the lack of real-time network state collection and key business trigger signal detection, the failure to generate a targeted resource allocation strategy based on business type identification and traffic prediction, the reliance on static configuration rather than dynamic mechanism to achieve public and private network resource isolation, and the failure to regularly adjust the resource allocation strategy. It achieves the technical effects of guaranteeing the bandwidth demand of key businesses in mines, avoiding resource contention between public and private networks, improving network resource utilization, and dynamically adapting resource allocation strategies to meet millisecond-level business response requirements.

[0028] Further, in a possible embodiment of the present disclosure, the resource allocation strategy is generated based on the business type identification results and the business traffic prediction data, including: identifying the business type of the data packet through deep packet inspection technology, and assigning a priority label to each business; using a machine learning model to analyze historical business data and predict the traffic peak in the future time period; according to the business priority label and the predicted traffic peak, calculate the resource allocation weight of each business, and generate strategy parameters containing private network slice bandwidth reservation and public network user bandwidth limitation.

[0029] Specifically, in the present embodiment, when generating the resource allocation strategy based on the service type identification result and the service traffic prediction data, first, the data packets transmitted in the network are parsed by deep packet inspection technology, which extracts the header field information (such as protocol type, port number) and payload characteristics (such as data transmission frequency, instruction encoding format) of the data packets layer by layer to distinguish the service types corresponding to the data packets, for example, accurately identifying the mining equipment cooperative control service, remote operation instruction service and ordinary public network video stream service in the mine scene, and according to the demand differences of various services for network delay and reliability, assigning corresponding priority labels to them, such as marking the mining equipment control service as the highest priority (such as P1) and marking the ordinary video stream service as the low priority (such as P3). Subsequently, a preset machine learning model (such as a long short-term memory network LSTM or an autoregressive integrated moving average model ARIMA) is called to input historical service data (including service traffic data and service type proportion data under different time periods and different working conditions in the past 1-3 months) into the model for training and iterative optimization, so that the model can learn the time distribution law and fluctuation characteristics of the service traffic, and then output the traffic peak data of various services in a specific time period (such as the next 1 hour or 4 hours). Finally, taking the weight coefficients corresponding to the service priority labels (such as the P1 priority weight is 0.6 and the P3 priority weight is 0.2) and the predicted traffic peak as the core parameters, the resource allocation weight of each service is calculated through a preset weighted calculation model (such as resource allocation weight = priority weight x predicted traffic peak / total predicted traffic), and the specific bandwidth reservation proportion of the private network slice (such as reserving 65% of the total bandwidth for the private network slice corresponding to the P1 service) and the maximum bandwidth limit proportion of the public network user (such as limiting the maximum occupied bandwidth of the public network user corresponding to the P3 service to not more than 20%) are determined based on the weight to form the final resource allocation strategy parameters.

[0030] The deep packet inspection technology realizes accurate identification and priority division of service types, avoids resource mismatch caused by misjudgment of service types; with the help of the machine learning model to analyze historical data, the accuracy of service traffic peak prediction is improved, providing a reliable basis for resource reservation; and the strategy parameters are generated based on the priority and the predicted peak value to calculate the weight, further enhancing the pertinence and rationality of the resource allocation strategy, which can more accurately guarantee the key business resources of the private network and effectively restrict the public network business resource occupation.

[0031] Further, in a possible embodiment of the present disclosure, the calculation of the resource allocation weight of each service includes: calculating the resource allocation weight of each service based on the service priority, real-time resource utilization rate and historical stability factor, and dynamically adjusting the resource allocation weight through a reinforcement learning mechanism to adapt to network state changes.

[0032] Specifically, in the present embodiment, when calculating the resource allocation weight of each service, first, the service priority, real-time resource utilization and historical stability factors are quantitatively processed respectively: for the service priority, the quantitative score is set according to the demand degree of the service for network reliability and delay, for example, the mining equipment control service priority in the mine scene is quantified as 10 points, the remote operation service is 8 points, and the ordinary public network data service is 5 points; for real-time resource utilization, the used proportion of the total bandwidth of the current base station is obtained through the network monitoring module, which is converted into a utilization coefficient between 0 and 1 (for example, if 60% is used, the coefficient is 0.6), and the higher the utilization is, the smaller the correction coefficient of the weight is; for the historical stability factor, the fluctuation amplitude of the traffic of the service in the past 30 days is calculated (for example, the difference between the maximum and minimum daily average traffic is proportional to the average value), and the stability coefficient is set to 0.9 when the fluctuation amplitude is less than 10%, and the stability coefficient is set to 0.5 when the fluctuation amplitude is greater than 30%. Subsequently, the initial weight is calculated using a weighted summation formula (for example, resource allocation weight = service priority quantitative value × 0.5 + (1-real-time resource utilization coefficient) × 0.3 + historical stability coefficient × 0.2). At the same time, a weight adjustment model is constructed by introducing a reinforcement learning mechanism, the current network load and the service delay compliance rate are used as the input state of the model, the fine-tuning amplitude (such as ± 0.05) of the weight coefficient is used as the action, and the condition that “resource utilization rate ≥ 80% and key service delay ≤ 50ms” is used as the reward function target. Through continuous iterative training, the model can dynamically adjust the weight proportion of each factor and the final resource allocation weight according to the network state change, for example, when the base station load increases suddenly, the model automatically increases the proportion of the real-time resource utilization factor and reduces the resource allocation weight of non-key services.

[0033] By calculating the weight through multiple factors (service priority, real-time resource utilization, historical stability), the weight deviation caused by a single factor is avoided, and the resource allocation is more in line with the actual demand of the service; and by introducing the reinforcement learning mechanism, the weight coefficient can be dynamically adapted to the network state, effectively dealing with network load fluctuations, further improving the flexibility and accuracy of resource allocation, and ensuring the stability of resource supply for key services in complex network environment.

[0034] Further, in a possible embodiment of the present disclosure, the adjusting the base station scheduling parameter and the core network traffic routing according to the resource allocation strategy comprises: creating an independent logical network slice for the private network user, and binding the logical network slice with the physical frequency band resource; realizing end-to-end slice data isolation through an encrypted channel, and setting a soft isolation boundary to allow the public network user to borrow the private network resource under a preset condition; using a weighted fair queue algorithm to dynamically allocate bandwidth on the base station side, and combining a traffic shaping mechanism to control the flow of public network users.

[0035] Specifically, in the present embodiment, when adjusting the base station scheduling parameters and core network traffic routing according to the resource allocation strategy, first, a network slice management module of the core network creates an independent logical network slice for the private network user, which contains exclusive computing, storage and bandwidth resource pools. At the same time, through the frequency band mapping configuration of the base station, the logical network slice is bound with the physical frequency band resource (such as the 5G millimeter wave frequency band for key control business in the coal mine scene), ensuring that private network business data is transmitted only in the bound physical frequency band. Then, an end-to-end encryption mechanism is used to build a data transmission channel. Specifically, an IPsec protocol is deployed between the user terminal and the core network gateway to encrypt and encapsulate data packets, and identity authentication nodes are set at the entrance and exit of the slice to allow only data packets carrying private network identifiers to pass through, realizing the isolation and protection of slice data. On this basis, soft isolation boundary rules are set, for example, when the real-time bandwidth utilization of the private network slice is less than a preset threshold (such as 20%), public network users are allowed to temporarily borrow part of the idle bandwidth without affecting private network business, and the upper limit of the borrowed bandwidth does not exceed 10% of the total reserved bandwidth of the private network, and the borrowed resources are automatically recovered when the private network business demand increases. At the same time, the weighted fair queue (WFQ) algorithm is enabled on the base station side for bandwidth scheduling, and corresponding weights are assigned to different priority business queues (such as the private network control business queue weight is 0.7, and the public network ordinary business queue weight is 0.3). The algorithm dynamically allocates time slot resources according to the queue weight, ensuring that high-priority business obtains bandwidth in priority; for public network users, in combination with the token bucket traffic shaping mechanism, by presetting the average rate and peak rate parameters of public network traffic (such as average rate 2Mbps, peak rate 5Mbps), public network data packets exceeding the limit are cached or discarded, realizing precise flow control of public network traffic.

[0036] The binding of logical network slice and physical frequency band strengthens the physical isolation basis of public and private network resources, and the encrypted channel guarantees the security of data transmission; the soft isolation boundary improves the resource utilization rate while ensuring the priority of private network business; the combination of weighted fair queue algorithm and traffic shaping mechanism not only realizes dynamic on-demand allocation of bandwidth, but also effectively restricts public network traffic, avoiding its occupation of private network resources, further improving the reliability and flexibility of resource isolation.

[0037] Further, in a possible embodiment of the present disclosure, the network state is periodically evaluated, and the resource allocation strategy is adjusted based on the evaluation result, including: collecting network performance indicators at a preset time interval, including frequency band interference coefficient and user business load; simulating network state through digital twin platform, preforming resource allocation optimization strategy, and preferentially migrating non-critical business to low priority slice; when detecting that the network state changes exceed the adjustment threshold, recalculating the resource allocation ratio and updating the resource allocation strategy.

[0038] Specifically, in the present embodiment, when periodically evaluating network state changes and adjusting resource allocation strategies, first, the network indicator collection process is started according to a preset time interval (such as every 5 minutes), and through the monitoring module and the core network performance statistics unit built in the base station, the key network performance indicators are obtained, among which the frequency band interference coefficient is obtained by calculating the ratio of the actual signal-to-noise ratio of the current working frequency band to the reference signal-to-noise ratio (such as if the ratio is lower than 0.7, it is determined that there is obvious interference), and the user service load is obtained by calculating the ratio of the actual occupied bandwidth to the allocated bandwidth of each service type (such as if the actual occupied bandwidth of a certain private network service reaches 95% of the allocated value, it is determined that the load is high). Subsequently, the collected real-time indicator data is imported into the digital twin platform, which constructs a 1:1 virtual network model based on the physical topology of the mine network, maps the real-time indicators to the virtual model to simulate the current network running state, and then inputs different resource allocation optimization strategies (such as adjusting the private network slice reservation ratio to 70% and the public network limit ratio to 15%) for pre-play. During the simulation process, non-critical services (such as public network file transfer services) are preferentially migrated from high-priority slices to low-priority slices, and the changes in network latency, bandwidth utilization, and other parameters in the pre-play results are observed to select the optimal pre-play scheme. Finally, set the adjustment threshold of network state changes (such as the frequency band interference coefficient being lower than 0.6 for 3 consecutive collection periods or the user service load being more than 90% for 2 consecutive collection periods), when the actual network state change exceeds the threshold, call the resource allocation weight calculation model, combine the latest network indicator data to recalculate the private network slice minimum bandwidth reservation ratio and the public network user maximum bandwidth limit ratio, generate the updated resource allocation strategy and synchronize it to the base station and the core network for execution.

[0039] By collecting key performance indicators at fixed intervals, it is ensured that network state changes can be captured in a timely manner; the pre-play optimization strategy of the digital twin platform can avoid network fluctuations caused by blind adjustments, and the migration of non-critical services further ensures the stability of critical slices; and the threshold-triggered strategy updating mechanism can quickly respond when the network state changes significantly, ensuring that the resource allocation strategy always accurately matches the actual network demand, improving the scientificity and timeliness of strategy adjustment.

[0040] Further, in a possible embodiment of the present disclosure, it further includes: when detecting a base station or network node failure, automatically switching to a backup network path and performing state recovery based on a strategy log record.

[0041] Specifically, in the present embodiment, the response process to the base station or network node failure is first pre-configured with a failure detection mechanism in the network deployment stage: a heartbeat detection module is set for each base station and core network key node (such as a routing node, a slice management node), which sends a heartbeat data packet to the network monitoring center at a frequency of 1 second / time, and the data packet contains the current running state of the node (such as CPU utilization, link connectivity); when the monitoring center does not receive the heartbeat data packet of a certain node for 3 times in a row, or receives a feedback link interruption, overload, and other failure information in the data packet, it is determined that the node has failed. Then trigger the automatic path switching process: the network monitoring center calls the pre-stored backup path configuration table (which has been generated according to the physical topology when the network is initialized, and at least one redundant backup path is matched for each primary path, such as switching to the adjacent backup base station when the primary base station fails, and switching to the backup node in the same area when the core network node fails), update the traffic routing rules through the core network routing control module, quickly guide the business data (including key business of private network and ordinary business of public network) originally flowing through the failed node to the backup path, and send a path switching notification to the related user terminal, ensuring uninterrupted business transmission. At the same time, start the state recovery process: the network system records resource allocation strategy logs in real time, and the log content includes key configuration information such as the currently effective private network slice bandwidth reservation ratio, public network bandwidth limitation ratio, and base station scheduling parameters, and a strategy snapshot is generated every 5 minutes and stored in the redundant database; when the failed node is repaired or the backup path is enabled, the system reads the last valid strategy log snapshot from the database, restores the base station scheduling rules and core network traffic control strategy according to the configuration parameters in the snapshot, and confirms the network running state after the strategy recovery through business connectivity detection (such as sending test data packets to verify the delay and packet loss rate), ensuring that the resource allocation strategy after recovery is consistent with that before failure.

[0042] The high-frequency heartbeat detection realizes the rapid identification of base station and network node failure, avoids long-time interruption of business caused by hidden failure, the pre-configured backup path and automatic switching mechanism greatly shortens the failure response time, and guarantees the continuous transmission of key business of private network; the state recovery based on the strategy log ensures the consistency of the resource allocation strategy after failure repair, avoids new resource isolation problems caused by strategy disorder, and significantly improves the reliability and anti-failure ability of network operation.

[0043] It should be noted that the embodiments of the present disclosure can include a plurality of steps, which are numbered for the convenience of description, but these numbers are not a limitation on the execution time slot and execution order between the steps; the steps can be implemented in any order, and the embodiments of the present disclosure do not limit this.

[0044] Corresponding to the dynamic resource isolation method described above, the present disclosure also proposes a dynamic resource isolation device. Since the device embodiment of the present disclosure corresponds to the method embodiment described above, for details not disclosed in the device embodiment, please refer to the method embodiment described above, which will not be described in detail in the present disclosure.

[0045] Figure 2 A structural schematic diagram of a dynamic resource isolation device provided by an embodiment of the present disclosure is shown in Figure 2 as shown, comprising: The acquisition unit 21 is configured to acquire network state information in real time, including base station load, user quantity and service throughput, and detect a key service trigger signal; The generation unit 22 is configured to generate a resource allocation strategy based on the service type identification result and the service traffic prediction data, wherein the resource allocation strategy includes a minimum bandwidth reservation ratio of a private network slice and a maximum bandwidth limitation ratio of a public network user; The isolation unit 23 is configured to adjust base station scheduling parameters and core network traffic routing according to the resource allocation strategy, and realize resource isolation of public and private network users through a dynamic bandwidth allocation mechanism and network slice mapping; The adjustment unit 24 is configured to periodically evaluate network state changes and adjust the resource allocation strategy based on the evaluation result.

[0046] The dynamic resource isolation device provided by the present disclosure can generate a resource allocation strategy containing a minimum bandwidth reservation ratio of a private network slice and a maximum bandwidth limitation ratio of a public network user based on the service type identification result and the service traffic prediction data, and can adjust base station scheduling parameters and core network traffic routing according to the strategy to realize resource isolation of public and private network users through a dynamic bandwidth allocation mechanism and network slice mapping, while periodically evaluating network state changes and adjusting the resource allocation strategy based on the evaluation result. Therefore, the problems of insufficient private network bandwidth, public network users occupying private network resources, and resource allocation failing to adapt to dynamic business needs caused by the lack of real-time network state acquisition and key service trigger signal detection, the failure to generate a targeted resource allocation strategy based on service type identification and traffic prediction, the reliance on static configuration rather than dynamic mechanism to realize public and private network resource isolation, and the failure to periodically adjust the resource allocation strategy can be solved, thereby achieving the technical effects of ensuring the bandwidth needs of mine key services, avoiding the competition for public and private network resources, improving network resource utilization, and dynamically adapting the resource allocation strategy to meet the millisecond-level business response needs.

[0047] It should be noted that the foregoing explanation and description of the method embodiment also apply to the device of the present embodiment, which has the same principle and will not be limited in the present embodiment.

[0048] According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0049] Figure 3 A schematic block diagram of an example electronic device 300 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present disclosure described and / or claimed in this document.

[0050] As shown in Figure 3 The electronic device 300 includes a computing unit 301 that can perform various appropriate actions and processes in accordance with a computer program stored in a ROM (Read-Only Memory) 302 or a computer program loaded into a RAM (Random Access Memory) 303 from a storage unit 308. Various programs and data required for the operation of the electronic device 300 can also be stored in the RAM 303. The computing unit 301, the ROM 302, and the RAM 303 are connected to each other through a bus 304. An I / O (Input / Output) interface 305 is also connected to the bus 304.

[0051] Various components in the electronic device 300 are connected to the I / O interface 305, including an input unit 306, such as a keyboard, a mouse, and the like; an output unit 307, such as various types of displays, a speaker, and the like; a storage unit 308, such as a magnetic disk, an optical disk, and the like; and a communication unit 309, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 309 allows the electronic device 300 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0052] The computing unit 301 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 301 include, but are not limited to, a CPU (Central Processing Unit), a GPU (Graphic Processing Units), various special-purpose AI (Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, a DSP (Digital Signal Processor), and any appropriate processor, controller, microcontroller, etc. The computing unit 301 performs various methods and processes described above, such as the dynamic resource isolation method. For example, in some embodiments, the dynamic resource isolation method can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as the storage unit 308. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 300 via the ROM 302 and / or the communication unit 309. When the computer program is loaded onto the RAM 303 and executed by the computing unit 301, one or more steps of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 301 can be configured to perform the aforementioned dynamic resource isolation method by any other appropriate means, such as by means of firmware.

[0053] Various implementations of the systems and techniques described above herein can be realized in digital electronic circuitry, integrated circuitry, a Field Programmable Gate Array (FPGA), an Application-Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a System on Chip (SOC), a Complex Programmable Logic Device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0054] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0055] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable storage medium can include but are not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include one or more lines of electrical wire, portable computer diskette, hard disk, RAM, ROM, EPROM (Electrically Programmable Read-Only-Memory) or flash memory, fiber optics, CD-ROM (Compact Disc Read-Only Memory), optical storage device, magnetic storage device, or any suitable combination of the foregoing.

[0056] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (Cathode-Ray Tube) or LCD (Liquid Crystal Display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0057] The systems and techniques described herein can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described herein), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a LAN (Local Area Network), a WAN (Wide Area Network), the Internet, and a blockchain network.

[0058] The computer system can include clients and servers. The clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server is one of communication and distribution, with the server generally providing communication and distribution services to the clients. The servers can be cloud servers, also known as cloud computing servers or cloud hosts, which are a host product in the cloud computing service system to solve the defects of large management difficulty and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or simply "VPS"). The servers can also be servers of a distributed system, or servers combined with a blockchain.

[0059] It should be noted that artificial intelligence is a discipline that studies enabling computers to simulate some thinking processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.) of people, both hardware and software technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing, etc.; artificial intelligence software technologies mainly include computer vision technology, speech recognition technology, natural language processing technology, and machine learning / deep learning, big data processing technology, knowledge graph technology, etc. several major directions.

[0060] The first, second, and various other numerical designations referred to in the present disclosure are only for the convenience of differentiation in the description and do not limit the scope of the embodiments of the present disclosure, nor represent the order of precedence.

[0061] At least one of the present disclosure can also be described as one or more, multiple can be two, three, four or more, the present disclosure does not make restrictions. In the embodiments of the present disclosure, for a technical feature, the technical features in the technical feature are distinguished by "first", "second", "third", "A", "B", "C" and "D" and the like. The technical features described by "first", "second", "third", "A", "B", "C" and "D" have no order or size order.

[0062] It should be understood that the steps shown above can be reordered, added or deleted using various forms of flow. For example, each step described in the present disclosure can be executed in parallel, sequentially or in different order, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, which is not limited herein.

[0063] The above specific embodiments do not constitute a limitation on the scope of protection of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements within the spirit and principles of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A dynamic resource isolation method, characterized in that, include: Real-time collection of network status information, including base station load, number of users and service throughput, and detection of key service trigger signals; Based on the business type identification results and business traffic prediction data, a resource allocation strategy is generated, which includes the minimum bandwidth reservation ratio for private network slices and the maximum bandwidth limit ratio for public network users. According to the resource allocation strategy, base station scheduling parameters and core network traffic routing are adjusted, and resource isolation between public and private network users is achieved through dynamic bandwidth allocation mechanism and network slice mapping. Periodically assess changes in network status and adjust the resource allocation strategy based on the assessment results.

2. The method according to claim 1, characterized in that, The resource allocation strategy generated based on the business type identification results and business traffic prediction data includes: The service type of data packets is identified through deep packet inspection technology, and priority labels are assigned to each service. Utilize machine learning models to analyze historical business data and predict peak business traffic in future time periods; Based on service priority labels and predicted peak traffic, calculate the resource allocation weight for each service and generate policy parameters that include reserved bandwidth for private network slices and bandwidth limits for public network users.

3. The method according to claim 2, characterized in that, The calculation of resource allocation weights for each service includes: Based on service priority, real-time resource utilization, and historical stability factors, the resource allocation weights for each service are calculated, and the resource allocation weights are dynamically adjusted through a reinforcement learning mechanism to adapt to changes in network state.

4. The method according to claim 1, characterized in that, The step of adjusting base station scheduling parameters and core network traffic routing according to the resource allocation strategy includes: Create independent logical network slices for private network users and bind the logical network slices to physical frequency band resources; End-to-end slice data isolation is achieved through encrypted channels, and soft isolation boundaries are set to allow public network users to borrow private network resources under preset conditions; The weighted fair queue algorithm is used to dynamically allocate bandwidth at the base station side, and a traffic shaping mechanism is combined to limit the traffic of public network users.

5. The method according to claim 1, characterized in that, The periodic assessment of network state changes and the adjustment of the resource allocation strategy based on the assessment results include: Network performance metrics, including frequency band interference coefficients and user service load, are collected at preset time intervals. By simulating network conditions through a digital twin platform, we can preview resource allocation optimization strategies and prioritize migrating non-critical services to low-priority slices. When a network status change is detected that exceeds the adjustment threshold, the resource allocation ratio is recalculated and the resource allocation strategy is updated and adjusted.

6. The method according to claim 1, characterized in that, Also includes: When a base station or network node failure is detected, the system automatically switches to a backup network path and performs state recovery based on policy log records.

7. A user dynamic resource isolation device, characterized in that, include: The data acquisition unit is used to collect network status information in real time, including base station load, number of users and service throughput, and to detect key service trigger signals. The generation unit is used to generate a resource allocation strategy based on the service type identification results and service traffic prediction data. The resource allocation strategy includes the minimum bandwidth reservation ratio of private network slices and the maximum bandwidth limit ratio of public network users. The isolation unit is used to adjust base station scheduling parameters and core network traffic routing according to the resource allocation strategy, and to achieve resource isolation of public and private network users through dynamic bandwidth allocation mechanism and network slice mapping. The adjustment unit is used to periodically assess changes in network status and adjust the resource allocation strategy based on the assessment results.

8. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method according to any one of claims 1-6.

Images