Network switching method and device, electronic equipment and readable storage medium

Abstract

The embodiment of the invention provides a network switching method and device, electronic equipment and a readable storage medium, the method is applied to a data transmission unit, and the method comprises the following steps: acquiring multi-source data corresponding to a current network of the data transmission unit; determining network state evolution in a future time window of the current network according to the multi-source data; when the network state evolution is lower than a preset network state evolution threshold value, if a switchable network exists in the data transmission unit, determining a target network from the switchable network; and switching from the current network to the target network, so that the data transmission unit performs data transmission through the target network. According to the embodiment of the invention, data transmission can be prevented from being continued through the current network with a relatively poor network state in the future, so that the problems of data transmission jitter, switching interruption and the like are avoided, and the high-reliability transmission requirement of data can be met.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of Internet, in particular to a network switching method, a network switching device, an electronic device and a computer readable storage medium. BACKGROUND

[0002] In an industrial Internet of Things scenario, a DTU (Data Transfer Unit) device is often deployed in an area with complex electromagnetic environment and dynamic network topology.

[0003] In related technologies, a network switching scheme of a DTU device relies on fixed operator binding or a single decision model based on a threshold of signal strength, which is difficult to meet the requirement of high-reliability data transmission. SUMMARY

[0004] In view of the above problems, a network switching method, device, electronic device and readable storage medium are provided to overcome the above problems or at least partially solve the above problems. The specific technical solutions are as follows:

[0005] Embodiments of the present application disclose a network switching method applied to a data transfer unit, the method comprising:

[0006] obtaining multi-source data corresponding to a current network of the data transfer unit;

[0007] determining network state evolution in a future time window of the current network according to the multi-source data;

[0008] if the data transfer unit has switchable networks when the network state evolution is lower than a preset network state evolution threshold, determining a target network from the switchable networks;

[0009] switching from the current network to the target network, so that the data transfer unit transmits data through the target network.

[0010] In an embodiment of the present application, the determining of the network state evolution in the future time window of the current network according to the multi-source data comprises:

[0011] generating a smooth quality curve according to the multi-source data;

[0012] extracting network quality trend features of the current network from the smooth quality curve;

[0013] inputting the network quality trend features into a preset prediction model to obtain network state evolution in the future time window of the current network output by the prediction model.

[0014] In an embodiment of the present application, the determining the target network from the switchable networks comprises:

[0015] Obtaining multi-source data corresponding to the switchable networks;

[0016] Determining the target network from the switchable networks according to the multi-source data corresponding to the switchable networks.

[0017] In an embodiment of the present application, the determining the target network from the switchable networks according to the multi-source data corresponding to the switchable networks comprises:

[0018] Discretizing the multi-source data corresponding to the switchable networks into data tuples according to preset service requirements;

[0019] Inputting the data tuples into a preset deep model to obtain predicted service performance data of the switchable networks output by the deep model;

[0020] Determining the target network from the switchable networks through a preset greedy strategy and the predicted service performance data of the switchable networks.

[0021] In an embodiment of the present application, the deep model is provided with a composite reward function, and the method further comprises:

[0022] Obtaining multi-source data corresponding to the target network and multi-source data corresponding to the current network;

[0023] Determining network performance change data according to the multi-source data corresponding to the target network and the multi-source data corresponding to the current network;

[0024] Inputting the network performance change data into a preset composite reward function to obtain a reward parameter, so as to update the deep model according to the reward parameter.

[0025] In an embodiment of the present application, the data transmission unit is a dual-baseband processor architecture, the dual-baseband processor architecture comprises a main processor and an auxiliary processor, and the switching from the current network to the target network comprises:

[0026] When the main processor performs data transmission with the target base station according to the current network, the auxiliary processor obtains context information of the target network and sends the context information to the target base station, so as to synchronize communication resources with the target base station; wherein the communication resources at least include a sequence number state, a superframe number and / or a robust header compression context;

[0027] After synchronizing the communication resources with the target base station, the data transmission with the target base station is switched to be performed through the target network.

[0028] In an embodiment of the present application, after the data transmission unit switches from the current network to the target network to make the data transmission unit transmit data through the target network, the method further comprises:

[0029] acquiring actual service performance data of the target network;

[0030] acquiring predicted service performance data output by the deep model;

[0031] determining a model error according to the actual service performance data and the predicted service performance data, and updating the deep model according to the model error.

[0032] In an embodiment of the present application, the multi-source data at least includes physical layer signal quality, transmission layer round trip time, application layer packet loss rate and / or network load state; the service performance data and the predicted service performance data at least include throughput and / or network delay.

[0033] In an embodiment of the present application, after the data transmission unit switches from the current network to the target network, the network address of the data transmission unit remains unchanged.

[0034] Embodiments of the present application also disclose a network switching device applied to a data transmission unit, the device comprising:

[0035] a multi-source data acquisition module, configured to acquire multi-source data corresponding to a current network of the data transmission unit;

[0036] a network state evolution determination module, configured to determine network state evolution in a future time window of the current network according to the multi-source data;

[0037] a target network determination module, configured to, when the network state evolution is lower than a preset network state evolution threshold, determine a target network from switchable networks if the data transmission unit exists in the switchable networks;

[0038] a target network switching module, configured to switch from the current network to the target network to make the data transmission unit transmit data through the target network.

[0039] Embodiments of the present application also disclose an electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other through the communication bus;

[0040] the memory is used to store a computer program;

[0041] the processor is used to execute the program stored on the memory to implement the method as described in embodiments of the present application.

[0042] The embodiment of the application further discloses a computer program product stored in a storage medium, and the computer program product is executed by at least one processor to implement the method according to the embodiment of the application.

[0043] The embodiment of the application further discloses a computer readable storage medium, and instructions are stored on the computer readable storage medium, and when the instructions are executed by one or more processors, the processors execute the method according to the embodiment of the application.

[0044] The embodiment of the application has the following advantages:

[0045] In the embodiment of the application, the multi-source data corresponding to the current network of the data transmission unit is acquired, and then the network state evolution in the future time window of the current network can be determined according to the multi-source data. When the network state evolution is lower than a preset network state evolution threshold, if the data transmission unit has a switchable network, a target network is determined from the switchable network and switched, so that the data transmission unit can perform data transmission through the target network, thereby avoiding continuing data transmission through the current network with poor future network state, and avoiding problems such as data transmission jitter and switching interruption, thereby meeting the high reliability transmission requirement of data. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 is a step flow chart of a network switching method provided in the embodiment of the application;

[0047] Figure 2 is a system structure diagram of network switching provided in the embodiment of the application;

[0048] Figure 3 is a structure block diagram of a network switching device provided in the embodiment of the application;

[0049] Figure 4 is a hardware structure schematic diagram of an electronic device for implementing various embodiments of the application. DETAILED DESCRIPTION

[0050] In order to make the above-mentioned purposes, features and advantages of the application more obvious and easy to understand, the application will be further described in detail below with reference to the drawings and specific embodiments.

[0051] Referring to Figure 1 , a step flow chart of a network switching method provided in the embodiment of the application is shown, which is applied to a data transmission unit, and the method can specifically include the following steps:

[0052] Step 101, multi-source data corresponding to a current network of the data transmission unit is acquired.

[0053] The data transmission unit can be a DTU device, which is an industrial Internet of Things communication device for remote data acquisition and transmission, responsible for transmitting data of terminal devices such as sensors through a network such as a wireless network and / or a wired network to a cloud server or other devices, and the DTU device can support at least one operator network (network). Specifically, a multi-standard radio frequency front end, such as 4G / 5G / NB-IoT, can be integrated on the data transmission unit, so that concurrent measurement of multiple operator networks can be supported.

[0054] In an embodiment of the present application, a cross-protocol layer probe is deployed on the data transmission unit, and based on the cross-protocol layer probe, multi-source data of the current network of the data transmission unit can be collected in real time. In an embodiment of the present application, the multi-source data can at least include physical layer signal quality, transmission layer round-trip time (Round-Trip Time, round-trip delay), application layer packet loss rate, and / or network load state (network layer (base station load) indicators) and the like. The physical layer signal quality can include RSRP (Reference Signal Received Power, Reference Signal Received Power) and SINR (Signal to Interference plus Noise Ratio, Signal to Interference plus Noise Ratio). Since the multi-source data involves data of multiple layers, it can also be referred to as cross-layer indicator data. After collecting the multi-source data, due to hardware delay, network transmission difference, or collection frequency, etc., the time stamp is deviated, at this time, the time stamp alignment technology can be used to eliminate the time sequence deviation of the multi-source data collection, and provide a reliable basis for subsequent analysis, decision or application.

[0055] In some embodiments, the data transmission unit can manage the data stream transmitted by it through a priority queue, so that the data transmission unit can transmit data in order according to the priority of the data, thereby ensuring the real-time performance of high-priority service data. For example, high-priority data can be device control instructions, and low-priority data can be environmental monitoring data. Based on the priority management of data transmission, the energy efficiency of the data transmission unit and the network resource utilization rate can be optimized.

[0056] Step 102, determining the network state evolution in a future time window of the current network according to the multi-source data.

[0057] In an embodiment of the present application, according to the multi-source data of the current network, the network state evolution in the future time window of the current network can be determined. For example, the network state evolution of the current network can be evolved in the direction of deterioration, or in the direction of optimization, of course, it can also be without any evolution.

[0058] Step 103, when the network state evolution is lower than the preset network state evolution threshold, if the data transmission unit exists a switchable network, a target network is determined from the switchable network.

[0059] Step 104, switching from the current network to the target network, so that the data transmission unit transmits data through the target network.

[0060] In the embodiment of the application, the preset network state evolution threshold can be set in advance. When the network state evolution is lower than the preset network state evolution threshold, it indicates that the network state evolution of the current network is evolving in the direction of deterioration. Although the current network can still normally transmit data at the moment, in order to avoid subsequent continuous data transmission through the current network, which may exist abnormality such as delay, data error, etc., when the data transmission unit exists a switchable network (candidate network), a target network is determined from the switchable network, and the current network is switched to the target network, so that the data transmission unit can continue to transmit data through the target network.

[0061] In a specific implementation, the data transmission unit can be set in a complex environment, such as environmental monitoring, energy management, oil pipeline, mine monitoring, etc. In the complex environment, the network fluctuation is very large. Therefore, if the network state evolution of the current network in the future time window can be predicted in time, the network can be switched in time to avoid long interruption of data transmission service of the data transmission unit when the current network appears abnormality, thereby improving the network selection robustness and service continuity of the data transmission unit in the complex environment.

[0062] In the embodiment of the application, the multi-source data corresponding to the current network of the data transmission unit is obtained, and then the network state evolution of the current network in the future time window can be determined according to the multi-source data. When the network state evolution is lower than the preset network state evolution threshold, if the data transmission unit exists a switchable network, a target network is determined from the switchable network and switched, so that the data transmission unit can transmit data through the target network. Therefore, the data transmission through the current network with poor future network state can be avoided, thereby avoiding problems such as data transmission jitter and switching interruption, so that the high reliability transmission requirement of data can be met.

[0063] In an embodiment of the application, the 102, determining the network state evolution of the current network in the future time window according to the multi-source data, can include:

[0064] Generating a smooth quality curve according to the multi-source data;

[0065] Extracting the network quality trend feature of the current network from the smooth quality curve;

[0066] input the network quality trend feature into a preset prediction model to obtain network state evolution in a future time window of the current network output by the prediction model.

[0067] In the embodiment of the application, each network interface can be activated in time according to a preset period to collect multi-source data, and then the collected multi-source data can be normalized to eliminate dimensional differences. Then, the multi-source data with the eliminated dimensional differences can be used to generate a smooth quality curve corresponding to the current network.

[0068] In some embodiments, a prediction network is deployed in a data transmission unit or a cloud server, wherein the prediction network is a light-weighted LSTM (Long Short-Term Memory Network) prediction model, and the network state evolution in a future time window can be predicted based on the prediction model. The prediction model is a cross-protocol layer quality perception model, which can fuse physical layer (RSRP / SINR), transport layer (RTT / packet loss rate), and network layer (base station load) indicators to construct a dynamic weighted evaluation system, thereby better predicting the network state evolution in the future time window.

[0069] In the embodiment of the application, an EWMA (Exponentially Weighted Moving Average) algorithm can be applied to generate a smooth quality curve according to multi-source data, and then a network quality trend feature of the current network can be extracted from the smooth quality curve and input into a prediction model, so as to obtain the network state evolution in a future time window of the current network output by the prediction model.

[0070] In an embodiment of the application, determining the target network from the switchable networks can include:

[0071] acquiring multi-source data corresponding to the switchable networks;

[0072] determining a target network from the switchable networks according to the multi-source data corresponding to the switchable networks.

[0073] In the embodiment of the present application, when it is determined that switching to other network is needed according to the network state evolution in the future time window of the current network, and the data transmission unit also has one or more switchable networks, when the data transmission unit has multiple switchable networks, the multi-source data of the switchable networks, i.e. physical layer signal quality, transmission layer round trip time, application layer packet loss rate and / or network load state, etc. can be obtained, and then the target network can be determined from the multi-source data of the switchable networks, and then the target network is switched to. When the data transmission unit has one switchable network, the switchable network can be taken as the target network, and then the target network is switched to.

[0074] In an embodiment of the present application, the determining of the target network from the switchable networks according to the multi-source data corresponding to the switchable networks can include:

[0075] Discretizing the multi-source data corresponding to the switchable networks into data tuples according to the preset service requirement;

[0076] Inputting the data tuples into a preset deep model to obtain the predicted service performance data of the switchable networks output by the deep model;

[0077] Determining the target network from the switchable networks through a preset greedy strategy and the predicted service performance data of the switchable networks.

[0078] In the embodiment of the present application, a service QoS (Quality of Service) policy library is constructed, and different weights are given to the mapping of the service requirement to the multi-source data. Specifically, the weights of the multi-source data can be dynamically adjusted based on the service type (such as low latency control, high throughput video), and the policy-based QoS guarantee is realized. Through dynamic adjustment of the weights, the dynamic matching of differentiated service requirements and network resources can be realized.

[0079] In the embodiment of the present application, in the data transmission unit or the cloud server, a Dueling DQN (Deep Q-Network) model, i.e. a deep model, can be deployed. The deep model is a model that can be faced to an embedded platform, and can combine offline pre-training and online incremental learning.

[0080] Based on multi-source data, the network state space of a switchable network can be discretized into a three-element data tuple: <signal level, latency level, load level>. Different weights can be assigned to the data tuples according to preset service requirements. Then, the data tuples are input into a preset deep model to obtain the predicted service performance data of the switchable network output by the deep model (the Q-value can be determined based on the predicted service performance data, and the target network can be determined based on the Q-value). The predicted service performance data can at least include throughput and / or network latency. At this point, the target network can be determined from the switchable networks based on the predicted service performance data; for example, the target network can be determined from switchable networks with high throughput and / or low network latency. For example, assuming the predicted latency in the multi-source data is 18ms, it can be discretized as "latency level 1"; assuming the SINR in the multi-source data is 10dB, it can be discretized as "signal level 3".

[0081] In some embodiments, a preset greedy strategy is introduced, wherein the preset greedy strategy is an ε-greedy strategy that balances exploration and utilization, which can avoid local optima traps. Specifically, when obtaining the predicted service performance data of the switchable network through deep model prediction, the target network can be determined from the switchable networks based on the ε-greedy strategy that balances exploration and utilization with the predicted service performance data.

[0082] In one embodiment of the present invention, the deep model is provided with a composite reward function, and the method may further include:

[0083] Acquire multi-source data corresponding to the target network and multi-source data corresponding to the current network;

[0084] Determine network performance change data based on the multi-source data corresponding to the target network and the multi-source data corresponding to the current network;

[0085] The network performance change data is input into a preset composite reward function to obtain reward parameters, and the depth model is updated according to the reward parameters.

[0086] In this embodiment of the invention, a composite reward function is also set for the deep model. The composite reward function can achieve multi-objective optimization. For example, it can combine switching benefits (such as quality improvement), stability costs (such as switching frequency), and service matching degree to perform multi-objective optimization. Specifically, the network performance change data is obtained and determined based on the multi-source data corresponding to the target network and the multi-source data corresponding to the current network. Then, the network performance change data can be input into the composite reward function to obtain reward parameters. The deep model is updated according to the reward parameters, so that the deep model can dynamically adjust the strategy and output results that meet multiple objectives, i.e., predict service performance data. Then, the target network that meets multiple objectives can be better determined from the switchable networks based on the results.

[0087] In one embodiment of the present invention, the data transmission unit is a dual-baseband processor architecture, which includes a main processor and a secondary processor. The switching from the current network to the target network may include:

[0088] When the main processor transmits data to the target base station according to the current network, the auxiliary processor obtains the context information of the target network and sends it to the target base station to synchronize communication resources with the target base station; wherein, the communication resources include at least sequence number status, superframe number and / or robust header shrinking context;

[0089] After synchronizing the communication resources with the target base station, the system switches to transmitting data with the target base station through the target network.

[0090] In this embodiment of the invention, the data transmission unit is a dual-baseband processor architecture, which may include a main processor and a secondary processor. The main processor maintains the transmission of the current service flow, and the secondary processor pre-synchronizes candidate networks (switchable networks).

[0091] Specifically, when the main processor is transmitting data between the current network and the target base station and is preparing to switch to the target network, the auxiliary processor can obtain the PDCP (Packet Data Convergence Protocol) layer context information of the target network and send it to the target base station to synchronize the communication resources between the data transmission unit and the target base station. These communication resources may include sequence number status, superframe number, and / or robust header shrinking context. After synchronizing the communication resources between the data transmission unit and the target base station, the system can switch to transmitting data through the target network. This embodiment of the invention achieves rapid context migration by parallelizing the main processor and auxiliary processor. The main processor can maintain the current network's services until the last frame of data. Thus, when switching to the target network, the service interruption lasts only one frame, a very short time. This embodiment of the invention minimizes the service interruption time during network switching by parallelly executing the release of resources in the current network and the reservation of resources in the target network, achieving seamless network switching.

[0092] The data transmission unit also integrates a lightweight security protocol, which can perform pre-authentication and key pre-negotiation of the target network. Pre-authentication is used to verify the identity of the target base station, and key pre-negotiation is used to generate a session key. Data transmission is carried out securely with the target base station through the session key.

[0093] In one embodiment of the present invention, after the data transmission unit switches from the current network to the target network, the network address (IP address) of the data transmission unit remains unchanged, thereby achieving seamless migration of the transport layer connection.

[0094] In one embodiment of the present invention, after switching from the current network to the target network so that the data transmission unit performs data transmission through the target network, the method further includes:

[0095] Obtain the actual service performance data of the target network;

[0096] Obtain the predicted business performance data output by the deep model;

[0097] The model error is determined based on the actual business performance data and the predicted business performance data, and the deep model is updated based on the model error.

[0098] In this embodiment of the invention, after the data transmission unit is switched to the target network, the actual service performance data of the target network, such as throughput and latency, can be collected. At the same time, the predicted service performance data output by the deep model can be obtained. Then, the model error can be determined by comparing the actual service performance data and the predicted service performance data. The parameters of the deep model can be updated through an online fine-tuning mechanism. The updated deep model can better adapt to complex environments, so that the deep model can more accurately predict the predicted service performance data of the network in the next prediction.

[0099] In this embodiment of the invention, after the data transmission unit is switched to the target network, the actual service performance data of the target network, such as throughput and latency, can be collected. At the same time, the predicted service performance data output by the deep model can be obtained. Then, the model error can be determined by comparing the actual service performance data and the predicted service performance data. The parameters of the deep model can be updated through an online fine-tuning mechanism. The updated deep model can better adapt to complex environments, so that the deep model can more accurately predict the predicted service performance data of the network in the next prediction.

[0100] In relevant technical solutions, network handover schemes rely on fixed operator binding or a single decision model based on signal strength thresholds, resulting in three core defects: lack of multi-dimensional perception, insufficient dynamic adaptability, and poor service continuity. For example, the LTE (Long-Term Evolution) cell reselection mechanism only focuses on physical layer signal quality, failing to integrate transport layer latency, network load, and service QoS requirements. This leads to frequent data transmission jitter, handover interruptions, and long service interruption times in industrial control scenarios, making it difficult to meet high-reliability transmission requirements. Furthermore, these technical solutions also suffer from high signaling overhead.

[0101] Therefore, this invention proposes a network handover method. By integrating a multi-dimensional dynamic evaluation system that considers physical layer signal quality, transport layer latency, network load status, and service QoS requirements, and combining a lightweight DQN model with a pre-synchronized dual-channel handover architecture, it achieves service-driven intelligent network handover, reducing data transmission jitter and handover interruption probability, effectively meeting the high-reliability transmission requirements of industrial control scenarios. Furthermore, since this invention accurately predicts the network state evolution within future time windows, it can reduce invalid signaling transmissions.

[0102] Reference Figure 2 This is a system architecture diagram of network handover provided in an embodiment of the present invention. The data transmission unit of this embodiment adopts a hierarchical collaborative architecture and includes the following core modules:

[0103] Multimodal sensing layer:

[0104] It integrates a multi-standard RF (multi-mode RF) front-end (4G / 5G / NB-IoT), enables multi-mode RF startup, and supports concurrent measurement of multiple operator networks;

[0105] Deploying cross-protocol layer probes allows for the simultaneous collection of physical layer, transport layer, and network layer metrics, enabling the simultaneous acquisition of multi-source data across layers, such as physical layer signal quality, transport layer round-trip time, application layer packet loss rate, and network load status.

[0106] Timestamp alignment technology is used to eliminate timing deviations in multi-source data acquisition.

[0107] Intelligent Analysis Layer:

[0108] Eliminating transient noise interference based on a sliding window filtering algorithm;

[0109] A smooth quality curve is generated based on multi-source data, and network quality trend features are extracted from the smooth quality curve.

[0110] By inputting network quality trend characteristics into a lightweight LSTM prediction model, the network state evolution of the current network within a future time window of the data transmission unit is predicted (quality prediction report).

[0111] Build a business QoS policy library to map business requirements into a combination of network indicator weights.

[0112] Dynamic decision-making level:

[0113] Deploy an improved Dueling DQN model (deep model) that discretizes the network state space of a switchable network into triples of <signal level, delay level, load level> based on multi-source data from the switchable network.

[0114] Design a composite reward function that integrates switching benefits (such as quality improvement), stability costs (such as switching frequency), and business matching to achieve multi-objective optimization of the deep model;

[0115] An ε-greedy strategy is introduced to balance exploration and utilization, avoiding the trap of local optima.

[0116] High-efficiency execution layer:

[0117] A dual-baseband processor architecture is adopted: the main processor maintains the current service flow of the data transmission unit, and the auxiliary processor pre-synchronizes the candidate networks of the data transmission unit;

[0118] Implement PDCP layer context fast migration technology to keep the IP address unchanged and preserve the transport layer connection state after switching target networks;

[0119] It integrates a lightweight security protocol to complete pre-authentication and key pre-negotiation for the target network.

[0120] Specifically, the network handover workflow based on the aforementioned hierarchical collaborative architecture can include:

[0121] Asynchronous data acquisition based on a multimodal sensing layer:

[0122] Each network interface is activated in a time-sharing manner according to a preset cycle to collect cross-layer indicator data (multi-source data);

[0123] Data flow is managed through priority queues to ensure the real-time performance of high-priority business metrics.

[0124] State assessment and prediction based on intelligent analysis layer:

[0125] Normalize multi-source data to eliminate dimensional differences;

[0126] The EWMA algorithm is applied to generate smooth quality curves based on multi-source data, and network quality trend features are extracted from the smooth quality curves.

[0127] The prediction model is invoked, and the network quality trend characteristics are input into the prediction model to output the network state evolution within the future time window of the current network.

[0128] Intelligent decision generation based on dynamic decision layer:

[0129] If, based on the network state evolution within a future time window, it is determined that the current network's overall score is below the dynamic threshold or a significant degradation is predicted, the decision-making process is triggered.

[0130] The optimal candidate network is selected as the target network based on the deep model.

[0131] Evaluate the pre-synchronization completion and service compatibility of candidate networks.

[0132] Achieve seamless execution switching based on a high-efficiency execution layer:

[0133] Migrate the PDCP layer context via a pre-established secure channel;

[0134] Keep the IP address of the data transmission unit unchanged to achieve seamless migration of the transport layer connection;

[0135] The resource release of the current network and the resource reservation of the target network are performed in parallel.

[0136] Closed-loop optimization iteration:

[0137] Collect actual service performance data (e.g., throughput, latency) after the switchover;

[0138] Compare the predicted business performance data with the actual business performance data to calculate the model error;

[0139] The parameters of the deep model are updated based on model error through an online fine-tuning mechanism.

[0140] In summary, by integrating transport layer latency, network load, and service QoS requirements, the embodiments of the present invention can solve problems such as frequent data transmission jitter and handover interruptions in complex scenarios such as industrial control, and achieve high-reliability transmission requirements.

[0141] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0142] Reference Figure 3 The diagram illustrates a structural block diagram of a network switching device provided in an embodiment of the present invention, applied to a data transmission unit. The device may specifically include the following modules:

[0143] The multi-source data acquisition module 301 is used to acquire multi-source data corresponding to the current network of the data transmission unit;

[0144] The network state evolution determination module 302 is used to determine the network state evolution of the current network within a future time window based on the multi-source data;

[0145] The target network determination module 303 is used to determine the target network from the switchable networks if the data transmission unit has a switchable network when the network state evolution is lower than a preset network state evolution threshold.

[0146] The target network switching module 304 is used to switch from the current network to the target network so that the data transmission unit can transmit data through the target network.

[0147] In one embodiment of the present invention, the network state evolution determination module 302 is used for:

[0148] A smooth quality curve is generated based on the multi-source data;

[0149] Extract the network quality trend features of the current network from the smoothed quality curve;

[0150] The network quality trend characteristics are input into a preset prediction model to obtain the network state evolution of the current network within a future time window, as output by the prediction model.

[0151] In one embodiment of the present invention, the target network determination module 303 is configured to:

[0152] Obtain multi-source data corresponding to the switchable network;

[0153] The target network is determined from the switchable networks based on the multi-source data corresponding to the switchable networks.

[0154] In one embodiment of the present invention, determining the target network from the switchable network based on the multi-source data corresponding to the switchable network includes:

[0155] According to the preset business requirements, the multi-source data corresponding to the switchable network is discretized into data tuples;

[0156] The data tuples are input into a preset deep model to obtain the predicted service performance data of the switchable network output by the deep model.

[0157] The target network is determined from the switchable network using a preset greedy strategy and the predicted service performance data of the switchable network.

[0158] In one embodiment of the present invention, the deep model is provided with a composite reward function, and the apparatus further includes: a deep model update module, used for:

[0159] Acquire multi-source data corresponding to the target network and multi-source data corresponding to the current network;

[0160] Determine network performance change data based on the multi-source data corresponding to the target network and the multi-source data corresponding to the current network;

[0161] The network performance change data is input into a preset composite reward function to obtain reward parameters, and the depth model is updated according to the reward parameters.

[0162] In one embodiment of the present invention, the data transmission unit is a dual-baseband processor architecture, the dual-baseband processor architecture including a main processor and a secondary processor, and the switching from the current network to the target network includes:

[0163] When the main processor transmits data to the target base station according to the current network, the auxiliary processor obtains the context information of the target network and sends it to the target base station to synchronize communication resources with the target base station; wherein, the communication resources include at least sequence number status, superframe number and / or robust header shrinking context;

[0164] After synchronizing the communication resources with the target base station, the system switches to transmitting data with the target base station through the target network.

[0165] In one embodiment of the present invention, the device further includes:

[0166] Obtain the actual service performance data of the target network;

[0167] Obtain the predicted business performance data output by the deep model;

[0168] The model error is determined based on the actual business performance data and the predicted business performance data, and the deep model is updated based on the model error.

[0169] In one embodiment of the present invention, the multi-source data includes at least physical layer signal quality, transport layer round-trip time, application layer packet loss rate and / or network load status; service performance data and predicted service performance data include at least throughput and / or network latency.

[0170] In one embodiment of the present invention, after the data transmission unit switches from the current network to the target network, the network address of the data transmission unit remains unchanged.

[0171] In this embodiment of the invention, multi-source data corresponding to the current network of the data transmission unit is obtained, and then the network state evolution within the future time window of the current network can be determined based on the multi-source data. When the network state evolution is lower than a preset network state evolution threshold, if the data transmission unit has a switchable network, a target network is determined from the switchable networks and switched to, so that the data transmission unit can transmit data through the target network. This avoids continuing to transmit data through the current network with a poor future network state, thereby avoiding problems such as data transmission jitter and switching interruption, and thus meeting the high reliability transmission requirements of data.

[0172] As the apparatus embodiment is basically similar to the method embodiment, it is described in a relatively simple manner. For relevant details, please refer to the description of the method embodiment.

[0173] In addition, this invention also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described network switching method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0174] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the network handover method embodiments described above and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0175] This invention also provides a computer program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the network handover method embodiments described above, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0176] Figure 4 A schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention.

[0177] The electronic device 400 includes, but is not limited to, components such as: a radio frequency unit 401, a network module 402, an audio output unit 403, an input unit 404, a sensor 405, a display unit 406, a user input unit 407, an interface unit 408, a memory 409, a processor 410, and a power supply 411. Those skilled in the art will understand that... Figure 4 The electronic device structures shown are not intended to limit the electronic device. An electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements. In embodiments of the present invention, the electronic device includes, but is not limited to, mobile phones, tablet computers, laptops, PDAs, in-vehicle terminals, wearable devices, and pedometers.

[0178] It should be understood that, in this embodiment of the invention, the radio frequency unit 401 can be used for receiving and transmitting signals during information transmission or calls. Specifically, it receives downlink data from the base station and processes it with the processor 410; additionally, it transmits uplink data to the base station. Typically, the radio frequency unit 401 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, etc. Furthermore, the radio frequency unit 401 can also communicate with networks and other devices through a wireless communication system.

[0179] The electronic device provides users with wireless broadband internet access through the network module 402, such as helping users send and receive emails, browse web pages, and access streaming media.

[0180] The audio output unit 403 can convert audio data received by the radio frequency unit 401 or the network module 402 or stored in the memory 409 into audio signals and output them as sound. Furthermore, the audio output unit 403 can also provide audio output related to specific functions performed by the electronic device 400 (e.g., call signal reception sound, message reception sound, etc.). The audio output unit 403 includes a speaker, a buzzer, and a receiver, etc.

[0181] Input unit 404 is used to receive audio or video signals. Input unit 404 may include a graphics processing unit (GPU) 4041 and a microphone 4042. GPU 4041 processes image data of still images or videos acquired by an image capture device (such as a camera) in video capture mode or image capture mode. The processed image frames can be displayed on display unit 406. The image frames processed by GPU 4041 can be stored in memory 409 (or other storage medium) or transmitted via radio frequency unit 401 or network module 402. Microphone 4042 can receive sound and process such sound into audio data. The processed audio data can be converted into a format that can be transmitted to a mobile communication base station via radio frequency unit 401 in telephone call mode.

[0182] The electronic device 400 also includes at least one sensor 405, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor includes an ambient light sensor and a proximity sensor. The ambient light sensor can adjust the brightness of the display panel 4061 according to the ambient light level, and the proximity sensor can turn off the display panel 4061 and / or backlight when the electronic device 400 is moved to the ear. As a type of motion sensor, an accelerometer sensor can detect the magnitude of acceleration in various directions (generally three axes). When stationary, it can detect the magnitude and direction of gravity and can be used to identify the posture of the electronic device (such as landscape / portrait switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc. The sensor 405 may also include a fingerprint sensor, pressure sensor, iris sensor, molecular sensor, gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc., which will not be described in detail here.

[0183] The display unit 406 is used to display information input by the user or information provided to the user. The display unit 406 may include a display panel 4061, which may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0184] User input unit 407 can be used to receive input numerical or character information, and generate key signal inputs related to user settings and function control of electronic devices. Specifically, user input unit 407 includes a touch panel 4071 and other input devices 4072. Touch panel 4071, also known as a touch screen, can collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near touch panel 4071). Touch panel 4071 may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch position and the signal generated by the touch operation, and transmits the signal to the touch controller; the touch controller receives touch information from the touch detection device, converts it into touch point coordinates, and sends it to the processor 410, which receives and executes commands from the processor 410. In addition, touch panel 4071 can be implemented using various types such as resistive, capacitive, infrared, and surface acoustic wave. Besides touch panel 4071, user input unit 407 may also include other input devices 4072. Specifically, other input devices 4072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, joysticks, etc., which will not be described in detail here.

[0185] Furthermore, the touch panel 4071 can cover the display panel 4061. When the touch panel 4071 detects a touch operation on or near it, it transmits the information to the processor 410 to determine the type of touch event. Subsequently, the processor 410 provides corresponding visual output on the display panel 4061 based on the type of touch event. Although in Figure 4 In this embodiment, the touch panel 4071 and the display panel 4061 are two independent components to realize the input and output functions of the electronic device. However, in some embodiments, the touch panel 4071 and the display panel 4061 can be integrated to realize the input and output functions of the electronic device. The specific implementation is not limited here.

[0186] Interface unit 408 serves as an interface for connecting external devices to electronic device 400. For example, external devices may include a wired or wireless headphone port, an external power supply (or battery charger) port, a wired or wireless data port, a memory card port, a port for connecting a device with an identification module, an audio input / output (I / O) port, a video I / O port, a headphone port, and so on. Interface unit 408 can be used to receive input from external devices (e.g., data, power, etc.) and transmit the received input to one or more components within electronic device 400, or it can be used to transmit data between electronic device 400 and external devices.

[0187] The memory 409 can be used to store software programs and various data. The memory 409 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function (such as sound playback, image playback, etc.), etc.; the data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory 409 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0188] The processor 410 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 409, and by calling data stored in the memory 409, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. The processor 410 may include one or more processing units; preferably, the processor 410 may integrate an application processor and a modem processor. The application processor mainly handles the operating system, user interface, and applications, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 410.

[0189] The electronic device 400 may also include a power supply 411 (such as a battery) for supplying power to various components. Preferably, the power supply 411 can be logically connected to the processor 410 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system.

[0190] In addition, the electronic device 400 includes some functional modules not shown, which will not be described in detail here.

[0191] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0192] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0193] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

[0194] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software 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 this invention.

[0195] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0196] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0197] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0198] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0199] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0200] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A network handover method, characterized in that, Applied to a data transmission unit, the method includes: Obtain multi-source data corresponding to the current network of the data transmission unit; The network state evolution within a future time window of the current network is determined based on the multi-source data. When the network state evolution is lower than a preset network state evolution threshold, if the data transmission unit has a switchable network, then the target network is determined from the switchable network. The current network is switched to the target network so that the data transmission unit can transmit data through the target network.

2. The method according to claim 1, characterized in that, Determining the network state evolution within a future time window based on the multi-source data includes: A smooth quality curve is generated based on the multi-source data; Extract the network quality trend features of the current network from the smoothed quality curve; The network quality trend characteristics are input into a preset prediction model to obtain the network state evolution of the current network within a future time window, as output by the prediction model.

3. The method according to claim 1, characterized in that, Determining the target network from the switchable networks includes: Obtain multi-source data corresponding to the switchable network; The target network is determined from the switchable networks based on the multi-source data corresponding to the switchable networks.

4. The method according to claim 3, characterized in that, The step of determining the target network from the switchable network based on the multi-source data corresponding to the switchable network includes: According to the preset business requirements, the multi-source data corresponding to the switchable network is discretized into data tuples; The data tuples are input into a preset deep model to obtain the predicted service performance data of the switchable network output by the deep model. The target network is determined from the switchable network using a preset greedy strategy and the predicted service performance data of the switchable network.

5. The method according to claim 4, characterized in that, The deep model is configured with a composite reward function, and the method further includes: Acquire multi-source data corresponding to the target network and multi-source data corresponding to the current network; Determine network performance change data based on the multi-source data corresponding to the target network and the multi-source data corresponding to the current network; The network performance change data is input into a preset composite reward function to obtain reward parameters, and the depth model is updated according to the reward parameters.

6. The method according to claim 4, characterized in that, The data transmission unit is a dual-baseband processor architecture, which includes a main processor and a secondary processor. The switching from the current network to the target network includes: When the main processor transmits data to the target base station according to the current network, the auxiliary processor obtains the context information of the target network and sends it to the target base station to synchronize communication resources with the target base station; wherein, the communication resources include at least sequence number status, superframe number and / or robust header shrinking context; After synchronizing the communication resources with the target base station, the system switches to transmitting data with the target base station through the target network.

7. The method according to claim 4, characterized in that, After switching from the current network to the target network so that the data transmission unit can transmit data through the target network, the method further includes: Obtain the actual service performance data of the target network; Obtain the predicted business performance data output by the deep model; The model error is determined based on the actual business performance data and the predicted business performance data, and the deep model is updated based on the model error.

8. The method according to claim 1, characterized in that, The multi-source data includes at least physical layer signal quality, transport layer round-trip time, application layer packet loss rate, and / or network load status; service performance data and predicted service performance data include at least throughput and / or network latency.

9. The method according to claim 1, characterized in that, After the data transmission unit switches from the current network to the target network, the network address of the data transmission unit remains unchanged.

10. A network switching device, characterized in that, Applied to a data transmission unit, the device includes: A multi-source data acquisition module is used to acquire multi-source data corresponding to the current network of the data transmission unit; The network state evolution determination module is used to determine the network state evolution of the current network within a future time window based on the multi-source data; The target network determination module is used to determine the target network from the switchable networks if the data transmission unit has a switchable network when the network state evolution is lower than a preset network state evolution threshold. The target network switching module is used to switch from the current network to the target network so that the data transmission unit can transmit data through the target network.

11. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method as described in any one of claims 1-9.

12. A computer-readable storage medium having instructions stored thereon that, when executed by one or more processors, cause the processors to perform the method as described in any one of claims 1-9.

Images