Cell selection method and device, terminal equipment and storage medium

Abstract

The embodiment of the invention discloses a cell selection method and device, terminal equipment and a storage medium. Relates to the technical field of communication. The method is applied to terminal equipment and comprises the following steps: acquiring currently acquired network characteristic parameters and service characteristic parameters; according to the currently collected network characteristic parameters and service characteristic parameters, obtaining the QoE (Quality of Experience) of each cell and a quantization result obtained by performing uncertainty quantization on the QoE of each cell; and selecting a target cell from each cell according to the QoE of each cell and a quantization result obtained by performing uncertainty quantization on the QoE of each cell. According to the scheme, the whole process of selecting the target cell not only considers various network characteristic parameters of the base station, but also considers the service characteristic parameters of the terminal equipment side, and the quantization result with context sensing is obtained based on uncertainty quantization, so that the determined target cell better conforms to the service type currently executed by the terminal equipment, and the user experience is improved. And the accuracy of determining the target cell is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of communication technology, and in particular to a cell selection method, apparatus, terminal equipment, and storage medium. Background Art

[0002] With the development of information technology and Internet technology, the use of terminal devices in daily life has become very common. During use, terminal devices need to establish communication connections with network devices such as base stations in advance.

[0003] For example, within the service range of a base station, a terminal device can establish a communication connection with the base station through its installed Subscriber Identity Module (SIM) card, thereby achieving communication interaction. In actual applications, as the terminal device moves, the network signal received by the terminal device from the base station will also change. When the network signal of the terminal device is poor, the network switching process of the terminal device (such as the process of switching to another base station) can be triggered. At present, most terminal devices are screened based on the Reference Signal Received Power (RSRP) received by each base station, and the base station with the largest RSRP is selected as the current re-access base station to complete the network switching. In this solution, the parameters of the network signal used are relatively simple, and the base station finally selected is prone to not meeting the current business needs of the terminal device, resulting in network instability. Summary of the Invention

[0004] In order to solve the problems of related technologies, more accurately screen out cells that meet the services of terminal devices from multiple cells detected in the surrounding area, and improve the stability of the network connection of the terminal device, the embodiments of the present application provide a cell selection method, apparatus, terminal device and storage medium. The technical solution is as follows:

[0005] In one aspect, an embodiment of the present application provides a cell selection method, applied to a terminal device, the method comprising:

[0006] Acquire currently collected network characteristic parameters and service characteristic parameters, the network characteristic parameters including at least signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters including at least a service type and service execution time currently executed by the terminal device;

[0007] Obtaining, based on the currently collected network characteristic parameters and the service characteristic parameters, a quality of experience (QoE) of each cell and a quantification result of uncertainty quantification of the QoE of each cell, wherein the uncertainty quantification is used to obtain a predicted value of the QoE of each cell after the service execution time;

[0008] A target cell is selected from each of the cells according to the QoE of each cell and a quantification result of uncertainty quantification of the QoE of each cell.

[0009] In another aspect, an embodiment of the present application provides a cell selection device, applied to a terminal device, the device comprising:

[0010] a first acquisition module, configured to acquire currently collected network characteristic parameters and service characteristic parameters, the network characteristic parameters including at least signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters including at least a service type and service execution time currently executed by the terminal device;

[0011] a second acquisition module, configured to acquire, based on the currently collected network characteristic parameters and the service characteristic parameters, a quality of experience (QoE) of each cell and a quantification result of uncertainty quantification of the QoE of each cell, wherein the uncertainty quantification is used to obtain a predicted value of the QoE of each cell after the service execution time;

[0012] The cell selection module is configured to select a target cell from each of the cells according to the QoE of each of the cells and a quantification result of uncertainty quantification of the QoE of each of the cells.

[0013] In another aspect, the present application provides a terminal device, which includes a processor and a memory, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, it implements the cell selection method as described in one aspect above.

[0014] In another aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the cell selection method as described in the above aspect.

[0015] On the other hand, an embodiment of the present application provides a computer program product. When the computer program product is run on a computer, it enables the computer to execute to implement the cell selection method as described in one aspect above.

[0016] On the other hand, an embodiment of the present application provides an application publishing platform, which is used to publish a computer program product, wherein when the computer program product runs on a computer, the computer is enabled to execute to implement the cell selection method as described in one aspect above.

[0017] The beneficial effects of the technical solutions provided in the embodiments of the present application include at least:

[0018] The terminal device obtains the currently collected network characteristic parameters and service characteristic parameters, where the network characteristic parameters include at least the signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters include at least the service type currently executed by the terminal device and the service execution time; based on the currently collected network characteristic parameters and service characteristic parameters, the terminal device obtains the experience quality QoE of each cell and the quantitative results of uncertainty quantification of the QoE of each cell, and the uncertainty quantification is used to obtain a predicted value of the QoE of each cell after the service execution time; based on the QoE of each cell and the quantitative results of the uncertainty quantification of the QoE of each cell, the target cell is selected from each cell. In the present application, the parameters used in selecting the target cell include network characteristic parameters and service characteristic parameters, and based on the network characteristic parameters and service characteristic parameters, the QoE of each cell is obtained, and the uncertainty of the QoE of each cell is quantified, and the predicted value of the QoE of each cell after the service execution time is obtained to obtain a quantified result. Combined with the QoE and quantification results of each cell, the target cell is selected from each cell. The entire process not only takes into account multiple network characteristic parameters of the base station, but also takes into account the service type and service execution time currently executed by the terminal device. Based on the uncertainty quantification, a context-aware quantification result is obtained, so that the determined target cell is more in line with the service type currently executed by the terminal device, thereby improving the accuracy of determining the target cell and increasing the stability of the network connection of the terminal device. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 This is a schematic diagram of an implementation environment involved in connecting a terminal device to a base station network according to an exemplary embodiment of the present application;

[0021] Figure 2 A schematic diagram of the internal structure of a terminal device provided as an exemplary embodiment of the present application;

[0022] Figure 3 A flowchart of a cell selection method provided by an exemplary embodiment of the present application;

[0023] Figure 4 A flowchart of a cell selection method provided by an exemplary embodiment of the present application;

[0024] Figure 5 This is a schematic diagram of location area division including various cells involved in an exemplary embodiment of the present application;

[0025] Figure 6 This is a structural diagram of a cell selection model involved in an exemplary embodiment of the present application;

[0026] Figure 7 A structural block diagram of a cell selection device provided by an exemplary embodiment of the present application;

[0027] Figure 8 A structural diagram of another example of a cell selection device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0029] In this document, "plurality" refers to two or more. "And / or" describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone, A and B exist simultaneously, or B exists alone. The character " / " generally indicates an "or" relationship between the associated objects.

[0030] It should be pointed out that the terms "first\second\third" involved in the embodiments of the present application are used to distinguish similar or different objects, and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described here can be implemented in an order other than that illustrated or described here.

[0031] The solution provided in this application can be used in real-life scenarios where people use terminal devices with network communication functions to establish network connections with base stations or switch base stations in their daily lives. For ease of understanding, the following is a brief introduction to some of the terms and application scenarios involved in the embodiments of this application.

[0032] Quality of Experience (QoE) refers to a user's subjective perception of the quality and performance of devices, networks, systems, applications, or services. It can represent a comprehensive indicator of multiple dimensions, including base station signal strength, network latency, throughput, and the user's own service quality, while using a terminal device.

[0033] Figure 1 This is a schematic diagram of an implementation environment involved in connecting a terminal device to a base station network, as shown in an exemplary embodiment of the present application. Figure 1 As shown, the implementation environment may include: several terminal devices 110 and a base station 120.

[0034] The terminal device 110 is a wireless communication device that can transmit data using wireless access technology. For example, the terminal device 110 can support cellular mobile communication technology, such as the fourth generation mobile communication technology (4G) and 5G technology. Alternatively, the terminal device 110 can also support the next generation mobile communication technology of 5G technology.

[0035] For example, the terminal device 110 may be a vehicle-mounted device, such as a driving computer with wireless communication function, or a wireless communication device externally connected to the driving computer.

[0036] Alternatively, the terminal device 110 may also be a roadside device, for example, a street lamp, a traffic light, or other roadside device with wireless communication function.

[0037] Alternatively, the terminal device 110 may also be a user terminal device, such as a mobile phone (or "cellular" phone) and a computer with a mobile terminal, for example, a portable, pocket-sized, handheld, computer-built-in, or vehicle-mounted mobile device. For example, a station (STA), a subscriber unit, a subscriber station, a mobile station, a mobile station, a remote station, an access point, a remote terminal, an access terminal, a user terminal, a user agent, a user device, or a user equipment (UE). Specifically, for example, the terminal device 110 may be a mobile terminal such as a smartphone, a tablet computer, or an e-book reader, or may be a smart wearable device such as smart glasses, a smart watch, or a smart bracelet.

[0038] Optionally, the terminal device 110 is a wireless communication device that supports half-duplex technology.

[0039] Optionally, wireless communication between several terminal devices 110 is supported through direct communication.

[0040] Please refer to Figure 2 , which shows a schematic diagram of the internal structure of a terminal device provided by an exemplary embodiment of the present application. Figure 2 The terminal device shown includes components such as a processor 210, a memory 220, a transceiver 230, a display unit 240, a sensor 250, and a battery module 260.

[0041] Processor 210 is the control center of the terminal device. It connects the various components of the entire terminal device using various interfaces and lines. By running or executing software programs and / or modules stored in memory 220 and accessing data stored in memory 220, it performs various terminal device functions and processes data, thereby monitoring the terminal device as a whole. Optionally, processor 210 may include one or more processing units; optionally, processor 210 may integrate an application processor, which primarily processes operating devices, user interfaces, and application programs. Of course, other processors may also be included, which are not listed here.

[0042] The memory 220 can be used to store software programs and modules. The processor 210 executes the various functional applications and data processing of the terminal device by running the software programs and modules stored in the memory 220. The memory 220 may mainly include a program storage area and a data storage area. The program storage area may store operating devices, at least one application required for a function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created based on the use of the terminal device (such as audio data, a phone book, etc.). In addition, the memory 220 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0043] The transceiver 230 can provide wireless communication solutions including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) network), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared (IR), etc., which are applied to the terminal device. The transceiver 230 can be one or more devices that integrate at least one communication processing module, for example, the antenna and the baseband processor are integrated into the transceiver 230, or the antenna and the modulation and demodulation processor are integrated into the transceiver 230, etc., which are not limited here. Optionally, the terminal device can communicate with the above-mentioned devices through its own transceiver 230. Figure 2 Establish a wireless communication connection with the base station in the network.

[0044] The display unit 240 can be used to display information input by the user or information provided to the user, as well as various menus of the terminal device. The display unit 240 can be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc., without limitation.

[0045] The terminal device may also include at least one sensor 250, such as a gyroscope, a motion sensor, an image sensor, or other sensors. The motion sensor may include an accelerometer, which detects the magnitude of acceleration in all directions and the magnitude and direction of gravity when stationary. This can be used in applications that identify the terminal device's posture, such as switching between landscape and portrait modes, related games, and magnetometer posture calibration. The image sensor can be used in modules such as cameras and video cameras in the terminal device to collect information about the external scene. As for other sensors that may be configured in the terminal device, such as pressure gauges, barometers, hygrometers, thermometers, and infrared sensors, they are not described in detail here.

[0046] The terminal device also includes a battery module 260 for supplying power to various components. Optionally, the battery module 260 can be logically connected to the processor 210 through a power management device, thereby managing charging, discharging, and power consumption management functions through the power management device.

[0047] Although not shown, the terminal device may further include a camera. Optionally, the camera may be positioned in the front or rear of the terminal device, which is not limited in this embodiment of the present application.

[0048] It is understood that the structures illustrated in the embodiments of the present application do not constitute specific limitations on the terminal device. In other embodiments of the present application, the terminal device may include more or fewer components than shown, or may combine or separate certain components, or arrange the components differently. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0049] Base station 120 may be a network-side device in a wireless communication system. The wireless communication system may be a fourth-generation mobile communication technology system, also known as a Long Term Evolution (LTE) system; or a 5G system, also known as a New Radio (NR) system. Alternatively, the wireless communication system may be a next-generation system to the 5G system.

[0050] Among them, the base station 120 can be an evolved base station (eNB) adopted in a 4G system. Alternatively, the base station 120 can also be a base station (gNB) adopting a centralized distributed architecture in a 5G system. When the base station 120 adopts a centralized distributed architecture, it usually includes a centralized unit (CU) and at least two distributed units (DU). The centralized unit is provided with a protocol stack of a packet data convergence protocol (PDCP) layer, a radio link layer control protocol (RLC) layer, and a media access control (MAC) layer; the distributed unit is provided with a physical (PHY) layer protocol stack. The embodiment of the present disclosure does not limit the specific implementation method of the base station 120.

[0051] A wireless connection can be established between the base station 120 and the terminal device 110 via a wireless air interface. In different implementations, the wireless air interface is a wireless air interface based on the fourth generation mobile communication network technology (4G) standard; or, the wireless air interface is a wireless air interface based on the fifth generation mobile communication network technology (5G) standard, for example, the wireless air interface is a new air interface; or, the wireless air interface can also be a wireless air interface based on the next generation mobile communication network technology standard of 5G.

[0052] Optionally, the wireless communication system may further include a network management device 130 .

[0053] Several base stations 120 are respectively connected to a network management device 130. The network management device 130 may be a core network device in a wireless communication system. For example, the network management device 130 may be a mobility management entity (MME) in an evolved packet core (EPC). Alternatively, the network management device may be other core network devices, such as a serving gateway (SGW), a public data network gateway (PGW), a policy and charging rules function (PCRF), or a home subscriber server (HSS). The embodiments of the present disclosure do not limit the implementation form of the network management device 130.

[0054] Optionally, the above-mentioned base station 120 usually provides network services to each terminal device within the service range. When there are too many devices that need to provide network services in the base station and the base station load is large, the base station is often required to reallocate some terminal devices. For example, the base station will be configured to trigger the terminal device to measure events of itself and other base stations, and measure the reference signal received power (RSRP), reference signal received quality (RSRQ), signal to interference and noise ratio (SINR) and other information of the terminal device. If the terminal device meets the threshold corresponding to the measurement event, the measurement report is reported to the base station, and the base station redirects or switches the terminal device to another base station. The measurement report may include information such as RSRP / RSRQ / SINR measured by the terminal device for each base station. This is the process of network reconnection or base station switching initiated by the base station. The switching triggered by the base station side is often based on whether its own load is overloaded and guides the terminal device to access a base station with a low load, so as to achieve the effect of reducing its own load.

[0055] Of course, in actual applications, the terminal device can also actively measure information such as RSRP and select the best base station for handover. For example, the base station with the largest RSRP is selected as the current base station to re-access, thereby completing the network handover.

[0056] For the two switching schemes existing in the above-mentioned terminal devices, the parameters of the network signals used are relatively simple, and the base station finally selected may not meet the current business needs of the terminal device. For example, after switching to another base station, it is found that the load of the other base station is still overloaded and it is necessary to switch again, or the base station after switching is not configured with the data link mapping required for the current business, or the base station after switching does not support the interface for the current business transmission and cannot connect / transmit the core network data between LTE and NR. Such processes will not only cause the terminal device to experience network freezes for a few seconds, but also cause network instability.

[0057] In order to solve the above problems existing in the relevant technology, and more accurately screen out cells that meet the services of the terminal device from multiple cells detected in the surrounding area, a cell selection method is provided in an embodiment of the present application. The quality of experience (QoE) of each cell and the quantification result of the uncertainty quantification of the QoE of each cell can be obtained based on the currently collected network characteristic parameters and service characteristic parameters, so as to select the target cell from each cell. The whole process considers more parameters, and the determined target cell is more in line with the service type currently executed by the terminal device, thereby improving the stability of the network connection of the terminal device.

[0058] Please refer to Figure 3 , which shows a method flow chart of a cell selection method provided by an exemplary embodiment of the present application, and the cell selection method can be applied to a terminal device. Figure 3 As shown, the cell selection method may include the following steps:

[0059] Step 301, obtain the currently collected network characteristic parameters and service characteristic parameters, the network characteristic parameters at least include the signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters at least include the service type and service execution time currently executed by the terminal device.

[0060] Optionally, in this solution, the terminal device may be a terminal device with an image sensor, for example, the terminal device may be but is not limited to a wearable device (such as a bracelet, a smart watch, smart glasses, etc.), a mobile phone, a tablet computer, a laptop computer, an MP3 player (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Group Audio Layer 3), an MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Group Audio Layer 4) player, a desktop computer, a laptop computer, etc.

[0061] Optionally, the terminal device can monitor the network characteristic parameters of the surrounding cells in real time, and can also monitor the service type and service execution time corresponding to the service currently being executed by itself in real time, and can obtain the currently collected network characteristic parameters and service characteristic parameters at any time. Among them, the network characteristic parameters take the signal quality and load quantity of the cell as an example. The terminal device measures the signal quality of multiple cells that can be received, and obtains the required signal quality of multiple cells for the load quantity. The terminal device can obtain the current load quantity by passing the identification parameter of the base station through the base station query application programming (Application Programming Interface, API) interface provided by the access operator or third-party service provider (such as the single base station query API provided by a developer platform). Among them, the methods of obtaining RSRP and load quantity here are exemplary. For example, the load quantity can also be replaced by the proportion of all current loads contained in the base station. In actual applications, it can also be obtained based on other methods, which is not limited here.

[0062] Optionally, taking RSRP as an example of representing the signal quality of a cell, the terminal device can measure the RSRP of each currently detected cell and obtain the RSRP of each of the multiple currently detected cells. Of course, the signal quality of a cell can also be represented by information such as RSRQ / SINR, and the terminal device can also obtain this information to represent the signal quality of the cell. In other words, the terminal device can apply various network key performance indicators (KPIs) that can represent the signal quality of the cell to this solution.

[0063] Optionally, in this solution, the developer can pre-define the service type based on the various services supported by the terminal device. For example, services for apps running video can be classified as video streaming, while services for apps running games can be classified as games. The specific service type is determined by the actual classification method and is not limited by this solution.

[0064] Optionally, the service execution time is the current time of the terminal device. For example, if the user is playing a game on the terminal device at 9 o'clock in the morning, and the service characteristic parameters are obtained at this time, the service type obtained is the game type, and the service execution time is the current 9 o'clock.

[0065] The signal quality of the cell mentioned above is represented by RSRP. The network characteristic parameters include RSRP and load quantity. The service characteristics include service type and service execution time. After the terminal device collects data at the current moment, it can obtain the RSRP of each cell detected around it, the load quantity of each cell, the service type currently being executed by itself, and the current time. i Indicates the currently collected network characteristic parameters and service characteristic parameters of the i-th cell. If the service type currently executed by the terminal device at 9 o'clock is the game type, and the currently detected cells are cell one, cell two, and cell three, then when executing step 301 of this solution, the network characteristic parameters and service characteristic parameters of cell one can be obtained: X1 is (RSRP1, load number one, game type, 9 o'clock), the network characteristic parameters and service characteristic parameters of cell two are: X2 is (RSRP2, load number two, game type, 9 o'clock), the network characteristic parameters and service characteristic parameters of cell three are: X3 is (RSRP3, load number three, game type, 9 o'clock). That is, the service characteristic parameters of the terminal device itself are the same for each cell, and the network characteristic parameters monitored by the terminal device in different cells are different. The terminal device needs to select the best cell from them.

[0066] Step 302: Based on the currently collected network characteristic parameters and service characteristic parameters, obtain the quality of experience (QoE) of each cell and the quantification result of uncertainty quantification of the QoE of each cell. The uncertainty quantification is used to obtain a predicted value of the QoE of each cell after the service execution time.

[0067] Optionally, the terminal device can obtain the QoE of each cell based on the currently collected network characteristic parameters and service characteristic parameters, and then quantify the uncertainty of the QoE of each cell to obtain a quantified result of the QoE of each cell.

[0068] Among them, uncertainty quantification can be achieved based on the Bayesian method. For example, this solution uses multiple neurons such as Bayesian neural networks and Monte Carlo to iteratively predict QoE, obtain the predicted value of QoE after the service execution time, quantify the uncertainty of the QoE of each cell, and obtain the predicted value of QoE of each cell. For example, for the network characteristic parameters and service characteristic parameters of cells 1, 2, and 3 collected above, the QoE of each cell 1, 2, and 3 can be obtained by calculation, and the uncertainty of the QoE of each cell can be quantified to obtain the predicted value of QoE after the service execution time. The quantified result obtained after the uncertainty quantification of the QoE of each cell is obtained through the predicted value of the QoE of each cell.

[0069] Optionally, the uncertainty quantification result may indicate the stability of each QoE prediction value obtained during the uncertainty quantification process. For example, the prediction variance of each QoE prediction value obtained may be used as the quantification result, or the range of each QoE prediction value obtained may be used as the quantification result, or the average of each QoE prediction value obtained may be used as the quantification result, etc., which are not limited here.

[0070] Step 303 : Select a target cell from each cell according to the QoE of each cell and the quantification result of uncertainty quantification of the QoE of each cell.

[0071] Optionally, the terminal device selects a target cell from each cell based on the QoE of each cell obtained above and the quantified result of the uncertainty quantification of the QoE of each cell. The target cell is the cell that is most suitable for the current service execution among the multiple cells detected by the terminal device. That is, the terminal device selects a cell that is more suitable for the current service type execution from each cell based on the QoE of each cell and the quantified result of the uncertainty quantification of the QoE of each cell, ensuring that the selected cell has the highest combined QoE value and stability when the terminal device executes the service of this type, thereby improving the stability of the network executing the current service in the terminal device and improving the efficiency of the terminal device when currently executing the service of this type.

[0072] To summarize, the terminal device obtains the currently collected network characteristic parameters and service characteristic parameters, where the network characteristic parameters include at least the signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters include at least the service type currently executed by the terminal device and the service execution time; based on the currently collected network characteristic parameters and service characteristic parameters, the terminal device obtains the quality of experience QoE of each cell and the quantitative results of uncertainty quantification of the QoE of each cell, and the uncertainty quantification is used to obtain the predicted value of the QoE of each cell after the service execution time; based on the QoE of each cell and the quantitative results of uncertainty quantification of the QoE of each cell, the target cell is selected from each cell. In the present application, the parameters used in selecting the target cell include network characteristic parameters and service characteristic parameters, and based on the network characteristic parameters and service characteristic parameters, the QoE of each cell is obtained, and the uncertainty of the QoE of each cell is quantified, and the predicted value of the QoE of each cell after the service execution time is obtained to obtain a quantified result. Combined with the QoE and quantification results of each cell, the target cell is selected from each cell. The entire process not only takes into account multiple network characteristic parameters of the base station, but also takes into account the service type and service execution time currently executed by the terminal device. Based on the uncertainty quantification, a context-aware quantification result is obtained, so that the determined target cell is more in line with the service type currently executed by the terminal device, thereby improving the accuracy of determining the target cell and increasing the stability of the network connection of the terminal device.

[0073] Below, we take the example of the terminal device needing to obtain the target comprehensive index as an intermediate variable in the process of selecting the target cell from each cell. The target comprehensive index is used to indicate the degree of matching between a cell and the service characteristic parameters. The target cell is selected by combining the optimal value of the target comprehensive index as an example to introduce the execution process of the above method. Please refer to Figure 4 , which shows a schematic diagram of a calibration process for obtaining each compensation matrix involved in an exemplary embodiment of the present application. The method can be applied to a terminal device, such as Figure 4 As shown, the calibration process may include the following steps:

[0074] Step 401, obtain the currently collected network characteristic parameters and service characteristic parameters, the network characteristic parameters at least include the signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters at least include the service type and service execution time currently executed by the terminal device.

[0075] Optionally, the collection and acquisition of network characteristic parameters and service characteristic parameters in the terminal device can refer to the description in the above step 301 and will not be repeated here.

[0076] In one possible implementation, in order to improve the privacy security of terminal devices, in this solution, local differential privacy (LDP) protection can be performed on the service type in the service feature parameters obtained locally by the terminal device, and noise is introduced during the data collection and processing stage to protect the type of the currently executed APP (that is, the user's personal privacy).

[0077] For example, the terminal device obtains the currently executed service type (AppType) in the following way:

[0078]

[0079] Add noise to the APP type, where Indicates the business type (AppType) obtained after the noise is introduced, Xapp indicates the type of the currently executed APP, represents the added noise.

[0080] Step 402: Acquire the Quality of Experience (QoE) of each cell based on the currently collected network characteristic parameters and service characteristic parameters.

[0081] Optionally, the terminal device may calculate the QoE corresponding to the current service executed in each cell during the service execution time based on the currently collected network characteristic parameters and service characteristic parameters. For example, the terminal device may be pre-set with the following first calculation formula:

[0082] QoE = f θ (a,b);

[0083] Among them, a represents the network characteristic parameter, and b represents the service characteristic parameter. θ Calculate the QoE of the cell. Among them, the function f θ After obtaining the currently collected network characteristic parameters and service characteristic parameters, the currently collected network characteristic parameters and service characteristic parameters can be brought into the above-mentioned first calculation formula to calculate the QoE corresponding to each cell.

[0084] For example, taking the above network characteristic parameters including the signal quality RSRP and load quantity (Load) of the cell, and the service characteristic parameters including the service type (AppType) and service execution time (Time) as an example, the signal quality and / or load quantity of multiple cells detected by the terminal device, the service type and service execution time currently executed by the terminal device are uniformly divided according to each cell, and X is used. i represents the network characteristic parameters and service characteristic parameters of the currently collected i-th cell, then, Xi The four characteristic parameters (RSRPi, Loadi, AppType, Time) are included in the first calculation formula to obtain QoE=f θ (Xi)=f θ (RSRPi, Loadi, AppType, Time), and then calculate the QoE of each cell.

[0085] For example, the multiple cells currently detected are cell 1, cell 2, and cell 3. Then, X1 represents the network characteristic parameters and service characteristic parameters of cell 1, X2 represents the network characteristic parameters and service characteristic parameters of cell 2, and X3 represents the network characteristic parameters and service characteristic parameters of cell 3. Through the above division, it can be seen that X1 is (RSRP1, Load1, AppType, Time), X2 is (RSRP2, Load2, AppType, Time), and X3 is (RSRP3, Load3, AppType, Time). By substituting X1, X2, and X3 into the above first calculation formula, the QoE of cell 1, the QoE of cell 2, and the QoE of cell 3 can be calculated.

[0086] Optionally, in order to facilitate the calculation of the terminal device, the service execution time can be a time feature encoding value obtained by encoding the time. For example, the terminal device can encode the current time in the following encoding method:

[0087]

[0088] Among them, t represents the currently acquired time (such as 9 o'clock, 8:30, etc.), Time(t) represents the time feature coding value obtained after encoding t. In the above-collected service feature parameters, Time(t) is used instead of the service execution time.

[0089] Optionally, in the method of calculating QoE based on network characteristic parameters and service characteristic parameters, the network characteristic parameters and the service characteristic parameters may correspond to respective weights. For example, the network characteristic parameters correspond to network weights, and the service characteristic parameters correspond to service weights. For example, the first calculation formula above may be expressed as follows:

[0090] QoE = f θ (W1*a,W2*b);

[0091] Where W1 represents the network weight corresponding to the network characteristic parameter, and W2 represents the service weight corresponding to the service characteristic parameter. Before the terminal device calculates the QoE, it can determine the network weight and service weight corresponding to the service characteristic parameter based on the currently collected service characteristic parameter; and obtain the QoE of each cell based on the network characteristic parameter and network weight, service characteristic parameter and service weight. The process of the terminal device obtaining the QoE of each cell based on the network characteristic parameter and network weight, service characteristic parameter and service weight can be represented as QoE = f θ (W1*a,W2*b).

[0092] For example, different network weights and service weights are set in advance for different service feature parameters in the terminal device. Please refer to Table 1, which shows a correspondence table between a service feature parameter and a network weight and service weight involved in an exemplary embodiment of the present application.

[0093] Table 1

[0094] Service characteristic parameters Network weights Business weight Business type 1, time 1 Network weight 1 Business Weight 1 Business type 2, time 1 Network Weight 2 Business Weight 2 Business type 1, time 2 Network Weight Three Business Weight Three Business Type 2, Time 2 Network Weight Four Business Weight Four Business type three, time one Network Weight Five Business Weight Five Business type three, time two Network Weight Six Business Weight Six …… …… ……

[0095] As shown in Table 1, different network weights and service weights are set for the currently executed service type and service execution time collected by the terminal device. If the currently executed service type and service execution time are service type 1 and time 2 in Table 1, respectively, then, by querying Table 1, the network weight corresponding to the network characteristic parameter is network weight 3, and the service weight corresponding to the service characteristic parameter is service weight 3. That is, the value of W1 in the above formula adopts the value of network weight 3 in Table 1, and the value of W2 adopts the value of service weight 3 in Table 1.

[0096] It should be noted that the above W1 and W2 are exemplary. In actual applications, different weights with finer granularity can be set based on RSRP and Load in the network characteristic parameters, and AppType and Time in the service characteristic parameters can also be set with finer granularity and different weights, which are not limited here.

[0097] Step 403: quantify the uncertainty of the QoE of each cell to obtain the quantified result of the QoE of each cell.

[0098] Optionally, after the terminal device obtains the QoE of each cell, it may perform uncertainty quantification on the QoE of each cell and obtain a quantified result of the QoE of each cell.

[0099] Among them, the QoE is iteratively predicted using multiple neurons based on Monte Carlo Dropout (MC Dropout), thereby completing the process of quantifying the uncertainty of QoE. Optionally, the quantification result of the QoE of each cell can be represented by the prediction variance of the predicted value of the QoE of each cell obtained in the process of uncertainty quantification. That is, the terminal device can quantify the uncertainty of the experience quality QoE of each cell, obtain the prediction variance of the QoE of each cell, and use the prediction variance of the QoE of each cell as the quantification result obtained after the uncertainty quantification of the QoE of each cell.

[0100] For example, in this solution, the terminal device can perform N forward propagations using MC Dropout, where N is an integer. Then, for each cell's QoE, the terminal device obtains N iteration values ​​after N iterations. Based on these N iteration values ​​and the average value of each cell's QoE, the predicted variance of each cell's QoE is obtained.

[0101] For example, for the QoE of cell one, the terminal device will perform N forward propagations on the QoE of cell one through MC Dropout, obtain N iteration values ​​after N iterative operations (that is, the predicted N QoEs after the service execution time), and calculate the corresponding prediction variance based on these N iteration values ​​and the average value of the QoE of each cell.

[0102] Optionally, the prediction variance may be calculated based on a preset second calculation formula. For example, the second calculation formula is as follows:

[0103]

[0104] Where σi represents the prediction variance of the QoE calculation for the i-th cell. The value of n ranges from 1 to N. It represents the nth iteration value of the QoE of the i-th cell. The QoE of each cell is represented by the average value. Based on the second calculation formula, the predicted variance of the QoE of each cell after the uncertainty of the QoE of each cell is quantified can be obtained, and the predicted variance of the QoE of each cell is used as the quantified result of the QoE of each cell.

[0105] Of course, in one possible implementation, the terminal device can also calculate the average of the prediction results using the outputs obtained from multiple forward propagations. For example, based on N iterations of the QoE for each cell, the average of these N iterations can be calculated and used as the quantized QoE result for each cell. This solution does not limit the specific mathematical parameter used as the corresponding quantized result.

[0106] Step 404: Obtain a target comprehensive index for each cell based on the QoE of each cell and the uncertainty quantification result of the QoE of each cell. The target comprehensive index for each cell is used to indicate the degree of matching between the network characteristic parameters and the service characteristic parameters of each cell.

[0107] Optionally, after obtaining the QoE of each cell and the quantified result of uncertainty quantification of the QoE of each cell, the terminal device can further calculate a target comprehensive index for each cell, and use the target comprehensive index of each cell to indicate the degree of match between the network characteristic parameters and the service characteristic parameters of each cell. The higher the target comprehensive index, the higher the degree of match between the network characteristic parameters and the service characteristic parameters of each cell, and the closer the match between the network characteristic parameters and the service characteristic parameters of each cell.

[0108] In one possible implementation, the target comprehensive index is the sum of the mathematical expectation corresponding to the QoE of each cell and the quantification result of the uncertainty quantification of the QoE of each cell; the terminal device executes the process of obtaining the target comprehensive index of each cell based on the QoE of each cell and the quantification result of the uncertainty quantification of the QoE of each cell as follows: according to the QoE of each cell, obtain the mathematical expectation corresponding to the QoE of each cell; according to the mathematical expectation corresponding to the QoE of each cell and the quantification result of the uncertainty quantification of the QoE of each cell, obtain the target comprehensive index of each cell.

[0109] That is, the target comprehensive index is obtained by summing the mathematical expectation of each cell's QoE with the quantified QoE result for each cell. For example, for the QoE calculated for cell one, the mathematical expectation of the QoE for cell one is obtained, and then the mathematical expectation of the QoE for cell one is summed with the quantified QoE result obtained for cell one to obtain target comprehensive index one. This target comprehensive index one indicates the degree of match between the network characteristic parameters and service characteristic parameters of cell one.

[0110] For example, taking the above quantification result as an example based on the prediction variance representation, the terminal device can calculate the target comprehensive index based on the third calculation formula, which is as follows:

[0111] MP i =E[f θ (Xi)]+α*σi;

[0112] Among them, MPi represents the target comprehensive index calculated for the i-th cell, E[f θ (Xi)] represents f θwhere α represents the coefficient of the prediction variance, which can be preset by the developer in the terminal device (e.g., a constant). σi represents the prediction variance calculated for the QoE of the i-th cell. For each cell, the terminal device calculates the corresponding target comprehensive indicator.

[0113] Optionally, the process for the terminal device to obtain the mathematical expectation corresponding to the QoE of each cell based on the QoE of each cell can be as follows: obtaining the predicted probability corresponding to the predicted value of the QoE of each cell; obtaining the probability of the QoE of each cell based on the predicted probability corresponding to the predicted value of the QoE of each cell; and obtaining the mathematical expectation corresponding to the QoE of each cell based on the QoE of each cell and the probability of the QoE of each cell. In other words, in this solution, the probability of the QoE used in the process of calculating the mathematical expectation of the QoE of each cell is obtained based on the predicted probability corresponding to each predicted value in the above-mentioned uncertainty quantification process.

[0114] In one possible implementation, the above-mentioned method of obtaining the probability of the QoE of each cell based on the predicted probability corresponding to the predicted value of the QoE of each cell can be as follows: the terminal device obtains the average value of the predicted probabilities corresponding to the predicted value of the QoE of each cell, and uses the average value as the probability of the QoE of each cell. For example, after performing N iterative calculations on the QoE of the i-th cell, the predicted probabilities corresponding to the N iterative values ​​(i.e., the predicted values ​​of each predicted QoE) are obtained as [P1, P2...PN]. The terminal device can obtain the average value corresponding to [P1, P2...PN], which is the probability of the QoE of the i-th cell. The mathematical expectation calculation method is used to calculate the mathematical expectation corresponding to the QoE of the i-th cell. Of course, in this process, the maximum or minimum value corresponding to each predicted probability can also be used as the probability of the QoE of the i-th cell. The specific method for calculating the probability of the QoE of the i-th cell is not limited here.

[0115] In addition, the above-mentioned method of calculating the target comprehensive index based on the sum of the mathematical expectation corresponding to the cell's QoE and the quantization result is also exemplary. In actual applications, the cell's QoE can also be used directly, or a preset coefficient can be added to the cell's QoE to calculate the target comprehensive index, or other mathematical parameters of the cell's QoE can be calculated to calculate the target comprehensive index. This solution is not limited to this.

[0116] Step 405 : Select a target cell from each cell according to the target comprehensive index of each cell. The target cell is the cell corresponding to the maximum value of the target comprehensive index of each cell.

[0117] In one possible implementation, the process of selecting a target cell from each cell by a terminal device based on the target comprehensive index of each cell can be as follows: based on the target comprehensive index of each cell, obtaining the target QoE corresponding to the maximum value of the target comprehensive index of each cell; based on the target QoE, obtaining the target network characteristic parameters and target service characteristic parameters used to calculate the target QoE; and selecting the cell corresponding to the target network characteristic parameters and target service characteristic parameters as the target cell. In other words, for each of the target comprehensive indicators obtained above, the terminal device can filter out the maximum target comprehensive index, obtain the target QoE corresponding to the maximum target comprehensive index, determine the target network characteristic parameters and target service characteristic parameters used for the target QoE based on the target QoE, and select the cell corresponding to the target network characteristic parameters and target service characteristic parameters as the target cell.

[0118] For example, the network characteristic parameters collected by the terminal device include multiple cells currently detected, namely cell one, cell two, and cell three. X1 represents the network characteristic parameters and service characteristic parameters of cell one, X2 represents the network characteristic parameters and service characteristic parameters of cell two, and X3 represents the network characteristic parameters and service characteristic parameters of cell three. Through the above process, the QoE of cell one, the QoE of cell two, and the QoE of cell three can be obtained, as well as the quantified results of the QoE of cell one, the quantified results of the QoE of cell two, and the quantified results of the QoE of cell three. In this step, the target comprehensive index of cell one, the target comprehensive index of cell two, and the target comprehensive index of cell three are calculated in the above manner. In order to screen out the target cell, the terminal device can compare the target comprehensive index of cell one, the target comprehensive index of cell two, and the target comprehensive index of cell three, and select the largest one. If the target comprehensive index of cell one is the largest, then the terminal device can determine the QoE of cell one corresponding to the target comprehensive index of cell one based on the target comprehensive index of cell one, and then obtain the target network characteristic parameters and target service characteristic parameters (i.e., X1) for calculating the QoE of cell one, and use cell one corresponding to X1 as the target cell.

[0119] In one possible implementation, a terminal device is provided with an objective function, the independent variable of which is the above-mentioned target comprehensive index, which is determined based on the mathematical expectation of the QoE of the cell and the quantified result of the cell. The objective function represents the mathematical expectation of the QoE of the cell and the quantified result of the cell that are used to obtain the maximum value of the target comprehensive index. The objective function is as follows:

[0120] y=arg max(E[f θ (Xi)]+α*σi);

[0121] From the above objective function, we can know that it is to obtain the θ (Xi)] + α*σi takes the maximum value, and the cell corresponding to Xi is finally determined as the target cell. That is, the terminal device can substitute the QoE of each cell and the quantized result of the QoE of each cell into the objective function and perform the operation to obtain the final Xi (including the target network characteristic parameters and the target service characteristic parameters), and then use the cell corresponding to Xi as the target cell.

[0122] In one possible implementation, the process of executing steps 402 to 405 by the terminal device in this solution can be performed by a pre-trained cell selection model. For example, the terminal device includes a cell selection model, which is used to obtain the quality of experience (QoE) of each cell and the quantification result of uncertainty quantification of the QoE of each cell based on network characteristic parameters and service characteristic parameters, and select a target cell from each cell based on the QoE of each cell and the quantification result of uncertainty quantification of the QoE of each cell; the cell selection model is trained based on historical characteristic data of the terminal device and historical target cells, and the historical characteristic data includes historical network characteristic parameters and historical service characteristic parameters; after obtaining the currently collected network characteristic parameters and service characteristic parameters, the currently collected network characteristic parameters and service characteristic parameters can be input into the cell selection model, and the subsequent process of steps 402 to 405 is executed.

[0123] For example, the terminal device can collect and record network characteristic parameters and service characteristic parameters in real time, and the developer or user can independently calibrate the optimal target cell. When model training is required, training is performed based on the recorded historical network characteristic parameters and historical service characteristic parameters, as well as the historical target cells, to obtain a cell selection model, so that the terminal device inputs the currently collected network characteristic parameters and service characteristic parameters into the cell selection model, and then outputs the target cell selected by the calculation. For example, the historical network characteristic parameters and historical service characteristic parameters are the network characteristic parameters and service characteristic parameters collected by the terminal device in the previous few days. The terminal device can send these data to the server, and the developer can calibrate the corresponding historical target cells for these historical network characteristic parameters and historical service characteristic parameters, and use these data for training, and then deploy the trained cell selection model to the terminal device. Of course, the terminal device can also be trained locally, and this application does not limit the specific model training process.

[0124] In one possible implementation, the terminal device includes cell selection models corresponding to different location areas. Please refer to Table 2, which shows a correspondence table between location areas and cell selection models involved in an exemplary embodiment of the present application.

[0125] Table 2

[0126] Location Area Cell selection model Location Area 1 Cell selection model 1 Location Area 2 Cell selection model 2 Location Area Three Cell selection model 3 …… ……

[0127] As shown in Table 2, different location areas correspond to different cell selection models.

[0128] In actual applications, the terminal device is located in different locations, and the names of multiple cells that can be detected around it and the services provided by each cell are generally different. In this solution, a cell selection model corresponding to the location area can be trained based on the historical feature data of different location areas and stored in the terminal device.

[0129] Optionally, taking the above model training executed in the terminal device as an example, during the process of training the cell selection model, the terminal device can also divide the historical feature data of the terminal device according to the location area of ​​the terminal device; and train based on the divided historical feature data and the historical target cells to obtain the cell selection model corresponding to each location area. For example, please refer to Figure 5 , which shows a schematic diagram of a location area division including various cells involved in an exemplary embodiment of the present application. Figure 5 As shown, it includes location area one 501 and location area two 502. When the terminal device is in location area one, the multiple cells it usually detects are cell set one (base station A, base station B and base station C). When the terminal device is in location area two, the multiple cells it usually detects are cell set two (base station D, base station E). The terminal device divides the historical feature data based on the location area to which each cell in the historical feature data belongs, and obtains the historical feature data under each location area. For example, if the network feature parameters in a certain historical feature data are the RSRP and Load of base station A, then the historical feature data belongs to location area one 501 to which base station A belongs. In this way, the historical feature data are divided, and training is performed based on the historical feature data under each location area and the corresponding historical target cell to obtain the cell selection model corresponding to each location area. It is stored in the terminal device in the manner of Table 2 above.

[0130] Optionally, when the terminal device inputs the currently collected network characteristic parameters and service characteristic parameters into the cell selection model, the currently collected network characteristic parameters and service characteristic parameters are input into the cell selection model corresponding to the location area to which the terminal device is currently located. That is, the terminal device can first obtain the area to which the terminal device is currently located, determine the location area to which the current area belongs, and then select the cell selection model corresponding to the location area to execute the process of this solution. For example, if the location area to which the current area belongs is location area two, by querying Table 2, it can be obtained that the cell selection model to be used is cell selection model two. When executing the subsequent process, the currently collected network characteristic parameters and service characteristic parameters can be input into cell selection model two, thereby realizing the selection of the target cell.

[0131] In one possible implementation, the cell selection model is divided into different modules, and the above steps are executed based on different modules. Figure 6 , which shows a schematic diagram of the structure of a cell selection model involved in an exemplary embodiment of the present application. Figure 6 As shown, the cell selection model 600 includes the following modules: a QoE calculation module 601 and a decision execution module 602. The QoE calculation module 601 includes a time coding unit for encoding the collected service execution time, and a calculation unit for calculating the QoE using the currently collected network characteristic parameters and service characteristic parameters. The time coding unit can be encoded using the coding method shown in the above-mentioned Time(t), and the calculation unit for calculating the QoE can be implemented using the first calculation formula in the book "Flower Girl". The decision execution module 602 includes an MC Dropout quantization algorithm for quantifying the uncertainty of the calculated QoE and obtaining the prediction variance, and a decision rule for target selection based on the calculated QoE and the prediction variance. The MC Dropout quantization algorithm can be calculated using the above-mentioned second calculation formula, and the decision rule can be implemented using the above-mentioned objective function.

[0132] Optionally, in an implementation method in which a cell selection model is included in a terminal device, after the cell selection model is trained and applied to the terminal device, the trained cell selection model can be compressed, for example, by using channel pruning to reduce the number of neural network parameters.

[0133] The terminal device includes the above Figure 6Taking the cell selection model shown as an example, when the terminal device executes the process of this solution, in step 401, the currently collected network characteristic parameters are obtained. The signal quality of each cell is: 5G1 RSRP: -78dBm, 4G1 RSRP: -85dBm, 4G2 RSRP: -88dBm, and the load proportion of each cell is: 5G1: 75%, 4G1: 60%, 4G2: 45%. The application currently running on the terminal device is application 1, and its type is a video application. Then, the currently executed service type obtained is video, and the service execution time is 12 noon. These parameters are input into the cell selection model, and based on the cell selection model, it is finally calculated that 4G2 is the target cell. The terminal device can establish a network connection with the 4G2 cell.

[0134] In one possible implementation, after a terminal device selects a target cell from each cell, the terminal device may further perform the following steps: establishing a network connection with the target cell; if a network anomaly is detected for the terminal device, retraining the cell selection model based on network characteristic parameters and service characteristic parameters collected within a preset time period; or updating the model parameters of the cell selection model. Specifically, for the trained cell selection model, after the terminal device establishes a network connection based on the selected target cell, the terminal device performs real-time monitoring of the terminal device. If a network anomaly is detected for the terminal device, the cell selection model is promptly retrained or the model parameters in the cell selection model are updated.

[0135] Among them, the network anomaly in this solution may refer to a scenario where the network is interrupted, the load of the cell suddenly changes, or the QoS requirement of the currently executed service suddenly changes from that of the network. Among them, the load of the cell suddenly changes from a low load situation to a high load situation, and the QoS requirement of the network suddenly changes from a low load situation to a high load situation after the currently executed service is switched. The terminal device monitors the above situations in real time. Once a network anomaly is found, the cell selection model can be retrained based on the network characteristic parameters and service characteristic parameters collected within a preset time period (such as the last 3 days, 1 day, or 8 hours, etc.); or, the model parameters of the cell selection model can be updated to ensure the accuracy of the cell selection model.

[0136] Optionally, the process of triggering the terminal device to select the optimal cell in this solution may be that the terminal device detects that the network provided by the cell to which it is currently connected cannot meet the service requirements, or that the signal quality of the cell to which it is currently connected is low (for example, the monitored RSRP is less than a preset RSRP threshold value), etc., thereby triggering the terminal device to execute the above Figure 4The method shown enables the terminal device to access the optimal cell (target cell) in the current surrounding area.

[0137] It should be noted that the above solution can also be applied to vehicle terminals, integrating vehicle speed, trajectory, and other characteristics into service characteristic parameters to optimize the V2X link selection process involved in the vehicle terminal. In the industrial Internet of Things, service criticality can also be incorporated into service characteristic parameters to implement hierarchical access control processes.

[0138] To summarize, the terminal device obtains the currently collected network characteristic parameters and service characteristic parameters, where the network characteristic parameters include at least the signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters include at least the service type currently executed by the terminal device and the service execution time; based on the currently collected network characteristic parameters and service characteristic parameters, the terminal device obtains the quality of experience QoE of each cell and the quantitative results of uncertainty quantification of the QoE of each cell, and the uncertainty quantification is used to obtain the predicted value of the QoE of each cell after the service execution time; based on the QoE of each cell and the quantitative results of uncertainty quantification of the QoE of each cell, the target cell is selected from each cell. In the present application, the parameters used in selecting the target cell include network characteristic parameters and service characteristic parameters, and based on the network characteristic parameters and service characteristic parameters, the QoE of each cell is obtained, and the uncertainty of the QoE of each cell is quantified, and the predicted value of the QoE of each cell after the service execution time is obtained to obtain a quantified result. Combined with the QoE and quantification results of each cell, the target cell is selected from each cell. The entire process not only takes into account multiple network characteristic parameters of the base station, but also takes into account the service type and service execution time currently executed by the terminal device. Based on the uncertainty quantification, a context-aware quantification result is obtained, so that the determined target cell is more in line with the service type currently executed by the terminal device, thereby improving the accuracy of determining the target cell and increasing the stability of the network connection of the terminal device.

[0139] In addition, in this solution, the terminal device can collect multiple features such as RSRP, cell load, APP service type, service execution time, etc. in real time, and generate a joint vector containing spatiotemporal-service features (for example, the joint vector is (RSRP1, Load1, AppType, Time)). The principle of context slot machine is combined with the neural network to optimize the acquisition process of multi-dimensional nonlinear QoE calculation, so that multi-modal features can be more deeply integrated, thereby improving the accuracy of the results of the cell selection model. Moreover, the network weight corresponding to the network feature parameters is determined based on the service feature parameters, while meeting differentiated service needs, the key correlation factors (network weight and service weight) are automatically captured to improve the accuracy of the calculated QoE. Through deep reinforcement learning and multi-dimensional feature fusion, the intelligent network selection performance of terminal devices in static scenarios is significantly improved, providing a core basic technology for smart connections in the 5G-A / 6G era.

[0140] In addition, this solution promptly detects whether a network anomaly occurs in the terminal device, and retrains or updates the parameters of the cell selection model in the event of a network anomaly, ensuring that the cell selection model is more accurate and reliable.

[0141] The following are device embodiments of the present application, which can be used to implement the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0142] Please refer to Figure 7 , which shows a structural block diagram of a cell selection device provided by an exemplary embodiment of the present application. The cell selection device 700 can be used in a terminal device to execute all or part of the steps performed by the terminal device in the methods provided in the above-mentioned embodiments. The cell selection device 700 includes:

[0143] A first acquisition module 701 is configured to acquire currently collected network characteristic parameters and service characteristic parameters, wherein the network characteristic parameters include at least signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters include at least a service type and service execution time currently executed by the terminal device;

[0144] A second acquisition module 702 is configured to acquire, based on the currently collected network characteristic parameters and the service characteristic parameters, a quality of experience (QoE) of each cell and a quantification result of uncertainty quantification of the QoE of each cell, wherein the uncertainty quantification is used to obtain a predicted value of the QoE of each cell after the service execution time;

[0145] The cell selection module 703 is configured to select a target cell from each of the cells according to the QoE of each of the cells and a quantification result of uncertainty quantification of the QoE of each of the cells.

[0146] To summarize, the terminal device obtains the currently collected network characteristic parameters and service characteristic parameters, where the network characteristic parameters include at least the signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters include at least the service type currently executed by the terminal device and the service execution time; based on the currently collected network characteristic parameters and service characteristic parameters, the terminal device obtains the quality of experience QoE of each cell and the quantitative results of uncertainty quantification of the QoE of each cell, and the uncertainty quantification is used to obtain the predicted value of the QoE of each cell after the service execution time; based on the QoE of each cell and the quantitative results of uncertainty quantification of the QoE of each cell, the target cell is selected from each cell. In the present application, the parameters used in selecting the target cell include network characteristic parameters and service characteristic parameters, and based on the network characteristic parameters and service characteristic parameters, the QoE of each cell is obtained, and the uncertainty of the QoE of each cell is quantified, and the predicted value of the QoE of each cell after the service execution time is obtained to obtain a quantified result. Combined with the QoE and quantification results of each cell, the target cell is selected from each cell. The entire process not only takes into account multiple network characteristic parameters of the base station, but also takes into account the service type and service execution time currently executed by the terminal device. Based on the uncertainty quantification, a context-aware quantification result is obtained, so that the determined target cell is more in line with the service type currently executed by the terminal device, thereby improving the accuracy of determining the target cell and increasing the stability of the network connection of the terminal device.

[0147] Optionally, the cell selection module includes: a first acquisition unit and a first selection unit;

[0148] The first acquisition unit is configured to acquire a target comprehensive indicator for each cell based on the QoE of each cell and a quantification result of uncertainty quantification of the QoE of each cell, where the target comprehensive indicator for each cell is used to indicate a degree of matching between the network characteristic parameters of each cell and the service characteristic parameters;

[0149] The first selection unit is configured to select a target cell from each of the cells according to the target comprehensive index of each cell, where the target cell is a cell corresponding to a maximum value among the target comprehensive indexes of each cell.

[0150] Optionally, the first selection unit is further configured to:

[0151] Obtaining, according to the target comprehensive indicator of each cell, a target QoE corresponding to a maximum value of the target comprehensive indicator of each cell;

[0152] According to the target QoE, obtaining target network characteristic parameters and target service characteristic parameters used to calculate the target QoE;

[0153] A cell corresponding to the target network characteristic parameter and the target service characteristic parameter is selected as the target cell.

[0154] Optionally, the target comprehensive indicator is the sum of a mathematical expectation corresponding to the QoE of each cell and a quantified result of uncertainty quantification corresponding to the QoE of each cell;

[0155] The first acquiring unit is further configured to:

[0156] Obtaining a mathematical expectation of the QoE of each cell according to the QoE of each cell;

[0157] The target comprehensive index of each cell is obtained according to the mathematical expectation corresponding to the QoE of each cell and the quantification result of uncertainty quantification of the QoE of each cell.

[0158] Optionally, obtaining, according to the QoE of each cell, a mathematical expectation corresponding to the QoE of each cell includes:

[0159] Obtaining a predicted probability corresponding to the predicted value of the QoE of each cell;

[0160] Obtaining a probability of the QoE of each cell according to a predicted probability corresponding to the predicted value of the QoE of each cell;

[0161] A mathematical expectation corresponding to the QoE of each cell is obtained according to the QoE of each cell and the probability of the QoE of each cell.

[0162] Optionally, the second acquisition module 702 includes: a second acquisition unit and a third acquisition unit;

[0163] The second acquiring unit is configured to acquire the QoE of each cell based on the currently collected network characteristic parameters and the service characteristic parameters;

[0164] The third acquisition unit is configured to quantify the uncertainty of the quality of experience (QoE) of each cell and obtain a prediction variance of the QoE of each cell, where the prediction variance of the QoE of each cell is used to indicate a quantization result obtained after the uncertainty quantification of the QoE of each cell.

[0165] Optionally, the third acquiring unit is further configured to:

[0166] For the QoE of each cell, obtain N iteration values ​​obtained after N iterations, where N is an integer;

[0167] A prediction variance of the QoE of each cell is obtained according to the N iterative values ​​of the QoE of each cell and an average value of the QoE of each cell.

[0168] Optionally, the second acquiring unit is further configured to:

[0169] Determining, based on the currently collected service characteristic parameters, a network weight and a service weight corresponding to the service characteristic parameters;

[0170] The QoE of each cell is obtained according to the network characteristic parameters and the network weight, the service characteristic parameters and the service weight.

[0171] Optionally, the terminal device includes a cell selection model, where the cell selection model is used to obtain the quality of experience (QoE) of each cell and a quantification result of uncertainty quantification of the QoE of each cell based on the network characteristic parameter and the service characteristic parameter, and select a target cell from each cell based on the QoE of each cell and the quantification result of uncertainty quantification of the QoE of each cell; the cell selection model is trained based on historical characteristic data of the terminal device and historical target cells, where the historical characteristic data includes historical network characteristic parameters and historical service characteristic parameters;

[0172] The device further comprises:

[0173] The first input module is used to input the currently collected network characteristic parameters and the service characteristic parameters into the cell selection model.

[0174] Optionally, the device further includes:

[0175] A first division module is configured to divide the historical feature data of the terminal device according to the location area of ​​the terminal device;

[0176] A third acquisition module is used to perform training based on the divided historical feature data and the historical target cells to obtain a cell selection model corresponding to each location area;

[0177] The first input module is used to:

[0178] The currently collected network characteristic parameters and the service characteristic parameters are input into a cell selection model corresponding to the location area to which the terminal device is currently located.

[0179] Optionally, the device further includes:

[0180] a cell connection module, configured to establish a network connection with the target cell after selecting the target cell from each of the cells;

[0181] The model update module is used to retrain the cell selection model based on the network characteristic parameters and service characteristic parameters collected within a preset time period when a network anomaly is detected in the terminal device; or to update the model parameters of the cell selection model.

[0182] Please refer to Figure 8 , which is a schematic diagram of another example of a cell selection device provided in an embodiment of the present application. The cell selection device 800 may be a terminal device capable of implementing the functions of the method provided in an embodiment of the present application. The cell selection device 800 may be a chip system. In the embodiment of the present application, the chip system may be composed of a chip, or may include a chip and other discrete devices.

[0183] In terms of hardware implementation, the above-mentioned communication module may be a transceiver, which is integrated into the cell selection device 800 to form the communication interface 803 .

[0184] The cell selection device 800 includes at least one processor 801, which is used to implement or support the cell selection device 800 to implement the functions of the terminal device in the method provided in the embodiment of the present application. Exemplarily, the processor 801 can obtain the currently collected network characteristic parameters and service characteristic parameters; based on the currently collected network characteristic parameters and service characteristic parameters, obtain the quality of experience (QoE) of each cell and the quantitative results of the uncertainty quantification of the QoE of each cell; based on the QoE of each cell and the quantitative results of the uncertainty quantification of the QoE of each cell, select the target cell from each cell, etc. For details, please refer to the detailed description in the method example, which will not be repeated here.

[0185] The cell selection device 800 may further include at least one memory 802 for storing program instructions and / or data. The memory 802 is coupled to the processor 801. Coupling in the embodiments of the present application refers to an indirect coupling or communication connection between devices, units, or modules, which may be electrical, mechanical, or other forms, and is used for information exchange between the devices, units, or modules. The processor 801 may operate in conjunction with the memory 802. The processor 801 may execute program instructions stored in the memory 802. At least one of the at least one memory may be included in the processor.

[0186] Cell selection apparatus 800 may further include a communication interface 803 for communicating with other devices via a transmission medium, thereby enabling the apparatus in cell selection apparatus 800 to communicate with the other devices. For example, the other devices may be network-side devices. Processor 801 may utilize communication interface 803 to transmit and receive data. Communication interface 803 may specifically be a transceiver.

[0187] The specific connection medium between the communication interface 803, the processor 801 and the memory 802 is not limited in the embodiment of the present application. Figure 8 In the embodiment, the memory 802, the processor 801 and the communication interface 803 are connected via a bus 804. Figure 8 The connections between the other components are shown in bold lines, which are only for illustration and not intended to be limiting. The bus can be divided into address bus, data bus, control bus, etc. Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0188] In the embodiments of the present application, the processor 801 may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and may implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present application may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0189] In an embodiment of the present application, the memory 802 may be a non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), or may be a volatile memory (volatile memory), such as a random-access memory (RAM). A memory is any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory in the embodiment of the present application may also be a circuit or any other device that can implement a storage function, for storing program instructions and / or data.

[0190] Optionally, an embodiment of the present application also provides a terminal device, which includes a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, it implements all or part of the steps performed by the terminal device in the cell selection method of each of the above embodiments.

[0191] Optionally, an embodiment of the present application further provides a computer-readable medium on which a computer program is stored. When the computer program is executed by a processor, it implements all or part of the steps performed by the terminal device in the cell selection method of each of the above embodiments.

[0192] Optionally, an embodiment of the present application further provides a chip, which contains an executable computer program. When the chip executes the computer program, it implements all or part of the steps performed by the terminal device in the cell selection method of each of the above embodiments.

[0193] Optionally, an embodiment of the present application further provides a computer program product, which, when executed on a computer, enables the computer to execute all or part of the steps of the cell selection method of each of the above embodiments, which are executed by the terminal device.

[0194] Optionally, an embodiment of the present application also provides an application publishing platform, which is used to publish a computer program product, wherein when the computer program product runs on a computer, the computer executes the cell selection method of each embodiment as described above, and all or part of the steps executed by the terminal device.

[0195] It should be noted that the apparatus provided in the above embodiments, when performing control of a terminal device, is illustrated only by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0196] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0197] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.

[0198] The above description is merely an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A cell selection method, characterized in that: Applied to a terminal device, the method includes: Acquire currently collected network characteristic parameters and service characteristic parameters, the network characteristic parameters including at least signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters including at least a service type and service execution time currently executed by the terminal device; Obtaining, based on the currently collected network characteristic parameters and the service characteristic parameters, a quality of experience (QoE) of each cell and a quantification result of uncertainty quantification of the QoE of each cell, wherein the uncertainty quantification is used to obtain a predicted value of the QoE of each cell after the service execution time; A target cell is selected from each of the cells according to the QoE of each cell and a quantification result of uncertainty quantification of the QoE of each cell.

2. The method according to claim 1, characterized in that The selecting a target cell from each cell according to the QoE of each cell and a quantization result of uncertainty quantification of the QoE of each cell includes: Obtaining a target comprehensive indicator for each cell according to the QoE of each cell and a quantified result of uncertainty quantification of the QoE of each cell, where the target comprehensive indicator for each cell is used to indicate a degree of matching between a network characteristic parameter of each cell and the service characteristic parameter; A target cell is selected from each of the cells according to the target comprehensive index of each cell, where the target cell is a cell corresponding to a maximum value among the target comprehensive indexes of each cell.

3. The method according to claim 2, characterized in that The selecting a target cell from each of the cells according to the target comprehensive index of each of the cells includes: Obtaining, according to the target comprehensive indicator of each cell, a target QoE corresponding to a maximum value of the target comprehensive indicator of each cell; According to the target QoE, obtaining target network characteristic parameters and target service characteristic parameters used to calculate the target QoE; A cell corresponding to the target network characteristic parameter and the target service characteristic parameter is selected as the target cell.

4. The method according to claim 2, characterized in that The target comprehensive indicator is the sum of the mathematical expectation corresponding to the QoE of each cell and the quantified result of uncertainty quantification corresponding to the QoE of each cell; The obtaining, based on the QoE of each cell and a quantified result of uncertainty quantification of the QoE of each cell, a target comprehensive indicator for each cell includes: Obtaining a mathematical expectation of the QoE of each cell according to the QoE of each cell; The target comprehensive index of each cell is obtained according to the mathematical expectation corresponding to the QoE of each cell and the quantification result of uncertainty quantification of the QoE of each cell.

5. The method according to claim 4, characterized in that The obtaining, according to the QoE of each cell, a mathematical expectation corresponding to the QoE of each cell includes: Obtaining a predicted probability corresponding to the predicted value of the QoE of each cell; Obtaining a probability of the QoE of each cell according to a predicted probability corresponding to the predicted value of the QoE of each cell; A mathematical expectation corresponding to the QoE of each cell is obtained according to the QoE of each cell and the probability of the QoE of each cell.

6. The method according to claim 1, wherein The obtaining, based on the currently collected network characteristic parameters and the service characteristic parameters, the quality of experience (QoE) of each cell and a quantification result of uncertainty quantification of the QoE of each cell includes: Obtaining the QoE of each cell according to the currently collected network characteristic parameters and the service characteristic parameters; Uncertainty quantification is performed on the quality of experience (QoE) of each cell to obtain a prediction variance of the QoE of each cell, where the prediction variance of the QoE of each cell is used to indicate a quantization result obtained after uncertainty quantification is performed on the QoE of each cell.

7. The method according to claim 6, characterized in that The step of quantifying uncertainty of the quality of experience (QoE) of each cell to obtain a predicted variance of the QoE of each cell includes: For the QoE of each cell, obtain N iteration values ​​obtained after N iterations, where N is an integer; A prediction variance of the QoE of each cell is obtained according to the N iterative values ​​of the QoE of each cell and an average value of the QoE of each cell.

8. The method according to claim 6, characterized in that The obtaining, according to the currently collected network characteristic parameters and the service characteristic parameters, the QoE of each cell includes: Determining, based on the currently collected service characteristic parameters, a network weight and a service weight corresponding to the service characteristic parameters; The QoE of each cell is obtained according to the network characteristic parameters and the network weight, the service characteristic parameters and the service weight.

9. The method according to any one of claims 1 to 8, characterized in that: The terminal device includes a cell selection model, where the cell selection model is used to obtain the quality of experience (QoE) of each cell and a quantification result of uncertainty quantification of the QoE of each cell based on the network characteristic parameter and the service characteristic parameter, and select a target cell from each cell based on the QoE of each cell and the quantification result of uncertainty quantification of the QoE of each cell; the cell selection model is trained based on historical characteristic data of the terminal device and historical target cells, where the historical characteristic data includes historical network characteristic parameters and historical service characteristic parameters; The method further comprises: The currently collected network characteristic parameters and service characteristic parameters are input into the cell selection model.

10. The method according to claim 9, characterized in that The method further comprises: Dividing the historical characteristic data of the terminal device according to the location area of ​​the terminal device; Training is performed based on the divided historical feature data and the historical target cells to obtain a cell selection model corresponding to each location area; The inputting the currently collected network characteristic parameters and the service characteristic parameters into the cell selection model includes: The currently collected network characteristic parameters and the service characteristic parameters are input into a cell selection model corresponding to the location area to which the terminal device is currently located.

11. The method according to claim 9, characterized in that After selecting a target cell from each of the cells, the method further includes: Establishing a network connection with the target cell; When a network anomaly is detected in the terminal device, the cell selection model is retrained based on the network characteristic parameters and service characteristic parameters collected within a preset time period; or, the model parameters of the cell selection model are updated.

12. A cell selection device, characterized in that: Applied to a terminal device, the device includes: a first acquisition module, configured to acquire currently collected network characteristic parameters and service characteristic parameters, the network characteristic parameters including at least signal quality and / or load quantity of multiple cells detected by the terminal device, and the service characteristic parameters including at least a service type and service execution time currently executed by the terminal device; a second acquisition module, configured to acquire, based on the currently collected network characteristic parameters and the service characteristic parameters, a quality of experience (QoE) of each cell and a quantification result of uncertainty quantification of the QoE of each cell, wherein the uncertainty quantification is used to obtain a predicted value of the QoE of each cell after the service execution time; The cell selection module is configured to select a target cell from each of the cells according to the QoE of each of the cells and a quantification result of uncertainty quantification of the QoE of each of the cells.

13. A terminal device, characterized in that: The terminal device includes a memory and a processor, the memory stores a computer program that can be run on the processor, and the processor implements the cell selection method according to any one of claims 1 to 11 when executing the computer program.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the cell selection method according to any one of claims 1 to 11 is implemented.

Images