Perception processing method, perception integrated system, electronic equipment and storage medium

Abstract

The invention provides a sensing processing method, a sensing integrated system, electronic equipment and a storage medium, relates to the technical field of communication, and is used for improving the sensing data processing accuracy of the sensing integrated system based on artificial intelligence and the convenience of data transmission between modules. The method is applied to a perception function entity, and comprises the following steps: acquiring a trained artificial intelligence model from a perception server; processing the sensing data based on an artificial intelligence model to obtain a sensing result; and sending the artificial intelligence perception result to the core network element.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technologies, and in particular, to a sensing processing method, a sensing integration system, an electronic device, and a storage medium. Background Art

[0002] With the wide application of the fifth-generation (5G) communication system, while 5G meets the needs of individual users, it is gradually penetrating into various industries and fields of society, thereby realizing the upgrade from consumption to industrial applications. Among them, communication and sensing integration is based on the existing network architecture and communication waveforms, and uses existing communication devices to implement sensing applications, which promotes the development of innovative application services and is an important research direction of 5G.

[0003] In the currently applied sensing functions, most use traditional algorithms for sensing. As the network scale becomes larger, the types of services increase, and terminal devices are complex and diverse, it is becoming increasingly difficult for technicians to optimize sensing algorithms. At the same time, with the development of software and hardware resources, artificial intelligence (AI) has also been gradually widely applied in communication systems, but the communication and sensing integration structure based on artificial intelligence has not been effectively developed and cannot be applied to complex and changing scenarios and applications, resulting in poor intelligent management capabilities. Summary of the Invention

[0004] Embodiments of the present disclosure provide a sensing processing method, a sensing integration system, an electronic device, and a storage medium, which are used to improve the accuracy of processing sensing data and the convenience of data transmission between modules in a communication and sensing integration system based on artificial intelligence.

[0005] In a first aspect, a sensing processing method is provided, which is applied to a sensing function entity. The method includes:

[0006] Obtain a trained artificial intelligence model from a sensing server;

[0007] Process sensing data based on the artificial intelligence model to obtain a sensing result;

[0008] Send the artificial intelligence sensing result to a core network element.

[0009] Based on the perception processing method provided by the embodiments of the present disclosure, based on the existing core network architecture, by obtaining the trained artificial intelligence model from a perception server with high storage capacity, the problem of large data volume processed by the core network caused by training the artificial intelligence model in the core network is avoided, and the speed of model acquisition is improved; at the same time, based on the trained artificial intelligence model to process the perception data, the obtained perception result is more accurate, and the automatic processing of the perception data realizes the intelligent management of the perception data, reducing the labor input; finally, the obtained perception result is sent to the core network element, enabling data sharing among the various modules in the system and realizing the integration of perception and communication.

[0010] In a second aspect, a perception processing method is provided, which is applied to a perception server. The artificial intelligence method includes:

[0011] Training an artificial intelligence model, where the artificial intelligence model is used to process perception data to obtain a perception result;

[0012] Sending the trained artificial intelligence model to a perception function entity.

[0013] Based on the perception processing method provided by the embodiments of the present disclosure, the artificial intelligence model is trained by a perception server with high storage capacity and low computing power requirements, and the training efficiency is high; then, based on the trained artificial intelligence model to process the perception data, the obtained perception result is more accurate, and the automatic processing of the perception data realizes the intelligent management of the perception data, reducing the labor input; finally, the trained artificial intelligence model is sent to the perception function entity, enabling the perception function entity to predict the perception data based on the trained artificial intelligence model and realizing data sharing among the various modules in the system.

[0014] In a third aspect, a communication and sensing integrated system is provided, including:

[0015] A perception server, which is used to train an artificial intelligence model;

[0016] A perception function entity, which is used to obtain the trained artificial intelligence model from the artificial intelligence perception server; process the perception data based on the artificial intelligence model to obtain a perception result, and send the artificial intelligence perception result to the core network element.

[0017] In a fourth aspect, a communication device is provided, including: a memory and a processor; the memory and the processor are coupled; the memory is used to store a computer program; when the processor executes the computer program, the perception processing method of any of the above embodiments is implemented.

[0018] In a fifth aspect, a computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the perception processing method of any of the above embodiments is implemented.

[0019] In a sixth aspect, there is provided a computer program product, which includes computer program instructions that, when executed by a processor, implement the perception processing method of any of the above embodiments.

[0020] For the specific descriptions of the third aspect to the sixth aspect and their various implementation manners in the present disclosure, reference may be made to the detailed descriptions in the first aspect, the second aspect and their various implementation manners; and, for the beneficial effects of the third aspect to the sixth aspect and their various implementation manners, reference may be made to the analysis of the beneficial effects in the first aspect, the second aspect and their various implementation manners, which will not be elaborated herein. Description of the Drawings

[0021] To more clearly illustrate the technical solutions in the present disclosure, the drawings required for use in some embodiments of the present disclosure will be briefly introduced below. Obviously, the drawings in the following description are only the drawings of some embodiments of the present disclosure, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0022] Figure 1 A schematic diagram of a perception scenario provided for some embodiments of the present disclosure;

[0023] Figure 2 A schematic diagram of the architecture of a communication and sensing integrated system provided for some embodiments of the present disclosure;

[0024] Figure 3 A schematic flowchart of a model training method provided for some embodiments of the present disclosure;

[0025] Figure 4 A schematic flowchart of a perception processing method provided for some embodiments of the present disclosure;

[0026] Figure 5 A schematic flowchart of another perception processing method provided for some embodiments of the present disclosure;

[0027] Figure 6 A schematic diagram of a range-Doppler map provided for some embodiments of the present disclosure;

[0028] Figure 7 A schematic diagram of the corresponding relationship of feature data provided for some embodiments of the present disclosure;

[0029] Figure 8 A schematic diagram of the architecture of a convolutional neural network provided for some embodiments of the present disclosure;

[0030] Figure 9 A schematic diagram of the structure of a long short-term memory neural network model provided for some embodiments of the present disclosure;

[0031] Figure 10 Schematic diagram of another long short - term memory neural network model provided by some embodiments of the present disclosure;

[0032] Figure 11 Flow diagram of another perception processing method provided by some embodiments of the present disclosure;

[0033] Figure 12 Schematic diagram of a perception processing device provided by some embodiments of the present disclosure;

[0034] Figure 13 Schematic diagram of another perception processing device provided by some embodiments of the present disclosure;

[0035] Figure 14 Schematic diagram of a communication device provided by some embodiments of the present disclosure. Detailed implementation manners

[0036] Next, the technical solutions in the present disclosure will be clearly and completely described in conjunction with the accompanying drawings in the present disclosure. Obviously, the described embodiments are only some embodiments of the present disclosure, rather than all embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present disclosure.

[0037] It should be noted that in the present disclosure, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the present disclosure should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, using words such as "exemplary" or "for example" aims to present related concepts in a specific manner.

[0038] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features.

[0039] In the description of the present disclosure, unless otherwise specified, " / " means "or". For example, A / B may represent A or B. The "and / or" herein is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, "at least one" means one or more, and "a plurality" means two or more.

[0040] The method provided by the embodiments of the present disclosure can be applied to various communication systems. For example, the communication system can be a 5G communication system, a Wi-Fi system, a 3GPP-related communication system, a future evolved communication system (such as a sixth-generation (6G) communication system, etc.), or a system integrating multiple systems, etc. The embodiments of the present disclosure are not limited thereto.

[0041] With the wide application of 5G, while meeting the needs of individual users, 5G is gradually penetrating into various industries and fields of society, thus realizing the upgrade from consumption to industrial application. Among them, the integration of communication and sensing is based on the existing network architecture and communication waveforms, and uses the existing communication equipment to implement sensing applications, which promotes the development of innovative application services and is an important research direction of 5G. As Figure 1 shown, it is a schematic diagram of a sensing scenario provided by the embodiments of the present disclosure. The sensing system in the sensing base station 1 can perform base station sensing on the drone 2, the human body 3, the vehicle 4, and the building 5. The sensing system in the satellite 6 can perform satellite sensing on the aircraft 7 (shown as an airplane). The sensing system in the drone 2 can perform drone sensing on the human body. The sensing terminal in the vehicle 4 can perform terminal sensing on the human body 3.

[0042] In some embodiments, the terminal performing terminal sensing can also be a device with full-duplex transmission capabilities. The terminal can be a mobile phone, a tablet computer (Pad), a computer with wireless transceiver functions, a virtual reality (VR) terminal, an augmented reality (AR) terminal, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, etc. The embodiments of the present disclosure do not limit the application scenarios. The terminal can sometimes also be referred to as a user, user equipment (UE), access terminal, UE unit, UE station, mobile station, mobile unit, remote station, remote terminal, mobile device, UE terminal, wireless communication device, UE agent, or UE device, etc. The embodiments of the present disclosure are not limited thereto.

[0043] In the currently applied sensing functions, most use traditional algorithms for sensing. As the network scale increases, the types of services grow, and terminal devices become complex and diverse, it is increasingly difficult for technicians to optimize sensing algorithms. At the same time, with the development of software and hardware resources, artificial intelligence has gradually been widely applied in communication systems, but the integrated communication and sensing structure based on artificial intelligence has not been effectively developed and cannot be applied to complex and changing scenarios and applications, resulting in poor intelligent management capabilities.

[0044] To address the above technical problems, the embodiments of the present disclosure provide a sensing processing method. The idea is as follows: Based on the existing core network, a sensing function entity is newly added, with less modification to the network architecture, which is easy to implement the sensing function. At the same time, since the sensing server has strong storage capabilities and low computing power requirements for the sensing function entity, the artificial intelligence model is trained by the sensing server, and the sensing data is processed based on the trained artificial intelligence model to obtain the sensing result, making the acquisition method of the sensing result more accurate, less prone to errors, and facilitating the opening of data to other modules or third parties in the system to achieve real-time communication among various modules in the integrated communication and sensing system and complete the sensing task.

[0045] See Figure 2 , which is a schematic diagram of the architecture of the integrated communication and sensing system provided by the embodiments of the present disclosure. As Figure 2 shown, the system includes: a core network, a sensing server, a sensing function entity, and a sensing node.

[0046] The core network includes at least one of the following network elements: Network Slice Selection Function (NSSF), Network Exposure Function (NEF), Network Repository Function (NRF), Policy Control Function (PCF), Unified Data Management (UDM), Application Function (AF), Authentication Server Function (AUSF), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), or other possible network functions (Network Function, NF).

[0047] In some embodiments, the NSSF virtualizes multiple end-to-end networks on a common hardware basis through slicing technology. Each network has different NFs to adapt to different types of service requirements.

[0048] In some embodiments, the NEF is located between the 5G core network and external third-party application functions and is responsible for managing the externally accessible network data. All external applications that want to access the internal data of the 5G core network must go through the NEF.

[0049] In some embodiments, the NRF is used for NF registration, management, and status detection to achieve automated management of all NFs. When each NF starts up, it must register with the NRF to provide services. The registration information includes the type, address, and service list of the NF, etc.

[0050] In some embodiments, the PCF is used to manage network behavior using a unified policy framework and cooperate with user information in the unified data repository (UDR) to execute relevant policies.

[0051] In some embodiments, the UDM is used to manage user identities, subscription data, and authentication data, and is also used to manage the registration of user service network elements.

[0052] In some embodiments, the AF refers to various services at the application layer, which can be applications within the operator or third-party AFs (such as video servers, game servers). If it is an AF within the operator, it is in a trusted domain with other NFs and can directly interact and access other NFs, while third-party AFs are not in the trusted domain and need to access other NFs through the NEF.

[0053] In some embodiments, the AUSF is used to receive requests from the AMF to authenticate the user equipment (UE), request a key from the UDM, and then forward the key sent by the UDM to the AMF for authentication processing.

[0054] In some embodiments, the AMF is used to provide a session management message transmission channel for the UE and the SMF, and provide authentication, authorization functions, and a core network control plane access point for the terminal and the radio when the user accesses.

[0055] In some embodiments, the SMF is used to support customized mobility management solutions together with the AMF, such as mobile initiated connection only (MICO) or radio access network (RAN) enhancements.

[0056] In some embodiments, the UPF is mainly used to send UE service data to a data network (DN), identify data and services, execute actions and policies, etc.

[0057] In some embodiments, a perception server is used to train an artificial intelligence model, and the artificial intelligence model is used to process perception data to obtain perception results.

[0058] Optionally, the training of the artificial intelligence model can be carried out in an offline training and online prediction manner, or in an online training and online prediction manner.

[0059] See Figure 3 , which is a schematic flowchart of a model training method provided by an embodiment of the present disclosure. As Figure 3 shown, taking the artificial intelligence model trained in an offline training and online prediction manner as an example, the training of the artificial intelligence model includes four stages: an initial stage, a training stage, a prediction stage, and a collection stage.

[0060] Among them, in the initial stage of the artificial intelligence model training, an initial data set using simulation data or measured data according to the scenario can be stored in the perception server, and an initial training model is obtained by performing initial training using the initial data set. In the training stage of the artificial intelligence model training, when the training data accumulates to a certain amount, periodic artificial intelligence model training is performed. In the prediction stage of the artificial intelligence model training, the perception function entity performs real-time online prediction, and transmits the preprocessed data with high confidence and the corresponding perception results to the perception server for training sample collection. In the collection stage of the artificial intelligence model training, the perception server continuously collects data for periodic training of the artificial intelligence model.

[0061] In some embodiments, the online training and online prediction manner of the artificial intelligence model specifically refers to the artificial intelligence function in the perception function entity, which can perform some lightweight online training and online prediction. And transmit the preprocessed data with high confidence and the corresponding perception results to the perception server for training sample collection.

[0062] In some embodiments, the perception server is further used to send the trained artificial intelligence model to the perception function entity.

[0063] In some embodiments, the perception server is further configured with at least one of the following functions: a service control function, a data preprocessing function, a data storage function, and a data opening function.

[0064] Among them, the service control function includes at least one of the following: management of artificial intelligence applications, management of model training time, management of data opening function, management of information transmission; data preprocessing function, used to extract or normalize feature data from the sensed data; data storage function, used to store the sensed data with relatively high confidence as the data set for subsequent periodic training of the artificial intelligence model; data opening function, used to externally provide at least one of the sensed data, sensing results, and artificial intelligence models.

[0065] In some embodiments, the management of artificial intelligence applications is used to determine whether the sensing server conducts training of the artificial intelligence model; the management of model training time is used to determine the training period of the artificial intelligence model; the management of the data opening function is used to determine whether to open and share the sensed data and sensing results; the management of information transmission is used to determine whether to transmit the open and shared sensed data and sensing results.

[0066] In some embodiments, the sensing server is also used to send the trained artificial intelligence model to the sensing function entity.

[0067] Optionally, the sensing server is also used to send at least one of the following to the core network element: sensed data, sensing results, artificial intelligence models.

[0068] Optionally, the sensing server is also used to receive the sensed data sent by the sensing node.

[0069] In some embodiments, the sensing function entity is used to obtain the trained artificial intelligence model from the sensing server.

[0070] In some embodiments, the sensing function entity is also used to process the sensed data based on the artificial intelligence model to obtain the sensing result.

[0071] Among them, the sensing result is used for at least one of the following: target detection, trajectory tracking, trajectory correction, target association, target recognition.

[0072] In some embodiments, the sensing function entity is also configured with at least one of the following functions: sensing control function, sensing management function, artificial intelligence function, sensing processing function, data preprocessing function.

[0073] In some embodiments, the sensing control function includes at least one of the following: selection of sensing nodes, management of sensing resources, management of sensing time, management of artificial intelligence applications, management of information transmission.

[0074] In some embodiments, the sensing management function includes at least one of the following: obtaining sensing service requirements, authenticating the sensing service requirements, determining sensing tasks based on the sensing service requirements.

[0075] In some embodiments, the artificial intelligence function includes at least one of the following: training an artificial intelligence model and using an artificial intelligence model.

[0076] In some embodiments, the perception processing function is used for object detection, false alarm suppression, object association, trajectory tracking, object recognition, and trajectory correction of perception data.

[0077] In some embodiments, the data preprocessing function is used for extracting or normalizing feature data from perception data.

[0078] In some embodiments, the perception function entity is further configured to process perception data based on an artificial intelligence model to obtain a perception result, and send the perception result to a core network element.

[0079] Optionally, the perception function entity may send at least one of the following to the core network element: a perception service request message, a perception service response message, and a perception result.

[0080] Among them, the perception service request message is used to indicate the perception service requirement.

[0081] Optionally, the perception function entity may further send a perception control instruction to the perception node, and the perception control instruction is used to select a perception node to perform a perception task.

[0082] Optionally, the perception function entity may further send at least one of the following to the perception server: first indication information for indicating whether the perception server performs model training, and second indication information for indicating a training period.

[0083] Specifically, the perception control function in the perception function entity sends at least one of the following to the perception server: first indication information for indicating whether the perception server performs model training, and second indication information for indicating a training period.

[0084] In some embodiments, the perception node is used to send a perception signal and obtain perception data.

[0085] Specifically, after receiving the perception service request message sent by the perception function entity, the perception node sends a perception signal according to the perception method indicated by the perception service request message, performs a perception task, and obtains perception data.

[0086] It should be noted that the perception data may be measurement data at the physical layer or perception result data.

[0087] In some embodiments, the perception node is further configured to send perception data to the perception function entity.

[0088] Specifically, after the sensing node obtains the sensing data, it sends the sensing data to the sensing function entity.

[0089] Optionally, referring to Figure 2 , there can be multiple sensing nodes, and the sensing nodes communicate with each other through the Uu interface (U is the user to network interface, and u is Universal), and each sensing node can be a (generation node B, gNB) base station or a UE.

[0090] In some embodiments, the sensing data or sensing result processed by the UE is transmitted to the sensing function entity through the Uu interface and the gNB base station.

[0091] In some embodiments, the sensing node is further used for sensing signal processing and communication signal processing.

[0092] It should be noted that the system architecture and application scenarios described in the embodiments of the present disclosure are for more clearly explaining the technical solutions of the embodiments of the present disclosure, and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. Those of ordinary skill in the art can know that with the evolution of the system architecture and the emergence of new service scenarios, the technical solutions provided by the embodiments of the present disclosure are equally applicable to similar technical problems.

[0093] Next, with reference to the accompanying drawings of the specification, a sensing processing method provided by the embodiments of the present disclosure will be described.

[0094] Referring to Figure 4 , which is a schematic flowchart of a sensing processing method provided by the embodiments of the present disclosure, applied to a sensing function entity. As Figure 4 shown, the method includes the following steps:

[0095] S101. The sensing function entity obtains the trained artificial intelligence model from the sensing server.

[0096] In some embodiments, after the sensing function entity receives the sensing service request message sent by the core network element, the sensing function entity obtains the trained artificial intelligence model from the sensing server.

[0097] Optionally, the sensing service request message may carry the sensing service type, sensing area, sensing report content, and sensing report type.

[0098] Among them, the sensing service type is the sensing of drones, vehicles, or other targets. The sensing area can be within a range of 1000 meters centered on the sensing node or other ranges. The sensing report content may include the position and speed of the target, and the sensing report type may be the reporting of sensing results.

[0099] It should be noted that the target can also be other people, vehicles or other objects with moving trajectories or state changes, and the present disclosure does not limit the type of the target.

[0100] S102. The sensing functional entity processes the sensing data based on the artificial intelligence model to obtain a sensing result.

[0101] In some embodiments, the sensing functional entity responds to the sensing service request message, selects a sensing node for sensing detection, and obtains the sensing data obtained by the sensing node during the sensing detection.

[0102] In some embodiments, the above-mentioned sensing functional entity responding to the sensing service request message, selecting a sensing node for sensing detection, and obtaining the sensing data obtained by the sensing node during the sensing detection can be specifically implemented as the following steps:

[0103] A1. The sensing functional entity determines a sensing task based on the sensing service request message.

[0104] Specifically, the sensing functional entity responds to the sensing service request message, authenticates the sensing service request message through the sensing management function in the sensing functional entity, and after the authentication of the sensing service request message passes, the sensing management function determines a sensing task based on the sensing service request message (such as sensing the unmanned aerial vehicle in the sensing area, mainly sensing the position and speed of the unmanned aerial vehicle).

[0105] A2. The sensing functional entity selects a sensing node for executing the sensing task and sends a sensing control instruction to the sensing node.

[0106] Among them, the sensing control instruction is used to instruct the sensing node to execute the sensing task, and the sensing control instruction includes sensing configuration information, and the sensing configuration information is used to configure at least one of sensing resources, sensing time, and sensing signals.

[0107] Specifically, after the sensing task is determined, the sensing management function in the sensing functional entity sends a sensing service request message to the sensing control function, and the sensing control function in the sensing functional entity selects a sensing node and a sensing method for executing the sensing task according to the sensing service request message, and sends a sensing control instruction to the sensing node.

[0108] Optionally, the sensing node can be a base station, and the sensing method can be A transmits and A receives or A transmits and B receives.

[0109] It should be noted that A transmits and A receives means that the sensing service request message is sent through the sensing node A (or base station A) and the sensing data is received through the sensing node A (or base station A), and A transmits and B receives means that the sensing service request message is sent through the sensing node A (or base station A) and the sensing data is received through the sensing node B (or base station B).

[0110] Further, the sensing node starts to perform sensing tasks, sense and detect the sensing area, obtain the sensing data obtained during the sensing detection process, and send the sensing data to the sensing functional entity.

[0111] In some embodiments, the sensing node receives the sensing signal reflected by the target according to the integrated sensing time-frequency resources of communication and sensing indicated by the sensing control function, performs baseband processing on the sensing signal, and obtains the range-doppler (RD) map data of the target.

[0112] A3. The sensing functional entity receives the sensing data sent by the sensing node.

[0113] In some embodiments, after the sensing node obtains the sensing data obtained during the sensing detection process, it sends the sensing data (such as RD map data) to the sensing functional entity, and the sensing functional entity receives the sensing data.

[0114] Further, after the sensing functional entity receives the sensing data sent by the sensing node, the sensing functional entity processes the sensing data based on the artificial intelligence model to obtain a sensing result.

[0115] S103. The sensing functional entity sends the sensing result to the core network element.

[0116] In some embodiments, after the sensing functional entity obtains the sensing result, it sends the sensing result to the core network element.

[0117] It can be understood that based on the sensing processing method provided in the embodiments of the present disclosure, based on the existing core network architecture, by obtaining the trained artificial intelligence model from a sensing server with higher storage capacity, the problem of large core network data processing volume caused by training the artificial intelligence model in the core network is avoided, and the speed of model acquisition is improved; at the same time, processing the sensing data based on the trained artificial intelligence model makes the obtained sensing result more accurate, and the automatic processing of the sensing data realizes the intelligent management of the sensing data, reducing the labor input; finally, sending the obtained sensing result to the core network element enables data sharing among the various modules in the system, realizing the integration of sensing and communication.

[0118] In some embodiments, the artificial intelligence model can be trained through a sensing server. After the sensing functional entity obtains the sensing data sent by the sensing node, the sensing processing function in the sensing functional entity preprocesses the sensing data and then sends the sensing data to the sensing server, and the sensing server trains the artificial intelligence model based on the preprocessed sensing data. Figure 5 It is a schematic flowchart of another sensing processing method provided in the embodiments of the present disclosure, applied to a sensing server, such as Figure 5As shown, the method includes the following steps:

[0119] S201. The sensing server trains an artificial intelligence model.

[0120] Among them, the artificial intelligence model is used to process sensing data to obtain a sensing result.

[0121] In some embodiments, in response to first indication information sent by the sensing control function for instructing the sensing server to perform model training, the sensing server trains the artificial intelligence model.

[0122] Refer to Figure 6 , Figure 6 A schematic diagram of a range-Doppler map provided by an embodiment of the present disclosure, as Figure 6 shown, P (m，n) represents the power of the grid where the m-th row and the n-th column are located in the RD map.

[0123] In some embodiments, at the initial stage of training the artificial intelligence model, the sensing server uses initial test data to train the artificial intelligence model. After determining the RD map of the sensing signal, according to the reference cells in the RD map, the cells to be inspected and the protection cells of the RD data are extracted from the RD map. Illustrated as the power data of 9 cells, the feature data [P0, P1, P2, P3, P4, P5, P6, P7, P8, P9] is formed. This set of feature data is used as a sample data, and its corresponding label is 0 or 1, where 0 represents no target and 1 represents a target. The corresponding relationship is as Figure 7 shown, Figure 7 which is a schematic diagram of the corresponding relationship of the feature data provided by an embodiment of the present disclosure.

[0124] In some embodiments, the present disclosure uses a convolutional neural (CNN) network to train the model. Figure 8 An architecture schematic diagram of a convolutional neural network provided by an embodiment of the present disclosure, as Figure 8 shown, Conv L j (P1, P1) where L j is the number of convolutional kernels, and (P1, P1) represents the size of the convolutional kernel.

[0125] It should be noted that network training is not limited to the CNN network, and other networks can also be used, such as the residual network (ResNet), etc.

[0126] In some embodiments, in addition to the input layer and the fully connected layer, the CNN network structure can also have J layers of networks in the middle, and each layer of network includes convolution, activation, and pooling operations.

[0127] It should be noted that the CNN network model is a type of feedforward neural network that includes convolutional calculations and has a deep structure. The CNN network model has the ability of feature learning and can perform translation-invariant classification on the input information according to its hierarchical structure.

[0128] Optionally, the activation function of the CNN network uses the rectified linear unit (ReLU), as shown in Equation (1):

[0129] f x = max (0，x) Equation (1)

[0130] It should be noted that the ReLU function is also called the rectified linear unit, which is a commonly used activation function in artificial neural networks and usually refers to non-linear functions represented by the ramp function and its variants.

[0131] Optionally, the activation function of the fully connected layer uses the Sigmoid function, as shown in Equation (2):

[0132]

[0133] It should be noted that the Sigmoid function is a common S-shaped function in biology, also known as the S-shaped growth curve. In information science, due to its properties such as monotonic increase and monotonic increase of the inverse function, the Sigmoid function is often used as the activation function of neural networks to map variables to the interval (0, 1).

[0134] Optionally, the loss function used by the fully connected layer is the cross-entropy function, as shown in Equation (3):

[0135]

[0136] where O s is the size of the output, y i is the true value of the sample, is the predicted value of the model output.

[0137] It should be noted that the cross-entropy function is mainly used to measure the difference information between two probability distributions. The meaning of cross-entropy is the difficulty of text recognition using this model, or from the perspective of compression, how many bits are required to encode each word on average.

[0138] In some embodiments, the initial test data can be segmented with 80% of the data as the training set and 20% of the data as the test set, and the training set is used to train the artificial intelligence model, and the test set is used to test the artificial intelligence model.

[0139] S202. The perception server sends the trained artificial intelligence model to the perception function entity.

[0140] In some embodiments, the sensing server sends the trained artificial intelligence model to the sensing functional entity. The sensing functional entity processes the sensing data based on the artificial intelligence model, extracts the data of the RD graph, and extracts the feature data [P0, P1, P2, P3, P4, P5, P6, P7, P8, P9] as shown above Figure 6 and uses this feature data as the input of the artificial intelligence model. The online prediction output is no target (0) or target (1).

[0141] In some embodiments, for the feature data with a target, other sensing detections are performed on the target, such as direction of arrival (DOA) estimation, locked signal-to-noise ratio (SNR) estimation, speed estimation, power calculation, etc.

[0142] It should be noted that DOA estimation is a positioning technology that obtains the distance information and azimuth information of the target by processing the received echo signal, and SNR refers to the ratio of the intensity of the received useful signal to the intensity of the received interference signals (noise and interference).

[0143] Furthermore, based on the sensing data of the above-mentioned sensing detections, target detection, trajectory tracking, trajectory correction, target association, and target recognition are performed on the target.

[0144] It should be noted that if it is multi-sensing node sensing, data fusion between multi-sensing nodes is performed to obtain more complete and accurate sensing results of target detection, trajectory tracking, trajectory correction, target association, and target recognition.

[0145] Optionally, the fusion of sensing results between multi-sensing nodes can be fused by means such as merging and taking the superior, or can also be fused with the help of video.

[0146] In some embodiments, the sensing functional entity transmits the sensing results with higher confidence and the corresponding labels to the sensing server for data collection. The sensing server performs periodic model training on the artificial intelligence model according to the second indication information sent by the sensing control function.

[0147] In some embodiments, the sensing server sends at least one of sensing data, sensing results, or the artificial intelligence model to the core network element.

[0148] In some embodiments, the sensing processing function in the sensing functional entity sends the response information of the sensing service request message to the sensing management function, and the sensing management function sends the response information of the sensing service request message to the core network and reports the sensing results according to the sensing service request message.

[0149] In some embodiments, the perception functional entity performs target detection, trajectory tracking, trajectory correction, target association, and target recognition based on information such as DOA estimation, SNR estimation, speed estimation, and power calculation reported by the perception nodes, and obtains the moving trajectory of the target.

[0150] In some embodiments, in wireless communication, especially in high-frequency band wireless communication, wireless signals are extremely easy to be blocked, resulting in signal blind spots, which leads to the loss of the target's trajectory. At this time, the missing part of the trajectory can be predicted based on the known historical trajectory information of the target.

[0151] In some embodiments, at the initial stage of training the artificial intelligence model, the perception server uses the initial test data to train the artificial intelligence model, and the initial test data can also be the actual test trajectory data of the target.

[0152] Exemplarily, the trajectory point sequence of the target can be expressed as [(Lo1, La1, H1, V1),

[0153] (Lo2, La2, H2, V2),..., (Lo n , La n , H n , V n )].

[0154] Among them, Lo t represents the longitude of the perceived target trajectory point at the t-th moment, La t represents the latitude of the perceived target trajectory point at the t-th moment, H t represents the height of the perceived target trajectory point at the t-th moment, V t represents the speed of the perceived target at the trajectory point (Lo t , La t , H t ) at the t-th moment, and t = 1, 2,..., n.

[0155] In some embodiments, after obtaining the actual test trajectory data of the target, it is necessary to perform data preprocessing on this actual test trajectory data, and normalize the actual test trajectory data between the minimum value and the maximum value within a certain range to achieve the standardization of the training data.

[0156] In some embodiments, the trajectory data at adjacent moments in the trajectory point sequence is differentiated, that is, x t = Lo t+1 - Lo t , y t = La t+1 - La t , z t = H t+1 - H t , vt = V t+1 -V t 。

[0157] Furthermore, (n - 1) trajectory sequences are obtained, namely [(x1, y1, z1, v1), (x2, y2, z2, v2),..., (x n-1 , y n-1 , z n-1 , v n-1 )].

[0158] In some embodiments, using the sklearn tool in the machine learning library, the differential trajectory sequence is normalized with the MinMaxScaler() function to scale it to the range [-1, 1], and the output parameter scaler of the function is saved. A supervised sequence is constructed for the normalized data. Table 1 is a schematic diagram of a method for constructing a supervised sequence provided by some embodiments of the present disclosure. Taking n = 13 as an example, as shown in Table 1:

[0159] Table 1

[0160]

[0161]

[0162] Exemplarily, if the constructed supervised sequence is [(x1, y1, z1, v1), (x2, y2, z2, v2),..., (x 12 , y 12 , z 12 , v 12 )], then the label of this supervised sequence is (x 13 , y 13 , z 13 , v 13 ); if the constructed supervised sequence is [(x2, y2, z2, v2), (x3, y3, z3, v3),..., (x 13 , y 13 , z 13 , v 13 )], then the label of this supervised sequence is (x 14 , y 14 , z 14 , v 14 ); if the constructed supervised sequence is [(x3, y3, z3, v3), (x4, y4, z4, v4),..., (x 14 , y 14 , z 14 , v 14 )], then the label of this supervised sequence is (x 15 , y 15 , z15 , v 15 ); If the constructed supervision sequence is [(x4, y4, z4, v4), (x5, y5, z5, v5),..., (x 15 , y 15 , z 15 , v 15 )], then the label of this supervision sequence is (x 16 , y 16 , z 16 , v 16 ); If the constructed supervision sequence is [(x t-12 , y t-12 , z t-12 , v t-12 ), (x t-11 , y t-11 , z t-11 , v t-11 ),..., (x t-1 , y t-1 , z t-1 , v t-1 )], then the label of this supervision sequence is

[0163] (x t , y t , z t , v t ).

[0164] In some embodiments, the processed data set is split into a training data set and a test data set at a ratio of 80% and 20%. The training data set is used to train the artificial intelligence model, and the test data set is used to evaluate the artificial intelligence model.

[0165] In some embodiments, a long short-term memory neural network model (long short term memory, LSTM) can be constructed, and the network model parameters can be configured to train the artificial intelligence model.

[0166] In some embodiments, the LSTM network model includes three gate information, namely, a forget gate, an input gate, and an output gate. Figure 9 For the structural schematic diagram of a long short-term memory neural network model provided by an embodiment of the present disclosure, as Figure 9 shown, the forget gate mainly controls the proportion of forgetting the data features of the previous layer network, that is, the input data features are weighted and summed with the previous sequence state and input to the forget gate to obtain the data feature output of the forget gate. Its calculation method is shown in formula (4):

[0167] f t =σ(W xf ·x t +W hf ·ht-1 +b f ) Formula (4)

[0168] Among them, f t is the data feature output of the forgetting gate, W xf is the weighting coefficient of the forgetting gate for the data at the current moment, x t is the input data feature, W hf represents the weighting coefficient of the forgetting gate for the state at the previous moment, h t-1 is the state of the previous sequence, b f is the data bias constant of the forgetting gate, and σ(x) represents the sigmod activation function.

[0169] Furthermore, the main memory data at the current moment is obtained by combining the memory system data of the previous moment, and the calculation method is shown in Formula (5):

[0170] c t1 = c t-1 ⊙f t Formula (5)

[0171] Among them, c t1 is the main memory data at the current moment, c t-1 is the memory system data of the previous moment.

[0172] In some embodiments, the input gate mainly filters the effective features of the current input data, and compensates and updates the control data features of the forgetting gate by combining the current data. The data feature extraction method of the input gate is shown in Formula (6):

[0173] i t = σ(W xi ·x t + W hi ·h t-1 + b i ) Formula (6)

[0174] Among them, i t is the data feature output of the input gate, W xi is the weighting coefficient of the input gate for the data at the current moment, W hi is the weighting coefficient of the input gate for the state at the previous moment, b i is the data bias constant of the input gate.

[0175] Furthermore, the input data features will compensate the memory data features to a certain extent, and the input gate feature parameters need to be filtered. The calculation method of the filtering control coefficient is shown in Formula (7):

[0176] c' t = tan h(W xc ·x t+W hc ·h t-1 +b c ) Formula (7)

[0177] where c' t is the feature screening control coefficient of the input gate, W xc is the input data screening control weight coefficient at the current moment, W hc is the input screening weight coefficient of the data state at the previous moment, b c is the screening control data bias constant, and tan h(x) is the activation function.

[0178] Furthermore, the input gate data is screened and controlled to obtain the memory system data at the next moment, and the calculation method is as shown in Formula (8):

[0179] c t2 = c' t ⊙i t Formula (8)

[0180] where c t2 is the memory system data at the next moment.

[0181] In some embodiments, the output value of the memory parameter at the current moment can be determined based on the main memory data at the current moment and the memory system data at the next moment, and the calculation method is as shown in Formula (9):

[0182] c t = c t1 + c t2 Formula (9)

[0183] where c t is the output value of the memory parameter at the current moment.

[0184] In some embodiments, the output gate mainly learns and trains the model based on the forgetting data characteristics and the current data characteristics, and its training method is as shown in Formula (10):

[0185] h t = σ(W xo ·x t + W ho ·h t-1 + b o ⊙tan h(c t )) Formula (10)

[0186] where h t is the output data of the output gate, W xo is the data weight coefficient at the current moment in the output gate, W ho is the weight coefficient of the state data at the previous moment in the output gate, b oIt is the data bias constant in the output gate.

[0187] Optionally, the present disclosure may adopt an architecture of two-layer LSTM plus a fully connected layer. Figure 10 The present disclosure provides a schematic diagram of the architecture of a long short-term memory neural network model. Table 2 shows a parameter configuration method for an architecture of two-layer LSTM plus a fully connected layer, as shown in Table 2:

[0188] Table 2

[0189] Parameter item Parameter value Number of LSTM layers 2 Number of hidden layers 64 Activation function ReLU Loss function MSE

[0190] Exemplarily, in combination with Figure 10 and Table 2, the number of hidden layers of the LSTM can be 64 layers, the activation function can be ReLU, and the loss function can be the mean square error function (MSE).

[0191] In some embodiments, after the artificial intelligence model is trained, the perception server sends the trained artificial intelligence model to the perception functional entity.

[0192] It should be noted that based on the perception processing method provided by the embodiments of the present disclosure, the artificial intelligence model is trained by a perception server with high storage capacity and low computing power requirements, and the training efficiency is high; then, based on the trained artificial intelligence model, the perception data is processed to make the obtained perception results more accurate, and the automatic processing of the perception data realizes the intelligent management of the perception data and reduces the labor input; finally, the trained artificial intelligence model is sent to the perception functional entity, so that the perception functional entity can predict the perception data based on the trained artificial intelligence model, realizing data sharing among various modules in the system.

[0193] In some embodiments, after receiving the trained artificial intelligence model, the perception functional entity preprocesses the actual test trajectory data (the preprocessing method is the same as the preprocessing method during training the model), inputs the preprocessed data set into the artificial intelligence model, and obtains predicted values (x t , y t , z t , v t ), and according to the scaler saved during data preprocessing, performs inverse normalization, and then adds (Lo n , La n , H n , V n ), to obtain the final predicted values (Lo n+1 , La n+1 , H n+1 ,

[0194] V n+1)。

[0195] In some embodiments, the perception functional entity transmits the actual test trajectory data with a relatively high confidence level and the corresponding labels to the perception server for data collection.

[0196] In some embodiments, the perception functional entity transmits the perception results with a relatively high confidence level and the corresponding labels to the perception server for data collection, and the perception server performs periodic model training on the artificial intelligence model according to the second indication information sent by the perception control function for indicating the training period.

[0197] In some embodiments, the perception processing function in the perception functional entity sends the response information of the perception service request message to the perception management function, and the perception management function sends the response information of the perception service request message to the core network, and reports the perception results according to the perception service request message, so as to display a complete running trajectory of the target.

[0198] In some embodiments, the above method can also be implemented through a process as Figure 11 shown, see Figure 11 , which is a schematic flowchart of another perception processing method provided by an embodiment of the present disclosure. As Figure 11 shown, taking the target as a drone as an example, the method includes the following steps:

[0199] S1. The core network element sends a perception service request message of the drone to the perception management function in the perception functional entity.

[0200] Among them, the perception service request message is used to indicate the perception service requirements, and the perception service request message may carry the perception service type, perception area, perception reporting content, and perception reporting type.

[0201] Among them, the perception service type is drone perception, the perception area can be within a range of 1000 meters centered on the perception node or other ranges, the perception reporting content may include the position and speed of the drone, and the perception reporting type may be perception result reporting.

[0202] S2. After obtaining the perception service request message, the perception management function authenticates the perception service request message.

[0203] S3. The perception management function sends the perception service request message passed through authentication to the perception control function in the perception functional entity.

[0204] S4. After obtaining the perception service request message, the perception control function selects the perception method, perception node, and processing method of the perception data for the drone according to the perception service request message.

[0205] S5. The perception control function sends a perception control instruction to the perception node.

[0206] Among them, the sensing control instruction is used to instruct the sensing node to perform a sensing task, and the sensing control instruction includes sensing configuration information, which is used to configure at least one of sensing resources, sensing time, and sensing signals.

[0207] S6. The sensing control function sends first indication information for model training to the sensing server.

[0208] S7. After the sensing node receives the sensing control instruction, according to the sensing control instruction, it sends a sensing signal to the UAV, performs a sensing task, and obtains sensing data of the UAV.

[0209] S8. The sensing node sends the sensing data of the UAV to the sensing processing function of the sensing functional entity.

[0210] S9. The sensing processing function performs preprocessing on the sensing data.

[0211] Among them, the preprocessing of the sensing data includes extracting or normalizing the feature data of the sensing data.

[0212] S10. After the sensing server receives the first indication information for model training sent by the sensing control function, it trains the artificial intelligence model and sends the trained artificial intelligence model to the sensing processing function.

[0213] It should be noted that for the training of the artificial intelligence model, please refer to the above step S201, which will not be elaborated here.

[0214] S11. The sensing processing function performs artificial intelligence prediction on the artificial intelligence model to obtain a predicted sensing result.

[0215] S12. The sensing processing function sends the sensing result to the sensing server for storage.

[0216] S13. The sensing server performs periodic training on the artificial intelligence model based on the second indication information for indicating the training period sent by the sensing control function and the sensing result.

[0217] S14. After obtaining the sensing result, the sensing processing function in the sensing functional entity sends a sensing service request message response to the sensing management function.

[0218] S15. The sensing management function sends the sensing service request message response to the core network element, and the core network element obtains the sensing result of the UAV.

[0219] Among them, the sensing result may include the position and speed of the UAV, showing a complete movement trajectory of the UAV.

[0220] The above mainly introduces the solutions of the embodiments of the present disclosure from the perspective of methods. It can be understood that in order to implement the above functions, the perception processing device includes at least one of the corresponding hardware structures and software modules for executing each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of the examples described in the embodiments disclosed herein, the embodiments of the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the embodiments of the present disclosure.

[0221] It can be understood that in order to implement the above functions, the perception processing device includes the corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should easily realize that, in combination with the algorithm steps of the examples described in the embodiments of the present disclosure, the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present disclosure.

[0222] The embodiments of the present disclosure can divide the functional modules of the perception processing device according to the above method embodiments. For example, each functional module can be divided corresponding to each function, or two or more functions can be integrated into one functional module. The above integrated module can be implemented in the form of hardware or software. It should be noted that the division of modules in the embodiments of the present disclosure is illustrative, only a logical functional division, and there can be other division methods in actual implementation. The following takes the example of dividing each functional module corresponding to each function for illustration.

[0223] Figure 12 It is a schematic structural diagram of a perception processing device provided by an embodiment of the present disclosure. The perception processing device is applied to a perception functional entity and can execute the perception processing method provided by the above method embodiment. As Figure 12 shown, the perception processing device 200 includes: an acquisition module 201, a processing module 202, and a sending module 203.

[0224] In some embodiments, the acquisition module 201 is configured to obtain a trained artificial intelligence model from a perception server.

[0225] In some embodiments, the processing module 202 is configured to process perception data based on the artificial intelligence model to obtain a perception result.

[0226] In some embodiments, the sending module 203 is configured to send the sensing result to a core network element.

[0227] In some embodiments, the obtaining module 201 is further configured to receive a sensing service request message sent by a core network element.

[0228] In some embodiments, the obtaining module 201 is further configured to, in response to the sensing service request message, select a sensing node for sensing detection and obtain sensing data obtained by the sensing node during the sensing detection process.

[0229] In some embodiments, the processing module 202 is further configured to determine a sensing task based on the sensing service request message.

[0230] In some embodiments, the sending module 203 is further configured to select a sensing node for executing the sensing task and send a sensing control instruction to the sensing node, where the sensing control instruction is used to instruct the sensing node to execute the sensing task.

[0231] In some embodiments, the sensing control instruction includes sensing configuration information, and the sensing configuration information is used to configure at least one of sensing resources, sensing time, and sensing signals.

[0232] In some embodiments, the obtaining module 201 is further configured to receive sensing data sent by the sensing node.

[0233] In some embodiments, the processing module 202 is further configured to authenticate the sensing service request message.

[0234] In some embodiments, the processing module 202 is further configured to, after the authentication of the sensing service request message passes, determine a sensing task based on the sensing service request message.

[0235] In some embodiments, the sensing functional entity configures at least one of the following functions: sensing control function, sensing management function, artificial intelligence function, sensing processing function, and data preprocessing function.

[0236] In some embodiments, the sensing control function includes at least one of the following: selection of sensing nodes, management of sensing resources, management of sensing time, management of artificial intelligence applications, and management of information transmission.

[0237] In some embodiments, the sensing management function includes at least one of the following: obtaining sensing service requirements, authenticating the sensing service requirements, and determining a sensing task based on the sensing service requirements.

[0238] In some embodiments, the artificial intelligence function includes at least one of the following: training an artificial intelligence model, and using an artificial intelligence model.

[0239] In some embodiments, the sending module 203 is further configured to send at least one of the following to the perception server: first indication information for indicating whether the perception server performs model training, and second indication information for indicating the training period.

[0240] Figure 13 It is a schematic structural diagram of another perception processing device provided by an embodiment of the present disclosure. The perception processing device is applied to a perception server and can execute the perception processing method provided by the above method embodiment. As Figure 13 shown, the perception processing device 300 includes: a training module 301, a sending module 302, and an obtaining module 303.

[0241] In some embodiments, the training module 301 is configured to train an artificial intelligence model, and the artificial intelligence model is used to process perception data to obtain a perception result.

[0242] In some embodiments, the sending module 302 is configured to send the trained artificial intelligence model to a perception function entity.

[0243] In some embodiments, the perception server is configured with at least one of the following functions: service control function, data preprocessing function, data storage function, data opening function.

[0244] In some embodiments, the service control function includes at least one of the following: management of artificial intelligence applications, management of model training time, management of data opening function, and management of information transmission.

[0245] In some embodiments, the data opening function is used to externally provide at least one of perception data, perception results, and artificial intelligence models.

[0246] In some embodiments, the sending module 302 is further configured to send at least one of the following to a core network element: perception data, artificial intelligence model.

[0247] In some embodiments, the obtaining module 303 is configured to receive perception data sent by a perception node.

[0248] In the case where the functions of the above integrated modules are implemented in the form of hardware, an embodiment of the present disclosure provides a possible structure of the communication device involved in the above embodiment. As Figure 14 shown, the communication device 400 includes: a processor 402, a bus 404. Optionally, the communication device 400 may further include a memory 401; optionally, the communication device 400 may further include a communication interface 403.

[0249] In some embodiments, the processor 402 may implement or execute various exemplary logical blocks, modules, and circuits described in connection with the embodiments of the present disclosure. The processor 402 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field-programmable gate array, or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It may implement or execute various exemplary logical blocks, modules, and circuits described in connection with the embodiments of the present disclosure. The processor 402 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0250] In some embodiments, the communication interface 403 is used to connect to other devices through a communication network. The communication network may be an Ethernet, a radio access network, a wireless local area network (WLAN), etc.

[0251] In some embodiments, the memory 401 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM), or other type of dynamic storage device that can store information and instructions. It may also be an electrically erasable programmable read-only memory (EEPROM), a magnetic disk storage medium, or other magnetic storage device, or 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.

[0252] As a possible implementation, the memory 401 may exist independently of the processor 402. The memory 401 may be connected to the processor 402 through a bus 404 and is used to store instructions or program code. When the processor 402 calls and executes the instructions or program code stored in the memory 401, the perception processing method provided by the embodiments of the present disclosure can be implemented.

[0253] In another possible implementation, the memory 401 may also be integrated with the processor 402.

[0254] In some embodiments, the bus 404 may be an extended industry standard architecture (EISA) bus, etc. The bus 404 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 14It is represented by only one thick line, but it does not mean that there is only one bus or one type of bus.

[0255] Some embodiments of the present disclosure provide a computer-readable storage medium (e.g., a non-transitory computer-readable storage medium), in which computer program instructions are stored. When the computer program instructions run on a computer, the computer is caused to execute the perception processing method of any one of the above embodiments.

[0256] Exemplarily, the above computer-readable storage medium may include, but is not limited to: magnetic storage devices (e.g., hard disks, floppy disks, or magnetic tapes, etc.), optical discs (e.g., Compact Disks (CDs), Digital Versatile Disks (DVDs), etc.), smart cards, and flash memory devices (e.g., Erasable Programmable Read-Only Memories (EPROMs), cards, sticks, or key drives, etc.). The various computer-readable storage media described in the present disclosure may represent one or more devices and / or other machine-readable storage media for storing information. The term "machine-readable storage medium" may include, but is not limited to, wireless channels and various other media capable of storing, containing, and / or carrying instructions and / or data.

[0257] Embodiments of the present disclosure provide a computer program product containing instructions. When the computer program product runs on a computer, the computer is caused to execute the perception processing method of any one of the above embodiments.

[0258] The above are only the specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present disclosure should be covered by the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A perception processing method, characterized in that, Applied to a perception functional entity, the method includes: Obtain a trained artificial intelligence model from a perception server; Process perception data based on the artificial intelligence model to obtain a perception result; Send the perception result to a core network element.

2. The method according to claim 1, wherein Before the step of processing perception data based on the artificial intelligence model to obtain a perception result, the method further includes: Receive a perception service request message sent by the core network element; In response to the perception service request message, select a perception node for perception detection, and obtain the perception data obtained by the perception node during the perception detection process.

3. The method according to claim 2, wherein The step of, in response to the perception service request message, selecting a perception node for perception detection and obtaining the perception data obtained by the perception node during the perception detection process includes: Determine a perception task based on the perception service request message; Select a perception node to execute the perception task, and send a perception control instruction to the perception node, where the perception control instruction is used to instruct the perception node to execute the perception task; Receive the perception data sent by the perception node.

4. The method according to claim 3, characterized in that, The perception control instruction includes perception configuration information, and the perception configuration information is used to configure at least one of perception resources, perception time, and perception signals.

5. The method according to claim 3, wherein The step of determining a perception task based on the perception service request message includes: Authenticate the perception service request message; After the authentication of the perception service request message passes, determine a perception task based on the perception service request message.

6. The method according to claim 1, wherein The perception functional entity is configured with at least one of the following functions: perception control function, perception management function, artificial intelligence function, perception processing function, and data preprocessing function.

7. The method according to claim 6, wherein The perception control function includes at least one of the following: selection of perception nodes, management of perception resources, management of perception time, management of artificial intelligence applications, and management of information transmission.

8. The method according to claim 6, characterized in that, The perception management function includes at least one of the following: obtaining perception service requirements, authenticating the perception service requirements, and determining a perception task based on the perception service requirements.

9. The method according to claim 6, wherein The artificial intelligence function includes at least one of the following: training the artificial intelligence model and using the artificial intelligence model.

10. The method according to claim 1, characterized in that, The method further includes: Send at least one of the following to the core network element: a perception service request message, a perception service response message, and the perception result; where the perception service request message is used to indicate perception service requirements.

11. The method according to claim 1, wherein Send a perception control instruction to a perception node; the perception control instruction is used to select the perception node to execute a perception task.

12. The method according to claim 1, wherein Send at least one of the following to the perception server: first indication information for indicating whether the perception server performs model training, and second indication information for indicating a training period.

13. A perception processing method, characterized in that, Applied to a perception server, the method includes: Train an artificial intelligence model, where the artificial intelligence model is used to process perception data to obtain a perception result; Send the trained artificial intelligence model to a perception functional entity.

14. The method according to claim 13, wherein The perception server is configured with at least one of the following functions: service control function, data preprocessing function, data storage function, and data opening function.

15. The method according to claim 14, wherein The service control function includes at least one of the following: management of artificial intelligence applications, management of model training time, management of the data opening function, and management of information transmission.

16. The method according to claim 14, wherein The data opening function is used to externally provide at least one of the sensing data, the sensing results, and the artificial intelligence model.

17. The method according to claim 13, wherein The method further includes: Sending at least one of the following to the core network element: the sensing data, the sensing results, and the artificial intelligence model.

18. The method according to claim 13, characterized in that, The method further includes: Receiving the sensing data sent by the sensing node.

19. A synaesthesia integration system, characterized in that, It includes: A sensing server for training an artificial intelligence model; A sensing function entity for obtaining the trained artificial intelligence model from the sensing server; Processing the sensing data based on the artificial intelligence model to obtain sensing results, and sending the sensing results to the core network element.

20. The system according to claim 19, wherein The system further includes: A sensing node for sending sensing signals, obtaining the sensing data; and sending the sensing data to the sensing function entity.

21. A communication device, characterized in that, It includes: A memory and a processor; The memory and the processor are coupled; The memory is used to store instructions executable by the processor; When the processor executes the instructions, it executes the method according to any one of claims 1-12, or the method according to any one of claims 13-18.

22. A computer-readable storage medium, characterized in that, Computer instructions are stored on the computer-readable storage medium, and when the computer instructions run on the computer, the computer is caused to execute the method according to any one of claims 1-12, or the method according to any one of claims 13-18.

Images