Communication method and device

Abstract

The invention discloses a communication method and device. The method comprises the following steps that: a fusion node can report the sensing data fusion capability of the fusion node to an agent node; after receiving the second information from the agent node, the fusion node may send the first perceptual data to the agent node. Wherein the second information is used for indicating a first mode, and the first mode is determined according to the sensing data fusion capability of the fusion node; the first sensing data is obtained by fusing the sensing data of the plurality of sensing nodes according to the first mode. Through the method, the agent node can determine the mode of fusing the sensing data for the fusion node according to the capability of fusing the sensing data of the fusion node, so that the mode of fusing the sensing data adaptive to the capability of the fusion node can be determined, and the efficiency and the quality of fusing the sensing data can be improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a communication method and apparatus. Background Technology

[0002] The concept of intelligent agents has been proposed in the field of computer science (CS). Any entity capable of independent thought and interaction with its environment can be abstracted as an intelligent agent. The basic characteristics of intelligent agents are: the ability to react to changes in the environment and automatically adjust their behavior and state; and the ability for different intelligent agents to interact with other intelligent agents according to their respective intentions.

[0003] However, the concept and application of intelligent agents are currently only proposed in the field of computer science. Further research is needed to explore how to apply intelligent agents in communication networks. Summary of the Invention

[0004] This application provides a communication method and apparatus for applying intelligent agents in a communication network.

[0005] In a first aspect, embodiments of this application provide a communication method. This method can be applied to a fusion node, which can be a terminal, access network device, core network device, or network management device; or it can be a module, communication module, circuit or chip responsible for communication functions (such as a modem chip, also known as a baseband chip, or a system-on-chip (SoC) chip or system-in-package (SIP) chip containing a modem core), chip system, or processor in the terminal, access network device, core network device, or network management device; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of the terminal, access network device, core network device, or network management device. The method may include: the fusion node sending first information to an intelligent agent node, the first information indicating the fusion node's ability to fuse sensing data; and after receiving second information from the intelligent agent node, the fusion node sending first sensing data to the intelligent agent node. The second information may indicate a first mode for fusing sensing data, the first mode being determined based on the fusion node's ability to fuse sensing data; the first sensing data may be obtained by fusing sensing data from multiple sensing nodes according to the first mode.

[0006] This method allows fusion nodes to report their ability to fuse sensing data to agent nodes. Agent nodes can then determine the fusion sensing data pattern for fusion nodes based on their fusion sensing data capabilities. This process helps determine a fusion sensing data pattern that is compatible with the capabilities of the fusion nodes, thereby improving the efficiency and quality of the fusion sensing data.

[0007] Furthermore, in this method, the fusion node can transmit fused sensing data, thereby reducing transmission overhead. Additionally, the sensing data fused by the fusion node can include sensing data from multiple modalities, thus reducing transmission overhead while broadening the scope of application.

[0008] Furthermore, in this method, both the fusion node and the agent node can be deployed with agents, thereby improving network intelligence and enhancing network maintenance and / or operational efficiency.

[0009] Secondly, embodiments of this application provide a communication method applicable to intelligent agent nodes. The intelligent agent node can be a terminal, access network device, core network device, or network management device; or it can be a module, communication module, circuit or chip responsible for communication functions (such as a modem chip, or a SoC chip or SIP chip containing a modem core), chip system, or processor within the terminal, access network device, core network device, or network management device; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of the terminal, access network device, core network device, or network management device. The method may include: the intelligent agent node receiving first information from a fusion node, the first information indicating the fusion node's ability to fuse sensing data; and after sending second information to the fusion node, the intelligent agent node receiving first sensing data from the fusion node. The second information may indicate a first mode for fusing sensing data, the first mode being determined based on the fusion node's ability to fuse sensing data; the first sensing data is obtained by fusing sensing data from multiple sensing nodes according to the first mode.

[0010] Based on the first or second aspect, in one possible design, the ability of a fusion node to fuse sensing data may include at least one of the following: the computing power of the fusion node; or at least one artificial intelligence (AI) model supported by the fusion node, which can be used to fuse sensing data. This design provides multiple possible ways for the fusion node to fuse sensing data, offering greater flexibility. Furthermore, by reporting the computing power of the fusion node and / or at least one AI model supported by the fusion node, the accuracy and effectiveness of the patterns determined by the agent nodes for the fusion node to fuse sensing data can be improved, thereby enhancing the quality and efficiency of the fusion node's fusion of sensing data.

[0011] Based on the first or second aspect, in one possible design, where the fusion node's ability to fuse sensing data includes at least one AI model, the first information can also be used to indicate at least one of the following: a modality supported by at least one AI model; or parameters of a modality supported by at least one AI model. In this design, the fusion node can report at least one modality supported by an AI model and its parameters to the agent node, thereby further improving the accuracy and effectiveness of the patterns determined by the agent node for the fusion node to fuse sensing data, and consequently improving the quality and efficiency of the fusion node's fusion of sensing data.

[0012] Based on the first or second aspect, in one possible design, the first mode may include at least one of the following: a modality of fused sensing data; a period of fused sensing data; or an AI model of fused sensing data. This design provides multiple possible ways to implement the first mode. Since the first mode is related to the parameters of the fused sensing data of the fusion node, or in other words, the first mode may include the parameters of the fused sensing data of the fusion node, this design allows the agent node to flexibly configure the parameters of the fused sensing data for the fusion node.

[0013] Based on the first or second aspect, in one possible design, the method may further include: an agent node sending a first request to a fusion node; correspondingly, the fusion node receiving the first request from the agent node, the first request being used to request the fusion node's ability to fuse sensing data. In this way, the fusion node only sends the first information to the agent node after receiving the first request, thereby avoiding unnecessary transmission of the first information and conserving transmission resources.

[0014] Based on the first aspect, in one possible design, the method may further include: a fusion node receiving third information from a first sensing node. The first sensing node belongs to multiple sensing nodes; the third information can be used to indicate a first modality, which is the modality of the sensing data of the first sensing node. If a first AI model in the fusion node supports the first modality, and the computational power required for the first AI model to fuse the sensing data of multiple sensing nodes is less than or equal to the computational power of the fusion node, the fusion node may send fourth information to the first sensing node. This fourth information can be used to instruct the first sensing node to send its own sensing data to the fusion node. Then, the fusion node can receive the first sensing node's sensing data from the first sensing node. Through this design, the first sensing node can send its own sensing data on demand according to the instructions of the fusion node, avoiding or reducing unnecessary transmission of its own sensing data, thereby saving transmission resources and avoiding resource waste.

[0015] Based on the first or second aspect, in one possible design, the method may further include: if the AI ​​model in the fusion node does not support the first modality, or if one or more AI models in the fusion node support the first modality, and the computational power required for the one or more AI models to fuse the perception data of multiple perception nodes is greater than the computational power of the fusion node, the fusion node may send a second request to the agent node; correspondingly, the agent node may receive the second request from the fusion node. The second request may be used to request a second AI model, which supports the first modality, and the computational power required for the second AI model to fuse the perception data of multiple perception nodes is less than or equal to the computational power of the fusion node. Through this design, after receiving the third information, if the AI ​​model in the fusion node cannot process the perception data of multiple perception nodes, the fusion node may request an AI model capable of processing the perception data of multiple perception nodes, thereby improving the performance of fusing perception data.

[0016] Based on the first or second aspect, in one possible design, if the AI ​​model in the fusion node does not support the first modality, the second request may include: information about the first modality; or, if one or more AI models in the fusion node support the first modality, and the computational power required for the one or more AI models to fuse the perception data of multiple perception nodes is greater than the computational power of the fusion node, the second request may include at least one of the following: parameters of each of the multiple perception nodes, and the number of the multiple perception nodes. Through this design, the agent node can determine a suitable second AI model for the fusion node based on the information in the second request.

[0017] Based on the first or second aspect, in one possible design, the method may further include: the agent node sending fifth information to the fusion node; correspondingly, the fusion node receiving the fifth information from the agent node. The fifth information may include a second AI model, or it may be used to indicate the download address of the second AI model. Through this design, the fusion node can quickly and accurately obtain the second AI model based on the fifth information.

[0018] Based on the first or second aspect, in one possible design, the method may further include: a fusion node sending sixth information to an agent node; correspondingly, the agent node receiving the sixth information from the fusion node. The sixth information may be auxiliary information required for fusing sensory data from multiple sensing nodes. Through this design, the fusion node can report the auxiliary information required for fusing sensory data from multiple sensing nodes, enabling the agent node to better manage the fusion node accordingly.

[0019] Based on the first or second aspect, in one possible design, the sixth information may include at least one of the following: the configuration of an AI model that fuses the perception data of multiple perception nodes; or the configuration of multiple perception nodes.

[0020] Thirdly, embodiments of this application provide a communication method. This method can be applied to a first sensing node, which can be a terminal, access network device, or core network device; or it can be a module, communication module, circuit or chip responsible for communication functions (such as a modem chip, or a SoC chip or SIP chip containing a modem core), chip system, or processor in the terminal, access network device, or core network device; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of the terminal, access network device, or core network device. The method may include: the first sensing node sending third information to a fusion node. The third information can be used to indicate a first modality, which may be the modality of the sensing data of the first sensing node. If a first AI model in the fusion node supports the first modality, and the computing power required for the first AI model to fuse the sensing data of multiple sensing nodes is less than or equal to the computing power of the fusion node, the first sensing node can receive fourth information from the fusion node. The fourth information can be used to instruct the first sensing node to send its sensing data to the fusion node. The multiple sensing nodes include the first sensing node. The first sensing node can send sensing data to the fusion node.

[0021] Fourthly, embodiments of this application provide a communication method that can be applied to a converged node. The converged node can be a terminal, access network device, core network device, or network management device; or it can be a module, communication module, circuit or chip (such as a modem chip, or a SoC chip or SIP containing a modem core), chip system, or processor within the terminal, access network device, core network device, or network management device; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of the terminal, access network device, core network device, or network management device. The method may include: the converged node receiving third information from a first sensing node. The first sensing node belongs to multiple sensing nodes; the third information can be used to indicate a first modality, which is the modality of the sensing data of the first sensing node. If a first AI model in the converged node supports the first modality, and the computing power required for the first AI model to fuse the sensing data of multiple sensing nodes is less than or equal to the computing power of the converged node, the converged node can send fourth information to the first sensing node. This fourth information can be used to instruct the first sensing node to send its own sensing data to the converged node. Then, the fusion node can receive the sensing data from the first sensing node.

[0022] Fifthly, embodiments of this application provide a communication method applicable to intelligent agent nodes. The intelligent agent node can be a terminal, access network device, core network device, or network management device; or it can be a communication module, circuit or chip responsible for communication functions (such as a modem chip, or a SoC chip or SIP chip containing a modem core), chip system, or processor within a terminal, access network device, core network device, or network management device; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of a terminal, access network device, core network device, or network management device. The method may include: when an AI model in a fusion node does not support a first modality, or when one or more AI models in a fusion node support the first modality, and the computational power required for the one or more AI models to fuse the perception data of multiple sensing nodes is greater than the computational power of the fusion node, the intelligent agent node can receive a second request from the fusion node. The second request can be used to request a second AI model, which supports the first modality, and the computational power required for the second AI model to fuse the perception data of multiple sensing nodes is less than or equal to the computational power of the fusion node. The first modality may be the modality of the perception data of a first sensing node.

[0023] Based on the fourth aspect, in one possible design, the method may further include: if the AI ​​model in the fusion node does not support the first modality, or if one or more AI models in the fusion node support the first modality, and the computational power required for the one or more AI models to fuse the perception data from multiple perception nodes is greater than the computational power of the fusion node, the fusion node may send a second request to the agent node. The second request may be used to request a second AI model, which supports the first modality, and the computational power required for the second AI model to fuse the perception data from multiple perception nodes is less than or equal to the computational power of the fusion node.

[0024] Based on the fourth or fifth aspect, in one possible design, if the AI ​​model in the fusion node does not support the first modality, the second request may include: information about the first modality; or, if one or more AI models in the fusion node support the first modality, and the computing power required for the one or more AI models to fuse the perception data of multiple perception nodes is greater than the computing power of the fusion node, the second request may include at least one of the following: parameters of each of the multiple perception nodes, and the number of the multiple perception nodes.

[0025] Based on the fourth or fifth aspect, in one possible design, the method may further include: the agent node sending fifth information to the fusion node; correspondingly, the fusion node receiving the fifth information from the agent node. The fifth information may include a second AI model, or it may be used to indicate the download address of the second AI model.

[0026] Based on the fourth or fifth aspect, in one possible design, the method may further include: the fusion node can send sixth information to the agent node; correspondingly, the agent node can receive the sixth information from the fusion node. The sixth information may be auxiliary information required for fusing sensory data from multiple sensing nodes.

[0027] Based on the fourth or fifth aspect, in one possible design, the sixth information may include at least one of the following: the configuration of an AI model that fuses the perception data of multiple perception nodes; or the configuration of multiple perception nodes.

[0028] Sixthly, embodiments of this application provide a communication method. This method can be applied to a fusion node, which can be a terminal, access network device, core network device, or network management device; or it can be a module, communication module, circuit or chip (such as a modem chip, or a SoC chip or SIP containing a modem core), chip system, or processor in a terminal, access network device, core network device, or network management device; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of a terminal, access network device, core network device, or network management device. The method may include: the fusion node receiving a third request from an intelligent agent node. The third request may be used to request sensing data for a sensing target. Then, the fusion node sending second sensing data to the intelligent agent node. The second sensing data may be obtained by fusing sensing data from multiple sensing nodes according to the sensing target.

[0029] This method allows the fusion node to fuse sensing data from multiple sensing nodes based on the sensing target, obtaining second sensing data, which is then transmitted. In this way, the sensing data fused and transmitted by the fusion node is related to the sensing target, and the fusion node can avoid fusing and transmitting sensing data unrelated to the sensing target. This reduces the complexity of the fusion node, lowers signaling overhead, and saves transmission resources.

[0030] Furthermore, in this method, both the fusion node and the agent node can be deployed with agents, thereby improving network intelligence and enhancing network maintenance and / or operational efficiency.

[0031] In a seventh aspect, embodiments of this application provide a communication method applicable to intelligent agent nodes. The intelligent agent node can be a terminal, access network device, core network device, or network management device; or it can be a module, communication module, circuit or chip responsible for communication functions (such as a modem chip, or a SoC chip or SIP chip containing a modem core), chip system, or processor within the terminal, access network device, core network device, or network management device; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of the terminal, access network device, core network device, or network management device. The method may include: the intelligent agent node sending a third request to a fusion node, the third request being used to request sensing data for a sensing target. Then, the intelligent agent can receive second sensing data from the fusion node, the second sensing data being obtained by fusing sensing data from multiple sensing nodes according to the sensing target.

[0032] Eighthly, embodiments of this application provide a communication method. This method can be applied to a first sensing node, which can be a terminal, access network device, or core network device; or it can be a module, communication module, circuit or chip responsible for communication functions (such as a modem chip, or a SoC chip or SIP chip containing a modem core), chip system, or processor within the terminal, access network device, or core network device; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of the terminal, access network device, or core network device. The method may include: after receiving fourth information from a fusion node, the first sensing node may send its sensing data to the fusion node. The fourth information can be used to instruct the first sensing node to send its sensing data to the fusion node.

[0033] Based on the sixth or eighth aspect, in one possible design, the first sensing node is any one of multiple sensing nodes. The method may further include: when the first sensing node is related to the sensing target, the fusion node may send fourth information to the first sensing node; correspondingly, when the first sensing node is related to the sensing target, the first sensing node may receive the fourth information from the fusion node. The fourth information can be used to instruct the first sensing node to send its sensing data to the fusion node. Then, the fusion node may receive the first sensing node's sensing data from the first sensing node. With this design, the fusion node only sends the fourth information to the first sensing node when the first sensing node is related to the sensing target. If the fourth information is received, the first sensing node may send its sensing data to the fusion node; if the fourth information is not received, the first sensing node does not send its sensing data to the fusion node. In this way, the first sensing node can send its sensing data as needed according to the fusion node's instructions, thereby avoiding unnecessary transmission of its sensing data when the first sensing node is unrelated to the sensing target, and thus saving transmission resources.

[0034] Based on the sixth or eighth aspect, in one possible design, the method may further include: the fusion node can send information indicating a sensing target to the first sensing node; correspondingly, the first sensing node can receive the information indicating the sensing target from the fusion node. Then, the first sensing node can send its sensing data to the fusion node; correspondingly, the fusion node can receive the first sensing node's sensing data from the first sensing node. The sensing data of the first sensing node may be related to the sensing target. Through this design, the first sensing node can send sensing data related to the sensing target to the fusion node, and may not send sensing data unrelated to the sensing target, thereby reducing signaling overhead and saving transmission resources.

[0035] Based on the sixth or seventh aspect, in one possible design, the method may further include: the fusion node sending sixth information to the agent node; correspondingly, the agent node receiving the sixth information from the fusion node. The sixth information may be auxiliary information required for fusing the sensory data of multiple sensing nodes. Through this design, the fusion node can report the auxiliary information required for fusing the sensory data of multiple sensing nodes, enabling the agent node to better manage the fusion node accordingly.

[0036] Based on the sixth or seventh aspect, in one possible design, the sixth information may include at least one of the following: the configuration of an AI model that fuses the perception data of multiple perception nodes; or the configuration of multiple perception nodes.

[0037] Ninthly, this application provides a communication device. This communication device can be a module, communication module, circuit or chip responsible for communication functions (such as a modem chip, or a SoC chip or SIP chip containing a modem core), chip system, or processor in a terminal, access network equipment, core network equipment, or network management equipment; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of a terminal, access network equipment, core network equipment, or network management equipment. The communication device has the functionality to implement any one of the first to eighth aspects described above.

[0038] In one possible design, the communication device includes modules, units, or means corresponding to the operations involved in any of the first to eighth aspects described above. These modules, units, or means can be implemented in software, hardware, or a combination of both. For example, the communication device includes an interface unit and a processing unit. The interface unit can be used to send and receive signals to enable communication between the communication device and other devices; the processing unit can be used to perform some internal operations of the communication device. The functions performed by the processing unit and the interface unit can correspond to the operations involved in any of the first to eighth aspects described above.

[0039] In one possible design, the communication device includes a processor. The processor is capable of executing computer programs or instructions that, when executed, cause the communication device to implement the methods in any of the possible designs of any of the first to eighth aspects described above.

[0040] In one possible design, the communication device includes a processor and a memory, the memory of which can store the necessary computer programs or instructions for implementing the functions described in the first aspect above. The processor can execute the computer programs or instructions stored in the memory, and when the computer programs or instructions are executed, cause the communication device to implement the methods in any possible design of any of the first to eighth aspects described above.

[0041] In one possible design, the communication device includes a processor and an interface circuit, wherein the processor is used to communicate with other devices through the interface circuit and to execute the methods in any possible design of any of the first to eighth aspects described above.

[0042] In a tenth aspect, this application provides a communication system that may include one or more of a fusion node, an agent node, and a first sensing node. The fusion node may execute the communication method provided in the first aspect, the agent node may execute the communication method provided in the second aspect, and the first sensing node may execute the communication method provided in the third aspect; or, the fusion node may execute the communication method provided in the fourth aspect, the agent node may execute the communication method provided in the fifth aspect, and the first sensing node may execute the communication method provided in the third aspect; or, the fusion node may execute the communication method provided in the sixth aspect, the agent node may execute the communication method provided in the seventh aspect, and the first sensing node may execute the communication method provided in the eighth aspect.

[0043] In one aspect, this application provides a computer-readable storage medium storing a computer program or instructions, wherein when the computer program or instructions are executed, a method in any possible design of any of the first to eighth aspects described above is implemented.

[0044] In a twelfth aspect, this application provides a computer program product comprising computer program code, wherein when the computer program code is run, a method in any possible design of any of the first to eighth aspects described above is implemented.

[0045] In a thirteenth aspect, this application provides a chip for reading a computer program stored in a memory to execute a method in any possible design of any of the first to eighth aspects described above.

[0046] The technical effects that can be achieved by any of the second to fifth aspects and the seventh to thirteenth aspects mentioned above can be described with reference to the technical effects that can be achieved by any possible design in the first or sixth aspect mentioned above. Where there is overlap, no further discussion will be given. Attached Figure Description

[0047] Figures 1 to 3 Architecture diagrams of several communication systems provided in the embodiments of this application;

[0048] Figure 4 This is a schematic diagram of the structure of an intelligent agent provided in an embodiment of this application;

[0049] Figures 5A to 5D Flowcharts of several methods for fusing sensing data provided in embodiments of this application;

[0050] Figures 6 to 9 Flowcharts of several communication methods provided in the embodiments of this application;

[0051] Figures 10 to 13Structural diagrams of several communication devices provided in the embodiments of this application. Detailed Implementation

[0052] The technical solutions in the embodiments of this application will be described below with reference to the accompanying drawings. The technical solutions in the embodiments of this application can be applied to various communication systems, such as wireless local area networks (WLANs), wireless fidelity (Wi-Fi or WiFi) systems, fourth-generation (4G) mobile communication systems (such as long-term evolution (LTE) systems), fifth-generation (5G) mobile communication systems (such as new radio (NR) systems), or future communication systems. The methods provided in the embodiments of this application can be applied to terrestrial network communication systems or non-terrestrial network (NTN) communication systems. NTN communication systems can be, for example, satellite communication systems, and may also include unmanned aerial vehicles (UAVs), high-altitude platform stations (HAPS), and other aerial access network equipment; this application does not limit these aspects.

[0053] This application will present various aspects, embodiments, or features relating to systems that may include multiple devices, components, modules, etc. It should be understood and appreciated that individual systems may include additional devices, components, modules, etc., and / or may not include all the devices, components, modules, etc. discussed in conjunction with the accompanying drawings. Furthermore, combinations of these approaches are also possible.

[0054] To facilitate understanding of the embodiments of this application, Figure 1 A possible, non-limiting system schematic diagram is shown. For example... Figure 1 As shown, the communication system 10 includes a radio access network (RAN) 100 and a core network (CN) 200. Optionally, the communication system 10 may also include an Internet 300.

[0055] RAN 100 includes at least one RAN node (such as...) Figure 1 110a and 110b (collectively referred to as 110) and at least one terminal (such as Figure 1 RAN 100, denoted as RAN 120a-120j, is collectively referred to as RAN 120. RAN 100 may also include other RAN nodes, such as wireless relay equipment and / or wireless backhaul equipment. Figure 1(Not shown in the image). Terminal 120 is connected to RAN node 110 wirelessly. RAN node 110 is connected to core network 200 wirelessly or via wired connection. The core network equipment in core network 200 and RAN node 110 in RAN 100 can be different physical devices, or they can be the same physical device integrating core network logical functions and radio access network logical functions.

[0056] RAN 100 can be a cellular system related to the 3rd Generation Partnership Project (3GPP), such as 4G, 5G mobile communication systems, or future-oriented evolution systems. RAN 100 can also be an open access network (O-RAN or ORAN), a cloud radio access network (CRAN), or a WiFi system. RAN 100 can also be a communication system that integrates two or more of the above systems.

[0057] RAN node 110, sometimes referred to as RAN entity or access node, constitutes part of the communication system and assists terminals in achieving wireless access. Multiple RAN nodes 110 in communication system 10 can be of the same type or different types. In some scenarios, the roles of RAN node 110 and terminal 120 are relative, for example... Figure 1 Network element 120i can be a helicopter or a drone, and it can be configured as a mobile base station. For terminals 120j that access RAN 100 through network element 120i, network element 120i is a base station; however, for base station 110a, network element 120i is a terminal. RAN node 110 and terminal 120 are sometimes referred to as communication devices, for example... Figure 1 Network elements 110a and 110b can be understood as communication devices with base station functions, while network elements 120a-120j can be understood as communication devices with terminal functions.

[0058] RAN nodes can also be described in different ways, such as access network equipment. Unless otherwise specified in this application, access network equipment will be used as the term.

[0059] Access network equipment can be devices or modules located on the network side of the aforementioned communication system and possessing corresponding communication functions. Access network equipment typically contains communication modules, circuits, or chips that perform the corresponding communication functions. Access network equipment may also be configured with programs or instructions for performing the corresponding communication functions, as well as the corresponding programs or instructions themselves.

[0060] In one possible scenario, access network equipment can be a base station (BS), an evolved NodeB (eNodeB), a transmitting point (TP), an access point (AP), a transmission reception point (TRP), a mobile switching center, a next-generation NodeB (gNB), a next-generation base station in a future communication system, or an access node in a WiFi system, etc. Access network equipment can also be a macro base station (such as...). Figure 1 110a), micro base stations or indoor stations (such as Figure 1 The access network device can be a relay node or donor node (as described in section 110b), or a wireless controller, satellite, drone, balloon, or aircraft in a CRAN scenario. Optionally, the access network device can also be a server, wearable device, vehicle, or in-vehicle equipment. For example, the access network device in vehicle-to-everything (V2X) technology can be a roadside unit (RSU). All or part of the functions of the access network device in this application can also be implemented through software functions running on hardware, or through virtualization functions instantiated on a platform (e.g., a cloud platform). The access network device in this application can also be a logical node, logical module, or software capable of implementing all or part of the access network device functions.

[0061] In another possible scenario, multiple access network devices collaborate to assist the terminal in achieving wireless access, with each device performing a portion of the base station's functions. For example, the access network devices can be a central unit (CU), a distributed unit (DU), a CU-control plane (CP), a CU-user plane (UP), or a radio unit (RU). The CU and DU can be separate entities or included in the same network element, such as a baseband unit (BBU). The RU can be included in radio frequency equipment or radio frequency units, such as a remote radio unit (RRU), an active antenna unit (AAU), or a remote radio head (RRH).

[0062] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called an open CU (O-CU), DU can also be called an open DU (O-DU), CU-CP can also be called an open CU-CP (O-CU-CP), CU-UP can also be called an open CU-UP (O-CU-UP), and RU can also be called an open RU (O-RU). Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software and hardware modules.

[0063] A terminal is a device or module that connects to the aforementioned communication system and possesses corresponding communication functions. A terminal can also be called a terminal device, user equipment (UE), mobile station, mobile terminal, wireless terminal device, subscriber unit, subscriber station, mobile station, remote station, user terminal, user agent, or user device, etc. A terminal typically contains communication modules, circuits, or chips that perform the corresponding communication functions. The terminal can also be configured with programs or instructions for performing these communication functions.

[0064] Terminals can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, and smart cities. Terminals can be mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices, vehicles, drones, helicopters, airplanes, ships, robots, robotic arms, smart home devices, etc. Wearable devices, also known as wearable smart devices or smart wearable devices, are a general term for devices that utilize wearable technology to intelligently design and develop everyday wearables. Terminals used in vehicles are called in-vehicle terminal devices, which include, for example, transportation vehicles with wireless communication capabilities, communication modules, or on-board units (OBUs).

[0065] For example, a terminal may include a mobile phone (or "cellular" phone), a computer with a mobile terminal device, or a portable, pocket-sized, handheld, or computer-embedded mobile device. For instance, a terminal may be a Personal Communication Service (PCS) phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA), or other similar devices. A terminal may also include restricted devices, such as devices with limited power consumption, limited storage capacity, or limited computing power. For example, a terminal may be an information sensing device such as a barcode scanner, radio frequency identification (RFID), a sensor, a global positioning system (GPS), or a laser scanner. The embodiments of this application do not limit the device form of the terminal.

[0066] Communication between access network devices and terminals follows a specific protocol layer structure. This protocol layer may include a control plane protocol layer and a user plane protocol layer. The control plane protocol layer may include at least one of the following: radio resource control (RRC) layer, packet data convergence protocol (PDCP) layer, radio link control (RLC) layer, media access control (MAC) layer, or physical (PHY) layer, etc. The user plane protocol layer may include at least one of the following: service data adaptation protocol (SDAP) layer, PDCP layer, RLC layer, MAC layer, or physical layer, etc.

[0067] The correspondence between network elements and their achievable protocol layer functions in the ORAN system can be found in Table 1 below.

[0068] Table 1

[0069] ORAN network elements 3GPP protocol layer functions O-CU-CP RRC+PDCP-Control Plane (PDCP-C) O-CU-UP SDAP+PDCP - User Plane (PDCP-U) O-DU RLC+MAC+PHY-high O-RU PHY-low

[0070] In this application, core network equipment refers to equipment in the core network that provides service support for terminals. For example, if CN200 is the core network of a future communication system, or a 5G core network, or an evolved 5G core network, then some examples of core network equipment include: access and mobility management function (AMF) entities, session management function (SMF) entities, user plane function (UPF) entities, policy control function (PCF) entities, etc., which are not listed here. Among them, the AMF entity can be responsible for terminal access management and mobility management; the SMF entity can be responsible for session management, such as user session establishment; the UPF entity can be a user plane functional entity, mainly responsible for connecting to external networks. For example, if CN200 is a 4G core network, some core network devices include: Mobility Management Entity (MME), Home Subscriber Server (HSS), Serving Gateway (S-GW), Policy and Charging Rules Function (PCRF), Public Data Network Gateway (PDN Gateway, P-GW), etc., which will not be listed here. It should be noted that in this application, entities can also be referred to as network elements or functional entities. For example, an AMF entity can also be called an AMF network element or AMF functional entity, and an SMF entity can also be called an SMF network element or SMF functional entity, etc. The above-mentioned core network devices can work independently or be combined to implement certain control functions. For example, AMF, SMF, and PCF can be combined into a single core network device.

[0071] The communication system 10 provided in this application may also include AI network elements for implementing some or all AI-related operations. AI network elements may also be referred to as AI nodes, AI devices, AI entities, AI modules, AI models, or AI units, etc. In some examples, AI network elements may be built into the network elements of the communication system. For example, an AI network element may be an AI module built into access network equipment, core network equipment, cloud servers, or operation, administration and maintenance (OAM) management systems to implement AI-related functions. OAM may be the network management system of the core network equipment and / or the network management system of the access network equipment. In other examples, AI network elements may be independently configured network elements within the communication system. In still other examples, AI network elements may be included in a terminal or a chip built into the terminal to implement AI-related functions.

[0072] Figure 2 This is a schematic diagram illustrating another communication system provided in an embodiment of this application. For example... Figure 2 As shown, devices in a communication system are connected via interfaces (e.g., NG, Xn) or air interfaces. These devices, such as core network devices, access network devices, terminals, or one or more devices in the OAM, are equipped with one or more AI modules (for clarity, ...). Figure 2 (Only one is shown in the image). The access network device can be a single RAN node or can include multiple RAN nodes, such as CU and DU. One or more AI modules can also be configured in the CU and / or DU. Optionally, the CU can also be split into CU-CP and CU-UP. One or more AI modules can be configured in the CU-CP and / or CU-UP.

[0073] It should be understood that AI modules deployed in different network elements can be the same or different. The models of AI modules are configured with different parameters, enabling them to perform different functions. The models of AI modules can be configured based on one or more of the following parameters: structural parameters (e.g., at least one of the following: number of neural network layers, neural network width, inter-layer connections, neuron weights, neuron activation function, or biases in the activation function), input parameters (e.g., type and / or dimension of input parameters), or output parameters (e.g., type and / or dimension of output parameters). The biases in the activation function can also be referred to as neural network biases.

[0074] An AI module can have one or more models (or AI models). A model can produce an output, which may include one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or they can be deployed on the same node or device.

[0075] Figure 3 A schematic diagram of yet another communication system provided in an embodiment of this application is shown. For example... Figure 3 As shown, this network architecture can be implemented through a hierarchical structure of artificial general intelligence (AGI) within the network, with parameters / capabilities decreasing in size and functions progressing from high to low levels. In practical applications, this application does not limit the number of agents or the number of node levels included in this network architecture.

[0076] like Figure 3 As shown in (a), taking an example where all four levels of nodes possess an agent, the order of these four levels of nodes from upper to lower level is: Level 0 (L0), Level 1 (L1), Level 2 (L2), and Level 3 (L3). From left to right, it represents the process of the upper-level node instructing the lower-level node on decisions (e.g., configuration refinement); from right to left, it represents the process of the lower-level node providing perception data to the upper-level node. This perception data may include the original perception results and / or the processing results of the perception data by the lower-level node.

[0077] like Figure 3 As shown in (b), the network architecture includes a central agent (C-agent), access network equipment, and terminals. The access network equipment deploys an agent (which may be referred to as the BS agent, A-BS), and the terminals also deploy agents (which may be referred to as the UE agent, A-UE). The terminal can sense data of at least one modality; in other words, the terminal can acquire sensed data of at least one modality, such as sensed data of at least one modality from text, audio, or images. The A-UE can fuse the sensed data acquired by the terminal and report it to the access network equipment. The A-BS can fuse the sensed data reported by the terminal with the sensed data sensed by the access network equipment (e.g., sensed data of modalities such as video and / or point clouds), and report the fused sensed data to the C-agent so that the C-agent can make decisions based on the fused sensed data.

[0078] In a communication network, different communication nodes, from upper-level nodes to lower-level nodes, may include the following order: network management equipment (e.g., network management OAM equipment and / or over-the-top (OTT) vendor equipment), core network equipment, access network equipment, and terminals. Optionally, in addition to these implementations, future communication networks may also include other device forms, which can also be applied to… Figure 3 The architecture shown.

[0079] Optionally, the core network equipment may include nodes of different levels, which may be ordered as follows from the upper-level node to the lower-level node: PCF network element, SMF network element, UPF.

[0080] Optionally, the access network equipment may include nodes of different levels, and the order from the upper-level node to the lower-level node may be as follows: CU, DU, RU; or, in the access network equipment of the O-RAN architecture, the order from the upper-level node to the lower-level node may be as follows: O-CU, O-DU, O-RU.

[0081] For example, taking the terminal as a UE, Figure 3 The fourth-level node shown in (a) can be represented as any row in Table 2.

[0082] Table 2

[0083] L0 L1 L2 L3 CU DU RU UE CN CU DU UE PCF SMF RAN UE PCF SMF UPF Access network equipment OAM / OTT CU DU UE OAM / OTT CN RAN UE OAM / OTT CN1 CN2 Access network equipment

[0084] It should be understood that the above naming is defined solely for the purpose of distinguishing different functions and should not constitute any limitation on this application. This application does not preclude the possibility of using other naming conventions in 5G networks and other future networks. For example, in future communication networks, some or all of the above-mentioned network elements may use the terminology from 5G, or they may use other names, etc.

[0085] Understandable. Figures 1 to 3 This is merely an example and does not constitute a limitation on the scope of protection of this application. The communication method provided in the embodiments of this application may also involve... Figures 1 to 3 Network elements not shown may also include wireless relay devices and wireless backhaul devices; each device may also include different functional units; the communication method provided in this application embodiment may also only include Figures 1 to 3 Some of the network elements are shown.

[0086] The communication systems and service scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new service scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0087] The relevant terms used in the embodiments of this application will be explained below. It should be noted that these explanations are for the purpose of making the embodiments of this application easier to understand, and should not be regarded as a limitation on the scope of protection claimed by this application.

[0088] 1. AI: AI enables machines to possess human-like intelligence, for example, allowing machines to use computer hardware and software to simulate certain intelligent human behaviors. To achieve artificial intelligence, machine learning methods can be employed. In machine learning, machines learn (or train) models using training data. This model represents the mapping between inputs and outputs. The learned model can be used for reasoning (or prediction), that is, it can be used to predict the output corresponding to a given input. This output can also be called the reasoning result (or prediction result).

[0089] 2. AI Model:

[0090] AI models can represent the mapping relationship between the model's input and output. AI models can be neural networks, linear regression models, decision tree models, support vector machines (SVMs), Bayesian networks, Q-learning models, or other machine learning models. Among these, neural networks can be deep neural networks, such as convolutional neural networks, recurrent neural networks, fully connected neural networks, or transformer networks.

[0091] AI models can be used to implement AI functions, which can be functions of functional modules. For example, an AI function may include at least one of the following: data collection (collecting training data and / or inference data), data preprocessing, model training (or model learning), model information dissemination (configuring model information), model validation, model inference, or inference result dissemination. Inference can also be referred to as prediction.

[0092] When an AI model is located on a device (e.g., a terminal, access network device, or core network device), for a specific functional module, one or more AI models within that device may be able to implement the function of that module. For example, for the channel state information (CSI) measurement result prediction function, one or more AI models in the terminal can provide prediction results.

[0093] In this application, the AI ​​model may be replaced with other names, such as AI algorithm, fusion model or fusion algorithm, without limitation.

[0094] 3. Large-scale models: These refer to neural network models containing an extremely large number of parameters (usually over one billion), and have the following characteristics:

[0095] 1) Huge scale: Large models can contain billions of parameters, and their size can reach hundreds of gigabytes (GB) or even larger. This huge model scale provides powerful expressive and learning capabilities.

[0096] 2) Multi-task learning: Large models typically learn one or more of various natural language processing (NLP) tasks, such as machine translation, text summarization, or question answering systems. This allows the model to learn broader and more generalized language understanding capabilities. Optionally, large models may also include multimodal models or domain-specific models. Multimodal models can take data from modalities such as video, images, speech, or point clouds as input; domain-specific models are used to perform tasks in a specific domain (such as tasks in the communications domain) and can be trained or fine-tuned using domain-specific data.

[0097] 3) Powerful computing resources: Training large models typically requires hundreds or even thousands of graphics processing units (GPUs) and a significant amount of time, usually ranging from weeks to months. This can accelerate the training process while preserving the ability to train large models.

[0098] 4) Abundant data: Large models require a large amount of data for training, and a large amount of training data can leverage the advantages of the parameter scale of large models.

[0099] Large models are widely used in the field of natural language processing (NLP) and are transforming NLP tasks, giving rise to more powerful and intelligent language technologies. Large models are also a key direction in the development of AI technology. Currently, large models demonstrate outstanding capabilities in various NLP tasks, such as text classification, sentiment analysis, summarization, and translation. Furthermore, large models can be used in multiple fields, including automated writing, chatbots, virtual assistants, voice assistants, and automated translation.

[0100] 4. Intelligent Agent: This is a concept in the field of AI. Any entity capable of independent thought and interaction with its environment can be abstracted as an intelligent agent. Intelligent agents are inseparable from AI; they possess some human-like intelligent abilities and behaviors, such as learning, reasoning, decision-making, and execution capabilities. The basic characteristics of an intelligent agent are: it can react to changes in its environment and automatically adjust its behavior and state; different intelligent agents can also interact with other intelligent agents according to their own intentions. Intelligent agents can be considered a type of AI module.

[0101] For example, an intelligent agent can be centered around an AI model (e.g., a large language model (LLM)) and include a memory module, a tool module, a planning module, and an action module. The memory module can be used to implement long-term and / or short-term memory functions; the tool module can contain multiple callable external tools; the planning module can contain various planning algorithms, for example, the agent can plan externally input tasks based on the content in its memory; and the action module can support the agent in performing actions based on the planning results, such as calling up tools.

[0102] Figure 4 A schematic diagram of the structure of an intelligent agent is shown. For example... Figure 4 As shown, in an LLM-supported autonomous agent system, the LLM can act as the brain of the agent and includes the following components:

[0103] 1) Planning:

[0104] Planning includes one or more of the following: reflection, self-criticism, chain of thoughts (CoT), and subgoal decomposition.

[0105] Self-reflection and self-criticism can be used for reflection and improvement. For example, an intelligent agent can engage in self-criticism and self-reflection on past behaviors, learn from mistakes, and refine future steps to improve the quality of the final result.

[0106] Thought chains and sub-goal decomposition can be used for task management. For example, an agent can break down large or difficult tasks into smaller, manageable sub-goals (or sub-tasks) based on sub-task decomposition, thereby effectively handling complex tasks. As another example, an agent can "think step-by-step" based on thought chains to break down large or difficult tasks into smaller, simpler steps.

[0107] 2) Memory: including short-term memory and / or long-term memory.

[0108] Short-term memory: Learning can be achieved by utilizing the short-term memory of models.

[0109] Long-term memory: Provides an agent with the ability to retain and recall (unlimited) information for a long time, usually by utilizing external vector storage and fast retrieval.

[0110] 3) Tool usage:

[0111] The agent can call external application programming interfaces (APIs) to obtain additional information missing from the model weights (which is usually difficult to change after pre-training), including current information, code execution capabilities, and access to proprietary information sources.

[0112] For example, the tool may include, but is not limited to, at least one of the following: calendar, calculator, code interpreter, or search tool.

[0113] 4) Task execution (action): The agent performs a specific task and records the results.

[0114] It should be understood that the intelligent agent in this application may have other names, such as AGI, AI, AI controller, AI agent, agent, intelligent agent, intelligent unit or intelligent entity, etc. As long as they have the same function, they are all within the protection scope of this application.

[0115] Optionally, in a communication device, the intelligent agent can be integrated into the existing hardware / software of the communication device, or the intelligent agent can be independent of the existing hardware / software of the communication device. For example, the existing hardware / software may include chips, baseband chips, modem chips, system-on-chip (SoC) chips containing modem cores, system-in-package (SIP) chips, communication modules, chip systems, processors, logic modules, or software, etc.

[0116] 5. In this application, "instruction" or "for instruction" can include both direct and indirect instruction. When describing information as being used to instruct A, it can include whether the information directly or indirectly instructs A, but does not necessarily mean that the information carries A.

[0117] The indication methods involved in the embodiments of this application should be understood to cover various methods that enable the party to be indicated to obtain the information to be indicated. The information to be indicated can be sent as a whole or divided into multiple sub-information and sent separately. Moreover, the sending period and / or sending time of these sub-information can be the same or different. This application does not limit the sending method, for example.

[0118] In the embodiments of this application, "information" can be an explicit indication, that is, a direct indication through signaling, or obtained by combining other rules or parameters with parameters indicated by signaling, or by deduction. It can also be an implicit indication, that is, obtained based on rules or relationships, or based on other parameters, or by deduction. This application does not specifically limit it in this regard.

[0119] 6. In this application, communication between different devices can refer to direct communication between different devices (i.e., without the need for relaying or forwarding by other devices), or communication between different devices through other devices (i.e., requiring relaying or forwarding by other devices), or communication between a functional unit within a device and other devices through another functional unit. For example, "sending information to…(terminal)" can be understood as the destination of the information being the terminal, and may include sending information directly or indirectly to the terminal. "Receiving information from…(terminal)" can be understood as the source of the information being the terminal, and may include receiving information directly or indirectly from the terminal. Information may undergo necessary processing between the source and destination, such as format changes, digital-to-analog conversion, amplification, filtering, etc., but the destination can understand the valid information from the source. Similar expressions in this application can be understood in a similar way, and will not be elaborated further here.

[0120] 7. In this application, "modality" can also be replaced with (or understood as) "type," etc. For example, the modality of the perceived data may include at least one of the following: text, video, image, voice, point cloud, radar data, or channel data. Here, a point cloud can be a dataset of points in space, which can be used to represent a three-dimensional image or object, and is typically acquired through a 3D scanner. Radar data can be referred to as radar perceived data, and can be acquired through a radar sensor. If the radar sensor is a lidar sensor, the radar data can also be referred to as lidar data.

[0121] 8. In this application, any two of the programs, instructions, and code can be substituted for one another.

[0122] As mentioned above Figure 3 As the description shows, in a communication system that includes devices equipped with intelligent agents, lower-level nodes can provide sensing data to higher-level nodes. How to implement this sensing data provision process requires further investigation.

[0123] To facilitate understanding of this application, the implementing entity (or operating entity) involved in this application will be explained below.

[0124] A sensing node can be a node that performs (or conducts) sensing. A sensing node may also have other names, such as sensor, without limitation. A sensing node can sense data of at least one modality. For example, the at least one modality may include one or more of text, video, images, voice, point cloud, or channel data (e.g., CSI). A sensing node may or may not have an agent deployed. Optionally, a sensing node may be a terminal, access network device, or core network device; or it may be a module, communication module, circuit or chip responsible for communication functions (such as a modem chip, or a SoC chip or SIP chip containing a modem core), chip system, or processor within a terminal, access network device, or core network device; or it may be a logical node, logical module, or software capable of implementing all or part of the functions of a terminal, access network device, or core network device.

[0125] A fusion node is a node that fuses perceived data. The perceived data fused by a fusion node can include at least one of the following: data perceived by the fusion node, or perceived data received by the fusion node from other perceived nodes. Specifically, the perceived data received by the fusion node from other perceived nodes can be the original data perceived by those other perceived nodes, or the data after processing the original data by those other perceived nodes (e.g., dimension alignment, upsampling, downsampling, or one or more of these processes), or features obtained after processing the original data through a neural network, or perceived data fused by other perceived nodes.

[0126] In some examples, the sensing data fused by the fusion node may include sensing data received from other sensing nodes. For example, such as Figure 5A As shown, in S501a, the fusion node can receive sensing data from sensing node #1. In S502a, the fusion node can receive sensing data from sensing node #2. In S503a, the fusion node can fuse the sensing data from sensing node #1 and sensing data from sensing node #2. The order of S501a and S502a is not important.

[0127] In other examples, the sensing data fused by the fusion node may include: data sensed by the fusion node itself, and sensing data received from other sensing nodes. For example, such as... Figure 5B As shown, in S501b, the fusion node can receive the sensing data from sensing node #1. In S502b, the fusion node can acquire its sensed data. In S503b, the fusion node can fuse the sensing data from sensing node #1 and the data sensed by the fusion node. The order of S501b and S502b is not important.

[0128] In other examples, the sensed data fused by the fusion node may include data from multiple modalities sensed by the fusion node. For example, such as Figure 5C As shown, in S501c, the fusion node can sense data from multiple modalities. In S502c, the fusion node can fuse this data from multiple modalities.

[0129] Intelligent agents can be deployed in the fusion node.

[0130] Optionally, the converged node may be a terminal, access network equipment, core network equipment, or network management equipment, or it may be a module, communication module, circuit or chip responsible for communication function (such as a modem chip, or a SoC chip or SIP chip containing a modem core), chip system or processor in the terminal, access network equipment, core network equipment, or network management equipment, or it may be a logical node, logical module or software that can realize all or part of the functions of the terminal, access network equipment, core network equipment, or network management equipment.

[0131] Intelligent agent nodes can manage converged nodes and / or acquire converged perception data from the converged nodes. Intelligent agent nodes may include intelligent agents. Optionally, intelligent agent nodes may be terminals, access network devices, core network devices, or network management devices; or they may be modules, communication modules, circuits or chips responsible for communication functions (such as modem chips, or SoC chips or SIP chips containing modem cores), chip systems, or processors within terminals, access network devices, core network devices, or network management devices; or they may be logical nodes, logical modules, or software capable of implementing all or part of the functions of terminals, access network devices, core network devices, or network management devices.

[0132] The sensing node and the fusion node can be located in the same device, in which case the sensing node can be the fusion node; or, the sensing node and the fusion node can be located in different devices, for example, the sensing node can be the next-level node of the fusion node. When the fusion node merges sensing data from multiple sensing nodes, these multiple sensing nodes can be located in the same device or in different devices; and / or, some or all of the multiple sensing nodes and the fusion node can be located in the same device, or each of the multiple sensing nodes and the fusion node can be located in different devices.

[0133] In some examples, the fusion node and the agent node can reside in the same device, thus the fusion node can also be an agent node. In this case, Figures 5A to 5C The fusion node can be an agent node. After fusing the sensing data, the fusion node may choose not to send the fused sensing data back. Optionally, in this example, Figure 5A and Figure 5BThe method shown is applicable to the following scenarios: the sensing node sending sensing data cannot perform sensing data fusion (e.g., multimodal sensing data). Here, the sensing node sending sensing data is, for example, a... Figure 5A The sensing nodes #1 and #2 in the middle, or are Figure 5B The sensing node #1 in the example. For instance, the computing power of the sensing node sending the sensing data is insufficient for fusion; in other words, the computing power of the sensing node sending the sensing data is less than the computing power required to fuse the sensing data. Another example is that the AI ​​model in the sensing node sending the sensing data cannot fuse the sensing data; or, the sensing node sending the sensing data does not include an AI model capable of fusing sensing data. And / or, in this example, Figure 5A The method shown can be applied to the following scenarios: sensing nodes cannot communicate with each other, or in other words, sensing nodes cannot connect with each other.

[0134] In other examples, the fusion node and the agent node may reside in different devices; for example, the fusion node may be a next-level node in the agent node's hierarchy. In this case, Figures 5A to 5C After fusing the sensing data, the fusion node can send the fused sensing data to the agent node. For example, such as... Figure 5D As shown, in S501a, the fusion node can receive sensing data from sensing node #1. In S502a, the fusion node can receive sensing data from sensing node #2. In S503a, the fusion node can fuse the sensing data from sensing node #1 and sensing data from sensing node #2. In S504a, the fusion node can send the fused sensing data to the agent node. Optionally, this example can be applied to at least one of the following scenarios: 1. The sensing node has local fusion capability; in other words, the sensing node can fuse sensing data. For example, Figure 5B and Figure 5C In this context, the fusion node can be a sensing node, meaning there exists a sensing node capable of fusing sensing data. 2. A sensing node includes multiple sensors. For example, in... Figure 5C In this system, fusion nodes can perceive multimodal sensing data through multiple sensors. 3. Sensing nodes can communicate with each other. For example, in… Figure 5B In this context, the fusion node can be a sensing node, and the fusion node can communicate with sensing node #1.

[0135] In addition, if the fusion node is located in a different device from the agent node and the perception node, the fusion node can be a device independent of the agent node and the perception node.

[0136] Intelligent agent nodes and sensing nodes can be located in different devices.

[0137] Below are some possible examples of perception nodes, fusion nodes, and agent nodes. For example, such as... Figure 3 As shown in (b), the sensing node can be a terminal (e.g., A-UE) and an access network device (e.g., A-BS), the convergence node can be an access network device (e.g., A-BS), and the agent node can be a device containing a central agent (e.g., access network device, core network device, or network management device). For example, Figure 3 In (b) of the diagram, the access network device can communicate with multiple terminals (one terminal is shown as an example in the figure). The sensing node can be one of these multiple terminals, the fusion node can be an access network device (e.g., A-BS), and the agent node can be a device containing a central agent (e.g., an access network device, a core network device, or a network management device). For example, Figure 3 In (b) of this example, the access network device can communicate with multiple terminals. The sensing node can be one of these multiple terminals, and the fusion node and the agent node can be one of these multiple terminals. For example, Figure 3 In (b) of this document, the access network device can communicate with multiple terminals. The sensing node can be one of these multiple terminals, the fusion node can be one of these multiple terminals, and the agent node can be the access network device (e.g., A-BS). For example, the sensing node may include... Figure 2 The terminal and DU#1 in the middle, the fusion node can be Figure 2 In DU#1, the agent node can be Figure 2 CU#1 or DU#1 in it. For example, Figure 2 The CU#1 in the diagram can communicate with multiple DUs. The sensing node can include these multiple DUs (the diagram shows one DU as an example). The fusion node can be CU#1, and the agent node can be CU#1, a core network device, or OAM.

[0138] This application provides a communication method. Figure 6 This is a flowchart illustrating the communication method provided in an embodiment of this application. Figure 6 As shown, the method includes:

[0139] S601: The fusion node sends the first information to the agent node; correspondingly, the agent node receives the first information from the fusion node.

[0140] The first piece of information can be used to indicate the ability of the fusion node to fuse sensing data; in other words, the fusion node can report its fused sensing data to the agent node.

[0141] In some possible ways, the ability of a fusion node to fuse perceived data may include at least one of the following; in other words, the first information may be used to indicate at least one of the following:

[0142] 1. Computational power of the fusion node: Optionally, the computational power of the fusion node may include at least one of the following: the current computational power of the fusion node, or the predicted computational power of the fusion node within the first time period. For example, the current computational power of the fusion node may be 1 Tera (T, i.e., 10^10). 12 Floating-point operations per second (FLOPS). For example, the computing power of a fused node in the first predicted time period could be 1.2T FLOPS. The computing power of a fused node can be its total computing power, available computing power, or remaining computing power. Computing power can be simply referred to as computing power.

[0143] In some implementations, the first information may include the computing power of the fusion node. For example, if the computing power of the fusion node includes the computing power of the current fusion node, and the value of the first information is 1, in units of T FLOPS, then the computing power of the current fusion node may be 1T FLOPS.

[0144] In other implementations, the first information may indicate a first capability range, to which the computing power of the fusion node belongs; in other words, the first information may indicate the capability range to which the computing power of the fusion node belongs, or the first information may indicate which capability range the computing power of the fusion node belongs to. For example, there are computing capability ranges #1 and #2. The computing power in computing capability range #1 is greater than or equal to 1.2T FLOPS. The computing power in computing capability range #2 is less than 1.2T FLOPS. If the computing power of the fusion node includes the computing power of the current fusion node, and the first information indicates computing capability range #1, then the computing power of the current fusion node may be greater than or equal to 1.2T FLOPS. In this implementation, each capability range may be predefined, for example, specified by the protocol; or it may be reported by the fusion node; or it may be notified to the fusion node by the agent node.

[0145] 2. The fusion node supports (or includes) at least one AI model that can be used to fuse sensing data: For example, if the fusion node supports AI models #1 to AI models #3, and AI models #1 and AI models #2 can be used to fuse sensing data, then the at least one AI model may include AI model #1 and AI model #2.

[0146] For example, the first information may include the identity (ID) or index of the at least one AI model. For instance, if the first information may include the IDs or indexes of AI model #1 and AI model #2, then the at least one AI model includes AI model #1 and AI model #2.

[0147] This approach offers multiple possibilities for enabling fusion nodes to fuse sensing data, providing greater flexibility. Furthermore, by reporting the computing power of the fusion node and / or at least one AI model supported by the fusion node, the accuracy and effectiveness of the patterns determined by the agent nodes for fusing sensing data can be improved, thereby enhancing the quality and efficiency of the fusion node's fusion of sensing data.

[0148] Optionally, where the ability of the fusion node to fuse sensing data includes the at least one AI model, or where the first information is available to indicate the at least one AI model, the first information may also be used to indicate at least one of the following:

[0149] 1. Modalities supported by at least one AI model: Optionally, the modalities supported by at least one AI model can be the modalities supported by each of the at least one AI models. For example (hereinafter referred to as Example 1), at least one AI model includes AI model #1 and AI model #2. If the modalities supported by AI model #1 include: text and images, and the modalities supported by AI model #2 include: text, video, images, speech, and point clouds, then the first information can indicate: The modalities supported by AI model #1 include: text and images, and the modalities supported by AI model #2 include: text, video, images, speech, and point clouds.

[0150] In some implementations, the first information may explicitly indicate the modalities supported by at least one AI model.

[0151] In some examples, the first information may include the ID or index of the modality supported by each AI model in at least one AI model. For example, the indices for text, video, image, speech, and point cloud are 1, 2, 3, 4, and 5, respectively. The first information includes field #1 and field #2, where field #1 indicates the modality supported by AI model #1, and field #2 indicates the modality supported by AI model #2. If field #1 contains indices 1 and 3, and field #2 contains indices 1 to 5, then the modalities supported by AI model #1 include text and image, and the modalities supported by AI model #2 include text, video, image, speech, and point cloud.

[0152] In other examples, the first information may include at least one bitmap, each bitmap indicating a modality supported by one of at least one AI models. Optionally, the first bitmap is one of at least one bitmaps, and the first bit is any bit in the first bitmap. If the first bit is a first value (e.g., 1 or 0), the AI ​​model corresponding to the first bitmap supports the modality corresponding to the first bit; if the first bit is a second value (e.g., 0 or 1), the AI ​​model corresponding to the first bitmap does not support the modality corresponding to the first bit. The first value and the second value are different. For example, the first information includes bitmap #1 and bitmap #2. The first to fifth bits in bitmap #1 and bitmap #2 correspond to text, video, image, speech, and point cloud, respectively. If bitmap #1 is 10100 and bitmap #2 is 11111, with the first value being 1 and the second value being 0, then the modalities supported by AI model #1 include: text and images, and the modalities supported by AI model #2 include: text, video, images, speech, and point clouds.

[0153] In some other examples, the first information may include the index of at least one bitmap. The specific content of this at least one bitmap can be found in the previous example and will not be repeated here. For example, the first information includes the index of bitmap #1. The first to fifth bits of bitmap #1 correspond to text, video, image, speech, and point cloud, respectively. If the index of bitmap #1 is 20, and the bitmap with index 20 is 10100, the first value is 1, and the second value is 0, then the modalities supported by AI model #1 include: text and image.

[0154] Optionally, when the first information explicitly indicates a modality supported by at least one AI model, the information indicating the modality supported by at least one AI model may be combined with the information indicating the at least one AI model. For example, at least one bitmap indicating the modality supported by at least one AI model may be included in the ID of the at least one AI model; or, the information indicating the modality supported by at least one AI model may be independent of the information indicating the at least one AI model.

[0155] In other implementations, the first information may implicitly indicate the modalities supported by at least one AI model. Optionally, the first information may include information corresponding to the modalities supported by the at least one AI model. For example, the first information may include the ID of the at least one AI model. The agent node may determine the modalities supported by the at least one AI model based on the first correspondence and the ID of the at least one AI model. The first correspondence is the correspondence between the ID of the AI ​​model and the modalities supported by the AI ​​model. The first correspondence may be pre-defined, for example, as specified by a protocol; or it may be reported by the fusion node to the agent node; or it may be notified by the agent node to the fusion node. The form of the first correspondence is not limited. For example, the first correspondence may be represented in tabular form, for example, the first correspondence may be included in a lookup table (LUT) maintained by the fusion node and the agent node.

[0156] Optionally, the modality supported by at least one AI model may be replaced with one of the following: the modality that at least one AI model can process, the modality of the input data of at least one AI model, or the modality of the perceptual data fused by at least one AI model.

[0157] 2. Parameters of at least one modality supported by the AI ​​model:

[0158] The following examples illustrate the parameters of a modality. For instance, if the modality includes an image, the modality parameter could include the image resolution. Similarly, if the modality includes video, the modality parameter could include the video frame rate. And if the modality includes text, the modality parameter could include the size of the file containing that text. It should be understood that the above examples can exist individually or in combination.

[0159] Optionally, the parameters of the modality supported by at least one AI model can be the parameters of the modality supported by each AI model in at least one AI model. For example, at least one AI model includes AI model #1 and AI model #2. If AI model #1 can be used to fuse images with a resolution of less than or equal to 640×480 pixels, and AI model #2 can be used to fuse images with a resolution of 4096×2160 pixels, then the first information can indicate that: the resolution of the images supported by AI model #1 is less than or equal to 640×480 pixels, and the resolution of the images supported by AI model #2 is 4096×2160 pixels. As another example, at least one AI model includes AI model #1 and AI model #2. If AI model #1 can be used to fuse videos with a frame rate of 30 frames per second (FPS), and AI model #2 can be used to fuse videos with a frame rate of 60 FPS, then the first information can indicate that: the frame rate of the videos supported by AI model #1 is 30 FPS, and the frame rate of the videos supported by AI model #2 is 60 FPS.

[0160] In some implementations, the first information may explicitly indicate the parameters of at least one modality supported by the AI ​​model.

[0161] In some examples, the first information may include parameters of the modalities supported by each AI model in at least one AI model. For example, the first information includes fields #3 and #4, where field #3 indicates the parameters of the modalities supported by AI model #1, and field #4 indicates the parameters of the modalities supported by AI model #2. If field #3 includes 640×480 pixels and field #4 includes 4096×2160 pixels, then the resolution of the image supported by AI model #1 is less than or equal to 640×480 pixels, and the resolution of the image supported by AI model #2 is 4096×2160 pixels.

[0162] In other examples, the first information may include M bitmaps, where M is a positive integer. These M bitmaps can be used to indicate parameters of at least one modality supported by an AI model. Optionally, each bitmap in the M bitmaps may indicate parameters of a modality supported by one of the at least one AI models. For example, the third bitmap is one of the M bitmaps, and the first set of bits is a set of bits within the third bitmap. This first set of bits indicates parameters of a modality supported by the AI ​​model corresponding to the first bitmap. For instance, the first set of bits indicates the resolution of images supported by AI model #1. If the value of the first set of bits is 011, then the resolution of images supported by AI model #1 may include 640×480 pixels and 1024×768 pixels, and the resolution of images not supported by AI model #1 includes 4096×2160 pixels. This example illustrates a method of indicating parameters of a modality using a set of bits; parameters of other modalities can be indicated in a similar manner, and will not be elaborated further.

[0163] In some other examples, the first information may include the index of an M bitmap. The specific content of these M bitmaps can be found in the previous example and will not be repeated here. For example, the first information includes the index of bitmap #4. The first set of bits in bitmap #4 indicates the resolution of images supported by AI model #1. If the index of bitmap #4 is 10, and the first set of bits in the bitmap with index 10 is 011, then the resolutions of images supported by AI model #1 may include 640×480 pixels and 1024×768 pixels, and the resolutions of images not supported by AI model #1 include 4096×2160 pixels.

[0164] In other implementations, the first information may implicitly indicate the parameters of the modality supported by at least one AI model. Optionally, the first information may include information that corresponds to the parameters of the modality supported by at least one AI model. For example, the first information may include the ID of the at least one AI model. The agent node may determine the parameters of the modality supported by at least one AI model based on the second correspondence and the ID of the at least one AI model. The second correspondence is the correspondence between the ID of the AI ​​model and the parameters of the modality supported by the AI ​​model. The second correspondence may be pre-defined, for example, as specified by a protocol; or it may be reported by the fusion node to the agent node; or it may be notified by the agent node to the fusion node. The form of the first correspondence is not limited; for example, the second correspondence may be represented in tabular form, for example, the second correspondence may be included in a LUT maintained by the fusion node and the agent node.

[0165] Optionally, the parameters of at least one modality supported by the AI ​​model may be replaced with one of the following: parameters of the modality that the AI ​​model can process, parameters of the input data of each modality of the AI ​​model, or parameters of the perceptual data of each modality fused by the AI ​​model.

[0166] In this approach, the fusion node can report at least one modality and its parameters supported by the AI ​​model to the agent node, thereby further improving the accuracy and effectiveness of the patterns determined by the agent node for the fusion node to fuse sensing data, and thus improving the quality and efficiency of the fusion node's fusion sensing data.

[0167] The first piece of information can be carried in a traditional message or a new message, without restriction. The first piece of information can have other names, such as "fusion capability" information, without restriction. The message carrying the first piece of information can have multiple names, such as "fusion capability message" or "capability message," without restriction.

[0168] There are several possible times when a fusion node sends its first message to an agent node.

[0169] In some possible ways, the fusion node can proactively send initial information to the agent node. For example, the fusion node can send initial information to the agent node upon power-on or registration.

[0170] In other possible approaches, the fusion node can send first information to the agent node based on a request from the agent node. For example, the agent node can send a first request to the fusion node; correspondingly, the fusion node can receive the first request from the agent node. This first request can be used to request the fusion node's ability to fuse sensing data; or, the first request can be used to request the fusion node's ability to send fusion sensing data to the agent node; or, the first request can request first information; or, the first request can be used to request (or instruct or trigger) the fusion node to send first information to the agent node. In this way, the fusion node sends the first information to the agent node only after receiving the first request, thereby avoiding unnecessary transmission of first information and saving transmission resources. The first request can be a traditional message or a new message, without limitation. The first request can have other names, such as a fusion capability enquiry message or a fusion capability request message, without limitation.

[0171] S602: The agent node sends the second information to the fusion node; correspondingly, the fusion node receives the second information from the agent node.

[0172] The second information can be used to indicate a first mode for fusing sensing data. The first mode can be determined based on the ability of the fusion node to fuse sensing data; in other words, the second information is determined based on the ability of the fusion node to fuse sensing data; or, the first mode is related to the ability of the fusion node to fuse sensing data; or, the agent node can determine the first mode based on the ability of the fusion node to fuse sensing data, thereby determining a first mode adapted to the ability of the fusion node; or, the agent node can determine the second information based on the ability of the fusion node to fuse sensing data.

[0173] Among some possible approaches, the first mode may include at least one of the following; in other words, the second information may be used to indicate at least one of the following:

[0174] 1. AI Model for Fusion Sensing Data: For example, the AI ​​model for fusion sensing data is AI model #1. The AI ​​model for fusion sensing data can be an AI model configured (or determined) by the agent node for the fusion node to fuse sensing data, and the fusion node can fuse sensing data based on (or use or through) this AI model.

[0175] In some implementations, the second information may include the ID or index of the AI ​​model that integrates the sensory data. For example, if the second information may include the ID or index of AI model #1, then the AI ​​model that integrates the sensory data may be AI model #1.

[0176] In other implementations, the second information may include the AI ​​model fused from the sensing data. Optionally, in this implementation, the fusion node does not support (or does not include) the AI ​​model indicated by the second information before sending the first information. For example, the first information indicates that the AI ​​models supported by the fusion node include AI model #1 and AI model #2. If the agent node determines that the fusion node uses AI model #4 to fuse the sensing data, then the second information may include AI model #4.

[0177] In other implementations, the second information may include the download address of the AI ​​model for the fused sensing data, allowing the fusion node to download the AI ​​model from that address. Optionally, in this implementation, the fusion node does not support (or does not include) the AI ​​model indicated by the second information before sending the first information. For example, the first information indicates that the AI ​​models supported by the fusion node include AI model #1 and AI model #2. If the agent node determines that the fusion node uses AI model #4 to fuse the sensing data, then the second information may include the download address of AI model #4, from which the fusion node can download AI model #4.

[0178] 2. Period of fused sensing data: For example, the period of fused sensing data can be 30 milliseconds (ms). The period of fused sensing data can be the period configured (or determined) by the agent node for the fusion node, and the fusion node can use this period to fuse sensing data.

[0179] For example, the second information may include field #5, which indicates the period of the fused sensing data. For instance, if the value of field #5 is 30 in milliseconds (ms), then the period of the fused sensing data may be 30 ms.

[0180] 3. Modalities of Fusion Perceptual Data: For example, the modalities of fusion perceptual data may include images and text. The modalities of fusion perceptual data can be configured (or determined) by the agent node for the fusion node, and the fusion node can fuse perceptual data from these modalities.

[0181] In some implementations, the second information can explicitly indicate the modality of the fused sensing data.

[0182] In some examples, the second information may include the ID or index of the modality of the fused perceptual data. For example, the indices for text, video, image, speech, and point cloud are 1, 2, 3, 4, and 5, respectively. The second information includes field #6, which indicates the modality of the fused perceptual data. If field #6 contains indices 1 and 3, then the modalities of the fused perceptual data include text and image.

[0183] In other examples, the second information may include a second bitmap, which may indicate the modality of the fused perceptual data. Optionally, the second bit can be any bit in the second bitmap. If the second bit is a third value (e.g., 1 or 0), the modality corresponding to the second bit belongs to the modality of the fused perceptual data; if the second bit is a fourth value (e.g., 0 or 1), the modality corresponding to the second bit does not belong to the modality of the fused perceptual data. The third and fourth values ​​are different. For example, the second information includes bitmap #3. Bits 1 through 5 in bitmap #3 correspond to text, video, image, speech, and point cloud, respectively. If bitmap #3 is 10100, the third value is 1, and the fourth value is 0, then the modalities of the fused perceptual data include: text and image.

[0184] In some other examples, the second information may include the index of the second bitmap. The specific content of this second bitmap can be found in the previous example and will not be repeated here. For example, the second information includes the index of bitmap #3. The first to fifth bits of bitmap #3 correspond to text, video, image, speech, and point cloud, respectively. If the index of bitmap #3 is 20, and the bitmap with index 20 is 10100, the third value is 1, and the fourth value is 0, then the modalities of the fused perceptual data include: text and image.

[0185] Optionally, when the second information explicitly indicates the modality of the fused sensing data, the modality used to indicate the fused sensing data may be combined with the information used to indicate the model of the fused sensing data. For example, the second bitmap may be included in the ID of the model of the fused sensing data; or, the information used to indicate the modality of the fused sensing data may be independent of the information used to indicate the model of the fused sensing data.

[0186] In other implementations, the second information may implicitly indicate the modality of the fused sensing data. Optionally, the second information may include information corresponding to the modality of the fused sensing data. For example, the second information may include the ID of the AI ​​model of the fused sensing data. The fusion node may determine the modality supported by the AI ​​model of the fused sensing data based on the first correspondence and the ID of the at least one AI model, and determine the modality supported by the AI ​​model of the fused sensing data as the modality of the fused sensing data. The first correspondence may be a correspondence between the ID of the AI ​​model and the modalities supported by the AI ​​model. The specific content of the first correspondence can be found in the description of the first correspondence in S601, and will not be repeated here.

[0187] This approach provides multiple possible methods for the first mode. Since the first mode is related to the parameters of the fused sensing data of the fusion node, or in other words, the first mode can include the parameters of the fused sensing data of the fusion node, the agent node can flexibly configure the parameters of the fused sensing data for the fusion node through this method.

[0188] As mentioned earlier, the first mode can be determined based on the ability of the fusion node to fuse sensing data. The following explanation will focus on the content of the first mode.

[0189] 1. AI Model for Fusion Sensing Data: The AI ​​model for fusion sensing data can be determined based on at least one AI model supported by the fusion node; or, in other words, the AI ​​model for fusion sensing data is related to at least one AI model supported by the fusion node. In some examples, the AI ​​model for fusion sensing data may belong to that at least one AI model. For example, if the at least one AI model supported by the fusion node may include AI model #1 and AI model #2, then the AI ​​model for fusion sensing data may be AI model #1 or AI model #2. In other examples, if the at least one AI model supported by the fusion node does not meet the sensing requirements, then the AI ​​model for fusion sensing data may not belong to that at least one AI model. The sensing requirements can be obtained by the agent node from other devices (e.g., a central agent node or network management device). For example, if the at least one AI model supported by the fusion node includes AI model #1 and AI model #2, and AI model #1 and AI model #2 do not meet the sensing requirements, then the AI ​​model for fusion sensing data may be an AI model other than AI model #1 and AI model #2, such as AI model #4. The sensing requirements may be related to the sensing task. For example, the larger the amount of sensory data corresponding to a sensing task, the higher the sensing requirement; the smaller the amount of sensory data corresponding to a sensing task, the lower the sensing requirement.

[0190] 2. Regarding the modality of the fused perceptual data: The modality of the fused perceptual data can be determined based on the modalities supported by at least one AI model; or, in other words, the modality of the fused perceptual data is related to the modalities supported by at least one AI model. In some examples, the modality of the fused perceptual data may belong to the modalities supported by at least one AI model; in other words, the modalities supported by at least one AI model may include the modality of the fused perceptual data. For example, in at least one AI model, the modalities supported by AI model #1 include: text and images, and the modalities supported by AI model #2 include: text, video, images, speech, and point clouds. In the first mode, the modality of the fused perceptual data may include: text and images. In other examples, if the modality supported by the at least one AI model does not meet the perceptual requirements, the modality of the fused perceptual data may include modalities other than those supported by the at least one AI model. For example, at least one AI model includes AI model #1, and the modalities supported by AI model #1 include: text and images. If the modalities corresponding to the perceptual requirements include: text, video, and images, then the modality of the fused perceptual data may include: text, video, and images.

[0191] Optionally, the modalities of the fused sensing data can also be determined based on the computing power of the fusion nodes; or, in other words, the modalities of the fused sensing data can also be related to the computing power of the fusion nodes. For example, the number of modalities of the fused sensing data can be directly proportional to the computing power of the fusion nodes. The stronger the computing power of the fusion nodes, the more modalities of the fused sensing data; the weaker the computing power of the fusion nodes, the fewer modalities of the fused sensing data.

[0192] 3. Regarding the period of fused sensing data: The period of fused sensing data can be determined based on the computing power of the fusion nodes; or, in other words, the period of fused sensing data is related to the computing power of the fusion nodes. Optionally, the period of fused sensing data can be inversely proportional to the computing power of the fusion nodes. Specifically, the higher the computing power of the fusion nodes, the shorter the period of fused sensing data; the lower the computing power of the fusion nodes, the longer the period of fused sensing data.

[0193] Optionally, the period of fused sensing data can be determined based on the computing power and sensing requirements of the fusion nodes; or, in other words, the period of fused sensing data is related to the computing power and sensing requirements of the fusion nodes. For example, the period of fused sensing data can be inversely proportional to the computing power and sensing requirements of the fusion nodes. Specifically, the higher the computing power of the fusion nodes and the higher the sensing requirements, the shorter the period of fused sensing data; conversely, the lower the computing power of the fusion nodes and the lower the sensing requirements, the longer the period of fused sensing data.

[0194] The second information can be carried in a traditional message or a new message, without restriction. The second information can have other names, such as "fusion mode indication" message, without restriction. The message carrying the second information can have multiple names, such as "fusion mode indication message" or "mode indication message," without restriction.

[0195] S603: The fusion node sends the first perception data to the agent node; correspondingly, the agent node receives the first perception data from the fusion node.

[0196] The first perception data can be obtained by fusing perception data from multiple perception nodes according to a first mode; or, in other words, the fusion node can fuse perception data from multiple perception nodes according to the first mode to obtain the first perception data. For example, if in the first mode, the AI ​​model for fusing perception data is AI model #1, the modalities of the fused perception data include images and text, and the period of the fused perception data is 30ms, then the fusion node can fuse perception data in the modal of images and text every 30ms according to (or using or through) AI model #1. Another example: if in the first mode, the AI ​​model for fusing perception data is AI model #1, then the fusion node can fuse perception data in the modal of images and text according to (or using or through) AI model #1. Yet another example: if in the first mode, the AI ​​model for fusing perception data is AI model #1, and the modalities of the fused perception data include images and text, then the fusion node can fuse perception data in the modal of images and text according to (or using or through) AI model #1. For example, if in the first mode, the AI ​​model for fusing the sensing data is AI model #1, and the period for fusing the sensing data is 30ms, then the fusion node can fuse the sensing data according to (or using or through) AI model #1 every 30ms.

[0197] pass Figure 6 The method shown allows the fusion node to report its ability to fuse sensing data to the agent node. The agent node can then determine the fusion sensing data pattern for the fusion node based on its fusion sensing data fusion capabilities, thereby determining a fusion sensing data pattern that is compatible with the fusion node's capabilities and improving the efficiency and quality of the fusion sensing data.

[0198] Furthermore, in this method, the fusion node can transmit fused sensing data, thereby reducing transmission overhead. Additionally, the sensing data fused by the fusion node can include sensing data from multiple modalities, thus reducing transmission overhead while broadening the scope of application.

[0199] Furthermore, in this method, both the fusion node and the agent node can be deployed with agents, thereby improving network intelligence and enhancing network maintenance and / or operational efficiency.

[0200] Among some possible ways, Figure 6 The method shown also includes S604 to S606:

[0201] S604: The first sensing node sends third information to the fusion node; correspondingly, the fusion node receives third information from the first sensing node.

[0202] In this context, the first sensing node belongs to multiple sensing nodes. The third information can be used to indicate the first modality. The method of indication is similar to the description in S601 of "the first information can explicitly indicate at least one modality supported by an AI model," except that the first information is replaced with the third information, and the modality supported by at least one AI model is replaced with the first modality. This will not be elaborated further here. The first modality can be the modality of the sensing data of the first sensing node. The sensing data of the first sensing node can include at least one of the following: the original data sensed by the first sensing node, the data obtained by processing the original data by the first sensing node (e.g., one or more of dimension alignment, upsampling, downsampling, and neural network processing), or the sensing data fused by the first sensing node.

[0203] Third-party information can be carried in traditional messages or new messages without restriction. Third-party information can have other names, such as sensor registration information or registration information, without restriction. The message carrying third-party information can have multiple names, such as sensor registration message or registration message, without restriction.

[0204] There are several possible times when the first sensing node sends third information to the fusion node.

[0205] In some possible ways, the first sensing node can proactively send third information to the fusion node. For example, the first sensing node can send third information to the fusion node when it powers on or registers.

[0206] In other possible approaches, the first sensing node may send third information to the fusion node based on a request from the fusion node. For example, the fusion node may send a fourth request to the first sensing node; correspondingly, the first sensing node may receive the fourth request from the fusion node. This fourth request can be used to request third information; or, the fourth request can be used to instruct the first sensing node to send third information to the fusion node. Thus, after receiving the fourth request, the first sensing node can send the third information to the fusion node. The fourth request can be a traditional message or a new message, without limitation. The fourth request can have other names, such as a modal query message or a modal request message, without limitation.

[0207] S605: If the first AI model in the fusion node supports the first modality, and the computing power required for the first AI model to fuse the sensing data of multiple sensing nodes is less than or equal to the computing power of the fusion node, the fusion node may send fourth information to the first sensing node; correspondingly, the first sensing node may receive fourth information from the fusion node. In other words, if the fusion node is capable of fusing the sensing data of multiple sensing nodes, including the first sensing node, the fusion node may send fourth information to the first sensing node. The fourth information may be used to instruct (or activate or trigger) the first sensing node to send its sensing data to the fusion node; or, the fourth information may be used to instruct (or activate or trigger) the first sensing node to enable its sensing or sensor functions.

[0208] For example, the fusion node includes AI model #1, and the modalities supported by AI model #1 include text and images. If the first modality includes text and images, the computing power of the fusion node is 1.2T FLOPS, and the computing power required for AI model #1 to fuse the perception data of multiple perception nodes is 1T FLOPS, then the fusion node can send fourth information to the first perception node.

[0209] The fourth piece of information can be carried in traditional messages or new messages, without restriction. The fourth piece of information can have other names, such as sensor register response information, registration response information, sensor activate information, or sensor enable information, without restriction. The message carrying the fourth piece of information can have multiple names, such as sensor register response message, registration response message, sensor activate message, or sensor enable message, without restriction.

[0210] S606: The first sensing node sends its sensing data to the fusion node; correspondingly, the fusion node can receive the sensing data from the first sensing node.

[0211] Optionally, S604 to S606 can precede S603. The order of S604 to S606 and S601 to S602 is not limited.

[0212] In this approach, the first AI model in the fusion node supports the first modality, and the computational power required for the first AI model to fuse the perception data of multiple perception nodes is less than or equal to the computational power of the fusion node—that is, when the fusion node can fuse the perception data of multiple perception nodes, including the first perception node—is instructed by the fusion node to send its own perception data to the fusion node. In this way, the first perception node can send its own perception data as needed according to the instructions of the fusion node, avoiding or reducing unnecessary transmission of its own perception data, thereby saving transmission resources and preventing resource waste.

[0213] Optional, Figure 6 The method shown may also include S607:

[0214] S607: In case 1 or case 2, the fusion node may send a second request to the agent node; correspondingly, the agent node may receive the second request from the fusion node.

[0215] Scenario 1: The AI ​​model in the fusion node does not support the first modality; or, in other words, none of the AI ​​nodes in the fusion node support the first modality.

[0216] For example, if the fusion node only includes AI model #1, and the modalities supported by AI model #1 include text and images, and the first modality includes video, then the fusion node can send a second request to the agent node; correspondingly, the agent node can receive the second request from the fusion node.

[0217] For example, if the fusion node includes AI model #1 and AI model #2, and the modalities supported by AI model #1 include text and images, and the first modality includes video, then the fusion node can send a second request to the agent node; correspondingly, the agent node can receive the second request from the fusion node.

[0218] Scenario 2: One or more AI models in the fusion node support the first modality, and the computing power required for the one or more AI models to fuse the perception data of multiple perception nodes is greater than the computing power of the fusion node.

[0219] In some examples, each AI model in the one or more AI models can independently fuse perceptual data from multiple sensing nodes. The computational power required by each AI model in the one or more AI models is greater than the computational power of the fusion node. In other words, for each AI model in the fusion node that supports the first modality, the computational power required for that AI model to fuse perceptual data from multiple sensing nodes is greater than the computational power of the fusion node itself.

[0220] For example, if the fusion node only includes AI model #1, and the modalities supported by AI model #1 include text and images, with the first modality including text, and the computing power required for AI model #1 to fuse the perception data of multiple perception nodes is greater than the computing power of the fusion node, then the fusion node can send a second request to the agent node; correspondingly, the agent node can receive the second request from the fusion node.

[0221] For example, if the fusion node only includes AI model #1 and AI model #2, AI model #1 supports modalities including text and images, and AI model #2 supports modalities including text, video, images, speech, and point clouds, with text as the first modality, and the computing power required for AI model #1 to fuse the perception data from multiple perception nodes is greater than the computing power of the fusion node, and the computing power required for AI model #2 to fuse the perception data from multiple perception nodes is greater than the computing power of the fusion node, then the fusion node can send a second request to the agent node; correspondingly, the agent node can receive the second request from the fusion node.

[0222] In other examples, the sensing data from multiple sensing nodes can be jointly (or jointly) fused in one or more AI models. The total computing power required in the one or more AI models is greater than the computing power of the fusion nodes.

[0223] For example, if the fusion node includes AI model #1 and AI model #2, AI model #1 supports modalities including text and images, and AI model #2 supports modalities including text, video, images, speech, and point clouds, and the first modality includes text, and the total computing power required for AI model #1 and AI model #2 to fuse the perception data of multiple perception nodes is greater than the computing power of the fusion node, then the fusion node can send a second request to the agent node; correspondingly, the agent node can receive the second request from the fusion node.

[0224] The second request can be used to request a second AI model. This second AI model may support the first modality, and the computational power required for the second AI model to fuse perceptual data from multiple sensing nodes is less than or equal to the computational power of the fusion nodes. For example, if AI model #4 supports modalities including text and images, the first modality includes text, and the computational power required for AI model #4 to fuse perceptual data from multiple sensing nodes is less than or equal to the computational power of the fusion nodes, then the second AI model can be AI model #4. The second AI model and the first AI model may be the same or different.

[0225] In some implementations, under case 1, the second request may include: information about the first modality (hereinafter referred to as information #1). Information #1 can also be referred to as indication information for the first modality. Information #1 indicates the specific content of the first modality; refer to the description in S604 regarding the use of third information to indicate the first modality, only replacing the third information with information #1, which will not be repeated here. In this way, the agent node can determine the second AI model supporting the first modality based on the first modality.

[0226] In other implementations, under case 2, the second request may include at least one of the following: parameters of each of the multiple sensing nodes, or the number of the multiple sensing nodes; or, the second request may include at least one of the following: indication information of the parameters of each of the multiple sensing nodes, or indication information of the number of the multiple sensing nodes. The parameters of each of the multiple sensing nodes are, for example, parameters of the sensors in the sensing node, such as dimension or throughput. The parameters of each of the multiple sensing nodes may be obtained by the fusion node from the multiple sensing nodes. For example, the fusion node may obtain the parameters of the first sensing node from the first sensing node. Optionally, the parameters of the first sensing node may be included in the third information. The number of the multiple sensing nodes may be determined by the fusion node based on the indication information (e.g., ID or index) of the multiple sensing nodes. For example, the fusion node may obtain the indication information of the first sensing node from the first sensing node. Optionally, the indication information of the first sensing node may be included in the third information.

[0227] The second request can be either a traditional message or a new message; there are no restrictions. The second request can also have other names, such as a fusion model request message or a model request message; there are no restrictions.

[0228] Optionally, S607 can be installed after S604. When Figure 6 When the method shown includes S607, S605 and S606 may be optional steps.

[0229] In this way, after receiving third information, if the AI ​​model in the fusion node cannot process the perception data of multiple perception nodes, the fusion node can request an AI model that can process the perception data of multiple perception nodes, thereby improving the performance of fusion perception data.

[0230] Optional, Figure 6 The method shown may also include S608:

[0231] S608: The agent node sends the fifth information to the fusion node; correspondingly, the fusion node receives the fifth information from the agent node.

[0232] The fifth piece of information may include a second AI model. In this way, the fusion node can obtain the second AI model from the fifth piece of information and thus fuse the perception data from multiple perception nodes based on the second AI model. Alternatively, the fifth piece of information can be used to indicate the download address of the second AI model. In this way, the fusion node can download the second AI model from that download address and thus fuse the perception data from multiple perception nodes based on the second AI model.

[0233] The fifth piece of information can be carried in traditional messages or new messages, without restriction. The fifth piece of information can have other names, such as fusion model indication information, without restriction. The message carrying the fifth piece of information can have multiple names, such as fusion model indication message, without restriction.

[0234] Optionally, after receiving the fifth information, the fusion node may send the fourth information to the first sensing node; correspondingly, the first sensing node may receive the fourth information from the fusion node. The fourth information can be used to instruct (or activate or trigger) the first sensing node to send sensing data to the fusion node. The specific content of the fourth information can be found in the description of the fourth information in S605, and will not be repeated here.

[0235] Optionally, S608 can be installed after S607.

[0236] Using this method, the fusion node can quickly and accurately obtain the second AI model based on the fifth piece of information.

[0237] In some possible approaches, the second AI model can be replaced by an adaptation layer. This adaptation layer can support the first modality; or, in other words, it can be a neural network adapted to the first modality. The third AI model can be an AI model supported (or included) by the fusion node. The computational power required to fuse the perception data of multiple perception nodes through the third AI model and the adaptation layer is less than or equal to the computational power of the fusion node. For example, before receiving the third information from the first perception node (i.e., S604), the fusion node can fuse the perception data of modality #1 and modality #2 through the third AI model. After receiving the third information from the first perception node (i.e., S604), if the AI ​​model in the fusion node cannot fuse the first modality, the fusion node can request the adaptation layer from the agent node. In this way, the perception data of the first modality can be input to the third AI model for fusion through the adaptation layer; that is, the fusion node can continue to fuse the perception data of modality #1, modality #2, and the first modality through the third AI model and the adaptation layer.

[0238] Optionally, after S604, Figure 6 The method shown may also include step A1:

[0239] Step A1: In case 1, 2, or 3, the fusion node does not send the fourth information to the first sensing node. The fourth information can be used to instruct the first sensing node to send its sensing data to the fusion node. Alternatively, in case 1, 2, or 3, the fusion node sends the seventh information to the first sensing node; correspondingly, the first sensing node receives the seventh information from the fusion node. The seventh information is used to instruct the first sensing node not to send its sensing data to the fusion node; or, the seventh information is used to indicate that the first sensing node's registration failed.

[0240] For details on cases 1 and 2, please refer to the descriptions of cases 1 and 2 in S607, which will not be repeated here.

[0241] Scenario 3: After the fusion node sends the second request to the agent node (i.e., S607), the fusion node does not receive the fifth message. In other words, the fusion node does not acquire the second AI model capable of fusing perception data from multiple perception nodes. In Scenario 3, the agent node may not have sent the fifth message; or, the agent node may have sent the fifth message, but due to a link failure between the fusion node and the agent node, the fusion node did not receive the fifth message.

[0242] The specific content of the fourth message can be found in the description of the fourth message in S605, and will not be repeated here. The seventh message can be carried in a traditional message or a new message, without restriction. The seventh message can have other names, such as registration failure message, without restriction. The message carrying the seventh message can have other names, such as registration failure message, without restriction.

[0243] Optionally, after the fusion node sends the fourth information to the first sensing node, Figure 6 The method shown may also include S609:

[0244] S609: The fusion node sends the eighth information to the first sensing node; correspondingly, the first sensing node receives the eighth information from the fusion node.

[0245] Specifically, the eighth message is used to instruct the first sensing node not to send its sensing data to the fusion node; or the eighth message is used to deactivate the first sensing node from sending its sensing data to the fusion node; or the eighth message is used to instruct (or activate or trigger) the first sensing node to disable its sensing function or sensor function. Thus, upon receiving the eighth message, the first sensing node can stop (or cease) sending its sensing data to the fusion node.

[0246] The eighth message can be carried in a traditional message or a new message, without restriction. The eighth message can have other names, such as sensor de-activation information, without restriction. The message carrying the eighth message can have other names, such as sensor de-activation message, without restriction.

[0247] Optionally, S609 can be performed after S605, or after the fusion node in S608 can send the fourth information to the first sensing node.

[0248] This step allows the fusion node to promptly instruct the first sensing node to stop sending sensing data to the fusion node, thereby avoiding unnecessary transmission of sensing data and saving transmission resources.

[0249] It should be understood that steps S604 to S609 and step A1 are explained using the first sensing node as an example. The operation of other sensing nodes among multiple sensing nodes can be referred to the explanation of the first sensing node, and will not be repeated here.

[0250] Among some possible ways, Figure 6 The method shown also includes S610:

[0251] S610: The fusion node sends the sixth information to the agent node; correspondingly, the agent node receives the sixth information from the fusion node.

[0252] The sixth piece of information can be auxiliary information required for fusing sensing data from multiple sensing nodes.

[0253] Optionally, the sixth piece of information may include at least one of the following:

[0254] 1. Configuration of the AI ​​model for fusing sensing data from multiple sensing nodes (hereinafter referred to as Configuration #1): For example, Configuration #1 may include at least one of the following: indication information (e.g., ID or index) of the AI ​​model for fusing sensing data from multiple sensing nodes, indication information of the modality of the sensing data fused by the AI ​​model, or indication information of the sensing target corresponding to the fused sensing data. The specific content of the modality of the sensing data fused by the AI ​​model can be referred to in the above description of "modality of fused sensing data" in S602, and will not be repeated here. The sensing target may be referred to as (or can be replaced by) a task target, a task, or a sensing task. For example, the sensing target is to construct a radio frequency (RF) map of one or more areas.

[0255] 2. Configuration of Multiple Sensing Nodes (hereinafter referred to as Configuration #2): For example, Configuration #2 may include at least one of the following: indication information (e.g., ID or index) of the multiple sensing nodes, activation information of at least one sensing node, or indication information of a filtering algorithm for the multiple sensing nodes. Wherein, at least one sensing node may include the multiple sensing nodes. The activation information of at least one sensing node may be referred to as (or can be replaced by) trigger information of at least one sensing node, and can be used to indicate whether to activate the at least one node, or to indicate whether to instruct (or activate or trigger) the at least one sensing node to send sensing data to the fusion node. The filtering algorithm for the multiple sensing nodes may be referred to as a selection algorithm for the multiple sensing nodes, and can be used to filter (or screen or select) the sensing data sent by the multiple sensing nodes to the fusion node.

[0256] The sixth piece of information can be carried in a traditional message or a new message, without restriction. The sixth piece of information can have other names, such as perception assistance information, without restriction. The sixth piece of information and the first perception data can be carried in the same message or in different messages. When the sixth piece of information and the first perception data are carried in different messages, the transmission order of the sixth piece of information and the first perception data is not restricted.

[0257] In this way, the fusion node can report the auxiliary information needed to fuse the sensing data of multiple sensing nodes, so that the intelligent agent node can better manage the fusion node accordingly.

[0258] The order of S603, S609 and S610 is not limited in this application.

[0259] This application provides another communication method. Figure 7 This is a flowchart illustrating the communication method provided in an embodiment of this application. Figure 7 As shown, the method includes:

[0260] S701: The first sensing node sends third information to the fusion node; correspondingly, the fusion node receives the third information from the first sensing node. The third information can be used to indicate a first mode for fusing sensing data. The first mode can be the mode of the sensing data from the first sensing node.

[0261] S702: If the first AI model in the fusion node supports the first modality, and the computing power required for the first AI model to fuse the perception data of multiple perception nodes is less than or equal to the computing power of the fusion node, the fusion node may send fourth information to the first perception node; correspondingly, the first perception node may receive the fourth information from the fusion node. The fourth information can be used to instruct (or activate or trigger) the first perception node to send its own perception data to the fusion node. The multiple perception nodes include the first perception node.

[0262] S703: The first sensing node sends its sensing data to the fusion node; correspondingly, the fusion node can receive the sensing data from the first sensing node.

[0263] For details on S701 to S703, please refer to S604 to S606, which will not be repeated here.

[0264] Among some possible ways, Figure 7 The method shown may also include:

[0265] S704: In case 1 or case 2, the fusion node may send a second request to the agent node; correspondingly, the agent node may receive the second request from the fusion node. The second request may be used to request a second AI model. The second AI model may support the first modality, and the computational power required for the second AI model to fuse the perception data from multiple perception nodes is less than or equal to the computational power of the fusion node.

[0266] For details on S704, please refer to S607; it will not be repeated here.

[0267] Optional, Figure 7 The method shown may also include:

[0268] S705: The agent node sends the fifth information to the fusion node; correspondingly, the fusion node receives the fifth information from the agent node. The fifth information may include the second AI model, or it may be used to indicate the download address of the second AI model.

[0269] For details on S705, please refer to S608; it will not be repeated here.

[0270] Optionally, after the fusion node sends the fourth information to the first sensing node, Figure 7 The method shown may also include:

[0271] S706: The fusion node sends the eighth information to the first sensing node; correspondingly, the first sensing node receives the eighth information from the fusion node. The eighth information is used to instruct the first sensing node not to send its sensing data to the fusion node; or the eighth information is used to deactivate the first sensing node from sending its sensing data to the fusion node.

[0272] For details on S706, please refer to S609; it will not be repeated here.

[0273] Among some possible ways, Figure 7 The method shown also includes:

[0274] S707: The fusion node sends the first perception data to the agent node; correspondingly, the agent node receives the first perception data from the fusion node. The first perception data can be data obtained by the fusion node fusing perception data from multiple perception nodes.

[0275] For details on S707, please refer to S603; it will not be repeated here.

[0276] Among some possible ways, Figure 7 The method shown also includes:

[0277] S708: The fusion node sends the sixth information to the agent node; correspondingly, the agent node receives the sixth information from the fusion node. The sixth information may be auxiliary information required for fusing the sensory data from multiple sensing nodes.

[0278] For details on S708, please refer to S610; it will not be repeated here.

[0279] pass Figure 7 The method described herein instructs the first sensing node to send its own sensing data to the fusion node only when the first AI model in the fusion node supports the first modality, and the computing power required for the first AI model to fuse the sensing data of multiple sensing nodes is less than or equal to the computing power of the fusion node—that is, when the fusion node is capable of fusing the sensing data of multiple sensing nodes, including the first sensing node. In this way, the first sensing node can send its own sensing data as needed according to the instructions of the fusion node, avoiding or reducing unnecessary transmission of its own sensing data, thereby saving transmission resources and preventing resource waste.

[0280] This application provides yet another communication method. Figure 8 This is a flowchart illustrating the communication method provided in an embodiment of this application. Figure 8 As shown, the method includes:

[0281] S801: The agent node sends a third request to the fusion node; correspondingly, the fusion node receives the third request from the agent node.

[0282] The third request can be used to request sensing data for a sensing target. Optionally, the third request may include information indicating the sensing target. The sensing target may be referred to as (or can be replaced by) a task objective, task, or sensing task. For example, the sensing target is the construction of an RF map of one or more areas.

[0283] The third request can be a traditional message or a new message, without restriction. The third request can also have other names, such as a sensing request message, without restriction.

[0284] S802: The fusion node sends the second perception data to the agent node; correspondingly, the agent node receives the second perception data from the fusion node.

[0285] The second sensing data can be obtained by fusing sensing data from multiple sensing nodes according to the sensing target; or, in other words, the fusion node can fuse sensing data from multiple sensing nodes according to the sensing target to obtain the second sensing data. Optionally, the second sensing data can be obtained by fusing data related to the sensing target from the sensing data of multiple sensing nodes. For example, the sensing target is to construct an RF map of one or more areas. The sensing data of the multiple sensing nodes includes sensing data from sensing nodes #1 to #3. Among them, the sensing data of sensing nodes #1 and #2 are related to constructing an RF map of one or more areas, or in other words, the sensing data of sensing nodes #1 and #2 can be used to construct an RF map of one or more areas. The fusion node can fuse the sensing data of sensing nodes #1 and #2 to obtain the second sensing data.

[0286] Optionally, the data related to the sensing target can be sensing data used to achieve the sensing target, or in other words, sensing data capable of achieving the sensing target. For example, the sensing target is to construct an RF map of one or more areas; the data related to the sensing target can be sensing data capable of being used to construct an RF map of one or more areas.

[0287] pass Figure 8 The method shown allows the fusion node to fuse sensing data from multiple sensing nodes according to the sensing target, obtaining second sensing data, and then sending the second sensing data. In this way, the sensing data fused and sent by the fusion node is related to the sensing target. The fusion node can avoid fusing and sending sensing data unrelated to the sensing target, thereby reducing the complexity of the fusion node, reducing signaling overhead, and saving transmission resources.

[0288] Furthermore, in this method, both the fusion node and the agent node can be deployed with agents, thereby improving network intelligence and enhancing network maintenance and / or operational efficiency.

[0289] The first sensing node is any one of the multiple sensing nodes; or, in other words, the first sensing node belongs to multiple sensing nodes. The following explanation uses the first sensing node as an example to illustrate the communication between the fusion node and the multiple sensing nodes.

[0290] Among some possible ways, Figure 8 The method shown also includes:

[0291] S803: When the first sensing node is related to the sensing target, the fusion node can send fourth information to the first sensing node; correspondingly, the first sensing node can receive the fourth information from the fusion node.

[0292] The fourth piece of information can be used to instruct (or activate or trigger) the first sensing node to send its sensing data to the fusion node. This fourth piece of information can be carried in a traditional message or a new message, without limitation. The fourth piece of information can have other names, such as sensor activate information or sensor enable information, without limitation. The message carrying the fourth piece of information can have multiple names, such as sensor activate message or sensor enable message, without limitation.

[0293] Optionally, the fusion node can determine the relationship between the first sensing node and the sensing target based on the correspondence between the sensing target and the sensing nodes (hereinafter referred to as the third correspondence). The third correspondence can be represented in various ways, for example, through a table.

[0294] S804: The first sensing node can send its sensing data to the fusion node; correspondingly, the fusion node can receive the sensing data from the first sensing node.

[0295] For example, the sensing objective is to construct an RF map of one or more areas. There are multiple sensing nodes in the network, designated as sensing nodes #1 to #3. Sensing nodes #1 and #2 are involved in constructing the RF map of one or more areas. A fusion node can send a fourth message to sensing node #1, after which sensing node #1 can send its sensing data to the fusion node. Similarly, a fusion node can send a fourth message to sensing node #2, after which sensing node #2 can send its sensing data to the fusion node. Sensing node #3, not receiving the fourth message, may not send its sensing data to the fusion node.

[0296] Optionally, S803 to S804 can be after S801 and before S802.

[0297] In this method, the fusion node only sends the fourth information to the first sensing node if the first sensing node is related to the sensing target. If the fourth information is received, the first sensing node can send its sensing data to the fusion node; if the fourth information is not received, the first sensing node does not send its sensing data to the fusion node. In this way, the first sensing node can send its sensing data as needed according to the instructions of the fusion node, thereby avoiding unnecessary transmission of its sensing data when it is unrelated to the sensing target, and thus saving transmission resources.

[0298] Among some possible ways, Figure 8 The method shown also includes:

[0299] S805: The fusion node may send information (hereinafter referred to as information #3) to the first sensing node to indicate the sensing target; correspondingly, the first sensing node receives the information to indicate the sensing target from the fusion node.

[0300] For details on the specific content of the sensing target, please refer to the description of the sensing target in S801, which will not be repeated here.

[0301] Message #3 can be carried in a traditional message or a new message, without restriction. Message #3 can have other names, such as sensor activate message or sensor enable message, without restriction. The message carrying message #3 can have multiple names, such as sensor activate message or sensor enable message, without restriction. Message #3 and the fourth message can be carried in the same message or in different messages. When message #3 and the fourth message are carried in different messages, the transmission order of message #3 and the fourth message is not restricted.

[0302] S806: The first sensing node can send its sensing data to the fusion node; correspondingly, the fusion node can receive the sensing data from the first sensing node. The sensing data of the first sensing node may be related to the sensing target.

[0303] For example, the sensing objective is to construct an RF map of one or more regions. There are multiple sensing nodes in the network, designated as sensing node #1 to sensing node #2. Sensing node #1's sensing data includes sensing data #1 and sensing data #2. Sensing data #1 is related to constructing the RF map of one or more regions, while sensing data #2 is not. Sensing node #2's sensing data includes sensing data #3, which is related to constructing the RF map of one or more regions. A fusion node can send information #3 to sensing node #1, and then sensing node #1 can send sensing data #1 to the fusion node; the fusion node can send information #3 to sensing node #2, and then sensing node #2 can send sensing data #3 to the fusion node.

[0304] Optionally, in this approach, the information indicating the sensing target can be replaced with information indicating an algorithm for filtering (or selecting, screening, or determining) sensing data related to the sensing target. In this way, the first sensing node can filter out sensing data related to the sensing target according to the algorithm.

[0305] Optionally, S805 to S806 can be after S801 and before S802.

[0306] In this way, the first sensing node can send sensing data related to the sensing target to the fusion node, and can choose not to send sensing data unrelated to the sensing target to the fusion node, thereby reducing signaling overhead and saving transmission resources.

[0307] Optionally, S803 to S804 and S805 to S806 can be combined. In this case, the order of S803 and S805 is not limited, and the fourth information and information #3 can be carried in the same message or in different messages. S804 and S806 can be combined as follows: the first sensing node can send its sensing data to the fusion node; correspondingly, the fusion node can receive the sensing data from the first sensing node. The sensing data of the first sensing node can be related to the sensing target.

[0308] Optionally, after the fusion node sends the fourth information to the first sensing node, Figure 8 The method shown may also include:

[0309] S807: The fusion node sends the eighth information to the first sensing node; correspondingly, the first sensing node receives the eighth information from the fusion node. The eighth information is used to instruct the first sensing node not to send its sensing data to the fusion node; or the eighth information is used to deactivate the first sensing node from sending its sensing data to the fusion node.

[0310] For details on S807, please refer to S609; it will not be repeated here.

[0311] Among some possible ways, Figure 8 The method shown also includes:

[0312] S808: The fusion node sends the sixth information to the agent node; correspondingly, the agent node receives the sixth information from the fusion node. The sixth information may be auxiliary information required for fusing the sensory data from multiple sensing nodes.

[0313] For details on S808, please refer to S610; it will not be repeated here.

[0314] Optionally, the order of S802, S807 and S808 is not limited.

[0315] Among some possible ways, Figure 8 The method shown and Figure 6 The methods shown can be combined.

[0316] In some examples, S603 and S802 can be combined into: the fusion node sends first perception data to the agent node; correspondingly, the agent node receives the first perception data from the fusion node. The first perception data can be obtained by fusing perception data from multiple perception nodes according to a first mode and a perception target.

[0317] In other examples, S605 and S803 can be combined as follows: If the first AI model in the fusion node supports the first modality, the computing power of the fusion node is greater than or equal to the computing power required for the first AI model to fuse the sensing data from multiple sensing nodes, and the first sensing node is related to the sensing target, the fusion node can send fourth information to the first sensing node; correspondingly, the first sensing node can receive the fourth information from the fusion node. The fourth information can be used to instruct the first sensing node to send sensing data to the fusion node.

[0318] In some other examples, S601 to S602 may precede S802, and the order of S601 to S602 and S801 is not limited.

[0319] In some other examples, S604 to S606 may precede S802, and the order of S604 to S606 and S801 is not limited.

[0320] In some other examples, S607 and / or S608 may precede S802, and the order of S607 and / or S608 with S801 is not limited.

[0321] It should be understood that Figure 8 The method shown and Figure 6 The above examples of methods shown can be used individually or in combination.

[0322] This application provides yet another communication method. The method illustrates... Figures 6 to 8The illustrated method is a possible example. The implementing entities involved in this method may include: UE1 to UE3, BS, and a central agent node. UE1 and UE2 may be located in the same area, and a sidelink may be established between UE1 and UE2; UE3 is in a different area from UE1 and UE2. Agents may be deployed in UE1, UE3, BS, and the central agent node, while no agent may be deployed in UE2. Therefore, UE1, UE3, BS, and the central agent node can act as agent nodes. UE1 and BS can act as fusion nodes. UE1 to UE3 can act as sensing nodes. UE1 may include a LiDAR sensor, therefore, the data sensed by UE1 may include LiDAR data; UE2 may include a camera sensor, therefore, the data sensed by UE2 may include image data; UE3 has integrated sensing and communication (ISAC) capabilities, therefore, the data sensed by UE3 may include point cloud data (or sensed point cloud data). UE1 to UE3 can all measure channel data; in other words, the data sensed by UE1 to UE3 may include channel data (e.g., CSI data). Figure 9 This is a flowchart illustrating the communication method provided in an embodiment of this application. Figure 9 As shown, the method includes:

[0323] S901: UE1 sends fusion capability information to BS; correspondingly, BS receives fusion capability information from UE1.

[0324] Fusion capability information can be used to indicate the ability of UE1 to fuse sensing data.

[0325] S902: The BS sends a convergence mode indication message to the UE1; correspondingly, the UE1 receives the convergence mode indication message from the BS.

[0326] The fusion mode indication information can be used to indicate the first mode #1 used for fusion sensing data.

[0327] The specific content of S901 to S902 can be found in S601 to S602, except that the fusion node is replaced with UE1, the agent node is replaced with BS, the first information is replaced with fusion capability information, and the second information is replaced with fusion mode indication information. These details will not be repeated here.

[0328] S903: UE2 sends registration information #1 to UE1; correspondingly, UE1 receives registration information #1 from UE2.

[0329] Registration information #1 can be used to indicate a first modality #1, which can be the modality of the UE2's sensing data. For example, the first modality #1 may include: image and channel data.

[0330] S904: If the AI ​​model in UE1 does not support the first mode #1, UE1 may send a fusion model request to BS; accordingly, BS receives the fusion model request from UE1.

[0331] S905: The BS sends a fusion model indication message to the UE1; correspondingly, the UE1 receives the fusion model indication message from the BS.

[0332] The fusion model indication information may include a second AI model that supports the first modality #1, or include a download address for the second AI model.

[0333] S906: UE1 can send sensor activation information #1 to UE2; correspondingly, UE2 can receive sensor activation information #1 from UE1. Sensor activation information #1 can be used to instruct UE2 to send its sensing data to UE1.

[0334] For details on S903 to S906, please refer to S604, S607 to S608. The only difference is that the fusion node is replaced with UE1, the first sensing node is replaced with UE2, the third information is replaced with registration information #1, the second request is replaced with fusion model request, the fifth information is replaced with fusion model indication information, and the fourth information is replaced with sensor activation information #1. These details will not be repeated here.

[0335] For example, in S903, the first modality #1 includes an image. In S904, if the AI ​​model in UE1 does not support images, UE1 can send a fusion model request to the BS. In S905, the BS can send the UE an AI model that supports images or a download address for that AI model.

[0336] S907: UE3 sends registration information #2 to BS; correspondingly, BS receives registration information #2 from UE3.

[0337] Registration information #2 can be used to indicate a first mode #2, which can be the mode of the UE3's sensing data. For example, the first mode #2 may include point cloud and channel data.

[0338] S908: If the first AI model in the BS supports the first mode #2, and the computing power required for the first AI model to fuse the sensing data from multiple sensing nodes is less than or equal to the computing power of the BS, the BS may send sensor activation information #2 to the UE3; correspondingly, the UE3 may receive sensor activation information #2 from the BS. The sensor activation information #2 can be used to instruct the UE3 to send its sensing data to the BS.

[0339] For details on S907 to S908, please refer to S604 to S605, except that the fusion node is replaced with BS, the first sensing node is replaced with UE3, the third information is replaced with registration information #2, and the fourth information is replaced with sensor activation information #2. These will not be elaborated on here.

[0340] This application does not restrict the execution order of S901 to S902, S903 to S906, and S907 to S908.

[0341] Among some possible ways, Figure 9 The method shown also includes S909 to S913:

[0342] S909: UE2 sends its perception data to UE1; correspondingly, UE1 receives UE2's perception data from UE2.

[0343] Among them, the perception data of UE2 can be the raw data perceived by UE2.

[0344] S910: UE1 can fuse the perception data of UE1 and the perception data of UE2 to obtain the first perception data #1.

[0345] For example, the perception data of UE1 may include the LiDAR data acquired by UE1 and the CSI data corresponding to UE1, and the perception data of UE2 may include the image (e.g., photo) data acquired by UE2 and the CSI data corresponding to UE1. The UE may fuse the following data to obtain the first perception data #1: the LiDAR data acquired by UE1, the CSI data corresponding to UE1, the image data acquired by UE2, and the CSI data corresponding to UE1.

[0346] Optionally, the first perception data #1 can be obtained by fusing the perception data of UE1 and the perception data of UE2 according to the first mode #1. For details, please refer to the explanation in S603 that "the first perception data can be obtained by fusing the perception data of multiple perception nodes according to the first mode", which will not be repeated here.

[0347] Optionally, UE1 may periodically fuse the perception data of UE1 and UE2. The period for fusing the perception data may be included in the first mode #1. For details on the period for fusing the perception data, please refer to the description of the period for fusing the perception data in S601, which will not be repeated here.

[0348] S911: UE1 sends first sensing data #1 to BS; correspondingly, BS receives first sensing data #1 from UE1.

[0349] S912: UE3 sends its perception data to BS; correspondingly, BS receives UE3's perception data from UE3.

[0350] Among them, the perception data of UE3 can be the raw data perceived by UE3.

[0351] This application does not specify the order of S909 to S911 and S912.

[0352] S913: BS can fuse the first perception data #1 and the perception data of UE3 to obtain the first perception data #2.

[0353] For details of S913, please refer to S910, except that UE1 is replaced with BS, the perception data of UE1 and UE2 are replaced with the first perception data #1 and the perception data of UE3, and the first perception data #1 is replaced with the first perception data #2. This will not be elaborated here.

[0354] Optionally, the BS can perform subsequent tasks based on the first perception data #2, the specific content of which is not limited.

[0355] S909 to S913 can be considered as a stage of periodic fusion of sensing data.

[0356] Among some possible ways, Figure 9 The method shown also includes S914 to S920:

[0357] S914: The central agent node sends a task instruction to the BS; correspondingly, the BS receives the task instruction from the central agent node.

[0358] This task indication can be used to indicate a sensing task. For example, the sensing task could be to improve the spectral efficiency of the system. Alternatively, the sensing task could be at least one subtask to improve the spectral efficiency of the system, the at least one subtask including at least one of the following: constructing an RF map, or using the RF map for channel prediction.

[0359] This task instruction can be carried in a traditional message or a new message, without restriction. This task instruction can also have other names, such as "perception request," without restriction.

[0360] The following explanation uses the example of building an RF map (hereinafter referred to as subtask 1) to illustrate the process.

[0361] S915: The BS sends a perception request #1 to the UE1; correspondingly, the UE1 receives the perception request #1 from the BS. The perception request #1 can be used to request perception data for subtask 1.

[0362] S916: UE1 obtains second perception data #1 by fusing the perception data of UE1 according to subtask 1.

[0363] The sensing data fused by UE1 may include LiDAR data and CSI data corresponding to UE1.

[0364] UE2 is unrelated to subtask 1, therefore UE1 may not send information to UE2 to instruct (or activate or trigger) UE2 to send sensing data to UE1.

[0365] S917: UE1 sends second sensing data #1 to BS; correspondingly, BS receives second sensing data #1 from UE1.

[0366] For details on S915 to S917, please refer to S801 to S802, except that the fusion node is replaced with UE1, the agent node is replaced with BS, the third request is replaced with perception request #1, the perception target is replaced with subtask 1, and the second perception data is replaced with second perception data #1. These details will not be repeated here.

[0367] S918: BS sends sensor activation information #3 to UE3; correspondingly, UE3 receives sensor activation information #3 from BS.

[0368] Sensor activation information #3 can be used to instruct UE3 to send its sensing data to BS. Optionally, sensor activation information #3 may also include information for instructing subtask 1.

[0369] S919: UE3 sends its perception data to BS; correspondingly, BS receives UE3's perception data from UE3.

[0370] Among them, the perception data of UE3 is related to subtask 1; in other words, the perception data sent by UE3 is the perception data after being filtered (or selected or screened) according to subtask 1.

[0371] Optionally, the perception data sent by UE3 may include point cloud data and CSI data corresponding to UE3.

[0372] For details on S918 to S919, please refer to S805 to S806, except that the fusion node is replaced with BS, the first sensing node is replaced with UE3, information #3 is replaced with sensor activation information #3, and the sensing target is replaced with subtask 1. These details will not be repeated here.

[0373] This application does not specify the order of S915 to S917 and S918 to S919.

[0374] S920: The BS can fuse the second perception data #1 and the perception data of UE3 according to the sub-task to obtain the second perception data #2.

[0375] Optionally, the BS can execute subtask 1 based on the first perception data #2, the specific content of which is not limited.

[0376] S914 to S920 can be considered as the mission execution phase.

[0377] Figure 9 The effects of the method shown may include Figures 6 to 8 The effects of the method shown will not be elaborated here.

[0378] Based on the same technical concept as the above-described method embodiments, this application provides a corresponding communication device that can be used to perform the functions of the relevant steps in the above-described method embodiments. This function can be implemented in hardware, software, or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions. The communication device can be a terminal, access network equipment, core network equipment, or network management equipment; or it can be a module, communication module, circuit or chip responsible for communication functions (such as a modem chip, or a SoC chip or SIP chip containing a modem core), chip system, or processor in a terminal, access network equipment, core network equipment, or network management equipment; or it can be a logical node, logical module, or software capable of implementing all or part of the functions of a terminal, access network equipment, core network equipment, or network management equipment.

[0379] In one possible implementation, the communication device provided in this application embodiment has the following structure: Figure 10 As shown, the communication device includes a processing unit 1002. Optionally, the communication device may also include an interface unit 1001. The functions of each unit in the communication device 1000 are described below.

[0380] Interface unit 1001 is used for inputting and / or outputting information. Input information can be replaced by received information, and output information can be replaced by transmitted information. When outputting information, interface unit 1001 can output information to other devices outside of communication device 1000, or to other units within communication device 1000. In some embodiments, interface unit 1001 can be implemented through at least one of a physical interface, a communication module, a communication interface, and an input / output interface. In other embodiments, interface unit 1001 can be implemented through interface circuitry, such as a mobile communication module. The mobile communication module may include one or more of at least one antenna, at least one filter, a switch, a power amplifier, a low-noise amplifier (LNA), etc.

[0381] In this application, the interface unit 1001 may also have other names, such as transceiver unit or communication unit. Optionally, the interface unit 1001 may include a receiving unit and a transmitting unit, used for inputting information and outputting information, respectively.

[0382] The processing unit 1002 can be used to support the communication device 1000 in performing the processing actions in the above method embodiments. The processing unit 1002 can be implemented by one or more processors. For example, the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), microprocessors (MCUs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor can be a microprocessor or any conventional processor.

[0383] In one embodiment, the communication device 1000 is applied to Figure 6 The fusion node shown in this embodiment of the application is illustrated below. The specific functions of the processing unit 1002 in this embodiment will be described below.

[0384] The processing unit 1002 is configured to: send first information to the agent node via the interface unit 1001, the first information being used to indicate the ability of the fusion node to fuse sensing data; receive second information from the agent node via the interface unit 1001, the second information being used to indicate a first mode for fusing sensing data, the first mode being determined based on the ability of the fusion node to fuse sensing data; and send first sensing data to the agent node via the interface unit 1001, the first sensing data being obtained by fusing sensing data from multiple sensing nodes according to the first mode.

[0385] In some possible ways, the processing unit 1002 is also configured to: receive a first request from the agent node via the interface unit 1001, the first request being available to request the fusion node's ability to fuse sensing data.

[0386] In some implementations, the processing unit 1002 is further configured to: receive third information from the first sensing node through the interface unit 1001, wherein the first sensing node belongs to multiple sensing nodes, and the third information can be used to indicate a first modality, wherein the first modality is the modality of the sensing data of the first sensing node; when the first AI model in the fusion node supports the first modality, and the computing power required for the first AI model to fuse the sensing data of multiple sensing nodes is less than or equal to the computing power of the fusion node, send fourth information to the first sensing node through the interface unit 1001, wherein the fourth information can be used to instruct the first sensing node to send the sensing data of the first sensing node to the fusion node; and receive the sensing data of the first sensing node from the first sensing node through the interface unit 1001.

[0387] Optionally, the processing unit 1002 is further configured to: send a second request to the agent node through the interface unit 1001 when the AI ​​model in the fusion node does not support the first modality, or when one or more AI models in the fusion node support the first modality, and the computing power required for one or more AI models to fuse the perception data of multiple perception nodes is greater than the computing power of the fusion node. The second request may be used to request a second AI model, which supports the first modality, and the computing power required for the second AI model to fuse the perception data of multiple perception nodes is less than or equal to the computing power of the fusion node.

[0388] Optionally, the processing unit 1002 is further configured to: receive fifth information from the agent node via the interface unit 1001. The fifth information includes the second AI model, or the fifth information is used to indicate the download address of the second AI model.

[0389] In some examples, the processing unit 1002 is also used to send sixth information to the agent node through the interface unit 1001. The sixth information is auxiliary information required for fusing the perception data of multiple perception nodes.

[0390] In another embodiment, the communication device 1000 is applied to Figure 6 The intelligent agent node shown in this embodiment of the application is illustrated below. The specific functions of the processing unit 1002 in this embodiment are described below.

[0391] The processing unit 1002 is configured to: receive first information from the fusion node through the interface unit 1001, the first information indicating the fusion node's ability to fuse sensing data; send second information to the fusion node through the interface unit 1001, the second information indicating a first mode for fusing sensing data, the first mode being determined based on the fusion node's ability to fuse sensing data; and receive first sensing data from the fusion node through the interface unit 1001, the first sensing data being obtained by fusing sensing data from multiple sensing nodes according to the first mode.

[0392] In some possible ways, the processing unit 1002 is also used to: send a first request to the fusion node through the interface unit 1001, the first request being used to request the fusion node's ability to fuse sensing data.

[0393] In some implementations, the first sensing node belongs to multiple sensing nodes, and the first modality is the modality of the sensing data of the first sensing node. If the AI ​​model in the fusion node does not support the first modality, or if one or more AI models in the fusion node support the first modality, and the computing power required for one or more AI models to fuse the sensing data of multiple sensing nodes is greater than the computing power of the fusion node, the processing unit 1002 is further configured to: receive a second request from the fusion node through the interface unit 1001, the second request being used to request a second AI model, the second AI model supporting the first modality, and the computing power required for the second AI model to fuse the sensing data of multiple sensing nodes being less than or equal to the computing power of the fusion node.

[0394] Optionally, the processing unit 1002 is further configured to: send fifth information to the fusion node through the interface unit 1001, the fifth information including the second AI model, or the fifth information being used to indicate the download address of the second AI model.

[0395] In some examples, the processing unit 1002 is also configured to: receive sixth information from the fusion node via the interface unit 1001, the sixth information being auxiliary information required for fusing the sensing data of multiple sensing nodes.

[0396] In yet another embodiment, the communication device 1000 is applied to Figure 6 or Figure 7 The first sensing node in this embodiment of the application is shown. The specific functions of the processing unit 1002 in this embodiment are described below.

[0397] Processing unit 1002 is configured to: send third information to the fusion node via interface unit 1001, the third information indicating a first modality, the first modality being the modality of the perception data of the first perception node; if a first artificial intelligence (AI) model in the fusion node supports the first modality, and the computing power required for the first AI model to fuse the perception data of multiple perception nodes is less than or equal to the computing power of the fusion node, receive fourth information from the fusion node via interface unit 1001, the fourth information instructing the first perception node to send perception data to the fusion node; and send perception data to the fusion node via interface unit 1001. The multiple perception nodes include the first perception node.

[0398] In yet another embodiment, the communication device 1000 is applied to Figure 7The fusion node shown in this embodiment of the application is illustrated below. The specific functions of the processing unit 1002 in this embodiment will be described below.

[0399] The processing unit 1002 is configured to: receive third information from a first sensing node via an interface unit 1001, wherein the first sensing node belongs to multiple sensing nodes, and the third information can be used to indicate a first modality, wherein the first modality is the modality of the sensing data of the first sensing node; when a first AI model in a fusion node supports the first modality, and the computing power required for the first AI model to fuse the sensing data of multiple sensing nodes is less than or equal to the computing power of the fusion node, send fourth information to the first sensing node via the interface unit 1001, wherein the fourth information can be used to instruct the first sensing node to send the sensing data of the first sensing node to the fusion node; and receive the sensing data of the first sensing node from the first sensing node via the interface unit 1001.

[0400] Optionally, the processing unit 1002 is further configured to: send a second request to the agent node through the interface unit 1001 when the AI ​​model in the fusion node does not support the first modality, or when one or more AI models in the fusion node support the first modality, and the computing power required for one or more AI models to fuse the perception data of multiple perception nodes is greater than the computing power of the fusion node. The second request may be used to request a second AI model, which supports the first modality, and the computing power required for the second AI model to fuse the perception data of multiple perception nodes is less than or equal to the computing power of the fusion node.

[0401] Optionally, the processing unit 1002 is further configured to: receive fifth information from the agent node via the interface unit 1001. The fifth information includes the second AI model, or the fifth information is used to indicate the download address of the second AI model.

[0402] In some examples, the processing unit 1002 is also used to send sixth information to the agent node through the interface unit 1001. The sixth information is auxiliary information required for fusing the perception data of multiple perception nodes.

[0403] In another embodiment, the communication device 1000 is applied to Figure 7 The intelligent agent node shown in this embodiment of the application is illustrated below. The specific functions of the processing unit 1002 in this embodiment are described below.

[0404] Processing unit 1002 is configured to: receive a second request from the fusion node via interface unit 1001 when the AI ​​model in the fusion node does not support the first modality, or when one or more AI models in the fusion node support the first modality, and the computational power required for the one or more AI models to fuse the perception data of multiple perception nodes is greater than the computational power of the fusion node. The second request requests a second AI model, which supports the first modality, and the computational power required for the second AI model to fuse the perception data of multiple perception nodes is less than or equal to the computational power of the fusion node. The first perception node belongs to multiple perception nodes, and the first modality is the modality of the perception data of the first perception node.

[0405] Optionally, the processing unit 1002 is further configured to: send fifth information to the fusion node through the interface unit 1001, the fifth information including the second AI model, or the fifth information being used to indicate the download address of the second AI model.

[0406] In some examples, the processing unit 1002 is also configured to: receive sixth information from the fusion node via the interface unit 1001, the sixth information being auxiliary information required for fusing the sensing data of multiple sensing nodes.

[0407] In yet another embodiment, the communication device 1000 is applied to Figure 8 The fusion node shown in this embodiment of the application is illustrated below. The specific functions of the processing unit 1002 in this embodiment will be described below.

[0408] The processing unit 1002 is configured to: receive a third request from the agent node through the interface unit 1001, the third request being for requesting perception data for the perception target; and send second perception data to the agent node through the interface unit 1001, the second perception data being obtained by fusing perception data from multiple perception nodes according to the perception target.

[0409] In some possible ways, the first sensing node is any one of multiple sensing nodes, and the processing unit 1002 is further configured to: send fourth information to the first sensing node through the interface unit 1001 when the first sensing node is related to the sensing target, the fourth information being used to instruct the first sensing node to send the sensing data of the first sensing node to the fusion node; and receive the sensing data of the first sensing node from the first sensing node through the interface unit 1001.

[0410] Optionally, the processing unit 1002 is further configured to: send information for indicating a sensing target to the first sensing node through the interface unit 1001; and receive sensing data from the first sensing node through the interface unit 1001, wherein the sensing data of the first sensing node is related to the sensing target.

[0411] In some examples, the processing unit 1002 is also used to send sixth information to the agent node through the interface unit 1001. The sixth information is auxiliary information required for fusing the perception data of multiple perception nodes.

[0412] In yet another embodiment, the communication device 1000 is applied to Figure 8 The intelligent agent node shown in this embodiment of the application is illustrated below. The specific functions of the processing unit 1002 in this embodiment are described below.

[0413] The processing unit 1002 is configured to: send a third request to the fusion node through the interface unit 1001, the third request being used to request sensing data for the sensing target; and receive second sensing data from the fusion node through the interface unit 1001, the second sensing data being obtained by fusing sensing data from multiple sensing nodes according to the sensing target.

[0414] In some examples, the processing unit 1002 is also configured to: receive sixth information from the fusion node via the interface unit 1001, the sixth information being auxiliary information required for fusing the sensing data of multiple sensing nodes.

[0415] In yet another embodiment, the communication device 1000 is applied to Figure 8 The first sensing node in this embodiment of the application is shown. The specific functions of the processing unit 1002 in this embodiment are described below.

[0416] The processing unit 1002 is configured to: receive fourth information from the fusion node through the interface unit 1001, the fourth information being used to instruct the first sensing node to send its sensing data to the fusion node; and send the first sensing node's sensing data to the fusion node through the interface unit 1001.

[0417] Optionally, the processing unit 1002 is specifically used to: receive fourth information from the fusion node through the interface unit 1001 when the first sensing node is related to the sensing target.

[0418] Optionally, the processing unit 1002 is further configured to: receive information for indicating a sensing target from the fusion node via the interface unit 1001; the sensing data of the first sensing node is related to the sensing target.

[0419] In one possible design, when the communication device 1000 is a communication equipment or a communication module within a communication equipment, the functionality of the processing unit 1002 can be implemented by one or more processors. For example, the processor may include a modem chip, or a system-on-a-chip (SoC) or SIP chip containing a modem core. The functionality of the interface unit 1001 can be implemented by transceiver circuitry.

[0420] In one possible design, when the communication device 1000 is a circuit or chip responsible for communication functions in a communication device, such as a modem chip or a system-on-a-chip (SoC) or SIP chip containing a modem core, the function of the processing unit 1002 can be implemented by a circuit system in the aforementioned chip that includes one or more processors or processor cores. The function of the interface unit 1003 can be implemented by the interface circuit or data transceiver circuit on the aforementioned chip.

[0421] The communication equipment can be a terminal, access network equipment, core network equipment, or network management equipment.

[0422] For a more detailed description of the processing unit 1002 and the interface unit 1001, please refer to [link / reference]. Figures 6 to 9 The relevant descriptions in the method embodiments shown are directly obtained and will not be repeated here.

[0423] It should be noted that the module division in the above embodiments of this application is illustrative and only represents a logical functional division. In actual implementation, there may be other division methods. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, exist as separate physical units, or have two or more units integrated into one unit. The integrated units can be implemented in hardware, as software functional units, or in a combination of hardware and software. Whether a function is executed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0424] For example, the functional unit in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as one or more ASICs, one or more CPUs, one or more MCUs, one or more DSPs, or one or more FPGAs, or a combination of at least two of these integrated circuit forms.

[0425] If the integrated units described above are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0426] In one possible implementation, the communication device provided in this application embodiment is described below. Figure 11 As shown, the communication device 1100 includes a processor 1102. Optionally, the communication device 1100 may also include an interface circuit 1101 and a memory 1103. The interface circuit 1101, the processor 1102, and the memory 1103 are coupled to each other.

[0427] Optionally, the interface circuit 1101, processor 1102, and memory 1103 are coupled to each other via bus 1104. Bus 1104 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 11 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0428] Interface circuit 1101 is used for inputting and / or outputting information. Input information can be replaced by received information, and output information can be replaced by transmitted information. When outputting information, interface circuit 1101 can output information to other devices outside of communication device 1100, or to other units within communication device 1100. For example, interface circuit 1101 can be implemented through at least one of a physical interface, a communication module, a communication interface, an input / output interface, and a mobile communication module. The mobile communication module may include one or more of at least one antenna, at least one filter, a switch, a power amplifier, an LNA, etc.

[0429] Interface circuit 1101 may also have other names, such as transceiver circuit, communication circuit, interface, communication interface, or input / output interface. Interface circuit 1101 may include input interface circuit and output interface circuit, used for inputting information and outputting information, respectively.

[0430] Processor 1102 can be used to support communication device 1100 in performing the processing actions in the above method embodiments. When communication device 1100 is used to implement the above method embodiments, processor 1102 can also be used to implement the functions of processing unit 1002. Processor 1102 can be a CPU, or other general-purpose processors, DSPs, ASICs, FPGAs, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. General-purpose processors can be microprocessors or any conventional processor.

[0431] In one embodiment, the communication device 1100 is applied to Figure 6 The fusion node shown in this embodiment of the application is illustrated below. The specific functions of the processor 1102 in this embodiment are described below.

[0432] The processor 1102 is configured to: send first information to an agent node via an interface circuit 1101, the first information being used to indicate the ability of a fusion node to fuse sensing data; receive second information from the agent node via the interface circuit 1101, the second information being used to indicate a first mode for fusing sensing data, the first mode being determined based on the ability of the fusion node to fuse sensing data; and send first sensing data to the agent node via the interface circuit 1101, the first sensing data being obtained by fusing sensing data from multiple sensing nodes according to the first mode.

[0433] In another embodiment, the communication device 1100 is applied to Figure 6 The intelligent agent node shown in this embodiment of the application is illustrated below. The specific functions of the processor 1102 in this embodiment are described below.

[0434] The processor 1102 is configured to: receive first information from a fusion node via an interface circuit 1101, the first information indicating the fusion node's ability to fuse sensing data; send second information to the fusion node via the interface circuit 1101, the second information indicating a first mode for fusing sensing data, the first mode being determined based on the fusion node's ability to fuse sensing data; and receive first sensing data from the fusion node via the interface circuit 1101, the first sensing data being obtained by fusing sensing data from multiple sensing nodes according to the first mode.

[0435] In yet another embodiment, the communication device 1100 is applied to Figure 6 or Figure 7The first sensing node in this embodiment of the application is shown below. The specific functions of the processor 1102 in this embodiment are described below.

[0436] The processor 1102 is configured to: send third information to the fusion node via the interface circuit 1101, the third information indicating a first modality, the first modality being the mode of the perception data of the first perception node; receive fourth information from the fusion node via the interface circuit 1101 when the first artificial intelligence (AI) model in the fusion node supports the first modality and the computing power of the fusion node is greater than or equal to the computing power required by the first AI model; and send perception data to the fusion node via the interface circuit 1101.

[0437] In yet another embodiment, the communication device 1100 is applied to Figure 7 The fusion node shown in this embodiment of the application is illustrated below. The specific functions of the processor 1102 in this embodiment are described below.

[0438] The processor 1102 is configured to: receive third information from a first sensing node via an interface circuit 1101, wherein the first sensing node belongs to multiple sensing nodes, and the third information can be used to indicate a first modality, wherein the first modality is the modality of the sensing data of the first sensing node; when a first AI model in a fusion node supports the first modality, and the computing power required for the first AI model to fuse the sensing data of multiple sensing nodes is less than or equal to the computing power of the fusion node, send fourth information to the first sensing node via the interface circuit 1101, wherein the fourth information can be used to instruct the first sensing node to send the sensing data of the first sensing node to the fusion node; and receive sensing data from the first sensing node via the interface circuit 1101.

[0439] In another embodiment, the communication device 1100 is applied to Figure 7 The intelligent agent node shown in this embodiment of the application is illustrated below. The specific functions of the processor 1102 in this embodiment are described below.

[0440] Processor 1102 is configured to: receive a second request from the fusion node via interface circuit 1101 when, in the case that the AI ​​model in the fusion node does not support the first modality, or when one or more AI models in the fusion node support the first modality and the computational power required for the one or more AI models to fuse the perception data of multiple perception nodes is greater than the computational power of the fusion node; the second request is used to request a second AI model, the second AI model supporting the first modality, and the computational power required for the second AI model to fuse the perception data of multiple perception nodes is less than or equal to the computational power of the fusion node. The first perception node belongs to multiple perception nodes, and the first modality is the modality of the perception data of the first perception node.

[0441] In yet another embodiment, the communication device 1100 is applied to Figure 8 The fusion node shown in this embodiment of the application is illustrated below. The specific functions of the processor 1102 in this embodiment are described below.

[0442] The processor 1102 is configured to: receive a third request from an agent node via an interface circuit 1101, the third request being for requesting perception data for a perception target; and send second perception data to the agent node via the interface circuit 1101, the second perception data being obtained by fusing perception data from multiple perception nodes according to the perception target.

[0443] In yet another embodiment, the communication device 1100 is applied to Figure 8 The intelligent agent node shown in this embodiment of the application is illustrated below. The specific functions of the processor 1102 in this embodiment are described below.

[0444] The processor 1102 is configured to: send a third request to the fusion node via the interface circuit 1101, the third request being used to request sensing data for a sensing target; and receive second sensing data from the fusion node via the interface circuit 1101, the second sensing data being obtained by fusing sensing data from multiple sensing nodes according to the sensing target.

[0445] In yet another embodiment, the communication device 1100 is applied to Figure 8 The first sensing node in this embodiment of the application is shown below. The specific functions of the processor 1102 in this embodiment are described below.

[0446] The processor 1102 is configured to: receive fourth information from the fusion node via the interface circuit 1101, the fourth information being used to instruct the first sensing node to send the sensing data of the first sensing node to the fusion node; and send the sensing data of the first sensing node to the fusion node via the interface circuit 1101.

[0447] The specific functions of processor 1102 can be found in the descriptions of the communication methods provided in the embodiments and examples of this application above. Figure 10 The specific functional description of the communication device 1000 shown in the embodiments of this application will not be repeated here.

[0448] Memory 1103 is used to store program instructions and / or data. Specifically, program instructions may include program code, which includes computer operation instructions. Memory 1103 may include RAM and may also include non-volatile memory, such as at least one disk storage device. Processor 1102 executes the program instructions stored in memory 1103 and uses the data stored in memory 1103 to implement the above-mentioned functions, thereby realizing the communication method provided in the embodiments of this application. Memory 1103 may be integrated with processor 1102 or may be a memory outside the communication device.

[0449] It is understood that this application Figure 11 The memory 1103 can be volatile memory or non-volatile memory, or may include both. The non-volatile memory can be ROM, programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be RAM, which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0450] In one possible implementation, the communication device provided in this application embodiment is described below. Figure 12 As shown. Figure 12 The communication device 1200 shown can be an access network device or a component in the access network device (such as a chip or communication module), and can be used to perform the operations of the fusion node, intelligent agent node or first sensing node in the above method embodiments.

[0451] The communication device 1200 may include at least one processor 1211 and at least one network interface 1214. Optionally, the communication device may further include at least one of the following: at least one memory 1212, at least one transceiver 1213, and one or more antennas 1215. The processor 1211, memory 1212, transceiver 1213, and network interface 1214 may be connected, for example, via various interfaces, transmission lines, or buses. The antenna 1215 is connected to the transceiver 1213. The network interface 1214 enables the communication device 1200 to communicate with other communication devices via a communication link. For example, the network interface 1214 may include a network interface between the communication device 1200 and other communication devices (e.g., other access network devices or core network devices), such as one or more of an S1 interface, an X2 interface, or an Xn interface.

[0452] Processor 1211 can be used to perform at least one of the following operations: processing communication protocols and communication data, controlling communication device 1200, executing computer programs (or software programs), or processing data of computer programs. Exemplarily, processor 1211 can be used to support communication device 1200 in performing the actions described in the above embodiments. Optionally, processor 1211 may include a baseband processor and / or a central processing unit (CPU); or, processor 1211 may integrate the functions of a baseband processor and a CPU. The baseband processor is mainly used for processing communication protocols and communication data; the CPU is mainly used for controlling communication device 1200, executing computer programs, and processing data of computer programs. It should be understood that the baseband processor and the CPU can also be independent processors interconnected via a bus, etc. It should also be understood that the access network device may include multiple baseband processors to adapt to different network standards; and / or, the access network device may include multiple CPUs to enhance its processing capabilities. The baseband processor can also be described as a baseband processing circuit or a baseband processing chip. The CPU can also be described as a central processing circuit or a central processing chip. The function of processing communication protocols and communication data can be built into the processor or stored in memory as a software program, which is then executed by the processor to implement the baseband processing function.

[0453] The memory 1212 can be used to store computer programs and / or data. The memory 1212 can be independent of the processor 1211, or connected to the processor 1211; alternatively, the memory 1212 can be integrated with the processor 1211, for example, integrated within a single chip. The memory 1212 is capable of storing computer programs that execute the technical solutions of the embodiments of this application, and its execution is controlled by the processor 1211. Optionally, the various types of computer programs being executed can also be considered as drivers for the processor 1211. The memory 1212 may also have other names, such as storage medium or storage device.

[0454] Figure 12 Only one memory and one processor are shown. In actual access network devices, there may be multiple processors and multiple memories. The memory can be a storage element on the same chip as the processor, i.e., an on-chip storage element, or a separate storage element; this application does not limit this.

[0455] Transceiver 1213 can be used to support the reception or transmission of radio frequency (RF) signals between a communication device and a terminal. Transceiver 1213 may include a transmitter Tx and a receiver Rx. Exemplarily, one or more antennas 1215 can receive RF signals. The receiver Rx of transceiver 1213 is used to receive the RF signals from the antennas, convert the RF signals into digital baseband signals or digital intermediate frequency (IF) signals, and provide the digital baseband signals or IF signals to the processor 1211 so that the processor 1211 can perform further processing on the digital baseband signals or IF signals, such as demodulation and decoding. Furthermore, the transmitter Tx in transceiver 1213 is also used to receive modulated digital baseband signals or IF signals from processor 1211, convert the modulated digital baseband signals or IF signals into RF signals, and transmit the RF signals through one or more antennas 1215. Specifically, the receiver Rx can selectively perform one or more stages of downmixing and analog-to-digital conversion on the radio frequency signal to obtain a digital baseband signal or a digital intermediate frequency (IF) signal. The order of these downmixing and IF conversion processes is adjustable. The transmitter Tx can selectively perform one or more stages of upmixing and digital-to-analog conversion on the modulated digital baseband signal or digital IF signal to obtain a radio frequency signal. The order of these upmixing and IF conversion processes is also adjustable. The digital baseband signal and the digital IF signal can be collectively referred to as digital signals.

[0456] The transceiver 1213 can also be referred to as a transceiver unit, transceiver, or transceiver device. Optionally, the device in the transceiver 1213 used to implement the receiving function can be regarded as a receiving unit, and the device in the transceiver 1213 used to implement the transmitting function can be regarded as a transmitting unit. That is, the transceiver 1213 includes a receiving unit and a transmitting unit. The receiving unit can also be referred to as a receiver, receiver circuit, or receiving device, and the transmitting unit can be referred to as a transmitter, transmitter, or transmitting circuit.

[0457] In one possible implementation, the communication device provided in this application embodiment is described below. Figure 13 As shown. It is understood that the communication device 1300 includes means of the necessary form, such as modules, units, elements, circuits, or interfaces, to be appropriately configured together to perform this solution. Figure 13The communication device 1300 shown can be a terminal, a component in a terminal (e.g., a chip or communication module), an access network device, or a component in an access network device (e.g., a chip or communication module), and can be used to perform the operations of the fusion node, intelligent agent node, or first sensing node in the above method embodiments. The communication device 1300 includes one or more processors 1301. The processor 1301 can be a general-purpose processor or a dedicated processor, etc. Optionally, the processor 1301 may include a baseband processor and / or a central processing unit; or, the processor 1301 may integrate the functions of a baseband processor and a central processing unit. The specific details of the processor 1301 can be found in the description of the processor 1211 above, and will not be repeated here.

[0458] Optionally, in one possible design, processor 1301 may include program 1303. Program 1303 can be executed on processor 1301, causing communication device 1300 to perform the methods described in the above method embodiments. In another possible design, communication device 1300 includes circuitry (…). Figure 13 (Not shown), this circuit is used to perform the methods in the above method embodiments; or, in other words, this circuit can be used to indicate the functions of the fusion node, agent node, or first sensing node in the above method embodiments.

[0459] Optionally, the communication device 1300 may include one or more memories 1302. The memories 1302 store a program 1304, which can be executed on the processor 1301 to cause the communication device 1300 to perform the methods described in the above method embodiments.

[0460] Optionally, processor 1301 may include AI module 1307, and / or memory 1302 may include AI module 1308. The AI ​​module can be used to implement AI-related functions. The AI ​​module can be implemented through software, hardware, or a combination of both. For example, the AI ​​module may include a radio intelligence control (RIC) module. For example, the AI ​​module may be a near real-time RIC or a non-real-time RIC.

[0461] Optionally, data may also be stored in the processor 1301 and / or the memory 1302. The processor and memory may be configured separately or integrated together.

[0462] Optionally, the communication device 1300 may also include a transceiver 1305 and / or an antenna 1306. The transceiver 1305 may also be referred to as a transceiver unit, transceiver, transceiver circuit, or transceiver, etc., and can be used to realize the transmission and reception functions of the communication device through the antenna 1306.

[0463] Based on the above embodiments, this application also provides a computer program product including computer-executable instructions, which, when run, causes the methods provided in the above embodiments to be executed.

[0464] Based on the above embodiments, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a computer, causes the computer to perform the methods provided in the above embodiments.

[0465] The storage medium can be any available medium that a computer can access. For example, but not limited to, a computer-readable medium can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.

[0466] Based on the above embodiments, this application also provides a chip for reading a computer program stored in a memory and implementing the method provided in the above embodiments.

[0467] Based on the above embodiments, this application provides a chip system including a processor for supporting a computer device in implementing the functions involved in the devices in the above embodiments. In one possible design, the chip system further includes a memory for storing necessary programs and data of the computer device. The chip system may be composed of chips or may include chips and other discrete components.

[0468] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of different embodiments are consistent and can be referenced by each other. The technical features of different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0469] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0470] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0471] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0472] In this application, the terms "system" and "network" are used interchangeably. "At least one item" refers to one or more items, and "more than one item" refers to two or more items. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. In the textual description of this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0473] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application. The order of the process numbers described above does not imply the order of execution; the execution order of each process should be determined by its function and internal logic.

[0474] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A communication method applied to a converged node, characterized in that, include: Send first information to the intelligent agent node, the first information being used to indicate the fusion node's ability to fuse sensing data; The agent node receives second information, which indicates a first mode for fusing sensing data, the first mode being determined based on the fusion node's ability to fuse sensing data. Send first perception data to the intelligent agent node. The first perception data is obtained by fusing the perception data of multiple perception nodes according to the first mode.

2. The method as described in claim 1, characterized in that, The ability of the fusion node to fuse sensing data includes at least one of the following: The computing power of the fusion node; or The fusion node supports at least one artificial intelligence (AI) model, which is used to fuse perceived data.

3. The method as described in claim 2, characterized in that, In the case that the fusion node's ability to fuse sensing data includes the at least one AI model, the first information is also used to indicate at least one of the following: The modalities supported by the at least one AI model; or The parameters of the modality supported by the at least one AI model.

4. The method according to any one of claims 1 to 3, characterized in that, The first mode includes at least one of the following: Modalities of fused sensory data; The cycle of fusion sensing data; or AI models that integrate sensory data.

5. The method according to any one of claims 1 to 4, characterized in that, Also includes: The first request is received from the agent node, the first request being used to request the fusion node's ability to fuse sensing data.

6. The method according to any one of claims 1 to 5, characterized in that, Also includes: The third information is received from the first sensing node, which belongs to the plurality of sensing nodes. The third information is used to indicate a first mode, which is the mode of the sensing data of the first sensing node. If the first AI model in the fusion node supports the first modality, and the computing power required for the first AI model to fuse the perception data of the multiple perception nodes is less than or equal to the computing power of the fusion node, then a fourth message is sent to the first perception node, the fourth message being used to instruct the first perception node to send the perception data of the first perception node to the fusion node. Receive sensing data from the first sensing node.

7. The method as described in claim 6, characterized in that, Also includes: If the AI ​​model in the fusion node does not support the first modality, or if one or more AI models in the fusion node support the first modality and the computing power required for the one or more AI models to fuse the perception data of the multiple perception nodes is greater than the computing power of the fusion node, a second request is sent to the agent node. The second request is used to request a second AI model, which supports the first modality and the computing power required for the second AI model to fuse the perception data of the multiple perception nodes is less than or equal to the computing power of the fusion node.

8. The method as described in claim 7, characterized in that, If the AI ​​model in the fusion node does not support the first modality, the second request includes: information about the first modality; or, If one or more AI models in the fusion node support the first modality, and the computing power required for the one or more AI models to fuse the perception data of the multiple perception nodes is greater than the computing power of the fusion node, the second request includes at least one of the following: parameters of each of the multiple perception nodes, and the number of the multiple perception nodes.

9. The method as described in claim 7 or 8, characterized in that, Also includes: The agent node receives fifth information, which includes the second AI model, or the fifth information is used to indicate the download address of the second AI model.

10. The method according to any one of claims 1 to 9, characterized in that, Also includes: A sixth message is sent to the intelligent agent node, which is auxiliary information required for fusing the perception data of the multiple perception nodes.

11. The method as described in claim 10, characterized in that, The sixth piece of information includes at least one of the following: Configuration of an AI model that fuses the sensing data from the multiple sensing nodes; or The configuration of the multiple sensing nodes.

12. A communication method applied to a fusion node, characterized in that, include: Receive a third request from the agent node, the third request being used to request perception data for the perception target; Send second perception data to the intelligent agent node. The second perception data is obtained by fusing the perception data of multiple perception nodes according to the perception target.

13. The method as described in claim 12, characterized in that, The first sensing node is any one of the plurality of sensing nodes, and further includes: When the first sensing node is related to the sensing target, a fourth message is sent to the first sensing node, the fourth message being used to instruct the first sensing node to send the sensing data of the first sensing node to the fusion node. Receive sensing data from the first sensing node.

14. The method as described in claim 12 or 13, characterized in that, Also includes: Send information indicating the sensing target to the first sensing node; The system receives sensing data from the first sensing node, and the sensing data from the first sensing node is related to the sensing target.

15. The method according to any one of claims 12 to 14, characterized in that, Also includes: A sixth message is sent to the intelligent agent node, which is auxiliary information required for fusing the perception data of the multiple perception nodes.

16. The method as described in claim 15, characterized in that, The sixth piece of information includes at least one of the following: Configuration of an artificial intelligence (AI) model that fuses the sensing data from the multiple sensing nodes; or The configuration of the multiple sensing nodes.

17. A communication device, characterized in that, Includes a unit for performing the method as described in any one of claims 1-16.

18. A communication device, characterized in that, Includes a processor for executing computer programs or instructions that cause the apparatus to perform the method as described in any one of claims 1-16.

19. A communication system, characterized in that, This includes one or more of the following: fusion node, agent node, and first perception node. The fusion node is used to perform the method as described in any one of claims 1-16, the agent node is used to perform the operation of the agent node as described in claims 1-16, and the first perception node is used to perform the operation of the first perception node as described in claims 1-16.

20. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program or instructions, which, when executed, implement the method as described in any one of claims 1-16.

21. A computer program product, characterized in that, The computer program product includes: computer program code, which, when the computer program code is run, implements the method as described in any one of claims 1-16.

Images