Communication method and communication device

Abstract

The invention provides a communication method and a communication device. The method comprises the following steps: a first node determines first training data, wherein the first training data comprises network experience information of a station STA; the first node sends the first training data to the second node, the first training data is used for model training of a first intelligent model, and the first intelligent model is used for reasoning a transmission strategy of a communication network where the first node is located. The model training of the intelligent model can be realized, so that the intelligent model is applied to a wireless network, a transmission strategy for improving STA experience can be obtained through reasoning, and the network communication performance is improved.

Description

Technical Field

[0001] The present application relates to the field of communications, and more specifically, to a communication method and a communication device. Background Art

[0002] Wireless communication is developing rapidly, and the fifth generation of mobile communication (5 th The 5G (5th generation) and the sixth-generation wireless fidelity (Wi-Fi) standards have been commercialized, and the next-generation wireless technology and standardization are in full swing. Wireless communication has been popularized in all aspects of production and life and has become an indispensable part. The number of smart terminals has increased dramatically, and various new wireless applications (such as virtual reality, augmented reality, and holographic imaging) have emerged in an endless stream, making wireless networks unprecedentedly complex.

[0003] In the complex wireless networks of the future, if the current access point (AP) determines the communication transmission strategy for the station (STA) based on the channel conditions, it may not match the network transmission environment, and thus fail to guarantee the transmission needs of the STA. With the development trend of high complexity of wireless networks, considering the application of artificial intelligence (AI) technology to assist in the management of wireless networks, how to implement AI technology in wireless networks to improve communication performance has become a hot topic of current research. Summary of the invention

[0004] The embodiments of the present application provide a communication method and a communication device, which can implement the application of intelligent models to wireless networks and improve communication performance.

[0005] In a first aspect, a data transmission method is provided, which is described below by taking a first node executing the method as an example. The first node may be a communication device or a module (such as a chip or a chip module) configured in (or used for) a communication device.

[0006] The method includes: a first node determines first training data, the first training data includes network experience information of multiple station STAs. The first node sends the first training data to a second node, the first training data is used for model training of a first intelligent model, and the first intelligent model is used to infer the transmission strategy of the communication network where the first node is located.

[0007] According to the above solution, the first node can send the first training data including the network experience information of the STA to the second node, so that through the transmission of the first training data in the network, the node (referred to as the intelligent node) maintaining the first intelligent model can perform model training on the first intelligent model based on the first training data, realizing the application of the first intelligent model to the wireless network, being able to infer the transmission strategy to improve the network experience of the STA, and achieving the improvement of network communication performance.

[0008] Combined with the first aspect, in some implementation manners of the first aspect, the first training data further includes one or more of the network state information corresponding to the network experience information, the transmission strategy information corresponding to the network experience information, or the acquisition time information. Wherein, the acquisition time information is used to indicate the acquisition time of the information in the first training data.

[0009] According to the above solution, the training data for training the first intelligent model may further include the network state information corresponding to the network experience information of the STA and the transmission strategy information of the STA. These training data may also be determined by the first node and provided to the intelligent node through network transmission to achieve the model training of the first intelligent model. However, this application is not limited thereto. The intelligent node can obtain these training data from other nodes (such as management nodes in the network) other than the first node, so as to achieve the model training of the first intelligent model.

[0010] Combined with the first aspect, in some implementation manners of the first aspect, the first node is the first STA, and the second node is the intelligent node, where the first STA is a STA with the ability to access the intelligent node, and the intelligent node is the node maintaining the first intelligent model.

[0011] According to the above solution, a STA with the ability to access the intelligent node can collect the network experience information of multiple STAs and provide it to the intelligent node to achieve the model training of the first intelligent model by the intelligent node, and further realize the application of the first intelligent model to the wireless network, being able to infer the transmission strategy to improve the network experience of the STA, and achieving the improvement of network communication performance.

[0012] Combined with the first aspect, in some implementation manners of the first aspect, the first node determines the first training data, including: the first STA sends a first request message to the access point AP, and the first request message is used to request the first training data; the first STA receives the first training data from the AP.

[0013] According to the above solution, the first STA can obtain the network experience information of multiple STAs collected by the AP through the AP, so as to provide it to the intelligent node to achieve the model training of the first intelligent model by the intelligent node.

[0014] In combination with the first aspect, in some implementations of the first aspect, the first node is an AP, and the second node is a smart node or a first STA, where the first STA is a STA capable of accessing the smart node, and the smart node is a node that maintains the first intelligent model.

[0015] According to the above solution, the first node can be an AP. The AP provides the collected first training data to the smart node, or provides the first training data to the first STA capable of accessing the smart node, so as to be transmitted to the smart node through the first STA, realizing the model training of the first intelligent model by the smart node.

[0016] In combination with the first aspect, in some implementations of the first aspect, the first node determines the first training data, including: the AP receives the first request information from the first STA, and the first request information is used to request the first training data; the AP receives multiple first information from multiple STAs, and each first information in the multiple first information includes the network experience information of each STA; the AP determines the first training data according to the multiple first information.

[0017] According to the above solution, the AP can, in response to the request of the first STA, collect the network experience information of each STA from multiple STAs and provide it to the first STA, so as to realize the model training of the first intelligent model by the smart node through the transmission of the first STA.

[0018] In one implementation, the first training data includes the first information of the multiple STAs, and the first information further includes at least one of the network state information corresponding to the network experience information of the STA, the transmission policy information of the STA corresponding to the network experience information of the STA, or the acquisition time information.

[0019] That is to say, the first information of the multiple STAs collected by the AP further includes the network state information and transmission policy information corresponding to the network experience information. The first training data determined by the AP includes this information in the first information of the multiple STAs. Through the transmission of the first STA, the smart node can obtain all the training data for training the first intelligent model to complete the model training without having to obtain training data from other nodes.

[0020] In another implementation, the first training data is determined by the AP according to at least one of the network state information, the transmission policy information of the STA, or the acquisition time information and the first information of the multiple STAs, where the acquisition time information is used to indicate the acquisition time of the information in the first training data.

[0021] That is to say, the first information obtained by the AP from multiple STAs includes the network experience information of the STAs. The AP can have a global observation ability to obtain the current network status information and the transmission policy information of the current STA, enabling the AP to determine the first training data including the network status information, the transmission policy information of the STA, and the network experience information of the STA based on this information and the first information collected from the STAs, and transmit it through the first STA, so that the intelligent node can obtain all the training data for training the first intelligent model to complete the model training without having to obtain training data from other nodes.

[0022] Combined with the first aspect, in some implementation manners of the first aspect, the determination of the first training data by the first node includes: the AP receives first request information from the first STA, and the first request information is used to request the first training data; the AP inputs the network status information and the transmission policy information of multiple STAs into a second intelligent model to obtain the first training data output by the second intelligent model.

[0023] Exemplarily, the second intelligent model is a reward model. The input of the reward model is the network status information and the transmission policy information of the STA, and the output of the reward model includes the network experience information of the STA.

[0024] According to the above solution, the AP can maintain an intelligent model for inferring the network experience information of the STA, so that each time the AP receives a request from the first STA, it can use the current network status information and the transmission policy information of multiple STAs as the input of the model to obtain the network experience information of multiple STAs inferred by the model. There is no need to collect the network experience information from each STA every time the first STA requests. It can reduce the signaling overhead and improve the efficiency of obtaining training data.

[0025] Combined with the first aspect, in some implementation manners of the first aspect, the method further includes: the AP performs model training according to the network experience information of multiple STAs, the transmission policy information corresponding to the network experience information, and the network status information to obtain a second intelligent model.

[0026] According to the above solution, the AP can obtain a second intelligent model capable of inferring the network experience information of the STA through model training based on the historical network experience information of the STAs in the network, the historical transmission policy information of the STAs, and the corresponding network status information. However, the present application is not limited thereto. The second intelligent model can also be configured on the AP after being trained by other network nodes.

[0027] In combination with the first aspect, in some implementations of the first aspect, the second node is an intelligent node, which is a node that maintains the first intelligent model; the method further includes: the first node sends second request information to the second node, where the second request information is used to request transmission policy information corresponding to the current network state information; the first node receives response information from the second node, where the response information is used to indicate first transmission policy information, and the first transmission policy information is the transmission policy information corresponding to the current network state information; the first node performs data transmission according to the first transmission policy information.

[0028] According to the above solution, after the intelligent node completes the model training of the first intelligent model, the first node can request a transmission policy from the intelligent node, and the intelligent node can infer a transmission policy that matches the current network state based on the first intelligent model, which can improve the transmission policy of the STA network experience and achieve the improvement of network communication performance.

[0029] In combination with the first aspect, in some implementations of the first aspect, the first node sending the second request information to the second node includes: when the number of STAs whose network experience does not meet the requirements is greater than or equal to a threshold number, the first node sends the second request information to the second node.

[0030] According to the above solution, when the number of STAs whose network experience does not meet the requirements reaches the threshold number, the first node requests a transmission policy from the intelligent node, so that the intelligent node can infer a transmission policy that matches the network environment based on the first intelligent model, improve the network experience of the STA, and thus achieve the improvement of network communication performance.

[0031] In combination with the first aspect, in some implementations of the first aspect, the first node is the first STA, and the first STA is an STA with the ability to access the intelligent node; the first node sending the second request information to the second node includes: when the network experience of the first STA does not meet the requirements, the first STA sends third request information to the AP, where the third request information is used to request the current network state information; the first STA receives the current network state information from the AP; the first STA sends the second request information to the AP, and the second request information includes the current network state information.

[0032] According to the above solution, the first STA can request a transmission policy from the intelligent node when its own network experience does not meet the requirements, so as to obtain a transmission policy that matches the network environment and can improve the STA network experience inferred by the intelligent node based on the first intelligent model.

[0033] In combination with the first aspect, in certain implementations of the first aspect, the second request information includes the current network status information represented by natural language; and / or, the response information includes the first transmission policy information represented by natural language.

[0034] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: the first node processes the current network transmission status based on the natural language representation method to obtain the current network transmission status information represented by natural language; and / or, the first node parses the first transmission policy information represented by natural language in the response information based on the natural language representation method to obtain the first transmission policy information.

[0035] According to the above solution, the first STA and the intelligent node can avoid the way of the protocol predefined information set (such as the network status information set, the transmission policy information set) to make both parties of the information interaction reach a consensus on the information set and select the corresponding information from the set for interaction. It can reduce the storage overhead of both parties of the interaction brought by storing the predefined information set, and the information type is not limited in the set, and the scalability is good.

[0036] In combination with the first aspect, in certain implementations of the first aspect, the first intelligent model is a large language model LLM, the input of the LLM is the first training data, and the output of the LLM is the transmission policy information.

[0037] According to the above solution, after the LLM trains the basic model, it only needs to introduce professional domain knowledge and fine-tune some parameters to be widely applied in all walks of life. It can reduce the implementation complexity of AI technology in wireless networks.

[0038] In a second aspect, a communication method is provided, and this method can be executed by a second node or a module (such as a chip or a chip module) configured in (or used for) the second node.

[0039] The method includes: the second node receives the first training data from the first node, and the first training data includes the network experience information of the STA; the second node performs model training on the first intelligent model according to the first training data to obtain the trained first intelligent model, and the first intelligent model is used to infer the transmission policy in the communication network where the first node is located.

[0040] In combination with the second aspect, in certain implementations of the second aspect, the first training data further includes one or more of the network status information corresponding to the network experience information, the transmission policy information corresponding to the network experience information, or the collection time information. Wherein, the collection time information is used to indicate the collection time of the information in the first training data.

[0041] In combination with the second aspect, in some implementations of the second aspect, the method further includes: the second node receives second request information from the first node, where the second request information is used to request transmission policy information corresponding to the current network state information; the second node sends response information to the first node, where the response information is used to indicate first transmission policy information, and the first transmission policy information is the transmission policy information corresponding to the current network state.

[0042] In combination with the second aspect, in some implementations of the second aspect, the method further includes: the second node inputs the current network state information into the first intelligent model to obtain the first transmission policy information inferred by the first intelligent model.

[0043] In combination with the second aspect, in some implementations of the second aspect, the second request information includes the current network state information.

[0044] In combination with the second aspect, in some implementations of the second aspect, the second request information includes the current network state information represented by natural language; and / or, the response information includes the first transmission policy information represented by natural language.

[0045] In combination with the second aspect, in some implementations of the second aspect, the first node is an access point AP or a station STA with the ability to access intelligent nodes, and the second node is an intelligent node.

[0046] In combination with the second aspect, in some implementations of the second aspect, the first intelligent model is a large language model LLM, the input of the LLM is the first training data, and the output of the LLM is the transmission policy.

[0047] In a third aspect, a communication device is provided. In one design, the device may include modules corresponding one by one to the methods / operations / steps / actions described in the first aspect or any implementation manner of the first aspect. The module may be a hardware circuit, software, or a combination of hardware circuit and software. In one design, the communication device is applied to the first node, and the device includes: a processing unit for determining first training data, where the first training data includes network experience information of multiple stations STA; a transceiver unit for sending the first training data to a second node, where the first training data is used for model training of a first intelligent model, and the first intelligent model is used to infer the transmission policy of the communication network where the first node is located.

[0048] In combination with the third aspect, in some implementations of the third aspect, the first training data further includes one or more of the network state information corresponding to the network experience information, the transmission policy information corresponding to the network experience information, or the collection time information. Wherein, the collection time information is used to indicate the collection time of the information in the first training data.

[0049] In combination with the third aspect, in some implementations of the third aspect, the first node is a first STA, and the second node is a smart node, where the first STA is a STA capable of accessing the smart node, and the smart node is a node that maintains the first intelligent model.

[0050] In combination with the third aspect, in some implementations of the third aspect, the transceiver unit is further configured to send a first request message to an access point AP, where the first request message is used to request the first training data; the transceiver unit is further configured to receive the first training data from the AP.

[0051] In combination with the third aspect, in some implementations of the third aspect, the first node is an AP, and the second node is a smart node or a first STA, where the first STA is a STA capable of accessing the smart node, and the smart node is a node that maintains the first intelligent model.

[0052] In combination with the third aspect, in some implementations of the third aspect, the transceiver unit is further configured to receive a first request message from the first STA, where the first request message is used to request the first training data; the transceiver unit is further configured to receive multiple first messages from multiple STAs, where each first message in the multiple first messages includes the network experience information of each STA; the processing unit is specifically configured to determine the first training data according to the multiple first messages.

[0053] In combination with the third aspect, in some implementations of the third aspect, the first training data includes the multiple first messages of the multiple STAs, and the first message further includes at least one of network state information corresponding to the network experience information of the STA, transmission policy information of the STA corresponding to the network experience information of the STA, or acquisition time information; or, the first training data is determined by the AP according to at least one of network state information, transmission policy information of the STA, or acquisition time information and the multiple first messages of the multiple STAs. The acquisition time information is used to indicate the acquisition time of the information in the first training data.

[0054] In combination with the third aspect, in some implementations of the third aspect, the transceiver unit is further configured to receive a first request message from the first STA, where the first request message is used to request the first training data; the processing unit is specifically configured to input network state information and transmission policy information of multiple STAs into a second intelligent model to obtain the first training data output by the second intelligent model.

[0055] In combination with the third aspect, in some implementation manners of the third aspect, the processing unit is further configured to perform model training according to the network experience information of multiple STAs, the transmission policy information corresponding to the network experience information, and the network status information, so as to obtain a second intelligent model.

[0056] In combination with the third aspect, in some implementation manners of the third aspect, the second intelligent model is a reward model, the input of the reward model is the network status information and the transmission policy information of the STA, and the output of the reward model includes the network experience information of the STA.

[0057] In combination with the third aspect, in some implementation manners of the third aspect, the second node is an intelligent node, and the intelligent node is a node that maintains the first intelligent model; the transceiver unit is further configured to send second request information to the second node, where the second request information is used to request the transmission policy information corresponding to the current network status information; the transceiver unit is further configured to receive response information from the second node, where the response information is used to indicate first transmission policy information, and the first transmission policy information is the transmission policy information corresponding to the current network status information; the processing unit is further configured to perform data transmission according to the first transmission policy information.

[0058] In combination with the third aspect, in some implementation manners of the third aspect, the transceiver unit is specifically configured to send the second request information to the second node when the number of STAs whose network experience does not meet the requirements is greater than or equal to a number threshold.

[0059] In combination with the third aspect, in some implementation manners of the third aspect, the first node is a first STA, and the first STA is a STA having the ability to access an intelligent node; the transceiver unit is specifically configured to send third request information to the AP when the network experience of the first STA does not meet the requirements, where the third request information is used to request the current network status information; the transceiver unit is specifically further configured to receive the current network status information from the AP; the transceiver unit is specifically further configured to send the second request information to the AP, and the second request information includes the current network status information.

[0060] In combination with the third aspect, in some implementation manners of the third aspect, the second request information includes the current network status information represented by natural language; and / or, the response information includes the first transmission policy information represented by natural language.

[0061] In combination with the third aspect, in some implementation manners of the third aspect, the processing unit is further configured to process the current network transmission status based on a natural language representation manner to obtain the current network transmission status information represented by natural language; and / or, the processing unit is further configured to parse the first transmission policy information represented by natural language in the response information based on a natural language representation manner to obtain the first transmission policy information.

[0062] In combination with the third aspect, in some implementations of the third aspect, the first intelligent model is a large language model (LLM), the input of the LLM is the first training data, and the output of the LLM is transmission policy information.

[0063] In a fourth aspect, a communication device is provided. In one design, the device may include modules corresponding one by one to the methods / operations / steps / actions described in the second aspect or any implementation manner of the second aspect. The module may be a hardware circuit, software, or a combination of a hardware circuit and software. In one design, the communication device is applied to a second node, and the device includes: a transceiver unit configured to receive first training data from a first node, where the first training data includes network experience information of a STA; a processing unit configured to perform model training on a first intelligent model according to the first training data to obtain the trained first intelligent model, where the first intelligent model is used to infer a transmission policy in the communication network where the first node is located.

[0064] In combination with the fourth aspect, in some implementations of the fourth aspect, the first training data further includes one or more of network state information corresponding to the network experience information, transmission policy information corresponding to the network experience information, or acquisition time information. The acquisition time information is used to indicate the acquisition time of the information in the first training data.

[0065] In combination with the fourth aspect, in some implementations of the fourth aspect, the transceiver unit is further configured to receive second request information from the first node, where the second request information is used to request transmission policy information corresponding to the current network state information; the transceiver unit is further configured to send response information to the first node, where the response information is used to indicate first transmission policy information, and the first transmission policy information is the transmission policy information corresponding to the current network state.

[0066] In combination with the fourth aspect, in some implementations of the fourth aspect, the processing unit is further configured to input the current network state information into the first intelligent model to obtain the first transmission policy information inferred by the first intelligent model.

[0067] In combination with the fourth aspect, in some implementations of the fourth aspect, the second request information includes the current network state information.

[0068] In combination with the fourth aspect, in some implementations of the fourth aspect, the second request information includes the current network state information represented by natural language; and / or, the response information includes the first transmission policy information represented by natural language.

[0069] In combination with the fourth aspect, in some implementations of the fourth aspect, the first node is an access point AP or a station STA capable of accessing intelligent nodes, and the second node is an intelligent node.

[0070] In combination with the fourth aspect, in some implementations of the fourth aspect, the first intelligent model is a large language model LLM, the input of the LLM is the first training data, and the output of the LLM is a transmission policy.

[0071] A fifth aspect provides a communication device including a processor. The processor can implement the methods in any of the possible implementations of the above first aspect to the second aspect and the first aspect to the second aspect. Optionally, the communication device further includes a memory, and the processor is coupled to the memory and can be used to execute instructions in the memory to implement the methods in any of the possible implementations of the above first aspect to the second aspect and the first aspect to the second aspect. Optionally, the communication device further includes a communication interface, and the processor is coupled to the communication interface. In the embodiments of the present application, the communication interface can be a transceiver, a pin, a circuit, a bus, a module, or other types of communication interfaces, without limitation.

[0072] In one implementation, the communication device is a communication device (such as an STA or an AP). When the communication device is a communication device, the communication interface can be a transceiver or an input / output interface.

[0073] In another implementation, the communication device is a chip configured in a communication device. When the communication device is a chip configured in a communication device, the communication interface can be an input / output interface.

[0074] Optionally, the transceiver can be a transceiver circuit. Optionally, the input / output interface can be an input / output circuit.

[0075] A sixth aspect provides a processor including: an input circuit, an output circuit, and a processing circuit. The processing circuit is used to receive signals through the input circuit and transmit signals through the output circuit, so that the processor executes the methods in any of the possible implementations of the above first aspect to the second aspect and the first aspect to the second aspect.

[0076] In the specific implementation process, the above-mentioned processor can be one or more chips, the input circuit can be an input pin, the output circuit can be an output pin, and the processing circuit can be transistors, gate circuits, flip-flops, and various logic circuits, etc. The input signal received by the input circuit can be received and input by, for example but not limited to, a receiver. The signal output by the output circuit can be output to, for example but not limited to, a transmitter and transmitted by the transmitter. Moreover, the input circuit and the output circuit can be the same circuit, which serves as the input circuit and the output circuit at different times respectively. The embodiments of the present application do not limit the specific implementation manners of the processor and various circuits.

[0077] In a seventh aspect, there is provided a computer program product, which includes: a computer program (which can also be referred to as code, or instruction). When the computer program is run, it causes the computer to execute the methods in the above-mentioned first aspect to the second aspect and any possible implementation manner in the first aspect to the second aspect.

[0078] In an eighth aspect, there is provided a computer-readable storage medium, which stores a computer program (which can also be referred to as code, or instruction). When it runs on a computer, it causes the computer to execute the methods in the above-mentioned first aspect to the second aspect and any possible implementation manner in the first aspect to the second aspect.

[0079] In a ninth aspect, there is provided a communication system, which includes at least one of the foregoing first nodes and at least one of the foregoing second nodes. Optionally, the first node is the above-mentioned first STA, the second node is the above-mentioned AP, and the communication system further includes at least one of the foregoing intelligent nodes. Description of the Drawings

[0080] Figure 1 is a schematic diagram of a wireless communication system applicable to the embodiments of the present application;

[0081] Figure 2 is a schematic flowchart of a communication method provided by the embodiments of the present application;

[0082] Figure 3 is another schematic flowchart of a communication method provided by the embodiments of the present application;

[0083] Figure 4 is a schematic structural diagram of a communication device of the present application;

[0084] Figure 5 is another schematic structural diagram of a communication device of the present application. Detailed Embodiments

[0085] Next, the technical solutions in the present application will be described in conjunction with the drawings.

[0086] In the embodiments of the present application, " / " may indicate that the objects associated before and after are in an "or" relationship. For example, A / B may indicate A or B; "and / or" may be used to describe three relationships of associated objects. For example, A and / or B may indicate: A exists alone, A and B exist simultaneously, and B exists alone. Herein, A and B may be singular or plural. For the convenience of describing the technical solutions of the embodiments of the present application, in the embodiments of the present application, terms such as "first" and "second" may be used for distinction. These terms such as "first" and "second" do not limit the quantity and execution order, and these terms such as "first" and "second" do not necessarily limit being different. In the embodiments of the present application, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Using words such as "exemplary" or "for example" aims to present relevant concepts in a specific manner for easy understanding. In the embodiments of the present application, at least one (kind) may also be described as one (kind) or more than one (kind). More than one (kind) may be two (kinds), three (kinds), four (kinds) or more. The present application does not make any restrictions.

[0087] The technical solutions provided by the embodiments of this application can be applied to various communication systems, such as: wireless local area network (WLAN) systems, such as wireless-fidelity (Wi-Fi), etc. For example, the solutions provided by the embodiments of this application can be applied to wireless local area network systems that support the Institute of Electrical and Electronics Engineers (IEEE) 802.11ax next-generation Wi-Fi protocol (such as 802.11bf, 802.11be, Wi-Fi 8, Extremely High Throughput (EHT), Ultra-High Reliability (UHR), Wi-Fi AI, etc., 802.11 series protocols), and can also be applied to wireless personal area network systems based on Ultra Wide Band (UWB) (such as 802.15 series standards), sensing systems (such as 802.11bf series standards). Among them, the 802.11ax standard can also be called the High-Efficient (HE) standard, and the 802.11be standard can also be called the Extremely High Throughput (EHT) standard. Among them, 802.11bf includes two major categories of standards: low frequency (for example, sub-7GHz) and high frequency (for example, 60GHz). The implementation of sub-7GHz mainly relies on standards such as 802.11ac, 802.11ax, 802.11be, and the next generation, etc. The implementation of 60GHz mainly relies on standards such as 802.11ad, 802.11ay, and the next generation, etc. Among them, 802.11ad can also be called the Directional Multi-Gigabit (DMG) standard, and 802.11ay can also be called the Enhanced Directional Multi-Gigabit (EDMG) standard.

[0088] All aspects related to the embodiments of this application can also be applied to other networks that adopt various standards or protocols, such as, High Performance Radio Local Area Network (HIPERLAN), Wireless Wide Area Network (WWAN), Wireless Personal Area Network (WPAN), or other currently known or future-developed networks.

[0089] The technical solution of the embodiment of the present application can also be applied to various communication systems, such as: WLAN communication systems, wireless fidelity (Wi-Fi) systems, long term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD), universal mobile telecommunication system (UMTS), worldwide interoperability for microwave access (WiMAX) communication systems, fifth generation (5G) systems or new radio (NR), sixth generation (6G) systems, internet of things (IoT) networks or vehicle to x (V2X), as well as new communication systems emerging in the future development of communications, etc.

[0090] Figure 1 It is a schematic diagram of the communication system 100 provided by the embodiment of the present application. The communication system 100 includes at least one network device, such as Figure 1 shown, the network device can be an access point (AP). And, the communication system 100 may further include at least one terminal, such as Figure 1 shown, the terminal can be a station (STA).

[0091] Optionally, the AP or STA provided by the present application may have certain intelligent model (or artificial intelligence (AI)) operation and maintenance capabilities, such as being able to use the AI model to infer communication transmission decisions, and being able to perform model training on the AI model to achieve model optimization. Exemplarily, the AI model can be a neural network model.

[0092] Optionally, the communication system 100 may further include an intelligent node, such as the intelligent node may be a server or the like. The intelligent node may have the ability to operate and maintain the AI model. The intelligent node may interact with the AP and / or STA. For example, the intelligent node may obtain model training data from the AP and / or STA to perform model training of the AI model. The intelligent node may also, in response to requests from the AP and / or STA, use the AI model to infer the communication transmission strategy of the communication network where the AP and / or STA are located, and feedback it to the AP and / or STA. To improve the system transmission performance.

[0093] Exemplarily, the AP may be understood as an access point entity, and the STA may also be understood as a station entity. Among them, the AP and STA may support the WLAN communication protocol, and the communication protocol may include protocols of the IEEE 802.11 series.

[0094] The AP provided in the embodiments of the present application may be a device with wireless communication functions, supporting communication using the WLAN protocol, and having the function of communicating with other devices (such as STA or other APs) in the WLAN network. Of course, it may also have the function of communicating with other devices. Alternatively, the AP is equivalent to a bridge connecting a wired network and a wireless network. Its main role is to connect various wireless network clients together and then connect the wireless network to the Ethernet. In the WLAN system, the access point may be referred to as an access point station AP STA. The device with wireless communication functions may be a whole-device device, or may also be a chip or a processing system installed in the whole-device device. The device installed with these chips or processing systems may, under the control of the chip or the processing system, implement the methods and functions of the embodiments of the present application. The AP in the embodiments of the present application is a device that provides services for the STA and may support 802.11 series protocols. For example, the access point may be an access point for a terminal device (such as a mobile phone) to enter a wired (or wireless) network, and may be deployed at home, inside a building, and inside a park, or may also be deployed outdoors. For another example, the AP may be a communication entity such as a communication server, a router, a switch, a bridge, etc.; the AP may include various forms of macro base stations, micro base stations, relay stations, etc., or the AP may be a chip and a processing system in these various forms of devices, so as to implement the methods and functions of the embodiments of the present application. The access point in the present application may be a high efficient (HE) AP or an extremely high throughput EHT AP, or may also be an access point applicable to future Wi-Fi protocols.

[0095] The STA provided by the embodiments of the present application is a device with wireless communication capabilities, supporting communication using the WLAN protocol and having the ability to communicate with other stations or access points in the WLAN network. In a WLAN system, the STA can be referred to as a non-access point station (non-AP STA). For example, the STA can communicate with other devices in the WLAN by communicating with the AP. The device with wireless communication capabilities can be a complete device, or can also be a chip or processing system installed in the complete device. The device installed with these chips or processing systems can implement the methods and functions of the embodiments of the present application under the control of the chips or processing systems. For example, the station can be a wireless communication chip, a wireless sensor, or a wireless communication terminal, etc., and can also be referred to as a user. For another example, the station can be a mobile phone supporting Wi-Fi communication function, a tablet computer supporting Wi-Fi communication function, a set-top box supporting Wi-Fi communication function, a smart TV supporting Wi-Fi communication function, a smart wearable device supporting Wi-Fi communication function, a vehicle-mounted communication device supporting Wi-Fi communication function, and a computer supporting Wi-Fi communication function, etc.

[0096] The above-mentioned AP or STA may include a transmitter, a receiver, a memory, a processor, etc. Among them, the transmitter and the receiver are respectively used for sending and receiving packet structures, the memory is used for storing signaling information and storing preset values agreed in advance, etc., and the processor is used for parsing signaling information, processing relevant data, etc.

[0097] With the continuous evolution of WLAN application scenarios, the WLAN system will be applied to more scenarios or industries. For example, it will be applied to the Internet of Things industry and the vehicle-to-everything industry. Devices supporting WLAN communication (such as AP or STA) can be sensor nodes in a smart city (such as smart water meters, smart electricity meters, smart air detection nodes), smart devices in a smart home (such as smart cameras, projectors, displays, televisions, speakers, refrigerators, washing machines, etc.), Internet of Things nodes, sensors, etc. in the Internet of Things, entertainment terminals (such as AR, VR and other wearable devices), smart devices in smart offices (such as printers, projectors, loudspeakers, speakers, etc.), infrastructure in daily life scenarios (such as vending machines, self-service navigation stations in shopping malls, self-service cashiers, self-service ordering machines, etc.), and devices in large sports and music venues. In the embodiments of the present application, the specific forms of the STA and the AP are not limited, and only exemplary descriptions are provided here.

[0098] To address the high - complexity development trend of supporting wireless networks, it is considered to apply artificial intelligence (AI) technology to assist in the management of wireless networks. How to implement the application of AI technology in wireless networks to improve communication performance has become a current research hotspot. This application proposes that the first node can send the first training data containing the network experience information of the STA to the second node. Through the transmission of the first training data in the network, the node maintaining the first intelligent model (referred to as the intelligent node) can perform model training of the first intelligent model based on the first training data, enabling the first intelligent model to be applied to the wireless network, inferring a transmission strategy to improve the network experience of the STA, and achieving an improvement in network communication performance.

[0099] Figure 2 It is a schematic flowchart of the communication method 200 provided by an embodiment of this application. The method 200 includes but is not limited to the following S201 and S202.

[0100] S201, the first node determines the first training data, and the first training data includes the network experience information of multiple STAs.

[0101] Among them, S201 is an optional step.

[0102] For example, the first node can collect the network experience information of multiple STAs and determine the first training data according to the network experience information of the multiple STAs.

[0103] The network experience information of the STA can indicate the experience evaluation of the STA for the communication network, such as reflecting the satisfaction of the STA with the service quality (and / or network quality). For example, the network experience information can indicate one of the multiple candidate evaluation levels of the STA for the service quality (and / or network quality). Exemplarily, the multiple candidate evaluation levels can include high, medium, and low. Or the multiple candidate evaluation levels can include excellent, good, average, and poor. This application does not make a limitation in this regard. Again, for example, the network experience information can indicate the gap between the service quality (and / or network quality) of the STA and the expected quality, or the satisfaction of the STA with the service quality (and / or network quality) and the expected quality, etc. The network experience information of the STA can also indicate the transmission delay, bit error rate, packet loss rate, throughput, etc. of the current service data. It should be understood that this application does not make a limitation on the specific name of the network experience information, and the network experience information can also be referred to as network evaluation information, quality of experience (QoE) information, or quality of service (QoS) information, etc.

[0104] S202, the first node sends the first training data to the second node. The first training data is used for training the first intelligent model, and the first intelligent model is used to infer the transmission strategy of the communication network where the first node is located.

[0105] After determining the first training data, the first node can send the first training data to the second node, so that the first training data can be transmitted to the intelligent node. The intelligent node is the node that maintains the first intelligent model. For example, the intelligent node can be a server, or a node in the network with AI capabilities (such as having a model training module and a model inference module). The intelligent node can train the first intelligent model based on the first training data, and the trained first intelligent model can infer the transmission strategy of the communication network where the first node is located.

[0106] In one implementation, the first node can be the first STA, and the second node can be the intelligent node. The first STA has the ability to access the intelligent node. The first STA can collect the first training data containing the network experience information of multiple STAs from the AP connected to the first STA, and then send it to the intelligent node, so that the intelligent node can perform model training on the first intelligent model based on the first training data.

[0107] In another implementation, the first node can be the AP, and the second node can be the first STA with the ability to access the intelligent node. The AP can collect network experience information from multiple STAs established communication connections with it, determine the first training data, and send the first training data to the first STA, so that the STA can send the first training data to the intelligent node, enabling the intelligent node to perform model training on the first intelligent model based on the first training data.

[0108] In yet another implementation, the first node can be the AP, and the second node can be the intelligent node. The AP has the ability to access the intelligent node. The AP can collect network experience information from multiple STAs established communication connections with it, determine the first training data, and send the first training data to the intelligent node, so that the intelligent node can perform model training on the first intelligent model based on the first training data.

[0109] Specifically, when the intelligent node performs model training on the first intelligent model, it can use the network experience information of these multiple STAs, network status information, and transmission strategy information of each STA in the multiple STAs as training data. Through model training, the trained first intelligent model is obtained. That is, the intelligent node updates the model parameters of the first intelligent model through model training based on the training data. The first intelligent model with updated model parameters is the trained first intelligent model, enabling the first intelligent model to infer the transmission strategy that conforms to the network status based on the network status information.

[0110] The network status information is used to indicate the communication network status corresponding to the network experience information of the STA. The network experience information of the STA specifically indicates the STA's evaluation of the communication network experience in this communication network status. The network status information may include, but is not limited to, one or more of the number of active devices in the communication network where the first node is located, communication throughput, transmission efficiency, bit error rate, packet loss rate, bandwidth, latency, number of transmitted / received packets, received energy, etc. The network status may be one or more of the measurement quantities defined in the IEEE 802.11k standard specification.

[0111] The transmission policy information of the STA can be used to indicate the transmission policy corresponding to the network experience information of the STA. The network experience information of the STA specifically indicates the STA's evaluation of the network experience when communicating using this transmission policy in the communication network. The transmission policy information may indicate one or more of the number of data streams (or data layers) when the STA communicates, the coding method of the channel coding used, the code rate of the channel coding, the modulation order, the bandwidth, and channel aggregation / binding.

[0112] Optionally, the first training data may further include the above network status information and / or the above STA's transmission policy information. Alternatively, the intelligent node may obtain the above network status information and / or the above STA's transmission policy information from other nodes.

[0113] The first training data further includes collection time information, which is used to indicate the collection time of other information in the first training data (i.e., information other than the collection time information). The intelligent node can determine the collection time of other information in the first training data according to this collection time information. For example, if the first training data does not include network status information and / or the STA's transmission policy information, and the first training data includes information indicating the collection time of the network experience information of the STA, the intelligent node can determine the network status information and the STA's transmission policy information at the corresponding time according to this collection time, and obtain the network status information and the STA's transmission policy information corresponding to the network experience information of the STA, so as to perform model training of the first intelligent model.

[0114] In one implementation, the first intelligent model may be a large language model (LLM).

[0115] Different from other AI technologies, large language models (LLMs) exhibit the emergence phenomenon. After training the basic model, by introducing domain knowledge and fine-tuning some parameters, LLMs can be widely applied in various industries. The model training of LLMs includes three training processes. First is the pre-training process. Second is the fine-tuning process based on domain knowledge to support downstream tasks. Finally, there is the refined fine-tuning process based on feedback, which can further align with downstream tasks during use. The pre-training process does not involve domain knowledge and is a basic model trained based on big data combined with context. Although the second training process is a fine-tuning process, it also belongs to the pre-training category. Only domain knowledge is used during training to enable the basic model to be better applied to downstream tasks. In current wireless systems, there are a large number of rule-based solutions, and these solutions or rules are obtained through a large number of simulations and expert experience. Therefore, these rules can be used as data for the second training process. Of course, the second training process can also be executed using the method of communication network simulation.

[0116] After the second training process, the LLM already has basic wireless domain knowledge and can infer transmission strategies based on the communication network state. However, the transmission strategies inferred at this time may not necessarily align with the preferences / experiences of the STAs. Therefore, for the first intelligent model obtained after the second training process, the intelligent node can obtain first training data including the network experience information of multiple STAs from the first node, thereby realizing the fine-tuning of the model parameters of the first intelligent model through model training, so that the first intelligent model can output transmission strategies that conform to the network state and the preferences / experiences of the STAs. The first intelligent model obtained after the second training process can be pre-configured in the intelligent node or obtained by the intelligent node through model training based on simulation. This application does not make any limitations in this regard.

[0117] Exemplarily, based on the first training data, the intelligent node uses the reinforcement learning from human feedback (RLHF) algorithm or the direct preference optimization (DPO) algorithm to perform model training on the first intelligent model. The intelligent node can also use other algorithms. This application does not make any limitations on the algorithms used by the intelligent node to perform model training on the first intelligent model.

[0118] The first intelligent model can also be other AI models, such as other neural network models. This application does not make any limitations in this regard.

[0119] According to the above solution, the intelligent node can obtain the first training data from the first node, enabling the intelligent node to train the first intelligent model based on the network experience information of multiple STAs obtained. By training the first intelligent model with reference to the network experience information of the STAs, the first intelligent model can be used for the inference of transmission policies to obtain transmission policies with a relatively high degree of matching with the network state, thereby improving the communication performance of the network and meeting the transmission requirements of the STAs.

[0120] Figure 3 It is a schematic flowchart of the communication method 300 provided by an embodiment of the present application. In this method 300, the first STA is a STA with the ability to access the intelligent node. The method 300 includes but is not limited to the following S301 to S305:

[0121] S301, the first STA sends a first request message to the AP, and this first request message is used to request the first training data.

[0122] Correspondingly, the AP receives this first request message from the first STA and determines that the first STA requests the first training data.

[0123] This first request message can be carried in a wireless frame sent by the first STA to the AP. This wireless frame can be a management frame, and this management frame can be a management frame in a currently defined frame format, such as the first request message multiplexing one of the indication fields or reserved fields. Or, this management frame can be a management frame in a frame format designed for the first request message. The present application does not make any limitations in this regard.

[0124] S302, the AP determines the first training data, and this first training data includes the network experience information of multiple STAs.

[0125] Among them, S302 is an optional step.

[0126] The manner in which the AP determines the first training data can include but is not limited to the following two implementation manners, which will be introduced separately below.

[0127] Implementation manner one, the AP receives the first information from multiple STAs, and each first information includes the network experience information of the corresponding STA.

[0128] In order to determine the first training data, the AP can send a second information to multiple STAs that have established a communication connection with the AP. This second information is used to request the first information, and the multiple STAs respond to the request of the AP and send the first information to the AP. Each first information includes the network experience information of the STA that sends this first information.

[0129] The second information may be carried in a radio frame for a status request. The first information may be carried in a radio frame for reporting a measurement report. The radio frame carrying the first information and / or the second information may be a radio frame that multiplexes a currently defined frame format or a newly defined frame format. This application does not limit this.

[0130] The AP may determine first training data according to the first information of multiple STAs. The first training data includes the network experience information of the multiple STAs. The first training data further includes the transmission policy information of the STA corresponding to the network experience information and the network status information corresponding to the network experience information.

[0131] It can be considered that the AP requests the real-time network experience information of the STA. Then this network experience information can be called the current network experience information of the STA. The transmission policy information of the STA corresponding to the network experience information can be called the current transmission policy of the STA, and the network status information can be called the current network status information.

[0132] In one example, the AP may have the ability of global network observation and be able to observe and obtain the current transmission policy information and the current network status information of the STA. After receiving the first information, the AP determines the first training data according to the network experience information of the STA in the first information and the transmission policy information and network status information of the STA observed by the AP.

[0133] In another example, the first information sent by the STA may further include the current transmission policy information and / or the current network status information of the STA. The second information sent by the AP and the multiple STAs is used not only to request the network experience information of the STA, but also to request the transmission policy information and / or the corresponding network status information corresponding to the network experience information of the STA. Then the first information sent by the multiple STAs to the AP includes the corresponding information requested by the second information, that is, the first information includes the network experience information of the STA, and the first information further includes the transmission policy information of the STA corresponding to the network experience information and / or the corresponding network experience information. The first training data includes the first information of the multiple STAs. If the first information only includes the transmission policy information (or the current network status information) of the STA, the AP combines the observed transmission policy information (or network status information) of the STA and the first information to determine the first training data.

[0134] Optionally, the first training data further includes acquisition time information, which is used to indicate the acquisition time of other information (i.e., other information other than the acquisition time information) in the first training data.

[0135] For example, the first information may include information indicating the collection time of the network experience information of the STA. The AP may determine the network status information corresponding to the collection time and / or the transmission policy information of the STA based on the collection time, and obtain the network status information and / or the transmission policy information of the STA corresponding to the network experience information of the STA. The first training data may also include the collection time information, so that after the intelligent node obtains the first training data, it can determine the time corresponding to the information in the first training data.

[0136] It should be understood that the first training data sent by the AP to the first STA does not carry the information of each STA, which can protect privacy.

[0137] In Embodiment 2, the AP may maintain a second intelligent model, and the AP may use the second intelligent model to infer the first training data. The input of the second intelligent model is the network status information and the transmission policy information of the STA, and the output is the first training data.

[0138] Exemplarily, the second intelligent model may be a reward model (RM).

[0139] After the AP receives the first request information of the first STA, the AP may input the current network status information and the current transmission policy information of multiple STAs into the second intelligent model, and the network experience information of multiple STAs is inferred by the second intelligent model. The network experience information of multiple STAs is specifically the network experience information of multiple STAs corresponding to the current network status information and the current transmission policy information of multiple STAs inferred by the second intelligent model.

[0140] The AP may determine the first training data according to the output of the second intelligent model. In one implementation, the output of the second intelligent model is the network experience information of multiple STAs inferred by the second intelligent model. The AP determines the first training data according to the output of the second intelligent model, the current network status information, and the current transmission policy information of multiple STAs. The first training data includes the network experience information of multiple STAs, the current network status information, and the current transmission policy information of multiple STAs. In another implementation, the second intelligent model outputs the first training data, that is, the second intelligent model not only outputs the network experience information of multiple STAs inferred by the second intelligent model, but also outputs the input information of the second intelligent model, that is, the current network status information and the transmission policy information of multiple STAs. The output of the second intelligent model obtained by the AP is the first training data.

[0141] In the second embodiment, the AP can infer the network experience information of the STA based on the second intelligent model, without collecting the network experience information from multiple STAs with which a connection is established every time the first STA requests the first training data, which can reduce the overhead of wireless resources and improve the efficiency of obtaining the first training data.

[0142] The second intelligent model can be pre-configured in the AP. For example, after a node (such as a server, etc.) in the network executes model training to obtain the second intelligent model, the second intelligent model can be configured for the AP. Alternatively, the second intelligent model can be obtained by the AP executing model training. The AP can execute model training. The AP can first collect the training data for the model training of the second intelligent model. For example, the AP can be based on network observations or obtain the network status information and the transmission policy information of the STA, and the AP can collect the network experience information of the STA corresponding to the network status information and the transmission policy information from the STAs with which it has established a communication connection. The AP can use the network status information and the transmission policy information of the STA as the training data, and use the network experience information of the STA as the verification information for model training, execute model training, and obtain the trained second intelligent model, so that the second intelligent model can infer the corresponding network experience information based on the input network status information and transmission policy information. After the AP obtains the second intelligent model through model training, it can use the network status information and the transmission policy information of the STA as the input of the second intelligent model, and obtain the network experience information of the STA inferred by the second intelligent model, without collecting the network experience information from multiple STAs with which a connection is established every time the first STA requests the first training data, which can reduce the overhead of wireless resources and improve the efficiency of obtaining the first training data.

[0143] S303. The AP sends the first training data to the first STA.

[0144] Correspondingly, the first STA receives the first training data from the AP.

[0145] S304. The first STA sends the first training data to the intelligent node.

[0146] After obtaining the first training data from the AP, the first STA can send the first training data to the intelligent node, so that the intelligent node can execute the model training of the first intelligent model based on the first training data.

[0147] S305. The intelligent node executes model training on the first intelligent model according to the first training data, and obtains the trained first intelligent model, which is used to infer the transmission policy of the communication network where the first node is located.

[0148] After the intelligent node obtains the first training data from the first STA, it performs model training. For example, it can use the RLHF algorithm or the DPO algorithm to perform model training. The first intelligent model with updated model parameters after training is the trained first intelligent model. This first intelligent model can be used to infer the transmission strategy of the communication network where the first node is located.

[0149] The first STA can request transmission strategy information from the intelligent node. The intelligent node can use the transmission strategy information inferred by the first intelligent model and feedback it to the first STA. Specifically, the first STA can send a second request message to the intelligent node. This second request message is used to request the transmission strategy information corresponding to the current network state information. After receiving this second request message, the intelligent node can input the current network state information into this first intelligent model, obtain the first transmission strategy information output by this first intelligent model, and send a response message to the first STA. This response message is used to indicate the first transmission strategy information.

[0150] The second request message sent by the first STA to the intelligent node may include the current network state information, or the intelligent node can obtain the current network state information in other ways. This application does not make any limitations in this regard.

[0151] In one implementation, the STA connected to the AP can report that the current network experience does not meet the requirements to the AP when the current network experience does not meet the requirements. When the number of STAs whose network experience does not meet the requirements is greater than or equal to the number threshold, the AP notifies the first STA. The first STA sends a second request message to the intelligent node to request the transmission strategy information suitable for the current network state information from the intelligent node. The intelligent node inputs the current network state information into the first intelligent model, and the first intelligent model infers the transmission strategy information corresponding to the current network state information and feedbacks it to the first STA. The first STA can send this transmission strategy information to the AP. The AP determines the transmission strategies of the AP and / or each STA according to this transmission strategy information. Specifically, the AP can directly adopt the transmission strategy indicated by the transmission strategy information fed back by the intelligent node, or the AP uses the transmission strategy information fed back by the intelligent node as a reference for the AP, and the AP makes overall adjustments and decisions to obtain the transmission strategies of the AP and / or each STA. This enables the AP and / or STA to adopt transmission strategies that match the current network state to meet the transmission requirements of the STA, improve the network experience of the STA, and thus improve the network communication performance.

[0152] In another implementation, when the network experience of the first STA does not meet the requirements, the first STA may send third request information to the AP. The third request information is used to request the current network status information. In response to the request of the first STA, the AP sends the current network status information to the first STA. After the first STA obtains the current network status information from the AP, the first STA sends second request information to the intelligent node to request the transmission policy information of the first STA corresponding to the current network status information. The second request information includes the current network status information obtained by the first STA from the AP. After receiving the second request information, the intelligent node may use the current network status information in the second request information as the input of the first intelligent model, and through inference of the first intelligent model, obtain the first transmission policy information. The first transmission policy is the transmission policy information of the first STA corresponding to the current network status. The intelligent node indicates the first transmission policy information through a response message, so that the first STA obtains the first transmission policy information through the response message. The first STA may perform data transmission according to the first transmission policy information. This enables the first STA to adopt a transmission policy that matches the current network status for data transmission, so as to meet the transmission requirements of the first STA, improve the network experience of the first STA, and further improve the network communication performance.

[0153] Optionally, the second request information sent by the first STA specifically includes the current network status information represented by natural language, and / or, the response information sent by the intelligent node specifically includes the first transmission policy information represented by natural language.

[0154] Specifically, the first STA may process the current network status information in a natural language representation manner to obtain the current network status information represented by natural language, and send it to the intelligent node through the second request information. After receiving the second request information, the intelligent node may parse the current network status information represented by natural language based on the natural language representation manner to obtain the current network status information. Similarly, the intelligent node may process the first transmission policy information in a natural language representation manner to obtain the first transmission policy information represented by natural language, and send it to the first STA through the response information. After receiving the response information, the first STA may parse the first transmission policy information represented by natural language based on the natural language representation manner to obtain the first transmission policy information. Optionally, the first STA / intelligent node may perform the conversion between information and information represented by natural language based on a model with natural language representation ability.

[0155] The first STA and the intelligent node can avoid the way of reaching a consensus on the information set by the pre-defined information set of the protocol (such as the network status information set and the transmission policy information set) through the natural language representation method, and select the corresponding information from the set for interaction. It can reduce the storage overhead of both parties in the interaction caused by storing the pre-defined information set, and the information type is not limited to the set, so the scalability is good.

[0156] In another implementation, the AP can have the ability to access the intelligent node. Then the AP can provide the first training data for the intelligent node. For example, the AP can determine the first training data and send the first training data to the intelligent node. The intelligent node can perform the model training of the first intelligent model according to the first training data from the AP to obtain the trained first intelligent model. Among them, the way for the AP to determine the first training data can refer to the introduction in S302 above, and the way for the intelligent node to perform model training can refer to the introduction in S305 above, which will not be elaborated here.

[0157] According to the above solution, the intelligent node can obtain the network experience information of the STA as the training data of the first intelligent model from the first node, so that the intelligent node can complete the model training of the first intelligent model based on the network experience information of the STA, so that the first intelligent model can infer the transmission policy information matching the current network state based on the current network state information, so that the STA can apply the transmission policy information to match the current network state, thereby meeting the transmission requirements of the STA, improving the network experience of the STA, and improving the network communication performance.

[0158] It can be understood that in order to implement the functions in the above embodiments, the first node, the second node, the STA, the AP, and the intelligent node include the corresponding hardware structures and / or software modules for performing various functions. Those skilled in the art should easily realize that, combined with the units and method steps of each example described in the embodiments disclosed in this application, this application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the way of hardware or computer software driving hardware depends on the specific application scenario and design constraints of the technical solution.

[0159] Figure 4 and Figure 5 FIG. is a schematic structural diagram of a possible communication device provided by an embodiment of the present application. These communication devices can be used to implement the functions of the first node or the second node in the above method embodiments, and thus can also achieve the beneficial effects possessed by the above method embodiments. In the embodiments of the present application, the communication device can be, for example, Figure 1 the STA shown in Figure 1The AP shown can also be an intelligent node for maintaining the first intelligent model, or a module (such as a chip or a chip system) applied to each node.

[0160] The communication device 400 includes a transceiver unit 420, which can be used to receive or send information. The communication device 400 may further include a processing unit 410, which can be used to process instructions or data to implement corresponding operations.

[0161] It should be understood that when the communication device 400 is a chip configured in (or for) a communication device, the transceiver unit 420 in the communication device 400 can be the input / output interface or circuit of the chip, and the processing unit 410 in the communication device 400 can be the processor in the chip.

[0162] Optionally, the communication device 400 may further include a storage unit 430, which can be used to store instructions or data. The processing unit 410 can execute the instructions or data stored in the storage unit to enable the communication device to implement corresponding operations.

[0163] The communication device 400 can be used to implement the functions of the first node or the second node in the method embodiments shown above. Figure 2 in the method embodiments shown above.

[0164] When the communication device 400 is used to implement Figure 2 the function of the terminal in the method embodiments shown: The processing unit 410 is used to determine first training data, which includes network experience information of multiple stations STA. The transceiver unit 420 is used to send the first training data to the second node. The first training data is used for model training of the first intelligent model, and the first intelligent model is used to infer the transmission strategy of the communication network where the communication device is located.

[0165] When the communication device 400 is used to implement Figure 2 the function of the network device in the method embodiments shown: The transceiver unit 420 is used to receive the first training data from the first node, and the first training data includes network experience information of the STA. The processing unit 410 is used to perform model training on the first intelligent model according to the first training data to obtain the trained first intelligent model, and the first intelligent model is used to infer the transmission strategy in the communication network where the first node is located.

[0166] For a more detailed description of the above processing unit 410 and transceiver unit 420, reference can be made to Figure 2 the relevant descriptions in the method embodiments shown.

[0167] It should be understood that the transceiver unit 420 in the communication device 400 can be implemented through a communication interface (such as a transceiver, a transceiver circuit, an input / output interface, or pins, etc.). When the communication interface is a transceiver, the transceiver can be composed of a receiver and / or a transmitter. The processing unit 410 in the communication device 400 can be implemented through at least one processor, and the processing unit 410 in the communication device 400 can also be implemented through at least one logic circuit. Optionally, the communication device 400 further includes a storage unit, and the storage unit can be implemented by a memory.

[0168] As Figure 5 shown, the communication device 500 includes a processor 510 and an interface circuit 520. The processor 510 and the interface circuit 520 are coupled to each other. It can be understood that the interface circuit 520 can be a transceiver or an input / output interface. Optionally, the communication device 500 can further include a memory 530 for storing instructions executed by the processor 510 or storing input data required for the processor 510 to run instructions or storing data generated after the processor 510 runs instructions.

[0169] In one implementation, the memory 530 can also be integrated in the processor 510 or be independent of the processor 510.

[0170] When the communication device 500 is used to implement Figure 2 the method shown, the processor 510 is used to implement the functions of the above-mentioned processing unit 410, and the interface circuit 520 is used to implement the functions of the above-mentioned transceiver unit 420.

[0171] When the above communication device is a chip applied to a STA, the STA chip can implement the functions of the STA in the above method embodiments. The STA chip receives information from other modules in the STA (such as a radio frequency module or an antenna), and this information is sent by the AP to the STA; or, the STA chip sends information to other modules in the STA (such as a radio frequency module or an antenna), and this information is sent by the STA to the AP.

[0172] When the above communication device is a module applied to an AP, the AP module can implement the functions of the AP in the above method embodiments. The AP module receives information from other modules in the AP (such as a radio frequency module or an antenna), and this information is sent by the STA to the AP; or, the AP module sends information to other modules in the AP (such as a radio frequency module or an antenna), and this information is sent by the AP to the STA. Here, the AP module can be the baseband chip of the AP, or a DU or other modules, and here the DU can be a DU under the open radio access network (O-RAN) architecture.

[0173] It can be understood that the processor in the embodiments of the present application may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0174] The method steps in the embodiments of the present application may be implemented in hardware or in software instructions executable by a processor. The software instructions may be composed of corresponding software modules, and the software modules may be stored in a random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disk, removable hard disk, CD-ROM, or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. The storage medium may also be a component of the processor. The processor and the storage medium may be located in an ASIC. Additionally, the ASIC may be located in an access network device or a terminal device. The processor and the storage medium may also exist as discrete components in the access network device or the terminal device.

[0175] According to the method provided by the embodiments of the application, the embodiments of the present application also provide a computer program product, which includes: computer program code, when the computer program code is executed by one or more processors, it causes a device including the processor to execute Figure 2 the method of the illustrated embodiment.

[0176] In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are executed in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user device, or other programmable devices.

[0177] According to the method provided by the embodiments of the present application, the embodiments of the present application further provide a computer-readable storage medium storing the above computer program or instructions. When the computer program or instructions are run by one or more processors, a device including the processor is caused to execute Figure 2 the method of the embodiment shown.

[0178] For example, the computer program or instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer program or instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center integrating one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video disc; or it can be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium can be a volatile or non-volatile storage medium, or can include both volatile and non-volatile types of storage media.

[0179] According to the method provided by the embodiments of the present application, the embodiments of the present application further provide a communication system including one or more of the foregoing terminals. The system can further include one or more of the foregoing network devices.

[0180] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the devices described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in an electrical, mechanical, or other form.

[0181] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of this solution.

[0182] In various embodiments of the present application, without special instructions and logical conflicts, the terms and / or descriptions between different embodiments are consistent and can be cross-referenced to each other. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0183] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claimed rights.

Claims

1. A communication method, characterized in that, Including: A first node determines first training data, where the first training data includes network experience information of multiple stations (STAs); The first node sends the first training data to a second node, where the first training data is used for model training of a first intelligent model, and the first intelligent model is used to infer a transmission policy of a communication network where the first node is located.

2. The method according to claim 1, wherein The first training data further includes one or more of the following: Network status information corresponding to the network experience information, transmission policy information corresponding to the network experience information, or acquisition time information, where the acquisition time information is used to indicate an acquisition time of information in the first training data.

3. The method according to claim 1 or 2, characterized in that The first node is a first STA, and the second node is an intelligent node, where the first STA is an STA capable of accessing the intelligent node, and the intelligent node is a node that maintains the first intelligent model.

4. The method according to claim 3, wherein The first node determines the first training data, including: The first STA sends first request information to an access point (AP), where the first request information is used to request the first training data; The first STA receives the first training data from the AP.

5. The method according to claim 1 or 2, characterized in that, The first node is an AP, and the second node is an intelligent node or a first STA, where the first STA is an STA capable of accessing the intelligent node, and the intelligent node is a node that maintains the first intelligent model.

6. The method according to claim 5, wherein The first node determines the first training data, including: The AP receives first request information from the first STA, where the first request information is used to request the first training data; The AP receives multiple first information from multiple STAs, and each piece of the multiple first information includes network experience information of each STA; The AP determines the first training data according to the multiple first information.

7. The method according to claim 6, wherein The first training data includes the multiple first information of the multiple STAs, and the first information further includes at least one of network status information corresponding to the network experience information of the STA, transmission policy information of the STA corresponding to the network experience information of the STA, or acquisition time information; Or, The first training data is determined by the AP according to at least one of network status information, transmission policy information of the STA, or acquisition time information and the multiple first information of the multiple STAs, where the acquisition time information is used to indicate an acquisition time of information in the first training data.

8. The method according to claim 5, characterized in that, The first node determines the first training data, including: The AP receives first request information from the first STA, where the first request information is used to request the first training data; The AP inputs network status information and transmission policy information of multiple STAs into a second intelligent model to obtain the first training data output by the second intelligent model.

9. The method according to claim 8, wherein The method further includes: The AP performs model training according to network experience information of multiple STAs, transmission policy information corresponding to the network experience information, and network status information to obtain a second intelligent model.

10. The method according to claim 8 or 9, characterized in that, The second intelligent model is a reward model. The input of the reward model is network state information and the transmission policy information of the STA, and the output of the reward model includes the network experience information of the STA.

11. The method according to any one of claims 1 to 10, characterized in that, The second node is an intelligent node, and the intelligent node is the node that maintains the first intelligent model; the method further includes: The first node sends second request information to the second node, and the second request information is used to request the transmission policy information corresponding to the current network state information. The first node receives response information from the second node, and the response information is used to indicate first transmission policy information, and the first transmission policy information is the transmission policy information corresponding to the current network state information. The first node performs data transmission according to the first transmission policy information.

12. The method according to claim 11, wherein The first node sending the second request information to the second node includes: In the case where the number of STAs whose network experience does not meet the requirements is greater than or equal to the number threshold, the first node sends the second request information to the second node.

13. The method according to claim 11 or 12, characterized in that, The first node is a first STA, and the first STA is a STA with the ability to access an intelligent node. The first node sending the second request information to the second node includes: When the network experience of the first STA does not meet the requirements, the first STA sends third request information to the AP, and the third request information is used to request the current network state information. The first STA receives the current network state information from the AP. The first STA sends the second request information to the AP, and the second request information includes the current network state information.

14. The method according to any one of claims 11 to 13, wherein the second request information includes the current network state information represented by natural language; and / or, the response information includes the first transmission policy information represented by natural language.

15. The method according to claim 14, wherein The method further includes: The first node processes the current network transmission state based on the natural language representation method to obtain the current network transmission state information represented by natural language; and / or, The first node parses the first transmission policy information represented by natural language in the response information based on the natural language representation method to obtain the first transmission policy information.

16. The method according to any one of claims 1 to 15, characterized in that, The first intelligent model is a large language model LLM. The input of the LLM is the first training data, and the output of the LLM is transmission policy information.

17. A data transmission method, characterized in that, including: The second node receives first training data from the first node, and the first training data includes the network experience information of the STA. The second node performs model training on the first intelligent model according to the first training data to obtain the trained first intelligent model, and the first intelligent model is used to infer the transmission policy in the communication network where the first node is located.

18. The method according to claim 17, wherein The first training data further includes one or more of the following: the network state information corresponding to the network experience information, the transmission policy information corresponding to the network experience information, or the collection time information Among them, the acquisition time information is used to indicate the acquisition time of the information in the first training data.

19. The method according to claim 17 or 18, characterized in that, The method further includes: The second node receives second request information from the first node, and the second request information is used to request transmission policy information corresponding to the current network state information; The second node sends response information to the first node, and the response information is used to indicate first transmission policy information, and the first transmission policy information is the transmission policy information corresponding to the current network state.

20. The method according to claim 19, wherein The method further includes: The second node inputs the current network state information into the first intelligent model to obtain the first transmission policy information inferred by the first intelligent model.

21. The method according to claim 19 or 20, characterized in that, The second request information includes the current network state information.

22. The method according to claim 21, wherein The second request information includes the current network state information represented by natural language; and / or, The response information includes the first transmission policy information represented by natural language.

23. The method according to any one of claims 17 to 22, characterized in that, The first node is an access point AP or a station STA capable of accessing an intelligent node, and the second node is an intelligent node.

24. The method according to any one of claims 17 to 23, characterized in that The first intelligent model is a large language model LLM, the input of the LLM is the first training data, and the output of the LLM is a transmission policy.

25. A communication device, characterized in that, It includes a processor, the processor is coupled with a memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the communication device executes the method according to any one of claims 1 to 16; or so that the communication device executes the method according to any one of claims 17 to 24.

26. A computer-readable storage medium, characterized in that, There are instructions that, when run on a computer, cause the computer to execute the method according to any one of claims 1 to 24.

Images