Model control and management of wireless communications
By introducing model control entities in the 5G wireless communication system to manage AI model description and deployment information, the control and management problems of AI/ML models in the 5G system are solved, the efficiency of channel state information and beam prediction is improved, and AI is realized Efficient application of models in 5G systems.
Patent Information
- Application Number
- CN202280094113.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-14
- Publication Date
- 2024-11-12
AI Technical Summary
The existing 5G wireless communication systems lack clear solutions in architecture and signaling design to control and manage artificial intelligence/machine learning models, resulting in limited application efficiency of AI/ML models in channel state information, beam prediction, etc.
By introducing model control entities in wireless communication systems, managing AI model description information and deployment information, including model functions, identifiers, input and output shapes, preprocessing and post-processing methods, using processors to execute code in computer-readable storage media Implement the storage and transmission of models to ensure effective control and management of AI models on the network side and user equipment side.
It realizes efficient control and management of AI models in 5G wireless communication systems, improves the accuracy of channel state information and beam prediction capabilities, improves the efficiency and flexibility of the system, and is suitable for various AI/ML model scenarios.
Smart Images

Figure CN118947172A_ABST
Abstract
Description
Technical Field
[0001] This document relates generally to digital wireless communications. Background Art
[0002] Mobile telecommunication technologies are driving the world towards an increasingly connected and networked society. Compared to existing wireless networks, next generation systems and wireless communication technologies will need to support a wider range of use case characteristics and provide a more complex and advanced range of access requirements and flexibility.
[0003] Long-Term Evolution (LTE) is a wireless communication standard for mobile devices and data terminals developed by the 3rd Generation Partnership Project (3GPP). LTE Advanced (LTE-A) is a wireless communication standard that enhances the LTE standard. The fifth generation wireless system, called 5G, advances the LTE and LTE-A wireless standards and is committed to supporting higher data rates, a large number of connections, ultra-low latency, high reliability, and other emerging business requirements. Summary of the invention
[0004] Techniques for controlling and / or managing models applied to wireless communication systems are disclosed, wherein the models may involve Artificial Intelligence (AI) and / or Machine Learning (ML).
[0005] A first example wireless communication method includes: sending a first message by a first wireless device to a second device located in a network requesting model information; and receiving a second message by the first wireless device in response to sending the first message, wherein the second message includes model description information, and the model description information describes one or more characteristics of a model to be used by the first wireless device.
[0006] In some embodiments, the model description information includes an identifier of the model. In some embodiments, the first message includes a field identifying the purpose for requesting the model information. In some embodiments, the first message includes any one or more of the following information: a physical cell identifier (PCI), a transmission and reception point (TRP) identifier, an absolute radio frequency channel number (ARFCN), a bandwidth of a cell, a subcarrier spacing of the cell, a number of beams, a direction of each of a plurality of beams, a number of ports, a time period of a reference signal, a subband granularity, a first expected number of opportunities for input of the model, a second expected number of opportunities for output of the model, and / or a resource mapping pattern in the frequency domain. In some embodiments, the method further includes: sending a third message requesting model deployment information to the second device in the network by the first wireless device, wherein the third message includes an identifier of the model; and in response to sending the third message, receiving a fourth message including the model deployment information, the model deployment information including one or more values corresponding to one or more parameters to be used by the model. In some embodiments, the second message includes model deployment information including one or more values corresponding to one or more parameters to be used by the model.
[0007] In some embodiments, the method further includes: receiving, by the first wireless device, a failure message in response to sending the first message, wherein the failure message indicates that the second device in the network does not have the model information. In some embodiments, the first wireless device is a base station. In some embodiments, the first wireless device is a communication device. In some embodiments, the first message is sent by the communication device, and the second message is received by the communication device using a first non-access stratum (NAS) message and a second NAS message, respectively. In some embodiments, the method further includes: sending, by the communication device, a request to obtain auxiliary data to a base station; and receiving, in response to sending the request, the auxiliary data, the auxiliary data including the line of sight direction of the reference signal when the base station sends the reference signal, and / or the beam width of the reference signal when the base station sends the reference signal.
[0008] In some embodiments, the method also includes: sending, by the communication device, a request to the base station for obtaining any one or more configurations of the following items: a physical cell identifier (PCI), a transmit and receive point (TRP) identifier, an absolute radio frequency channel number (ARFCN), a bandwidth of the cell, a subcarrier spacing of the cell, the number of beams, the direction of each of multiple beams, the number of ports, a time period of a reference signal, a subband granularity, a first expected number of opportunities for input to the model, a second expected number of opportunities for output of the model, and / or a resource mapping pattern in the frequency domain; and receiving, in response to sending the request, any one or more configurations included in the request.
[0009] A second example wireless communication method includes: receiving, by a second device located in a network, a first message requesting model information from a first wireless device; and sending, by the second device in response to receiving the first message, a second message including model description information describing one or more characteristics of a model to be used by the first wireless device.
[0010] In some embodiments, the model description information includes an identifier of the model. In some embodiments, the first message includes a field identifying the purpose for requesting the model information. In some embodiments, the first message includes any one or more of the following information: a physical cell identifier (PCI), a transmit and receive point (TRP) identifier, an absolute radio frequency channel number (ARFCN), a bandwidth of a cell, a subcarrier spacing of the cell, a number of beams, a direction of each of a plurality of beams, a number of ports, a time period of a reference signal, a subband granularity, a first expected number of opportunities for input of the model, a second expected number of opportunities for output of the model, and / or a resource mapping pattern in the frequency domain. In some embodiments, the method further includes: receiving, by the second device in the network, a third message requesting model deployment information from the first wireless device, wherein the third message includes an identifier of the model; and in response to receiving the third message, sending a fourth message including the model deployment information, the model deployment information including one or more values corresponding to one or more parameters to be used by the model.
[0011] In some embodiments, the second message includes model deployment information, the model deployment information including one or more values corresponding to one or more parameters to be used by the model. In some embodiments, the method further includes: sending, by the second device in response to receiving the first message, a failure message, wherein the failure message indicates that the second device in the network does not have the model information. In some embodiments, the first wireless device is a base station. In some embodiments, the first wireless device is a communication device.
[0012] A third example wireless communication method includes sending, by a base station, a system information message including model information, wherein the model information includes a plurality of model description information associated with a corresponding plurality of models, and wherein each model description information describes one or more characteristics of a model to be used by a communication device.
[0013] In some embodiments, one of the multiple model description information includes at least one identifier of a model. In some embodiments, the system information message is sent in a system information block (SIB); or when the communication device is in a radio resource control (RRC) connection state, the system information message is sent in an RRC message. In some embodiments, the system information message includes a configuration of multiple resources to be used by the communication device, and at least one of the multiple resources is mapped to at least one of the multiple models. In some embodiments, the multiple resources include multiple physical uplink control channel (PUCCH) resources or multiple random access channels (RACH). In some embodiments, the method further includes: sending auxiliary data to the communication device, wherein the auxiliary data includes the line of sight direction of the reference signal when the base station sends the reference signal, and / or the beam width of the reference signal when the base station sends the reference signal.
[0014] In some embodiments, the at least one identifier comprises any of: a first identifier associated with an encoder of the model, a second identifier associated with a decoder of the model, or the first identifier and the second identifier. In some embodiments, the method further comprises: receiving, by the base station, a message from the communication device, the message comprising an identifier associated with an encoder of the model used by the communication device.
[0015] A fourth example wireless communication method includes: receiving, by a communication device, a system information message including model information from a base station, wherein the model information includes a plurality of model description information associated with a corresponding plurality of models, and wherein each model description information describes one or more characteristics of a model to be used by the communication device.
[0016] In some embodiments, one of the plurality of model description information includes at least one identifier of a model. In some embodiments, the system information message is received in a system information block (SIB); or when the communication device is in a radio resource control (RRC) connected state, the system information message is received in an RRC message. In some embodiments, the system information message includes a configuration of a plurality of resources to be used by the communication device, and at least one of the plurality of resources is mapped to at least one of the plurality of models. In some embodiments, the plurality of resources include a plurality of physical uplink control channel (PUCCH) resources or a plurality of random access channels (RACH).
[0017] In some embodiments, the method further comprises: receiving auxiliary data, wherein the auxiliary data comprises the boresight direction of the reference signal when the base station transmits the reference signal, and / or the beam width of the reference signal when the base station transmits the reference signal. In some embodiments, the at least one identifier comprises any one of: a first identifier associated with an encoder of the model, a second identifier associated with a decoder of the model, or the first identifier and the second identifier. In some embodiments, the method further comprises: sending a message by the communication device to the base station, the message comprising an identifier associated with an encoder of the model used by the communication device.
[0018] In another exemplary aspect, the above method is implemented in the form of processor executable code and stored in a non-transitory computer readable storage medium. When the code included in the computer readable storage medium is executed by a processor, it causes the processor to implement the method described in this patent document.
[0019] In yet another exemplary embodiment, a device is disclosed, the device being configured or operable to perform the above method.
[0020] The above and other aspects and embodiments thereof are described in more detail in the drawings, the description and the claims. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Figure 1 A block diagram of a communication system including a plurality of communication devices is shown.
[0022] Figure 2An example flow chart for requesting model information is shown.
[0023] Figure 3 An example flow chart for sending model information is shown.
[0024] Figure 4A Another example flow chart for sending model information is shown.
[0025] Figure 4B An example flow chart for receiving model information is shown.
[0026] Figure 5 An exemplary block diagram of a hardware platform that may be part of a network device or a communication device is shown.
[0027] Figure 6 An example of wireless communication including a base station (BS) and a user equipment (UE) based on some embodiments of the disclosed technology is shown. DETAILED DESCRIPTION
[0028] I. Introduction
[0029] Artificial intelligence / machine learning has been studied and used in various fields. There are also some studies to improve the efficiency of wireless communication systems (especially the physical layer). For example, AI / ML models can be used to improve the accuracy of channel state information (CSI). In addition, AI / ML models can predict channel beam information in both spatial and time domains. In addition, positioning, channel estimation, energy saving, and mobility management are some other use cases. However, there is no clear solution on how to control and manage AI models in the architecture and signaling design of current 5G wireless communication systems. This patent document proposes some technical solutions for controlling and managing AI / ML models applied to wireless communication systems.
[0030] For ease of discussion, the following terms are given with some general descriptions:
[0031]
[0032]
[0033] The example titles of the following sections are used to facilitate understanding of the disclosed subject matter and do not limit the scope of the claimed subject matter in any way. Therefore, one or more features of one example section may be combined with one or more features of another example section. In addition, for clarity of explanation, 5G terminology is used, but the technology disclosed in this document is not limited to 5G or 5G advanced technology and can be used in wireless systems that implement other protocols. In addition, the AI / ML model is an exemplary scenario in which the technical solutions described in this patent document can be generalized to or applied to any model that determines the relationship between input and output.
[0034] II. Example Technical Solution
[0035] First, AI model information may include or may refer to AI model description information and / or AI model deployment information:
[0036] ○ The AI model description information may include information describing the characteristics of the AI model to be used, wherein the characteristics of the AI model included in the AI model description information may include any one or more of the following:
[0037] ■ Model function / purpose (e.g. it could be one of CSI compression, beam prediction in the spatial domain, or UE positioning)
[0038] ■Model identifier (ID) and / or version ID
[0039] For two-sided models, the Model ID can include any of the following:
[0040] ○ The first ID associated with the encoder of the AI model (or the generated model ID),
[0041] ○ A second ID associated with the decoder of the AI model (or the reconstructed model ID),
[0042] ○First ID and second ID.
[0043] Two AI models with the same model ID can have different version IDs, which can mean at least one of the following:
[0044] ○ Two AI models have the same AI model structure but different AI model parameters,
[0045] ○ One AI model is updated / fine-tuned from another AI model, so the two AI models may have some common parts in the AI model structure, or the two AI models may have some common parts in the AI model structure and AI model parameters.
[0046] ■Preprocessing of AI model input
[0047] ●For example, normalization methods for data used as input to AI models.
[0048] ○ For example, data should be normalized to the maximum value of the data before feeding it into the AI model input.
[0049] AI model input type
[0050] • For example, channel measurements and / or assistance data.
[0051] ○ For example, the channel measurement result may be a channel impulse response in the time domain / a channel frequency response in the frequency domain / a reference signal received power (RSRP).
[0052] ○ For example, the auxiliary data may be the boresight direction of the reference signal when the gNB sends the reference signal, and / or the beamwidth of the reference signal when the gNB sends the reference signal (e.g., 3 decibel (dB) beamwidth).
[0053] AI model input shape
[0054] For example, the number of dimensions and the size of each dimension of the data used as input to the AI model.
[0055] ■AI model input sequence
[0056] For example, the order of different dimensions to include in the data used for AI model input.
[0057] The following examples explain how to understand AI model input type, AI model input shape, and AI input order:
[0058] ○ The data includes channel measurement results (i.e., AI model input type), which has the following three dimensions (i.e., the number of dimensions and the size of each dimension). Now, the data is a three-dimensional matrix. Therefore, the AI model description information should include how to construct the three-dimensional matrix (or sort the three-dimensional matrix). An example can be: the spatial domain, frequency domain, and time domain correspond to the first dimension, second dimension, and third dimension (i.e., the order of different dimensions) of the data, respectively.
[0059] The number (or size) of ports in the airspace,
[0060] ■The number (or size) of frequency cells in the frequency domain,
[0061] ■ The number (or size) of opportunities in the time domain.
[0062] ●For example, quantization methods used for AI model inputs, which will use a limited number of bits to represent each element of the data.
[0063] AI model output type
[0064] ●For example, UE location, and different types of measurement results (e.g., RSRP, timing information), etc.
[0065] AI model output shape
[0066] ●For example, the number of dimensions output by the AI model and the size of each dimension.
[0067] ■AI model output sequence
[0068] ●For example, the order of different dimensions to include in the AI model output.
[0069] Post-processing of AI model output
[0070] ●For example, a quantization method for AI model output that uses a limited number of bits to represent each element of the AI model output.
[0071] AI model inference latency
[0072] ●For example, the time / delay that an AI model inference entity may spend performing AI model inference.
[0073] ■Applicable scenarios
[0074] ●For example, used to indicate which physical cell and / or carrier frequency the AI model can only be applied to.
[0075] ○ The AI model deployment information may include one or more values corresponding to one or more parameters to be used by the AI model. The one or more values of the AI model deployment information may be related to the model structure and / or model parameters. The AI model deployment information includes any one or more of the following:
[0076] ■AI model structure, which may include:
[0077] The number of layers,
[0078] ● The network at each layer (e.g., fully connected neural network or convolutional neural network),
[0079] ●The number of neurons in each layer,
[0080] ●The activation function used in each layer,
[0081] ●The order of each layer.
[0082] AI model parameters (e.g., the values / weights of each neuron in the AI model)
[0083] ■For example, the AI model deployment information may be a compiled file (e.g., a runtime binary image) that can be executed by an AI model inference entity (e.g., a UE or a network).
[0084] ■In another example, the AI model deployment information may be a compiled file, which may be an interpretable file that can be further compiled by the AI model inference entity for execution, wherein the interpretable file may use a unified representation format that can be interpreted by different AI model inference entities:
[0085] ■The AI model inference entity can update the AI model structure and / or AI model parameters of the AI model.
[0086] ■The AI model inference entity can receive updates to the AI model structure and / or AI model parameters of the traditional AI model.
[0087] In the following sections, model control and management will be discussed and proposed based on which entity will perform AI model inference, including:
[0088] ○Network side model,
[0089] ○UE side model,
[0090] ○ Bilateral model.
[0091] Figure 1 A block diagram of a communication system including multiple communication devices is shown. Figure 1 As shown, a user equipment (UE) that communicates with a base station (e.g., a gNB or a next generation radio access network (NG-RAN) node) and with a network device (e.g., a core network entity or a cloud server). The network side device (e.g., a core network entity) can also communicate with the base station. For ease of discussion, a model control entity is defined, which is responsible for model storage, transmission of AI model information, or data collection, etc. The model control entity can reside at a gNB, a core network entity, or a cloud server.
[0092] III. Case 1: Network side model
[0093] In this case, the gNB / NG-RAN node determines the AI model inference and the model control entity resides at the core network entity. In this case, the gNB / NG-RAN node can be considered as the first wireless device and the second device can be a device residing in the core network entity. The gNB can send a request for AI model information to the model control entity.
[0094] ● In some embodiments, the request may include information indicating the function / purpose of the requested AI model information. For example, for beam prediction, CSI prediction, or Line of Sight (LoS) / Non Line of Sight (NLoS) identification.
[0095] In some embodiments, the request may also include any one or more of the following information (which the model control entity may use to decide which AI model the gNB requires):
[0096] ○ Physical Cell ID (PCI),
[0097] ○ Transmit and Receive Point (TRP) ID,
[0098] ○ Absolute Radio Frequency Channel Number (ARFCN),
[0099] ○The bandwidth of the cell,
[0100] ○Subcarrier spacing of the cell,
[0101] o The number of beams (e.g., the number of CSI-RS resources in a CSI-RS resource set),
[0102] o the direction of each beam (for example, the direction may be the boresight direction of the beam),
[0103] o The number of ports (e.g., the number of ports used for CSI-RS resources),
[0104] ○The time period of the reference signal,
[0105] o subband granularity (e.g., the number of resource blocks included in a subband),
[0106] ○ The expected number of opportunities for AI model input (which can be used for beam prediction in time domain or CSI prediction in time domain),
[0107] ○ The expected number of opportunities for the AI model output (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0108] o Resource mapping pattern in the frequency domain (eg, used to indicate which resource elements will transmit a reference signal, which can be used for channel estimation or CSI prediction in the frequency domain).
[0109] The model control entity sends a response in response to the request from the gNB.
[0110] ●The response may indicate that the request failed (eg, the requested AI model information is not available in the model control entity).
[0111] ●In some embodiments, the response may include only the AI model description information of the AI model, where each AI model may be uniquely identified by an AI model ID.
[0112] ○ In some embodiments, in response to the gNB receiving the AI model description information, the gNB may send a request for AI model deployment information of the AI model to the model control entity, wherein the request may include the AI model ID.
[0113] ●In some embodiments, the response may include both AI model description information and AI deployment information for multiple AI models, where each AI model may be uniquely identified by an AI model ID.
[0114] Figure 2 An example flow chart for requesting model information is shown. Operation 202 includes sending, by a first wireless device, a first message requesting model information to a second device located in a network. Operation 204 includes receiving, by the first wireless device in response to sending the first message, a second message including model description information describing one or more characteristics of a model to be used by the first wireless device.
[0115] In some embodiments, the model description information includes an identifier of the model. In some embodiments, the first message includes a field identifying the purpose for requesting model information. In some embodiments, the first message includes any one or more of the following information: a physical cell identifier (PCI), a transmit and receive point (TRP) identifier, an absolute radio frequency channel number (ARFCN), a bandwidth of a cell, a subcarrier spacing of a cell, a number of beams, a direction of each of a plurality of beams, a number of ports, a time period of a reference signal, a subband granularity, a first expected number of opportunities for input of the model, a second expected number of opportunities for output of the model, and / or a resource mapping pattern in the frequency domain. In some embodiments, the method further includes: sending a third message requesting model deployment information to a second device in the network by the first wireless device, wherein the third message includes an identifier of the model; and in response to sending the third message, receiving a fourth message including model deployment information, the model deployment information including one or more values corresponding to one or more parameters to be used by the model. In some embodiments, the second message includes model deployment information, the model deployment information including one or more values corresponding to one or more parameters to be used by the model.
[0116] In some embodiments, the method further includes: receiving, by the first wireless device, a failure message in response to sending the first message, wherein the failure message indicates that the second device in the network does not have the model information. In some embodiments, the first wireless device is a base station. In some embodiments, the first wireless device is a communication device. In some embodiments, the first message is sent by the communication device, and the second message is received by the communication device using a first non-access stratum (NAS) message and a second NAS message, respectively. In some embodiments, the method further includes: sending, by the communication device, a request to obtain auxiliary data to the base station; and receiving, in response to sending the request, auxiliary data, the auxiliary data including the line of sight direction of the reference signal when the base station sends the reference signal, and / or the beam width of the reference signal when the base station sends the reference signal.
[0117] In some embodiments, the method also includes: sending, by the communication device, a request to the base station for obtaining any one or more configurations of the following items: a physical cell identifier (PCI), a transmit and receive point (TRP) identifier, an absolute radio frequency channel number (ARFCN), a bandwidth of the cell, a subcarrier spacing of the cell, the number of beams, the direction of each of multiple beams, the number of ports, a time period of a reference signal, a subband granularity, a first expected number of opportunities for input to a model, a second expected number of opportunities for output of the model, and / or a resource mapping pattern in the frequency domain; and in response to sending the request, receiving any one or more of the configurations included in the request.
[0118] Figure 3 An example flow chart for sending model information is shown. Operation 302 includes: receiving, by a second device located in a network, a first message from a first wireless device requesting model information. Operation 304 includes: sending, by the second device in response to receiving the first message, a second message, wherein the second message includes model description information that describes one or more characteristics of a model to be used by the first wireless device.
[0119] In some embodiments, the model description information includes an identifier of the model. In some embodiments, the first message includes a field identifying the purpose for requesting model information. In some embodiments, the first message includes any one or more of the following information: a physical cell identifier (PCI), a transmit and receive point (TRP) identifier, an absolute radio frequency channel number (ARFCN), a bandwidth of a cell, a subcarrier spacing of a cell, a number of beams, a direction of each of a plurality of beams, a number of ports, a time period of a reference signal, a subband granularity, a first expected number of opportunities for input of the model, a second expected number of opportunities for output of the model, and / or a resource mapping pattern in the frequency domain. In some embodiments, the method further includes: receiving, by a second device in the network, a third message requesting model deployment information from the first wireless device, wherein the third message includes an identifier of the model; and in response to receiving the third message, sending a fourth message including model deployment information, the model deployment information including one or more values corresponding to one or more parameters to be used by the model.
[0120] In some embodiments, the second message includes model deployment information including one or more values corresponding to one or more parameters to be used by the model. In some embodiments, the method further includes: sending, by the second device in response to receiving the first message, a failure message, wherein the failure message indicates that the second device in the network does not have the model information. In some embodiments, the first wireless device is a base station. In some embodiments, the first wireless device is a communication device.
[0121] IV. Case 2: UE side model
[0122] In this case, the UE can be considered as a first wireless device, and the second device can be a gNB / NG-RAN node or a device residing in a core network entity.
[0123] Case 2-1: UE can provide some AI model information to gNB / NG-RAN node
[0124] In some embodiments, the UE can download the AI model from the cloud server. However, the gNB does not have the AI model information of the AI model before obtaining some AI model information from the UE.
[0125] In some embodiments, the some AI model information only includes some of the AI model description information of the corresponding AI model.
[0126] In some embodiments, the UE is not required to provide AI deployment information to the gNB.
[0127] In some embodiments, the UE may send a request to the gNB for assistance data that may facilitate AI model inference at the UE, where the assistance data may include any one or more of the following:
[0128] The boresight direction of the reference signal when the gNB sends the reference signal,
[0129] ●The beamwidth of the reference signal when the gNB sends the reference signal (e.g., 3dB beamwidth).
[0130] In some embodiments, the UE may send a request to the gNB to provide a preferred configuration (measurement results / assistance data based on the preferred configuration may be used as AI model input), wherein the preferred configuration indicated in the request may include any one or more of the following:
[0131] ● Physical Cell ID (PCI),
[0132] ● Transmit and Receive Point (TRP) ID,
[0133] ●Absolute Radio Frequency Channel Number (ARFCN),
[0134] ●The bandwidth of the cell,
[0135] ●Subcarrier spacing of the cell,
[0136] The number of beams (e.g., the number of CSI-RS resources in a CSI-RS resource set),
[0137] The direction of each beam (for example, the direction may be the boresight direction of the beam),
[0138] ● The number of ports (e.g., the number of ports used for CSI-RS resources),
[0139] ●The time period of the reference signal,
[0140] The subband granularity (e.g., the number of resource blocks included in a subband),
[0141] The expected number of opportunities for AI model input (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0142] The expected number of opportunities for the AI model output (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0143] ● Resource mapping pattern in the frequency domain (eg, used to indicate which resource element will transmit a reference signal, which can be used for channel estimation or CSI prediction in the frequency domain).
[0144] Case 2-2: Model control entity resides in the gNB / NG-RAN node
[0145] In some embodiments, the gNB may send the AI model information in a system information message.
[0146] • System information messages may be sent via a System Information Block (SIB) in a broadcast channel.
[0147] ●When the UE is in the RRC connected state, the system information message may be sent to the UE in an RRC message.
[0148] ●The system information message may only include the AI model description information in the AI model information.
[0149] ○ The system information message may include multiple AI models, each AI model having its corresponding AI model description information. Each AI model may be uniquely identified by an AI model ID.
[0150] ●The system information message may also include a configuration of resources that can be used by the UE to request or indicate AI model information. For example, the base station may store a mapping between multiple resources and multiple AI models so that when the UE sends data on one of the resources, the base station can determine the AI model that the UE is requesting to use based on the resource being used by the UE to send data.
[0151] ○ The resource may be a PUCCH resource.
[0152] o The resource may be a Random Access Channel (RACH).
[0153] ○ A resource can be associated with at least one AI model.
[0154] ○ The UE can send PUCCH / RACH to request the gNB to provide the AI model information of the associated AI model.
[0155] In some embodiments, the gNB may provide assistance data to the UE (which may facilitate AI model inference at the UE), where the assistance data may include any one or more of the following:
[0156] The boresight direction of the reference signal when the gNB sends the reference signal,
[0157] ●The beamwidth of the reference signal when the gNB sends the reference signal (e.g., 3dB beamwidth).
[0158] Figure 4AAn example flow chart for sending model information is shown. Operation 402 includes sending, by a base station, a system information message including model information, wherein the model information includes a plurality of model description information associated with a corresponding plurality of models, and wherein each model description information describes one or more characteristics of a model to be used by a communication device.
[0159] In some embodiments, one of the multiple model description information includes at least one identifier of a model. In some embodiments, the system information message is sent in a system information block (SIB); or when the communication device is in a radio resource control (RRC) connected state, the system information message is sent in an RRC message. In some embodiments, the system information message includes a configuration of multiple resources to be used by the communication device, and at least one of the multiple resources is mapped to at least one of the multiple models. In some embodiments, the multiple resources include multiple physical uplink control channel (PUCCH) resources or multiple random access channels (RACH). In some embodiments, the method further includes: sending auxiliary data to the communication device, wherein the auxiliary data includes the line of sight direction of the reference signal when the base station sends the reference signal, and / or the beam width of the reference signal when the base station sends the reference signal.
[0160] In some embodiments, the at least one identifier comprises any of: a first identifier associated with an encoder of the model, a second identifier associated with a decoder of the model, or a first identifier and a second identifier. In some embodiments, the method further comprises: receiving, by the base station, a message from the communication device, the message comprising an identifier associated with an encoder of the model used by the communication device.
[0161] Figure 4B An example flow chart for receiving model information is shown. Operation 452 includes receiving, by the communication device, a system information message including model information from a base station, wherein the model information includes a plurality of model description information associated with a corresponding plurality of models, and wherein each model description information describes one or more characteristics of a model to be used by the communication device.
[0162] In some embodiments, one of the plurality of model description information includes at least one identifier of a model. In some embodiments, the system information message is received in a system information block (SIB); or when the communication device is in a radio resource control (RRC) connected state, the system information message is received in an RRC message. In some embodiments, the system information message includes a configuration of a plurality of resources to be used by the communication device, and at least one of the plurality of resources is mapped to at least one of the plurality of models. In some embodiments, the plurality of resources includes a plurality of physical uplink control channel (PUCCH) resources or a plurality of random access channels (RACH).
[0163] In some embodiments, the method further comprises: receiving auxiliary data, wherein the auxiliary data comprises a boresight direction of a reference signal when the base station sends a reference signal, and / or a beam width of the reference signal when the base station sends a reference signal. In some embodiments, the at least one identifier comprises any one of: a first identifier associated with an encoder of the model, a second identifier associated with a decoder of the model, or a first identifier and a second identifier. In some embodiments, the method further comprises: sending a message by the communication device to the base station, the message comprising an identifier associated with an encoder of the model used by the communication device.
[0164] Case 2-3: Model control entity resides in the core network entity
[0165] Case 2-3-1: AI model information is non-transparent to gNB / NG-RAN nodes
[0166] In this case, the UE determines / performs AI model inference and the model control entity resides at the core network entity. The AI model information is opaque to the gNB / NG-RAN node, which means that the gNB has both the AI model description information and the AI model deployment information of the AI model.
[0167] The transmission of AI model information between the gNB and the model control entity is similar to that discussed in Case 1 (except that the AI model inference is performed entirely at the UE), as shown below:
[0168] The gNB may send a request for AI model information to the model control entity.
[0169] ○ In some embodiments, the request may include information indicating the purpose of the AI model. For example, for beam prediction, CSI prediction, or LoS / NLoS identification.
[0170] ○ In some embodiments, the request may also include any one or more of the following information (which the model control entity may use to decide which AI model the gNB requires):
[0171] ■Physical Cell ID (PCI),
[0172] ■ Transmit and Receive Point (TRP) ID,
[0173] ■Absolute Radio Frequency Channel Number (ARFCN),
[0174] ■The bandwidth of the cell,
[0175] ■Subcarrier spacing of the cell,
[0176] ■ the number of beams (e.g., the number of CSI-RS resources in a CSI-RS resource set),
[0177] ■ the direction of each beam (for example, the direction may be the boresight direction of the beam),
[0178] ■ the number of ports (e.g., the number of ports used for CSI-RS resources),
[0179] ■The time period of the reference signal,
[0180] ■ Subband granularity (e.g., the number of resource blocks included in a subband)
[0181] ■ the expected number of opportunities for AI model input (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0182] ■ the expected number of opportunities for the AI model output (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0183] ■ Resource mapping pattern in the frequency domain (eg, used to indicate which resource element will transmit a reference signal, which can be used for channel estimation or CSI prediction in the frequency domain).
[0184] ○ The model control entity sends a response in response to the request from the gNB.
[0185] ■The response may indicate that the request failed (eg, the requested AI model information is not available in the model control entity).
[0186] ■In some embodiments, the response may include only AI model description information of multiple AI models, where each AI model may be uniquely identified by a model ID.
[0187] ●The gNB may send a request for AI model deployment information of the AI model to the model control entity, where the request may include the AI model ID.
[0188] ■In some embodiments, the response may include both AI model description information and AI deployment information for multiple AI models, where:
[0189] Each AI model can be uniquely identified by a model ID.
[0190] ○The gNB may send a request for AI model deployment information of the AI model to the model control entity, where the request may include the AI model ID.
[0191] The transmission of AI model information between gNB and UE is similar to the discussion in Case 2-2, as shown below:
[0192] ●In some embodiments, the gNB may send AI model information in a system information message.
[0193] o System information messages may be sent via a System Information Block (SIB) in a broadcast channel.
[0194] o When the UE is in the RRC connected state, the system information message may be sent to the UE in an RRC message.
[0195] ○The system information message may only include the AI model description information in the AI model information.
[0196] ■ The system information message may include multiple AI models, each AI model having its corresponding AI model description information. Each AI model may be uniquely identified by an ID.
[0197] ○ The system information message may also include the configuration of resources that can be used by the UE to request AI model information:
[0198] ■The resource may be a PUCCH resource.
[0199] ■ The resource may be a Random Access Channel (RACH).
[0200] ■A resource can be associated with at least one AI model.
[0201] ■UE can send PUCCH / RACH to request gNB to provide AI model deployment information of the associated AI model.
[0202] In some embodiments, the gNB may provide assistance data to the UE (which may facilitate AI model inference at the UE), where the assistance data may include any one or more of the following:
[0203] The boresight direction of the reference signal when the gNB sends the reference signal,
[0204] ●The beamwidth of the reference signal when the gNB sends the reference signal (e.g., 3dB beamwidth).
[0205] Case 2-3-2: AI model information is transparent to gNB / NG-RAN nodes
[0206] In this case, the UE determines / performs AI model inference and the model control entity resides at the core network entity. The gNB does not have AI model information for the AI model until it obtains some AI model information from the UE.
[0207] ●UE can send a request for AI model information to the model control entity.
[0208] ○ In some embodiments, the request may include information indicating the purpose of the AI model. For example, for beam prediction, CSI prediction, or LoS / NLoS identification.
[0209] ○ In some embodiments, the request may also include any one or more of the following information (which the model control entity may use to decide which AI model the UE requires):
[0210] ■Physical Cell Identifier (PCI),
[0211] ■ Transmitting and receiving point identifier (TRP ID),
[0212] ■Absolute Radio Frequency Channel Number (ARFCN),
[0213] ■The bandwidth of the cell,
[0214] ■Subcarrier spacing of the cell,
[0215] ■ the number of beams (e.g., the number of CSI-RS resources in a CSI-RS resource set),
[0216] ■ the direction of each beam (for example, the direction may be the boresight direction of the beam),
[0217] ■ the number of ports (e.g., the number of ports used for CSI-RS resources),
[0218] ■The time period of the reference signal,
[0219] ■ Subband granularity (e.g., the number of resource blocks included in a subband)
[0220] ■ the expected number of opportunities for AI model input (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0221] ■ the expected number of opportunities for the AI model output (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0222] ■ Resource mapping pattern in the frequency domain (eg, used to indicate which resource element will transmit a reference signal, which can be used for channel estimation or CSI prediction in the frequency domain).
[0223] • The model control entity sends a response in response to a request from the UE.
[0224] ○ The response may indicate the request (eg, the requested AI model information is not available in the model control entity).
[0225] ○ In some embodiments, the response may include only AI model description information of multiple AI models, where each AI model may be uniquely identified by an AI model ID.
[0226] ■The UE may send a request for AI model deployment information of the AI model to the model control entity, wherein the request may include the AI model ID.
[0227] ○ In some embodiments, the response may include both AI model description information and AI deployment information for multiple AI models, wherein:
[0228] Each AI model can be uniquely identified by a model ID.
[0229] • In some embodiments, the request and response are sent via Non-Access Stratum (NAS) messages between the UE and the model control entity.
[0230] ● In some embodiments, the UE may download the AI model from the model control entity. However, the gNB does not have the AI model information of the AI model before obtaining some AI model information from the UE, similar to case 2-1, as shown below:
[0231] ○ In some embodiments, the AI model information only includes some of the AI model description information of the corresponding AI model.
[0232] ○ In some embodiments, the UE is not required to provide AI deployment information to the gNB, or the UE determines not to provide AI deployment information to the gNB.
[0233] ○ In some embodiments, the UE may send a request to the gNB to obtain assistance data (which may assist in AI model inference at the UE), where the assistance data may include any one or more of the following:
[0234] ■The boresight direction of the reference signal when the gNB sends the reference signal,
[0235] ■The beamwidth of the reference signal when the gNB sends the reference signal (e.g., 3dB beamwidth).
[0236] ○ In some embodiments, the UE may send a request to the gNB to provide a preferred configuration (measurement results / assistance data based on the preferred configuration may be used as AI model input), where the preferred configuration indicated in the request may include any one of the following or
[0237] More than one:
[0238] ■Physical Cell ID (PCI),
[0239] ■ Transmit and Receive Point (TRP) ID,
[0240] ■Absolute Radio Frequency Channel Number (ARFCN),
[0241] ■The bandwidth of the cell,
[0242] ■Subcarrier spacing of the cell,
[0243] ■ the number of beams (e.g., the number of CSI-RS resources in a CSI-RS resource set),
[0244] ■ the direction of each beam (for example, the direction may be the boresight direction of the beam),
[0245] ■ the number of ports (e.g., the number of ports used for CSI-RS resources),
[0246] ■The time period of the reference signal,
[0247] ■ Subband granularity (e.g., the number of resource blocks included in a subband)
[0248] ■ the expected number of opportunities for AI model input (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0249] ■ the expected number of opportunities for the AI model output (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0250] ■ Resource mapping pattern in the frequency domain (eg, used to indicate which resource element will transmit a reference signal, which can be used for channel estimation or CSI prediction in the frequency domain).
[0251] Case 2-3-3: Partial AI model information is transparent to gNB (or NG-RAN node) In this case, the UE determines / performs AI model inference and the model control entity resides at the core network entity. Partial AI model information is transparent to gNB means that the gNB has AI model description information of the AI model and the gNB does not have AI model information of AI deployment information of the AI model.
[0252] The gNB may send a request for AI model description information to the model control entity.
[0253] ○ In some embodiments, the request may include information indicating the purpose of the AI model. For example, for beam prediction, CSI prediction, or LoS / NLoS identification.
[0254] ○ In some embodiments, the request may also include any one or more of the following information (the model control entity may use this
[0255] Some information to decide which AI model gNB needs):
[0256] ■Physical Cell ID (PCI),
[0257] ■ Transmit and Receive Point (TRP) ID,
[0258] ■Absolute Radio Frequency Channel Number (ARFCN),
[0259] ■The bandwidth of the cell,
[0260] ■Subcarrier spacing of the cell,
[0261] ■ the number of beams (e.g., the number of CSI-RS resources in a CSI-RS resource set),
[0262] ■ the direction of each beam (for example, the direction may be the boresight direction of the beam),
[0263] ■ the number of ports (e.g., the number of ports used for CSI-RS resources),
[0264] ■The time period of the reference signal,
[0265] ■ Subband granularity (e.g., the number of resource blocks included in a subband)
[0266] ■ the expected number of opportunities for AI model input (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0267] ■ the expected number of opportunities for the AI model output (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0268] ■ Resource mapping pattern in the frequency domain (eg, used to indicate which resource element will transmit a reference signal, which can be used for channel estimation or CSI prediction in the frequency domain).
[0269] ●The model control entity sends a response in response to the request from the gNB.
[0270] ○ The response may indicate that the request failed (e.g., the requested AI model is not available in the model control entity).
[0271] ○ In some embodiments, the response may include only AI model description information of multiple AI models, wherein each AI model may be represented by
[0272] Model ID uniquely identifies the model.
[0273] ● In some embodiments, the UE may obtain the AI model information in one of the following ways:
[0274] ○ The gNB can instruct the UE to download the AI model deployment information from the model control entity.
[0275] ■The indication may include the model ID of the AI model or AI model description information.
[0276] ■The UE may send a response to the gNB indicating that the UE has obtained the requested AI model.
[0277] ○ The gNB can request the model control entity to provide AI model deployment information to the UE.
[0278] ■The indication may include the model ID of the AI model or AI model description information.
[0279] ■The model control entity may send a response to the gNB indicating which AI model has been provided to the UE.
[0280] ○UE can request the model control entity to provide AI model information.
[0281] ○ In the above method, the transmission of AI model information between the model control entity and the UE is similar to the discussion in case 2-3-2, as shown below:
[0282] ■UE can send a request for AI model information to the model control entity.
[0283] ● In some embodiments, the request may include information indicating the purpose of the AI model. For example, for beam prediction, CSI prediction, or LoS / NLoS identification.
[0284] ● In some embodiments, the request may also include any one or more of the following information (which the model control entity may use to decide which AI model the UE requires):
[0285] ○ Physical Cell ID (PCI),
[0286] ○ Transmit and Receive Point (TRP) ID,
[0287] ○ Absolute Radio Frequency Channel Number (ARFCN),
[0288] ○The bandwidth of the cell,
[0289] ○Subcarrier spacing of the cell,
[0290] o The number of beams (e.g., the number of CSI-RS resources in a CSI-RS resource set),
[0291] o the direction of each beam (for example, the direction may be the boresight direction of the beam),
[0292] o The number of ports (e.g., the number of ports used for CSI-RS resources),
[0293] ○The time period of the reference signal,
[0294] o subband granularity (e.g., the number of resource blocks included in a subband),
[0295] ○ The expected number of opportunities for AI model input (which can be used for beam prediction in time domain or CSI prediction in time domain),
[0296] ○ The expected number of opportunities for the AI model output (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0297] o Resource mapping pattern in the frequency domain (eg, used to indicate which resource elements will transmit a reference signal, which can be used for channel estimation or CSI prediction in the frequency domain).
[0298] ■ The model control entity sends a response in response to a request from the UE.
[0299] ●The response may indicate that the request failed (eg, the requested AI model information is not available in the model control entity).
[0300] ●The response may include only AI model description information of multiple AI models, where each AI model may be uniquely identified by an AI model ID.
[0301] ●In some embodiments, the response may only include AI model description information of multiple AI models, where each AI model can be uniquely identified by a model ID.
[0302] ○The UE may send a request for AI model deployment information of the AI model to the model control entity, where the request may include the AI model ID.
[0303] ●In some embodiments, the response may include both AI model description information and AI deployment information for multiple AI models, where each AI model may be uniquely identified by an AI model ID.
[0304] ■ In some embodiments, the request and response are sent via Non-Access Stratum (NAS) messages between the UE and the model control entity.
[0305] In some embodiments, the gNB may provide assistance data to the UE (which may facilitate AI model inference at the UE), where the assistance data may include any one or more of the following:
[0306] The boresight direction of the reference signal when the gNB sends the reference signal,
[0307] ●The beamwidth of the reference signal when the gNB sends the reference signal (e.g., 3dB beamwidth).
[0308] V. Case 3: Two-Sided Model
[0309] Case 3-1: UE can provide some AI model information to gNB / NG-RAN node
[0310] In some embodiments, the UE can download the encoder part of the AI model from the cloud server. However, the gNB does not have the AI model information of the AI model until it obtains some AI model information from the UE.
[0311] In some embodiments, some AI model information includes only some of the AI model description information corresponding to the encoder parts of multiple AI models, where each encoder part can be uniquely identified by an ID. In some embodiments, the UE is not required to provide AI deployment information to the gNB.
[0312] In some embodiments, the UE may send a request to the gNB for assistance data that may facilitate AI model inference at the UE, where the assistance data may include any one or more of the following:
[0313] The boresight direction of the reference signal when the gNB sends the reference signal,
[0314] ●The beamwidth of the reference signal when the gNB sends the reference signal (e.g., 3dB beamwidth).
[0315] In some embodiments, the UE may send a request to the gNB to provide a preferred configuration (measurement results based on the preferred configuration may be used as AI model input), wherein the preferred configuration indicated in the request may include any one or more of the following:
[0316] ● Physical Cell ID (PCI),
[0317] ● Transmit and Receive Point (TRP) ID,
[0318] ●Absolute Radio Frequency Channel Number (ARFCN),
[0319] ●Cell bandwidth and subcarrier spacing,
[0320] The number of beams (e.g., the number of CSI-RS resources in a CSI-RS resource set),
[0321] The direction of each beam (for example, the direction may be the boresight direction of the beam),
[0322] ● The number of ports (e.g., the number of ports used for CSI-RS resources),
[0323] ●The time period of the reference signal,
[0324] The subband granularity (e.g., the number of resource blocks included in a subband),
[0325] The expected number of opportunities for AI model input (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0326] The expected number of opportunities for the AI model output (which can be used for beam prediction in the time domain or CSI prediction in the time domain),
[0327] ● Resource mapping pattern in the frequency domain (eg, used to indicate which resource element will transmit a reference signal, which can be used for channel estimation or CSI prediction in the frequency domain).
[0328] In some embodiments, the gNB may indicate which AI model (or encoder part) should be used for AI model inference at the UE, where the indication may include at least the AI model ID of the encoder part.
[0329] Case 3-2: gNB / NG-RAN node can provide some AI model information to UE
[0330] In some embodiments, the gNB may send the AI model information in a system information message.
[0331] • System information messages may be sent via a System Information Block (SIB) in a broadcast channel.
[0332] ●When the UE is in the RRC connected state, the system information message may be sent to the UE in an RRC message.
[0333] ●The system information message may only include the AI model description information in the AI model information.
[0334] ○ The system information message may include multiple AI models, each AI model having its corresponding AI model description information. Each AI model may be uniquely identified by an AI model ID. In some embodiments, the AI model ID may be one of the following:
[0335] ■Encoder Part ID,
[0336] ■Decoder Part ID,
[0337] ■ A pair of {encoder part ID, decoder part ID}.
[0338] In some embodiments, the system information message may also include configuration of resources that can be used by the UE to request AI model information.
[0339] ●The resource may be a PUCCH resource.
[0340] • The resource may be a Random Access Channel (RACH).
[0341] ●A resource can be associated with at least one AI model.
[0342] ●UE can send PUCCH / RACH to request gNB to provide AI model information of the encoder part of the AI model.
[0343] In some embodiments, the UE may inform the gNB (e.g., via an encoder part ID) which encoder part has been used for AI model inference.
[0344] Case 3-3: Model control entity resides in the core network entity
[0345] If there is no conflict, the transmission of AI model information related to the decoder part of the AI model between the model control entity and the gNB can reuse the process in Case 1 for the network side model.
[0346] If there is no conflict, the transmission of AI model information related to the encoder part of the AI model that occurs between the model control entity and the gNB or between the model control entity and the UE can reuse the procedures in Case 2-3 for the UE-side model.
[0347] Figure 5An exemplary block diagram of a hardware platform 500 that may be part of a network device (e.g., a base station) or a communication device (e.g., a user equipment (UE)) is shown. The hardware platform 500 includes at least one processor 510 and a memory 505 having instructions stored thereon. When the instructions are executed by the processor 510, the hardware platform 500 is configured to execute Figures 1 to 4B As well as the operations described in the various embodiments described in this patent document. Transmitter 515 transmits or sends information or data to another device. For example, a network device transmitter can send a message to a user device. Receiver 520 receives information or data transmitted or sent by another device. For example, a user device can receive a message from a network device.
[0348] The embodiments discussed above may be applied to wireless communications. Figure 6 An example of a wireless communication system (e.g., a 5G or NR (New Radio) cellular network) is shown, which includes a base station 620 and one or more user equipment (UE) 611, 612, and 613. In some embodiments, the UE uses a communication link to the network (sometimes referred to as an uplink direction, which is depicted by dashed arrows 631, 632, 633) to access the BS (e.g., the network), and then subsequent communication from the BS to the UE is achieved (e.g., shown as in the direction from the network to the UE, which is sometimes referred to as a downlink direction, which is shown by arrows 641, 642, 643). In some embodiments, the BS sends information to the UE (sometimes referred to as a downlink direction, depicted by arrows 641, 642, 643), and then subsequent communication from the UE to the BS is achieved (e.g., shown as in the direction from the UE to the BS, which is sometimes referred to as an uplink direction, which is shown by dashed arrows 631, 632, 633). The UE may be, for example, a smart phone, a tablet computer, a mobile computer, a machine to machine (M2M) device, an Internet of Things (IoT) device, etc.
[0349] In this document, the term "exemplary" is used to mean "an example of," and is not intended to refer to an ideal or preferred embodiment unless otherwise specified.
[0350] Some of the multiple embodiments described herein are described in the general context of methods or processes, which can be implemented in one embodiment by a computer program product, implemented in a computer-readable medium, which includes computer-executable instructions, such as program codes, executed by a computer in a network environment. Computer-readable media may include removable storage devices and non-removable storage devices, including but not limited to read-only memory (ROM), random access memory (RAM), compact disc (CD), digital versatile disc (DVD), etc. Therefore, computer-readable media may include non-transitory storage media. Generally, program modules may include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. Computer executable instructions or processor executable instructions, associated data structures, and program modules represent examples of program codes for executing the steps of the methods disclosed herein. A specific sequence of such executable instructions or associated data structures represents an example of corresponding actions for implementing the functions described in such steps or processes.
[0351] Some embodiments in the disclosed embodiments may be implemented as devices or modules using hardware circuits, software, or a combination thereof. For example, a hardware circuit implementation may include discrete analog components and / or digital components, for example, integrated as a part of a printed circuit board. Alternatively, or additionally, the disclosed components or modules may be implemented as application specific integrated circuits (Application Specific Integrated Circuit, ASIC) and / or field programmable gate arrays (Field Programmable Gate Array, FPGA) devices. Some embodiments may additionally or alternatively include a digital signal processor (Digital Signal Processor, DSP), which is a dedicated microprocessor with such an architecture: an architecture optimized for the operational requirements of digital signal processing associated with the disclosed functions of the present application. Similarly, various components or subcomponents within each module may be implemented in software, hardware, or firmware. Any of the connection methods and media known in the art (including but not limited to communication via the Internet, wired or wireless networks using suitable protocols) may be used to provide connections between modules and / or components within the modules.
[0352] Although this document contains many details, these details should not be interpreted as limitations on the scope of the claimed invention or the scope of the content that may be claimed, but should be interpreted as descriptions of features specific to a specific embodiment. Certain features described in the context of a separate embodiment in this document may also be implemented in combination in a single embodiment. Conversely, the various features described in the context of a single embodiment may also be implemented individually in multiple embodiments or in any suitable sub-combination. In addition, although features may be described above as working in certain combinations, and even initially claimed as such, in some cases, one or more features from the combination may be deleted from the claimed combination, and the claimed combination may involve a sub-combination or a variation of a sub-combination. Similarly, although each operation is depicted in a particular order in the accompanying drawings, this should not be understood as requiring such operations to be performed in the particular order shown or in order, or performing all the operations shown, to achieve the desired result.
[0353] Only a few implementations and examples are described, and other implementations, enhancements, and variations may be made based on what is described and illustrated in this disclosure.
Claims
1. A wireless communication method, comprising: The first wireless device sends a first message requesting model information to a second device located in the network; as well as receiving, by the first wireless device in response to sending the first message, a second message, The second message includes model description information, wherein the model description information describes one or more characteristics of a model to be used by the first wireless device.
2. The method according to claim 1, wherein: The model description information includes an identifier of the model.
3. The method according to claim 1, wherein: The first message includes a field identifying a purpose for requesting the model information.
4. The method according to claim 1, wherein: The first message includes any one or more of the following information: Physical Cell Identifier (PCI), Transmit and Receive Point (TRP) identifier, Absolute Radio Frequency Channel Number (ARFCN), The bandwidth of the cell, the subcarrier spacing of the cell, The number of beams, the direction of each of the plurality of beams, The number of ports, The time period of the reference signal, Subband granularity, a first expected number of opportunities for input to the model, a second expected number of opportunities for an output of the model, and / or Resource mapping pattern in frequency domain.
5. The method according to claim 1, further comprising: Sending, by the first wireless device, a third message requesting model deployment information to the second device in the network, wherein the third message includes an identifier of the model; and In response to sending the third message, a fourth message is received including the model deployment information, the model deployment information including one or more values corresponding to one or more parameters to be used by the model.
6. The method according to claim 1, wherein: The second message includes model deployment information including one or more values corresponding to one or more parameters to be used by the model.
7. The method according to claim 1, further comprising: A failure message is received by the first wireless device in response to sending the first message, wherein the failure message indicates that the second device in the network does not have the model information.
8. The method according to claim 1, wherein: The first wireless device is a base station.
9. The method according to claim 1, wherein: The first wireless device is a communication device.
10. The method according to claim 9, wherein: The first message is sent by the communication device and the second message is received by the communication device using a first non-access stratum (NAS) message and a second NAS message, respectively.
11. The method according to claim 9, further comprising: The communication device sends a request for obtaining assistance data to a base station; as well as In response to sending the request, the auxiliary data is received, the auxiliary data including the boresight direction of the reference signal when the base station sends the reference signal, and / or the beam width of the reference signal when the base station sends the reference signal.
12. The method according to claim 9, further comprising: The communication device sends a request to a base station for obtaining any one or more configurations of the following items: Physical Cell Identifier (PCI), Transmit and Receive Point (TRP) identifier, Absolute Radio Frequency Channel Number (ARFCN), The bandwidth of the cell, the subcarrier spacing of the cell, The number of beams, the direction of each of the plurality of beams, The number of ports, The time period of the reference signal, Subband granularity, a first expected number of opportunities for input to the model, a second expected number of opportunities for an output of the model, and / or a resource mapping pattern in the frequency domain; and In response to sending the request, any one or more of the configurations included in the request are received.
13. A wireless communication method, comprising: receiving, by a second device located in the network, a first message from a first wireless device requesting model information; as well as sending, by the second device in response to receiving the first message, a second message, The second message includes model description information, wherein the model description information describes one or more characteristics of a model to be used by the first wireless device.
14. The method according to claim 13, wherein: The model description information includes an identifier of the model.
15. The method according to claim 13, wherein: The first message includes a field identifying a purpose for requesting the model information.
16. The method according to claim 13, wherein: The first message includes any one or more of the following information: Physical Cell Identifier (PCI), Transmit and Receive Point (TRP) identifier, Absolute Radio Frequency Channel Number (ARFCN), The bandwidth of the cell, the subcarrier spacing of the cell, The number of beams, the direction of each of the plurality of beams, The number of ports, The time period of the reference signal, Subband granularity, a first expected number of opportunities for input to the model, a second expected number of opportunities for an output of the model, and / or Resource mapping pattern in frequency domain.
17. The method according to claim 13, further comprising: receiving, by the second device in the network, a third message from the first wireless device requesting model deployment information, wherein the third message includes an identifier of the model; and In response to receiving the third message, sending a fourth message including the model deployment information, the model deployment information including one or more values corresponding to one or more parameters to be used by the model.
18. The method according to claim 13, wherein: The second message includes model deployment information including one or more values corresponding to one or more parameters to be used by the model.
19. The method according to claim 13, further comprising: A failure message is sent by the second device in response to receiving the first message, wherein the failure message indicates that the second device in the network does not have the model information.
20. The method according to claim 13, wherein: The first wireless device is a base station.
21. The method according to claim 13, wherein: The first wireless device is a communication device.
22. A wireless communication method, comprising: The base station sends a system information message including model information, The model information includes a plurality of model description information associated with the corresponding plurality of models. Each model description information describes one or more characteristics of a model to be used by the communication device.
23. The method according to claim 22, wherein: One of the plurality of model description information includes at least one identifier of a model.
24. The method according to claim 22, wherein: The system information message is sent in a system information block (SIB); or when the communication device is in a radio resource control (RRC) connected state, the system information message is sent in an RRC message.
25. The method according to claim 22, in, The system information message includes a configuration of a plurality of resources to be used by the communication device, and At least one resource among the plurality of resources is mapped to at least one model among the plurality of models.
26. The method according to claim 25, wherein: The plurality of resources include a plurality of physical uplink control channel (PUCCH) resources or a plurality of random access channels (RACH).
27. The method of claim 22, further comprising: sending auxiliary data to the communication device, The auxiliary data includes the line of sight direction of the reference signal when the base station sends the reference signal, and / or the beam width of the reference signal when the base station sends the reference signal.
28. The method of claim 22, wherein: The at least one identifier includes any of the following: a first identifier associated with an encoder of said model, a second identifier associated with a decoder of the model, or the first identifier and the second identifier.
29. The method of claim 22, further comprising: A message is received by the base station from the communication device, the message including an identifier associated with an encoder of the model used by the communication device.
30. A wireless communication method, comprising: receiving, by the communication device, a system information message including model information from a base station, The model information includes a plurality of model description information associated with the corresponding plurality of models. Each model description information describes one or more characteristics of a model to be used by the communication device.
31. The method according to claim 30, wherein: One of the plurality of model description information includes at least one identifier of a model.
32. The method of claim 30, wherein: The system information message is received in a system information block (SIB); or when the communication device is in a radio resource control (RRC) connected state, the system information message is received in an RRC message.
33. The method according to claim 30, in, The system information message includes a configuration of a plurality of resources to be used by the communication device, and At least one resource among the plurality of resources is mapped to at least one model among the plurality of models.
34. The method of claim 33, wherein: The plurality of resources include a plurality of physical uplink control channel (PUCCH) resources or a plurality of random access channels (RACH).
35. The method of claim 30, further comprising: Receive auxiliary data, The auxiliary data includes the line of sight direction of the reference signal when the base station sends the reference signal, and / or the beam width of the reference signal when the base station sends the reference signal.
36. The method of claim 30, wherein: The at least one identifier includes any of the following: a first identifier associated with an encoder of said model, a second identifier associated with a decoder of the model, or the first identifier and the second identifier.
37. The method of claim 30, further comprising: A message is sent by the communication device to the base station, the message comprising an identifier associated with an encoder of the model used by the communication device.
38. An apparatus for wireless communication, comprising a processor configured to implement the method according to one or more of claims 1 to 37.
39. A non-transitory computer-readable program storage medium having codes stored thereon, which, when executed by a processor, cause the processor to implement the method according to one or more of claims 1 to 37.
