WO2023198094A1 - 模型输入的确定方法及通信设备 - Google Patents

模型输入的确定方法及通信设备 Download PDF

Info

Publication number
WO2023198094A1
WO2023198094A1 PCT/CN2023/087739 CN2023087739W WO2023198094A1 WO 2023198094 A1 WO2023198094 A1 WO 2023198094A1 CN 2023087739 W CN2023087739 W CN 2023087739W WO 2023198094 A1 WO2023198094 A1 WO 2023198094A1
Authority
WO
WIPO (PCT)
Prior art keywords
elements
model
communication device
domain
input
Prior art date
Application number
PCT/CN2023/087739
Other languages
English (en)
French (fr)
Inventor
杨昂
孙鹏
李佳林
孙布勒
Original Assignee
维沃移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 维沃移动通信有限公司 filed Critical 维沃移动通信有限公司
Publication of WO2023198094A1 publication Critical patent/WO2023198094A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/48Program initiating; Program switching, e.g. by interrupt
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08Configuration management of networks or network elements
    • H04L41/0803Configuration setting
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W24/00Supervisory, monitoring or testing arrangements
    • H04W24/02Arrangements for optimising operational condition

Definitions

  • the present application belongs to the field of communication technology, and specifically relates to a method for determining model input and communication equipment.
  • communication equipment such as terminals or network-side equipment
  • AI Artificial Intelligence, AI
  • the results of model predictions can assist communication equipment in channel estimation, signal processing, etc. .
  • the input to the model is usually fixed resource locations and/or resource numbers.
  • the resources that the communication device can use are not fixed resource locations and/or resource data. In this way, when the resources that the communication device can use are inconsistent with the input of the AI model, the AI model will not be able to be used. Model prediction leads to poor flexibility and generalization ability of AI models.
  • Embodiments of the present application provide a method for determining model input and a communication device, which can solve the problem of poor flexibility and generalization ability of current AI models.
  • a method for determining model input includes: a first communication device determines the input of the AI model according to configuration information of the AI model, the configuration information being used to indicate selecting N from the first domain. elements are used as the input of the AI model, N is an integer greater than or equal to 1, the first domain includes M elements, and M is an integer greater than N.
  • a device for determining model input includes: a determining module configured to determine the input of the AI model according to the configuration information of the AI model, the configuration information being used to indicate selection from the first domain.
  • N elements are used as inputs to the AI model, and N is an integer greater than or equal to 1.
  • the first domain includes M elements, and M is an integer greater than N.
  • a communication device in a third aspect, includes a processor and a memory.
  • the memory stores a program or instructions that can be run on the processor.
  • the program or instructions are implemented when executed by the processor. The steps of the method as described in the first aspect.
  • a communication device including a processor and a communication interface, wherein the processor is configured to determine the input of the AI model according to the configuration information of the AI model, and the configuration information is used to indicate the input from the first domain.
  • the processor is configured to determine the input of the AI model according to the configuration information of the AI model, and the configuration information is used to indicate the input from the first domain.
  • a readable storage medium is provided. Programs or instructions are stored on the readable storage medium. When the programs or instructions are executed by a processor, the steps of the method described in the first aspect are implemented.
  • a chip in a sixth aspect, includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the method described in the first aspect. .
  • a computer program/program product is provided, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement the method described in the first aspect. Steps in the determination method of model inputs.
  • the communication device can determine the input of the AI model according to the configuration information of the AI model.
  • the configuration information is used to indicate that N elements are selected from the M elements of the first domain as the input of the AI model, and N is greater than Or an integer equal to 1, M is an integer greater than N.
  • the communication device can flexibly select N elements among the M elements of the first domain as inputs to the AI model, the communication device can be enabled to use as few as possible under different resource locations and/or resource numbers.
  • the AI model improves the flexibility and generalization capabilities of the AI model, and reduces the cost of the communication system.
  • Figure 1 is a schematic diagram of a wireless communication system according to an embodiment of the present application.
  • Figure 2 is a schematic flow chart of a method for determining model input according to an embodiment of the present application
  • Figure 3 is a schematic structural diagram of a device for determining model input according to an embodiment of the present application
  • Figure 4 is a schematic structural diagram of a communication device according to an embodiment of the present application.
  • Figure 5 is a schematic structural diagram of a communication device according to an embodiment of the present application.
  • Figure 6 is a schematic structural diagram of a communication device according to an embodiment of the present application.
  • first, second, etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in sequences other than those illustrated or described herein, and that "first" and “second” are distinguished objects It is usually one type, and the number of objects is not limited.
  • the first object can be one or multiple.
  • “and/or” in the description and claims indicates at least one of the connected objects, and the character “/" generally indicates that the related objects are in an "or” relationship.
  • LTE Long-term evolution
  • LTE-Advanced, LTE-A Long-term evolution
  • LTE-Advanced, LTE-A Long-term evolution
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple Access
  • FDMA Frequency Division Multiple Access
  • OFDMA Orthogonal Frequency Division Multiple Access
  • SC-FDMA Single-carrier Frequency Division Multiple Access
  • NR New Radio
  • FIG. 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable.
  • the wireless communication system includes a terminal 11 and a network side device 12.
  • the terminal 11 can be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a handheld computer, a netbook, or a super mobile personal computer.
  • Tablet Personal Computer Tablet Personal Computer
  • laptop computer laptop computer
  • PDA Personal Digital Assistant
  • PDA Personal Digital Assistant
  • UMPC ultra-mobile personal computer
  • UMPC mobile Internet device
  • Mobile Internet Device MID
  • augmented reality augmented reality, AR
  • VR virtual reality
  • robots wearable devices
  • VUE vehicle-mounted equipment
  • PUE pedestrian terminal
  • smart home home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.
  • game consoles personal computers (personal computers, PC), teller machines or self-service Terminal devices
  • wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), Smart wristbands, smart clothing, etc.
  • the network side equipment 12 may include access network equipment or core network equipment, where the access network equipment 12 may also be called wireless access network equipment, radio access network (Radio Access Network, RAN), radio access network function or Wireless access network unit.
  • the access network device 12 may include a base station, a Wireless Local Area Network (WLAN) access point or a WiFi node, etc.
  • the base station may be called a Node B, an Evolved Node B (eNB), an access point, or a base transceiver station.
  • Base Transceiver Station BTS
  • radio base station radio transceiver
  • BSS Basic Service Set
  • ESS Extended Service Set
  • home B-node home evolved B-node
  • sending and receiving point Transmitting Receiving Point, TRP
  • TRP Transmitting Receiving Point
  • the embodiment of the present application provides a method 200 for determining model input.
  • the method can be executed by a first communication device.
  • the first communication device can be a terminal or a network side device in the embodiment shown in Figure 1.
  • the method can be executed by software or hardware installed on the terminal or network side device, and the method includes the following steps.
  • the first communication device determines the input of the AI model according to the configuration information of the AI model, and the configuration information is used to indicate Indicates that N elements are selected from the first domain as the input of the AI model.
  • N is an integer greater than or equal to 1.
  • the first domain includes M elements, and M is an integer greater than N.
  • the first communication device when using the AI model for model prediction, can select N elements from the M elements of the first domain as inputs to the AI model according to the configuration information of the AI model, where N is greater than or equal to 1. is an integer, M is an integer greater than N.
  • N is greater than or equal to 1.
  • M is an integer greater than N.
  • the AI model can be used in the first communication device.
  • the AI model and the configuration information of the AI model can both be configured by the first communication device.
  • the first communication device may be a terminal or a network side device.
  • the terminal when the terminal uses the AI model, the terminal can configure the AI model and the configuration information of the AI model.
  • the terminal uses the AI model, it can flexibly select N elements from the M elements in the first domain according to the configuration information. Elements serve as input to the AI model.
  • the network side device uses the AI model, the network side device can configure the AI model and the configuration information of the AI model.
  • the network side device uses the AI model, it can flexibly select from the M elements of the first domain according to the configuration information. N elements serve as input to the AI model.
  • the AI model can be used in the first communication device.
  • the AI model and the configuration information of the AI model can be configured by the second communication device to the first communication device.
  • the first communication device is a terminal
  • the second communication device is a network-side device
  • the first communication device is a network-side device
  • the second communication device is a terminal
  • the first communication device is a first terminal
  • the second communication device is a second terminal
  • the first communication device is a first network side device
  • the second communication device is a second network side device.
  • the AI model and the configuration information of the AI model can be configured to the terminal by the network side device, or configured to the terminal by other terminals.
  • the AI model and the configuration information of the AI model can be configured by the terminal to the network-side device, or by other network-side devices.
  • the second communication device can configure the AI model and the configuration information of the AI model to the first communication device at the same time.
  • the communication device may also separately configure the AI model and the configuration information of the AI model to the first communication device at different times.
  • the AI model and the configuration information of the AI model may be configured to the first communication device by the same second communication device, or may be configured to the first communication device by a different second communication device.
  • the network side device 1 can configure the AI model and the AI model configuration information to the terminal, or the network side device 1 can configure the AI model to the terminal, and the network side device 2 can configure the AI model to the terminal. Configure the configuration information of the AI model to the terminal.
  • the first communication device may receive the input from the second communication device before determining the input of the AI model based on the configuration information of the AI model.
  • the AI model and the configuration information of the AI model can first send the AI model and the configuration information of the AI model to the first communication device.
  • the first communication device uses the AI model to perform model prediction, it can make predictions based on the information received from the second communication device.
  • the configuration information of the AI model selects N elements from the M elements of the first domain as model inputs of the AI model.
  • the above-mentioned first domain may include at least one of the following:
  • Frequency domain time domain; spatial domain; Doppler domain; delay domain; beam domain.
  • the first communication device determines the model input of the AI model based on the configuration information, it can select from any one or more of the frequency domain, time domain, spatial domain, Doppler domain, delay domain, and beam domain. Select N elements from M elements as input to the AI model. Since the first communication device can select N elements from at least one domain of frequency domain, time domain, spatial domain, Doppler domain, delay domain, and beam domain as the input of the AI model, the efficiency of the AI model input can be further improved. Flexibility to improve the generalization ability of AI models.
  • OFDM Orthogonal Frequency Division Multiplexing
  • the M elements in the first domain can be determined in any of the following ways:
  • the first domain can be configured by the second communication device to the first communication device; if the first communication device is a network-side device, the second communication device If the first communication device is a terminal, the first domain can be reported to the first communication device by the second communication device; if the first communication device is a terminal or a network side device, the first domain can be customized by the first communication device, or agreed upon by the protocol, or Converted from the second domain.
  • the above-mentioned second domain may be provided by a designated module in the first communication device or provided by a designated module in the second communication device. That is to say, the input of the AI model is the output of other modules, the output of other modules is the second domain, and the other modules can be designated modules in the first communication device or designated modules in the second communication device. When other modules provide a second domain, the second domain needs to be converted into the first domain before it can be used as input to the AI model.
  • other modules provide air domain information and frequency domain information to the AI model, that is, the second domain is the air domain and frequency domain, and the first domain is the beam domain and time delay domain, so it is necessary to perform discrete Fourier analysis on the air domain of the second domain.
  • Discrete Fourier Transform DFT
  • IDFT the inverse transform of the discrete Fourier transform
  • the frequency domain that is, the spatial domain is converted into the beam domain and the frequency domain is converted into the delay domain before it can be used as Input to the AI model.
  • the selected N elements may include any of the following three situations:
  • N elements are N elements at the specified position in the first domain, or consecutive N elements at the specified position, or consecutive N elements at equal intervals at the specified position;
  • N elements are N elements at any position in the first domain, or consecutive N elements at any position, or consecutive N elements at equal intervals at any position;
  • N elements are N elements at any position in the designated area of the first domain, or consecutive N elements at any position, or consecutive N elements at equal intervals at any position; among them, the The total number of elements is greater than N and less than or equal to M.
  • the positions of the N elements in the first domain are specified and cannot be changed.
  • the first communication device may select N elements at corresponding positions from the first domain according to the specified position.
  • the positions of the N elements in the first domain may be indicated by configuration information.
  • the configuration information may include position information of each of the N elements in the first domain.
  • the first communication device selects N elements as inputs to the AI model according to the configuration information, it can select N elements at corresponding positions from the first domain according to the position information of the N elements indicated in the configuration information.
  • the configuration information may include the starting position information, and/or the intermediate position information and/or the ending position information of the N elements in the first domain, that is, the configuration
  • the information may include the position information of the first element among the N elements, and/or the position information of any one of the second to N-1 elements among the N elements, and/or the last element among the N elements. (i.e. the position information of the Nth element).
  • the first communication device selects N elements as input to the AI model according to the configuration information, it can select the element at that position and the consecutive N-1 elements after the element as input according to the starting position information indicated in the configuration information.
  • the input of the AI model, and/or, according to the intermediate position information indicated in the configuration information select N consecutive elements with the element at this position as the intermediate element as the input of the AI model, and/or, according to the end indicated in the configuration information
  • the position information selects the element at the position and the consecutive N-1 elements before the element as the input of the AI model.
  • the ID can be selected from the first domain.
  • the set of 5 consecutive RBs [1 2 3 4 5] is used as the input of the AI model.
  • the configuration information may include interval information of the N elements and starting position information and/or intermediate positions of the N elements in the first domain.
  • Information and/or end position information that is, the configuration information may include interval information of N elements, and position information of the first element among the N elements, and/or the second to N-1 elements among the N elements.
  • the first communication device selects N elements as input to the AI model based on the configuration information, it can select the element at that position and the consecutive equally spaced elements after the element based on the interval information and starting position information indicated in the configuration information.
  • N-1 elements are used as inputs to the AI model, and/ Or, according to the interval information and intermediate position information indicated in the configuration information, select N consecutive equally spaced elements with the element at that position as the intermediate element as the input of the AI model, and/or, according to the interval information indicated in the configuration information And the end position information selects the element at this position and the consecutive N-1 elements at equal intervals before the element as the input of the AI model.
  • the interval between N consecutive equally spaced elements can be a fixed value (such as 2), or any value in the preset interval (such as any integer in the [2,5] interval).
  • the configuration information indicates that the starting position information of the first RB among N RBs is 1, the interval information is 2 RBs, and N is 5, then you can select from the first domain.
  • the 5 equally spaced RBs with the ID set [1 3 5 7 9] are used as the input of the AI model.
  • the positions of the N elements in the first domain are arbitrary.
  • the selected N elements may be N elements at any position, or may be consecutive N elements at any position.
  • the selected N elements may be from the Starting from the X1th element of a domain, select the element at the position [X1X1+1X1+2...X1+N1-1] as the input of the AI model.
  • the value range of X1 is [1, N2-N1+1], or , the N elements selected from the first domain can also be N elements at equal intervals at any position, and the equal intervals can be fixed values or any values in the preset interval.
  • N elements are located in a designated area of the first domain.
  • the designated area may be indicated by configuration information.
  • the positions of the N elements are arbitrary.
  • the first communication device may select N elements from the designated area of the first domain.
  • the N elements may be N elements at any position in the designated area, or, alternatively, It can be N consecutive elements at any position in the specified area, or it can be N elements at equal intervals at any position in the specified area.
  • the intervals can be fixed values or any values in the preset interval. .
  • the first communication device can select any position from the designated area according to the configuration information of the AI model. 5 RBs, or 5 consecutive RBs, or 5 consecutive RBs at equal intervals are used as the input of the AI model.
  • the AI model in order to further improve the flexibility and generalization ability of the AI model, can cover the first domain.
  • the N elements need to meet at least one of the following conditions. item:
  • the N elements are N elements at any position in the first domain, or consecutive N elements at any position, or consecutive N elements at equal intervals at any position;
  • a collection of multiple N elements covers the first domain, and one of the N elements corresponds to one AI model.
  • any N elements can be selected from the M elements of the first domain as the input of the AI model.
  • the N elements It can be N elements at any position, or it can be consecutive N elements at any position, or it can be N consecutive equally spaced two elements at any position.
  • multiple AI models can be AI models with the same function or purpose.
  • the input of each AI model includes N elements.
  • the N elements can be N elements at specified positions or consecutive N elements or consecutive equally spaced N elements. elements, or it can be N elements at any position or N consecutive elements or N elements at consecutive equal intervals. It can also be N elements at any position in the specified area or N consecutive elements or N elements at consecutive equal intervals. elements, as long as the combination or collection of inputs from multiple AI models can cover the first domain.
  • the input of AI model 1 can be the 1st to 5th elements in the first domain
  • the input of AI model 2 can be the 1st to 5th element in the first domain. Any 5 consecutive elements among the 6th to 15th elements in a domain.
  • the input of AI model 3 can be any 5 elements among the 13th to 20th elements.
  • the set of input elements of these three AI models covers first domain.
  • the first communication device after the first communication device determines N elements, that is, after selecting N elements from the first domain according to the configuration information as inputs to the AI model, the first communication device can input the N elements into the AI model, so that the N elements can be input into the AI model based on the N elements. and AI models for model prediction.
  • the N elements when N elements are input into the AI model, the N elements can be sorted.
  • sorting the N elements at least one of the following can be included:
  • the N elements are sorted in order from large to small or in order from small to large.
  • the channel characteristics of an element may include at least one of the following: the power, amplitude, or phase of the information on the element; the correlation of the element with other elements.
  • the other elements may be other elements among the N elements or other elements among the M elements of the first domain.
  • the correlation can be cosine similarity, vector correlation, Euclidean distance, dispersion, etc., which are not listed here. This correlation can also be the correlation after the elements are projected into other spaces, such as high-dimensional space, kernel space, low-dimensional space, etc., which are not listed here.
  • the first communication device can perform model prediction based on the AI model.
  • the input of the AI model and the purpose of the AI model may include at least one of the following (1) to (9):
  • the AI model is used for signal processing.
  • the input of the AI model includes at least one of the following:
  • DMRS Demodulation Reference Signal
  • SRS Sounding Reference Signal
  • Synchronization signal and physical broadcast channel block Synchronization Signal and physical broadcast channel block (Synchronization Signal and PBCH block, SSB);
  • TRS Tracking Reference Signal
  • PTRS Phase Tracking Reference Signal
  • CSI-RS Channel State Information Reference Signal
  • the above signal may be an estimation result or detection result of the signal, and signal processing may include signal detection, filtering, equalization, etc.
  • the AI model is used for signal transmission, reception, demodulation or transmission.
  • the input of the AI model includes at least one of the following:
  • PDCCH Physical Downlink Control Channel
  • PDSCH Physical Downlink Shared Channel
  • PUCCH Physical Uplink Control Channel
  • PUSCH Physical Uplink Shared Channel
  • PRACH Physical Random Access Channel
  • PBCH Physical Broadcast Channel
  • the AI model is used to obtain channel state information.
  • the input of the AI model includes at least one of the following: Channel State Information (CSI); CSI-RS; SRS.
  • obtaining channel status information can include the following two scenarios:
  • the feedback channel state information may include channel related information, channel matrix related information, channel characteristic information, channel matrix characteristic information, precoding matrix indicator (Precoding Matrix Indicator, PMI), rank indicator (Rank Indication, RI), CSI-RS resources Indicator (CSI-RS Resource Indicator, CRI), channel quality indicator (Channel Quality Indicator, CQI), layer indicator (Layer Indicator, LI), etc.
  • Precoding Matrix Indicator, PMI Precoding Matrix Indicator, PMI
  • rank indicator Rank Indication, RI
  • CSI-RS resources Indicator CRI
  • channel quality indicator Channel Quality Indicator, CQI
  • Layer Indicator Layer Indicator, LI
  • FDD Frequency Division Duplexing
  • the network side device For FDD systems, based on the partial mutuality of uplink and downlink channels, the network side device obtains angle and delay information based on the uplink channel, and can notify the terminal of the angle information and delay information through CSI-RS precoding or direct instructions.
  • the terminal reports the channel state information according to the instructions of the network side device or selects and reports the channel state information within the instruction range of the network side device, thereby reducing the terminal's calculation amount and the overhead of CSI reporting.
  • the AI model is used for beam management, and the input of the AI model includes at least one of the following: beam quality; beam information.
  • Beam management may include beam measurement, beam reporting, beam prediction, beam failure detection, beam failure recovery, new beam indication in beam failure recovery, etc.
  • Beam quality can be the channel quality of various reference signals used for beam management, such as Reference Signal Received Power (RSRP), Reference Signal Received quality (Reference Signal Received) of SSB, CSI-RS, SRS and other reference signals. Quality (RSRQ), Signal-to-noise and Interference Ratio (SINR), etc.
  • the channel quality also includes the beam quality of layer 1 and/or the beam quality of layer 3.
  • the beam information may be beam ID, beam direction, beam precoding information (precoding vector, precoding matrix), etc.
  • Beam information can be divided by direction, such as horizontal beam ID, vertical beam ID, horizontal beam direction, vertical beam direction, etc.
  • the AI model is used for channel prediction.
  • the input of the AI model includes at least one of the following: channel information at historical moments; channel information at the current moment.
  • Channel prediction may include prediction of channel state information, beam prediction, etc.
  • the AI model is used for interference suppression.
  • the input of the AI model includes at least one of the following: signal; interference.
  • Interference may include intra-cell interference, inter-cell interference, out-of-band interference, intermodulation interference, etc.
  • the AI model is used for positioning.
  • the input of the AI model includes at least one of the following: channel information of the reference signal; information to assist position estimation or trajectory estimation.
  • Positioning can be the specific position (including horizontal position and/or vertical position) or possible future trajectories of the terminal estimated through reference signals (such as SRS, positioning reference signals, etc.), or information to assist position estimation or trajectory estimation (such as timing , timing advance, arrival time, arrival angle), etc.
  • reference signals such as SRS, positioning reference signals, etc.
  • information to assist position estimation or trajectory estimation such as timing , timing advance, arrival time, arrival angle, etc.
  • the AI model is used for prediction and management of high-level business and/or parameters.
  • the input of the AI model includes at least one of the following:
  • MAC Medium Access Control
  • High-level services and/or parameters may include throughput, required packet size, traffic requirements, movement speed, noise information, etc.
  • the AI model is used to parse control signaling.
  • the input of the AI model includes at least one of the following: signaling; control channel reception information.
  • the control signaling may be power control-related signaling, beam management-related signaling, etc.
  • the signaling included in the input of the AI model can be physical layer signaling, MAC layer signaling, Radio Resource Control (RRC) layer signaling, high-level signaling, etc.
  • the input of the AI model includes the control channel
  • the reception information may be reception information on PDCCH/PUCCH.
  • the communication device can determine the input of the AI model according to the configuration information of the AI model.
  • the configuration information is used to indicate that N elements are selected from the M elements of the first domain as the input of the AI model, and N is greater than Or an integer equal to 1, M is an integer greater than N.
  • the communication device can flexibly select N elements among the M elements of the first domain as inputs to the AI model, the communication device can be enabled to use as few as possible under different resource locations and/or resource numbers.
  • the AI model improves the flexibility and generalization capabilities of the AI model, and reduces the cost of the communication system.
  • the execution subject may be a model input determination device.
  • the method of determining the model input performed by the model input determining apparatus is used as an example to illustrate the model input determining apparatus provided by the embodiment of the present application.
  • Figure 3 is a schematic structural diagram of a device for determining model input according to an embodiment of the present application. This device may correspond to the first communication device in other embodiments. As shown in Figure 3, the device 300 includes the following modules.
  • Determination module 301 used to determine the input of the AI model according to the configuration information of the AI model.
  • the configuration information is used to indicate that N elements are selected from the first domain as the input of the AI model, and N is greater than or equal to 1. is an integer, the first domain includes M elements, and M is an integer greater than N.
  • the AI model is used in a first communication device, and the AI model and the configuration information are configured by the first communication device.
  • the first communication device is a terminal or a network. side equipment.
  • the AI model is used in the first communication device, and the AI model and the Configuration information is configured by the second communication device to the first communication device;
  • the first communication device is a terminal
  • the second communication device is a network side device
  • the first communication device is a network side device, and the second communication device is a terminal; or,
  • the first communication device is a first terminal
  • the second communication device is a second terminal
  • the first communication device is a first network side device
  • the second communication device is a second network side device.
  • the determination module 301 is also used to:
  • the first domain includes at least one of the following:
  • Frequency domain time domain; spatial domain; Doppler domain; delay domain; beam domain.
  • the first communication device supports the use of M elements in the first domain
  • the M elements in the first domain are configured by the second communication device to the first communication device, or reported to the first communication device by the second communication device, or automatically by the first communication device. It is defined, or agreed upon by a protocol, or converted from a second domain, which is provided by a designated module in the first communication device or provided by a designated module in the second communication device.
  • the N elements include any of the following:
  • the N elements are N elements at specified positions in the first domain, or consecutive N elements at designated positions, or consecutive N elements at equal intervals at designated positions;
  • the N elements are N elements at any position in the first domain, or consecutive N elements at any position, or consecutive N elements at equal intervals at any position;
  • the N elements are N elements at any position in the designated area of the first domain, or consecutive N elements at any position, or consecutive N elements at equal intervals at any position; wherein, in the designated area The total number of elements is greater than the N and less than or equal to the M.
  • the configuration information includes position information of each of the N elements in the first domain
  • the configuration information includes starting position information and/or intermediate position information of the N elements in the first domain. and/or end location information;
  • the configuration information includes interval information of the N elements and the location of the N elements in the first domain. start position information, and/or intermediate position information, and/or end position information.
  • the interval between the N consecutive equally spaced elements is a fixed value, or any value in a preset interval.
  • the N elements satisfy at least one of the following:
  • the N elements are N elements at any position in the first domain, or consecutive N elements at any position, or consecutive equally spaced elements at any position. N elements;
  • a collection of multiple N elements covers the first domain, and one N element corresponds to one AI model.
  • the determination module 301 is also used to:
  • the N elements are sorted, including at least one of the following:
  • the N elements in the first domain sort the N elements in order from large to small or in order from small to large according to the positions or identities;
  • the N elements are sorted in order from large to small or in order from small to large.
  • the channel characteristics of the elements include at least one of the following: Power, amplitude or phase; correlation of an element with other elements, other elements of the N elements or other elements of the M elements.
  • the AI model and the input of the AI model include at least one of the following:
  • the AI model is used for signal processing, and the input of the AI model includes at least one of the following: demodulation reference signal DMRS; sounding reference signal SRS; synchronization signal and physical broadcast channel block SSB; tracking reference signal TRS; phase tracking reference signal PTRS; channel state information reference signal CSI-RS;
  • the AI model is used for signal transmission, reception, demodulation or transmission.
  • the input of the AI model includes at least one of the following: physical downlink control channel PDCCH; physical downlink shared channel PDSCH; physical uplink control channel PUCCH; physical uplink shared channel PUSCH; physical random access channel PRACH; physical broadcast channel PBCH;
  • the AI model is used to obtain channel state information, and the input of the AI model includes at least one of the following: CSI; CSI-RS; SRS;
  • the AI model is used for beam management, and the input of the AI model includes at least one of the following: beam quality; beam information;
  • the AI model is used for channel prediction, and the input of the AI model includes at least one of the following: channel information at historical moments; channel information at the current moment;
  • the AI model is used for interference suppression, and the input of the AI model includes at least one of the following: signal; interference;
  • the AI model is used for positioning, and the input of the AI model includes at least one of the following: channel information of the reference signal; information to assist position estimation or trajectory estimation;
  • the AI model is used for prediction and management of high-level services and/or parameters, and the input of the AI model includes at least one of the following: high-level services and/or parameters; physical layer services and/or parameters; medium access control MAC layer business and/or parameters;
  • the AI model is used to parse control signaling, and the input of the AI model includes at least one of the following: signaling; control channel reception information.
  • the device 300 can refer to the process of the method 200 corresponding to the embodiment of the present application, and each unit/module in the device 300 and the above-mentioned other operations and/or functions are respectively intended to implement the corresponding steps in the method 200.
  • the process can achieve the same or equivalent technical effect. For the sake of simplicity, it will not be described again here.
  • the device for determining model input in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or may be a component in the electronic device, such as an integrated circuit or chip.
  • the electronic device may be a terminal or other devices other than the terminal.
  • terminals may include but are not limited to the types of terminals 11 listed above, and other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., which are not specifically limited in the embodiment of this application.
  • NAS Network Attached Storage
  • the model input determination device provided by the embodiment of the present application can implement each process implemented by the method embodiment in Figure 2 and achieve the same technical effect. To avoid duplication, the details will not be described here.
  • this embodiment of the present application also provides a communication device 400, which includes a processor 401 and a memory 402.
  • the memory 402 stores programs or instructions that can be run on the processor 401, for example.
  • the communication device 400 is a terminal, when the program or instruction is executed by the processor 401, each step of the above-mentioned model input determination method embodiment is implemented, and the same technical effect can be achieved.
  • the communication device 400 is a network-side device, when the program or instruction is executed by the processor 401, each step of the above-mentioned model input determining device method embodiment is implemented, and the same technical effect can be achieved. To avoid duplication, it will not be described again here. .
  • An embodiment of the present application also provides a communication device, including a processor and a communication interface.
  • the processor is configured to determine the input of the AI model according to the configuration information of the AI model.
  • the configuration information is used to indicate selecting N from the first domain. elements serve as inputs to the AI model, N is an integer greater than or equal to 1, the first domain includes M elements, and M is an integer greater than N.
  • This communication device embodiment corresponds to the above-mentioned first communication device method embodiment.
  • FIG. 5 is a schematic diagram of the hardware structure of a communication device that implements an embodiment of the present application.
  • the communication device 500 includes but is not limited to: radio frequency unit 501, network module 502, audio output unit 503, input unit 504, sensor 505, display unit 506, user input unit 507, interface unit 508, memory 509, processor 510, etc. at least some parts of it.
  • the communication device 500 may also include a power supply (such as a battery) that supplies power to various components.
  • the power supply may be logically connected to the processor 510 through a power management system, thereby managing charging, discharging, and function through the power management system. Consumption management and other functions.
  • the structure of the communication device shown in Figure 5 does not constitute a limitation on the communication device.
  • the communication device may include more or less components than shown in the figure, or combine certain components, or arrange different components, which will not be described again here. .
  • the input unit 504 may include a graphics processing unit (GPU) 5041 and a microphone 5042.
  • the GPU 5041 is used for recording data generated by an image capture device (such as a camera) in the video capture mode or the image capture mode. ) to process the image data of still pictures or videos obtained.
  • the display unit 506 may include a display panel 5061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
  • the user input unit 507 includes a touch panel 5071 and at least one of other input devices 5072 .
  • Touch panel 5071 also called touch screen.
  • the touch panel 5071 may include two parts: a touch detection device and a touch controller.
  • Other input devices 5072 may include, but are not limited to, physical keyboards, function keys (such as volume control keys, switch keys etc.), trackball, mouse, and joystick, which will not be described in detail here.
  • the radio frequency unit 501 after receiving downlink data from the network side device, the radio frequency unit 501 can transmit it to the processor 510 for processing; in addition, the radio frequency unit 501 can send uplink data to the network side device.
  • the radio frequency unit 501 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low noise amplifier, duplexer, etc.
  • Memory 509 may be used to store software programs or instructions as well as various data.
  • the memory 509 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, Image playback function, etc.) etc.
  • memory 509 may include volatile memory or non-volatile memory, or memory 509 may include both volatile and non-volatile memory.
  • non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically removable memory. Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (Synch link DRAM) , SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DRRAM).
  • RAM Random Access Memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM Double Data Rate SDRAM
  • DDRSDRAM double data rate synchronous dynamic random access memory
  • Enhanced SDRAM, ESDRAM enhanced synchronous dynamic random access memory
  • Synch link DRAM synchronous link dynamic random access memory
  • SLDRAM direct memory bus
  • the processor 510 may include one or more processing units; optionally, the processor 510 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 510.
  • the processor 510 is configured to determine the input of the AI model according to the configuration information of the AI model.
  • the configuration information is used to indicate that N elements are selected from the first domain as the input of the AI model, and N is greater than or An integer equal to 1, the first domain includes M elements, and M is an integer greater than N.
  • the communication device can determine the input of the AI model according to the configuration information of the AI model.
  • the configuration information is used to indicate that N elements are selected from the M elements of the first domain as the input of the AI model, and N is greater than Or an integer equal to 1, M is an integer greater than N.
  • the communication device can flexibly select N elements among the M elements of the first domain as inputs to the AI model, the communication device can be enabled to use as few as possible under different resource locations and/or resource numbers.
  • the AI model improves the flexibility and generalization capabilities of the AI model, and reduces the cost of the communication system.
  • the communication device 500 provided by the embodiment of the present application can also implement each process of the above-mentioned method embodiment 200, and can achieve the same technical effect. To avoid duplication, the details will not be described here.
  • An embodiment of the present application also provides a communication device, including a processor and a communication interface.
  • the processor is configured to determine the input of the AI model according to the configuration information of the AI model.
  • the configuration information is used to indicate selecting N from the first domain.
  • N is an integer greater than or equal to 1
  • the first domain includes M elements
  • M is an integer greater than N.
  • This communication device embodiment corresponds to the above-mentioned first communication device method embodiment.
  • Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this communication device embodiment, and can achieve the same technical effect.
  • the embodiment of the present application also provides a communication device.
  • the communication device 600 includes: an antenna 61 , a radio frequency device 62 , a baseband device 63 , a processor 64 and a memory 65 .
  • the antenna 61 is connected to the radio frequency device 62 .
  • the radio frequency device 62 receives information through the antenna 61 and sends the received information to the baseband device 63 for processing.
  • the baseband device 63 processes the information to be sent and sends it to the radio frequency device 62.
  • the radio frequency device 62 processes the received information and then sends it out through the antenna 61.
  • the method performed by the first communication device in the above embodiment can be implemented in the baseband device 63, which includes a baseband processor.
  • the baseband device 63 may include, for example, at least one baseband board on which multiple chips are disposed, as shown in FIG.
  • the program performs the first communication device operation shown in the above method embodiment.
  • the communication device may also include a network interface 66, such as a common public radio interface (CPRI).
  • a network interface 66 such as a common public radio interface (CPRI).
  • CPRI common public radio interface
  • the communication device 600 of the embodiment of the present invention also includes: instructions or programs stored in the memory 65 and executable on the processor 64.
  • the processor 64 calls the instructions or programs in the memory 65 to execute the modules shown in Figure 6 The implementation method and achieve the same technical effect will not be repeated here to avoid repetition.
  • Embodiments of the present application also provide a readable storage medium.
  • Programs or instructions are stored on the readable storage medium.
  • the program or instructions are executed by a processor, each process of the above-mentioned model input determination method embodiment is implemented, and can To achieve the same technical effect, to avoid repetition, we will not repeat them here.
  • the processor is the processor in the terminal described in the above embodiment.
  • the readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.
  • An embodiment of the present application further provides a chip.
  • the chip includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the method for determining the model input.
  • Each process in the example can achieve the same technical effect. To avoid repetition, we will not repeat it here.
  • chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.
  • Embodiments of the present application further provide a computer program/program product.
  • the computer program/program product is stored in a storage medium.
  • the computer program/program product is executed by at least one processor to implement the above method for determining model input.
  • Each process of the embodiment can achieve the same technical effect, so to avoid repetition, it will not be described again here.
  • the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation.
  • the technical solution of the present application can be embodied in the form of a computer software product that is essentially or contributes to the existing technology.
  • the computer software product is stored in a storage medium (such as ROM/RAM, disk , CD), including several instructions to cause a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of this application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种模型输入的确定方法及通信设备,属于通信技术领域,本申请实施例的模型输入的确定方法包括:第一通信设备根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N个元素作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。

Description

模型输入的确定方法及通信设备
交叉引用
本申请要求在2022年04月14日提交中国专利局、申请号为202210390701.5、名称为“模型输入的确定方法及通信设备”的中国专利申请的优先权,该申请的全部内容通过引用结合在本申请中。
技术领域
本申请属于通信技术领域,具体涉及一种模型输入的确定方法及通信设备。
背景技术
在通信系统中,通信设备(比如终端或网络侧设备)通常可以使用人工智能(Artificial Intelligence,AI)模型对输入的资源进行模型预测,模型预测的结果可以辅助通信设备进行信道估计、信号处理等。
目前,通信设备在进行使用AI模型进行模型预测时,模型的输入通常是固定的资源位置和/或资源数目。然而,在实际的通信系统中,通信设备可使用的资源并非是固定的资源位置和/或资源数据,这样,当通信设备可使用的资源与AI模型的输入不一致时,将无法使用AI模型进行模型预测,导致AI模型的灵活性和泛化能力较差。
发明内容
本申请实施例提供一种模型输入的确定方法及通信设备,能够解决目前的AI模型灵活性和泛化能力较差的问题。
第一方面,提供了一种模型输入的确定方法,该方法包括:第一通信设备根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N个元素作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。
第二方面,提供了一种模型输入的确定装置,该装置包括:确定模块,用于根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N个元素作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。
第三方面,提供了一种通信设备,该通信设备包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤。
第四方面,提供了一种通信设备,包括处理器及通信接口,其中,所述处理器用于根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N 个元素作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。
第五方面,提供了一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤。
第六方面,提供了一种芯片,所述芯片包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现如第一方面所述的方法。
第七方面,提供了一种计算机程序/程序产品,所述计算机程序/程序产品被存储在存储介质中,所述计算机程序/程序产品被至少一个处理器执行以实现如第一方面所述的模型输入的确定方法的步骤。
在本申请实施例中,通信设备可以根据AI模型的配置信息确定AI模型的输入,该配置信息用于指示从第一域的M个元素中选择N个元素作为AI模型的输入,N为大于或等于1的整数,M为大于N的整数。这样,由于通信设备可以在第一域的M个元素中灵活选择N个元素作为AI模型的输入,因此,可以使得通信设备能够在不同的资源位置和/或资源数目的情况下,使用尽量少的AI模型,提升了AI模型的使用灵活性和泛化能力,降低了通信系统的开销。
附图说明
图1是根据本申请实施例的无线通信系统的示意图;
图2是根据本申请实施例的模型输入的确定方法的示意性流程图;
图3是根据本申请实施例的模型输入的确定装置的结构示意图;
图4是根据本申请实施例的通信设备的结构示意图;
图5是根据本申请实施例的通信设备的结构示意图;
图6是根据本申请实施例的通信设备的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚描述,显然,所描述的实施例是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员所获得的所有其他实施例,都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象,而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换,以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施,且“第一”、“第二”所区别的对象通常为一类,并不限定对象的个数,例如第一对象可以是一个,也可以是多个。此外,说明书以及权利要求中“和/或”表示所连接对象的至少其中之一,字符“/”一般表示前后关联对象是一种“或”的关系。
值得指出的是,本申请实施例所描述的技术不限于长期演进型(Long Term Evolution, LTE)/LTE的演进(LTE-Advanced,LTE-A)系统,还可用于其他无线通信系统,诸如码分多址(Code Division Multiple Access,CDMA)、时分多址(Time Division Multiple Access,TDMA)、频分多址(Frequency Division Multiple Access,FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access,OFDMA)、单载波频分多址(Single-carrier Frequency Division Multiple Access,SC-FDMA)和其他系统。本申请实施例中的术语“系统”和“网络”常被可互换地使用,所描述的技术既可用于以上提及的系统和无线电技术,也可用于其他系统和无线电技术。以下描述出于示例目的描述了新空口(New Radio,NR)系统,并且在以下大部分描述中使用NR术语,但是这些技术也可应用于NR系统应用以外的应用,如第6代(6th Generation,6G)通信系统。
图1示出本申请实施例可应用的一种无线通信系统的框图。无线通信系统包括终端11和网络侧设备12。其中,终端11可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant,PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer,UMPC)、移动上网装置(Mobile Internet Device,MID)、增强现实(augmented reality,AR)/虚拟现实(virtual reality,VR)设备、机器人、可穿戴式设备(Wearable Device)、车载设备(VUE)、行人终端(PUE)、智能家居(具有无线通信功能的家居设备,如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer,PC)、柜员机或者自助机等终端侧设备,可穿戴式设备包括:智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是,在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备,其中,接入网设备12也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备12可以包括基站、无线局域网(Wireless Local Area Network,WLAN)接入点或WiFi节点等,基站可被称为节点B、演进节点B(eNB)、接入点、基收发机站(Base Transceiver Station,BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set,BSS)、扩展服务集(Extended Service Set,ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point,TRP)或所述领域中其他某个合适的术语,只要达到相同的技术效果,所述基站不限于特定技术词汇,需要说明的是,在本申请实施例中仅以NR系统中的基站为例进行介绍,并不限定基站的具体类型。
下面结合附图,通过一些实施例及其应用场景对本申请实施例提供的模型输入的确定方法及通信设备进行详细地说明。
如图2所示,本申请实施例提供一种模型输入的确定方法200,该方法可以由第一通信设备执行,该第一通信设备可以是图1所示实施例中的终端或网络侧设备,换言之,该方法可以由安装在终端或网络侧设备的软件或硬件来执行,该方法包括如下步骤。
S202:第一通信设备根据AI模型的配置信息确定AI模型的输入,配置信息用于指 示从第一域中选择N个元素作为AI模型的输入,N为大于或等于1的整数,第一域中包括M个元素,M为大于N的整数。
本实施例中,第一通信设备在使用AI模型进行模型预测时,可以根据AI模型的配置信息从第一域的M个元素中选择N个元素作为AI模型的输入,N为大于或等于1的整数,M为大于N的整数。这样,由于第一通信设备可以在第一域的M个元素中灵活选择N个元素作为AI模型的输入,使得AI模型的输入不再是固定的资源位置和/或资源数目,因此,可以使得通信设备能够在不同的资源位置和/或资源数目的情况下,使用尽量少的AI模型,提升了AI模型的使用灵活性和泛化能力,降低了通信系统的开销。
可选地,作为一个实施例,AI模型可以在第一通信设备中使用,在这种情况下,AI模型和AI模型的配置信息可以均由第一通信设备配置得到。其中,第一通信设备可以是终端或网络侧设备。
也就是说,在终端使用AI模型的情况下,终端可以配置AI模型和AI模型的配置信息,终端在使用AI模型时,可以根据配置信息灵活地从第一域的M个元素中选择N个元素作为AI模型的输入。在网络侧设备使用AI模型的情况下,网络侧设备可以配置AI模型和AI模型的配置信息,网络侧设备在使用AI模型时,可以根据配置信息灵活地从第一域的M个元素中选择N个元素作为AI模型的输入。
可选地,作为一个实施例,AI模型可以在第一通信设备中使用,在这种情况下,AI模型和AI模型的配置信息可以由第二通信设备配置给第一通信设备。其中,第一通信设备为终端,第二通信设备为网络侧设备;或,第一通信设备为网络侧设备,第二通信设备为终端;或,第一通信设备为第一终端,第二通信设备为第二终端;或,第一通信设备为第一网络侧设备,第二通信设备为第二网络侧设备。
也就是说,在终端使用AI模型的情况下,AI模型和AI模型的配置信息可以均由网络侧设备配置给终端,或由其他终端配置给该终端。在网络侧设备使用AI模型的情况下,AI模型和AI模型的配置信息可以均由终端配置给网络侧设备,或由其他网络侧设备配置给该网络侧设备。
需要说明的是,在AI模型和AI模型的配置信息由第二通信设备配置给第一通信设备的情况下,第二通信设备可以在同一时间将AI模型和AI模型的配置信息配置给第一通信设备,也可以在不同时间将AI模型和AI模型的配置信息分开配置给第一通信设备。此外,AI模型和AI模型的配置信息可以由同一个第二通信设备配置给第一通信设备,也可以由不同的第二通信设备配置给第一通信设备,比如,在AI模型和AI模型的配置信息由网络侧设备配置给终端的情况下,网络侧设备1可以将AI模型和AI模型配置信息配置给终端,或者,也可以是网络侧设备1将AI模型配置给终端,网络侧设备2将AI模型的配置信息配置给终端。
在AI模型和AI模型的配置信息由第二通信设备配置给第一通信设备的情况下,第一通信设备在根据AI模型的配置信息确定AI模型的输入之前,可以接收第二通信设备发送 的AI模型和AI模型的配置信息。也就是说,第二通信设备可以先将AI模型和AI模型的配置信息发送给第一通信设备,当第一通信设备在使用AI模型进行模型预测时,可以根据从第二通信设备接收到的AI模型的配置信息从第一域的M个元素中选择N个元素作为AI模型的模型输入。
可选地,作为一个实施例,上述第一域可以包括以下至少一项:
频域;时域;空域;多普勒域;时延域;波束域。
也就是说,第一通信设备在根据配置信息确定AI模型的模型输入时,可以从频域、时域、空域、多普勒域、时延域、波束域中的任一个或多个域的M个元素中选择N个元素作为AI模型的输入。由于第一通信设备可以从频域、时域、空域、多普勒域、时延域、波束域的至少一个域中选择N个元素作为AI模型的输入,因此,可以进一步提高AI模型输入的灵活性,提高AI模型的泛化能力。
本实施例中,第一通信设备支持使用第一域中的M个元素,即第一域中的M个元素为第一通信设备的工作范围。比如,若第一域为频域,第一通信设备在频域上支持50个资源块(Resource Blocks,RB),则M=50,再比如,若第一域为时域,第一通信设备在时域上支持12个正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)符号,则M=12。
可选地,作为一个实施例,第一域中的M个元素可以通过以下任一种方式确定得到:
由第二通信设备配置给第一通信设备;
由第二通信设备上报给第一通信设备;
由第一通信设备自定义;
由协议约定;
由第二域转换得到,第二域由第一通信设备中的指定模块提供或由第二通信设备中的指定模块提供。
比如,若第一通信设备为终端,第二通信设备为网络侧设备,则第一域可以由第二通信设备配置给第一通信设备;若第一通信设备为网络侧设备,第二通信设备为终端,则第一域可以由第二通信设备上报给第一通信设备;若第一通信设备为终端或网络侧设备,第一域可以由第一通信设备自定义,或由协议约定,或由第二域转换得到。
上述第二域可以由第一通信设备中的指定模块提供或由第二通信设备中的指定模块提供。也就是说,AI模型的输入是其他模块的输出,其他模块的输出为第二域,其他模块可以是第一通信设备中的指定模块或第二通信设备中的指定模块。在其他模块提供第二域的情况下,需要将第二域转换为第一域后才可以作为AI模型的输入。
比如,其它模块提供给AI模型的是空域信息和频域信息,即第二域是空域和频域,而第一域是波束域和时延域,则需要对第二域的空域做离散傅里叶变换(Discrete Fourier Transform,DFT)、频域做离散傅里叶变换的逆变换IDFT(Inverse Discrete Fourier Transform,IDFT),即将空域转为波束域以及将频域转为时延域后才能作为AI模型的输入。
本实施例中,第一通信设备在根据配置信息从第一域的M个元素中选择N个元素时,选择的N个元素可以包括以下三种情况中的任一项:
第一种情况:N个元素为第一域中指定位置的N个元素,或指定位置的连续N个元素,或指定位置的连续等间隔的N个元素;
第二种情况:N个元素为第一域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;
第三种情况:N个元素为第一域的指定区域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;其中,指定区域中的元素总数目大于N且小于等于M。
在上述第一种情况中,N个元素在第一域中的位置是指定的,不能变化的。第一通信设备可以根据该指定的位置从第一域中选择对应位置上的N个元素。其中,N个元素在第一域中的位置可以由配置信息指示。
具体地,在N个元素为指定位置的N个元素的情况下,配置信息中可以包括N个元素中每个元素在第一域中的位置信息。这样,第一通信设备在根据配置信息选择N个元素作为AI模型的输入时,可以根据配置信息中指示的N个元素的位置信息从第一域中选择对应位置上的N个元素。
在N个元素为指定位置的连续N个元素的情况下,配置信息中可以包括N个元素在第一域中的起始位置信息、和/或中间位置信息和/或结束位置信息,即配置信息中可以包括N个元素中第一个元素的位置信息,和/或N个元素中第二至第N-1个元素中任一元素的位置信息,和/或N个元素中最后一个元素(即第N个元素)的位置信息。这样,第一通信设备在根据配置信息选择N个元素作为AI模型的输入时,可以根据配置信息中指示的起始位置信息选择该位置上的元素及该元素之后的连续N-1个元素作为AI模型的输入,和/或,根据配置信息中指示的中间位置信息选择以该位置上的元素作为中间元素的连续N个元素作为AI模型的输入,和/或,根据配置信息中指示的结束位置信息选择该位置上的元素及该元素之前的连续N-1元素作为AI模型的输入。
比如,第一域中的元素为RB,配置信息指示N个RB中的第一个RB的起始位置信息(即起始ID)为1,N为5,则可以从第一域中选择ID集合为[1 2 3 4 5]的连续5个RB作为AI模型的输入。
在N个元素为指定位置的连续等间隔的N个元素的情况下,配置信息中可以包括N个元素的间隔信息以及N个元素在第一域中的起始位置信息、和/或中间位置信息和/或结束位置信息,即配置信息中可以包括N个元素的间隔信息,以及N个元素中第一个元素的位置信息,和/或N个元素中第二至第N-1个元素中任一元素的位置信息,和/或N个元素中最后一个元素(即第N个元素)的位置信息。这样,第一通信设备在根据配置信息选择N个元素作为AI模型的输入时,可以根据配置信息中指示的间隔信息和起始位置信息选择该位置上的元素及该元素之后的连续等间隔的N-1个元素作为AI模型的输入,和/ 或,根据配置信息中指示的间隔信息和中间位置信息选择以该位置上的元素作为中间元素的连续等间隔的N个元素作为AI模型的输入,和/或,根据配置信息中指示的间隔信息和结束位置信息选择该位置上的元素及该元素之前的连续等间隔的N-1元素作为AI模型的输入。其中,连续等间隔的N个元素之间的间隔可以是固定值(比如2),或是预设区间中的任意值(比如[2,5]区间内的任一整数)。
比如,第一域中的元素为RB,配置信息指示N个RB中的第一个RB的起始位置信息为1,间隔信息为2个RB,N为5,则可以从第一域中选择ID集合为[1 3 5 7 9]的等间隔的5个RB作为AI模型的输入。
在上述第二种情况中,N个元素在第一域中的位置是任意的。第一通信设备在从第一域中选择N个元素时,选择的N个元素可以是任意位置上的N个元素,或者,也可以是任意位置上的连续N个元素,比如,可以从第一域的第X1个元素开始,选择[X1X1+1X1+2…X1+N1-1]位置上的元素作为AI模型的输入,X1的取值范围是[1,N2-N1+1],或者,从第一域中选择的N个元素还可以是任意位置上的等间隔的N个元素,该等间隔可以是固定值,或是预设区间中的任意值。
在上述第三种情况中,N个元素位于第一域的指定区域中,指定区域可以由配置信息指示,在该指定区域中,N个元素的位置是任意的。第一通信设备在从第一域中选择N个元素时,可以从第一域的指定区域中选择N个元素,该N个元素可以是指定区域中任意位置上的N个元素,或者,也可以是指定区域中任意位置上的连续N个元素,或者,还可以是指定区域中任意位置上的等间隔的N个元素,该等间隔可以是固定值,或是预设区间中的任意值。
比如,第一域中有M个元素,从这M个元素中选一个指定区域,指定区域包含M1个元素,M1<=M。若第一域中包括20个RB,从20RB中选择第5个RB至第12个RB为指定区域,则第一通信设备可以根据AI模型的配置信息,从该指定区域中选择任意位置上的5个RB、或连续5个RB、或连续等间隔的5个RB作为AI模型的输入。
可选的,作为一个实施例,为了进一步提高AI模型的灵活性和泛化能力,AI模型可以覆盖(cover)第一域,在这种情况下,N个元素需要满足以下条件中的至少一项:
在AI模型的个数为1的情况下,N个元素为第一域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;
在AI模型的个数为多个的情况下,多个N个元素的合集覆盖第一域,一个所述N个元素对应一个AI模型。
也就是说,在AI模型的个数为1的情况下,为了使AI模型覆盖第一域,可以从第一域的M个元素中选择任意N个元素作为AI模型的输入,该N个元素可以是任意位置上的N个元素,或者,也可以是任意位置上的连续N个元素,或者,还可以是任意位置上的连续等间隔的N个二元素。
在AI模型的个数为多个的情况下,多个AI模型的输入的组合或合集可以覆盖第一域, 其中,多个AI模型可以是功能或用途相同的AI模型,每个AI模型的输入包括N个元素,该N个元素可以是指定位置的N个元素或连续N个元素或连续等间隔的N个元素,也可以是任意位置的N个元素或连续N个元素或连续等间隔的N个元素,还可以是指定区域中任意位置的N个元素或连续N个元素或连续等间隔的N个元素,只要保证多个AI模型的输入的组合或合集能够覆盖第一域即可。
比如,在AI模型的个数为3、第一域中包括20个元素的情况下,AI模型1的输入可以是第一域中的第1~5个元素,AI模型2的输入可以是第一域中的第6~15个元素中的任意5个连续元素,AI模型3的输入可以是第13~20个元素中的任意5个元素,这三个AI模型的输入元素的合集覆盖了第一域。
本实施例中,第一通信设备在确定N个元素后,即根据配置信息从第一域中选择N个元素作为AI模型的输入后,可以将N个元素输入AI模型,以便基于N个元素和AI模型进行模型预测。
可选地,作为一个实施例,在将N个元素输入AI模型时,可以对N个元素进行排序,在对N个元素进行排序时,可以包括以下至少一项:
根据N个元素在第一域中的位置或标识,按照位置或标识从大到小的顺序或从小到大的顺序对N个元素进行排序;
根据N个元素的信道特征,按照信道特征从大到小的顺序或从小到大的顺序对N个元素进行排序。
元素的信道特征可以包括以下至少一项:元素上信息的功率、幅度或相位;元素与其他元素的相关性。其中,该其他元素可以是N个元素中的其他元素或第一域的M个元素中的其他元素。该相关性可以是余弦相似度、向量相关性、欧氏距离、离散度等,在此不一一列举。该相关性,也可以是所述元素投影到其它空间后的相关性,如高维空间、核空间、低维空间等,在此不一一列举。
本实施例中,第一通信设备在将N个元素输入AI模型后,可以基于AI模型进行模型预测。其中,可选地,作为一个实施例,AI模型的输入和AI模型的用途可以包括以下(1)至(9)中的至少一项:
(1)AI模型用于信号处理,AI模型的输入包括以下至少一项:
解调参考信号(Demodulation Reference Signal,DMRS);
探测参考信号(Sounding Reference Signal,SRS);
同步信号和物理广播信道块((Synchronization Signal and PBCH block,SSB);
跟踪参考信号(Tracking Reference Signal,TRS);
相位跟踪参考信号(Phase Tracking Reference Signal,PTRS);
信道状态信息参考信号(Channel State Information Reference Signa,CSI-RS)。
其中,以上信号可以是信号的估计结果或检测结果,信号处理可以包括信号检测、滤波、均衡等。
(2)AI模型用于信号传输、接收、解调或发送,AI模型的输入包括以下至少一项:
物理下行控制信道(Physical Downlink Control Channel,PDCCH);
物理下行共享信道(Physical Downlink Shared Channel,PDSCH);
物理上行控制信道(Physical Uplink Control Channel,PUCCH);
物理上行共享信道(Physical Uplink Shared Channel,PUSCH);
物理随机接入信道(Physical Random Access Channel,PRACH);
物理广播信道(Physical Broadcast Channel,PBCH)。
(3)AI模型用于获取信道状态信息,AI模型的输入包括以下至少一项:信道状态信息(Channel State Information,CSI);CSI-RS;SRS。
其中,获取信道状态信息可以包括以下两种场景:
A)信道状态信息反馈。反馈的信道状态信息可以包括信道相关信息、信道矩阵相关信息、信道特征信息、信道矩阵特征信息、预编码矩阵指示(Precoding Matrix Indicator,PMI)、秩指示(Rank Indication,RI)、CSI-RS资源指示(CSI-RS Resource Indicator,CRI)、信道质量指示(Channel Quality Indicator,CQI)、层指示(Layer Indicator,LI)等。
B)频分双工(Frequency Division Duplexing,FDD)上下行部分互易性。
对于FDD系统,根据上下行信道的部分互异性,网络侧设备根据上行信道获取角度和时延信息,可以通过CSI-RS预编码或者直接指示的方式,将角度信息和时延信息通知给终端,终端根据网络侧设备的指示上报或者在网络侧设备的指示范围内选择并上报信道状态信息,从而减少终端的计算量和CSI上报的开销。
(4)AI模型用于波束管理,AI模型的输入包括以下至少一项:波束质量;波束信息。
波束管理可以包括波束测量、波束上报、波束预测、波束失败检测、波束失败恢复、波束失败恢复中的新波束指示等。
波束质量可以是各类用于波束管理的参考信号的信道质量,如SSB、CSI-RS、SRS等参考信号的参考信号接收功率(Reference Signal Received Power,RSRP)、参考信号接收质量(Reference Signal Received Quality,RSRQ)、信号与干扰加噪声比(Signal-to-noise and Interference Ratio,SINR)等,所述信道质量还包括层1的波束质量和/或层3的波束质量。
波束信息可以是波束ID、波束方向、波束的预编码信息(预编码向量、预编码矩阵)等。波束信息可以按方向划分,比如水平维波束ID、垂直维波束ID、水平维波束方向、垂直维波束方向等。
(5)AI模型用于信道预测,AI模型的输入包括以下至少一项:历史时刻的信道信息;当前时刻的信道信息。
信道预测可以包括信道状态信息的预测、波束预测等。
(6)AI模型用于干扰抑制,AI模型的输入包括以下至少一项:信号;干扰。
干扰可以包括小区内干扰、小区间干扰、带外干扰、交调干扰等等。
(7)AI模型用于定位,AI模型的输入包括以下至少一项:参考信号的信道信息;辅助位置估计或轨迹估计的信息。
定位可以是通过参考信号(例如SRS、定位参考信号等)估计出的终端的具体位置(包括水平位置和/或垂直位置)或未来可能的轨迹,或辅助位置估计或轨迹估计的信息(如定时、定时提前、到达时间、到达角度)等。
(8)AI模型用于高层业务和/或参数的预测和管理,AI模型的输入包括以下至少一项:
高层业务和/或参数;
物理层的业务和/或参数;
介质访问控制(Medium Access Control,MAC)层的业务和/或参数。
高层业务和/或参数可以包括吞吐量、所需数据包大小、业务需求、移动速度、噪声信息等等。
(9)AI模型用于解析控制信令,AI模型的输入包括以下至少一项:信令;控制信道的接收信息。
控制信令可以是功率控制的相关信令,波束管理的相关信令等。AI模型的输入中包括的信令可以是物理层信令、MAC层信令、无线资源控制(Radio Resource Control,RRC)层信令、高层信令等,AI模型的输入中包括的控制信道的接收信息可以是PDCCH/PUCCH上的接收信息。
在本申请实施例中,通信设备可以根据AI模型的配置信息确定AI模型的输入,该配置信息用于指示从第一域的M个元素中选择N个元素作为AI模型的输入,N为大于或等于1的整数,M为大于N的整数。这样,由于通信设备可以在第一域的M个元素中灵活选择N个元素作为AI模型的输入,因此,可以使得通信设备能够在不同的资源位置和/或资源数目的情况下,使用尽量少的AI模型,提升了AI模型的使用灵活性和泛化能力,降低了通信系统的开销。
本申请实施例提供的模型输入的确定方法,执行主体可以为模型输入的确定装置。本申请实施例中以模型输入的确定装置执行模型输入的确定方法为例,说明本申请实施例提供的模型输入的确定装置。
图3是根据本申请实施例的模型输入的确定装置的结构示意图,该装置可以对应于其他实施例中的第一通信设备。如图3所示,装置300包括如下模块。
确定模块301,用于根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N个元素作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。
可选的,作为一个实施例,所述AI模型在第一通信设备中使用,所述AI模型和所述配置信息由所述第一通信设备配置得到,所述第一通信设备为终端或网络侧设备。
可选的,作为一个实施例,所述AI模型在第一通信设备中使用,所述AI模型和所述 配置信息由第二通信设备配置给所述第一通信设备;
其中,所述第一通信设备为终端,所述第二通信设备为网络侧设备;或,
所述第一通信设备为网络侧设备,所述第二通信设备为终端;或,
所述第一通信设备为第一终端,所述第二通信设备为第二终端;或,
所述第一通信设备为第一网络侧设备,所述第二通信设备为第二网络侧设备。
可选的,作为一个实施例,所述确定模块301,还用于:
接收所述第二通信设备发送的所述AI模型和所述配置信息。
可选的,作为一个实施例,所述第一域包括以下至少一项:
频域;时域;空域;多普勒域;时延域;波束域。
可选的,作为一个实施例,第一通信设备支持使用所述第一域中的M个元素;
其中,所述第一域中的M个元素由第二通信设备配置给所述第一通信设备,或由第二通信设备上报给所述第一通信设备,或由所述第一通信设备自定义,或由协议约定,或由第二域转换得到,所述第二域由所述第一通信设备中的指定模块提供或由第二通信设备中的指定模块提供。
可选的,作为一个实施例,所述N个元素包括以下任一项:
所述N个元素为所述第一域中指定位置的N个元素,或指定位置的连续N个元素,或指定位置的连续等间隔的N个元素;
所述N个元素为所述第一域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;
所述N个元素为所述第一域的指定区域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;其中,所述指定区域中的元素总数目大于所述N且小于等于所述M。
可选的,作为一个实施例,包括以下任一项:
在所述N个元素为所述指定位置的N个元素的情况下,所述配置信息中包括所述N个元素中每个元素在所述第一域中的位置信息;
在所述N个元素为所述指定位置的连续N个元素的情况下,所述配置信息中包括所述N个元素在所述第一域中的起始位置信息、和/或中间位置信息和/或结束位置信息;
在所述N个元素为所述指定位置的连续等间隔的N个元素的情况下,所述配置信息中包括所述N个元素的间隔信息以及所述N个元素在所述第一域中的起始位置信息、和/或中间位置信息和/或结束位置信息。
可选的,作为一个实施例,所述连续等间隔的N个元素之间的间隔为固定值,或为预设区间中的任意值。
可选的,作为一个实施例,所述N个元素满足以下至少一项:
在所述AI模型的个数为1的情况下,所述N个元素为所述第一域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;
在所述AI模型的个数为多个的情况下,多个所述N个元素的合集覆盖所述第一域,一个所述N个元素对应一个所述AI模型。
可选的,作为一个实施例,所述确定模块301,还用于:
在确定所述N个元素后,在将所述N个元素输入所述AI模型时,对所述N个元素进行排序;
其中,对所述N个元素进行排序,包括以下至少一项:
根据所述N个元素在所述第一域中的位置或标识,按照所述位置或标识从大到小的顺序或从小到大的顺序对所述N个元素进行排序;
根据所述N个元素的信道特征,按照所述信道特征从大到小的顺序或从小到大的顺序对所述N个元素进行排序,元素的信道特征包括以下至少一项:元素上信息的功率、幅度或相位;元素与其他元素的相关性,所述其他元素为所述N个元素中的其他元素或所述M个元素中的其他元素。
可选的,作为一个实施例,所述AI模型和所述AI模型的输入包括以下至少一项:
所述AI模型用于信号处理,所述AI模型的输入包括以下至少一项:解调参考信号DMRS;探测参考信号SRS;同步信号和物理广播信道块SSB;跟踪参考信号TRS;相位跟踪参考信号PTRS;信道状态信息参考信号CSI-RS;
所述AI模型用于信号传输、接收、解调或发送,所述AI模型的输入包括以下至少一项:物理下行控制信道PDCCH;物理下行共享信道PDSCH;物理上行控制信道PUCCH;物理上行共享信道PUSCH;物理随机接入信道PRACH;物理广播信道PBCH;
所述AI模型用于获取信道状态信息,所述AI模型的输入包括以下至少一项:CSI;CSI-RS;SRS;
所述AI模型用于波束管理,所述AI模型的输入包括以下至少一项:波束质量;波束信息;
所述AI模型用于信道预测,所述AI模型的输入包括以下至少一项:历史时刻的信道信息;当前时刻的信道信息;
所述AI模型用于干扰抑制,所述AI模型的输入包括以下至少一项:信号;干扰;
所述AI模型用于定位,所述AI模型的输入包括以下至少一项:参考信号的信道信息;辅助位置估计或轨迹估计的信息;
所述AI模型用于高层业务和/或参数的预测和管理,所述AI模型的输入包括以下至少一项:高层业务和/或参数;物理层的业务和/或参数;介质访问控制MAC层的业务和/或参数;
所述AI模型用于解析控制信令,所述AI模型的输入包括以下至少一项:信令;控制信道的接收信息。
根据本申请实施例的装置300可以参照对应本申请实施例的方法200的流程,并且,该装置300中的各个单元/模块和上述其他操作和/或功能分别为了实现方法200中的相应 流程,并且能够达到相同或等同的技术效果,为了简洁,在此不再赘述。
本申请实施例中的模型输入的确定装置可以是电子设备,例如具有操作系统的电子设备,也可以是电子设备中的部件,例如集成电路或芯片。该电子设备可以是终端,也可以为除终端之外的其他设备。示例性的,终端可以包括但不限于上述所列举的终端11的类型,其他设备可以为服务器、网络附属存储器(Network Attached Storage,NAS)等,本申请实施例不作具体限定。
本申请实施例提供的模型输入的确定装置能够实现图2的方法实施例实现的各个过程,并达到相同的技术效果,为避免重复,这里不再赘述。
可选的,如图4所示,本申请实施例还提供一种通信设备400,包括处理器401和存储器402,存储器402上存储有可在所述处理器401上运行的程序或指令,例如,该通信设备400为终端时,该程序或指令被处理器401执行时实现上述模型输入的确定方法实施例的各个步骤,且能达到相同的技术效果。该通信设备400为网络侧设备时,该程序或指令被处理器401执行时实现上述模型输入的确定装置方法实施例的各个步骤,且能达到相同的技术效果,为避免重复,这里不再赘述。
本申请实施例还提供一种通信设备,包括处理器和通信接口,处理器用于根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N个元素作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。该通信设备实施例与上述第一通信设备方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该通信设备实施例中,且能达到相同的技术效果。具体地,图5为实现本申请实施例的一种通信设备的硬件结构示意图。
该通信设备500包括但不限于:射频单元501、网络模块502、音频输出单元503、输入单元504、传感器505、显示单元506、用户输入单元507、接口单元508、存储器509以及处理器510等中的至少部分部件。
本领域技术人员可以理解,通信设备500还可以包括给各个部件供电的电源(比如电池),电源可以通过电源管理系统与处理器510逻辑相连,从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。图5中示出的通信设备结构并不构成对通信设备的限定,通信设备可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件布置,在此不再赘述。
应理解的是,本申请实施例中,输入单元504可以包括图形处理单元(Graphics Processing Unit,GPU)5041和麦克风5042,GPU 5041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元506可包括显示面板5061,可以采用液晶显示器、有机发光二极管等形式来配置显示面板5061。用户输入单元507包括触控面板5071以及其他输入设备5072中的至少一种。触控面板5071,也称为触摸屏。触控面板5071可包括触摸检测装置和触摸控制器两个部分。其他输入设备5072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键 等)、轨迹球、鼠标、操作杆,在此不再赘述。
本申请实施例中,射频单元501接收来自网络侧设备的下行数据后,可以传输给处理器510进行处理;另外,射频单元501可以向网络侧设备发送上行数据。通常,射频单元501包括但不限于天线、放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器509可用于存储软件程序或指令以及各种数据。存储器509可主要包括存储程序或指令的第一存储区和存储数据的第二存储区,其中,第一存储区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外,存储器509可以包括易失性存储器或非易失性存储器,或者,存储器509可以包括易失性和非易失性存储器两者。其中,非易失性存储器可以是只读存储器(Read-Only Memory,ROM)、可编程只读存储器(Programmable ROM,PROM)、可擦除可编程只读存储器(Erasable PROM,EPROM)、电可擦除可编程只读存储器(Electrically EPROM,EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory,RAM),静态随机存取存储器(Static RAM,SRAM)、动态随机存取存储器(Dynamic RAM,DRAM)、同步动态随机存取存储器(Synchronous DRAM,SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM,DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM,ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM,SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM,DRRAM)。本申请实施例中的存储器509包括但不限于这些和任意其它适合类型的存储器。
处理器510可包括一个或多个处理单元;可选的,处理器510集成应用处理器和调制解调处理器,其中,应用处理器主要处理涉及操作系统、用户界面和应用程序等的操作,调制解调处理器主要处理无线通信信号,如基带处理器。可以理解的是,上述调制解调处理器也可以不集成到处理器510中。
其中,处理器510,用于根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N个元素作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。
在本申请实施例中,通信设备可以根据AI模型的配置信息确定AI模型的输入,该配置信息用于指示从第一域的M个元素中选择N个元素作为AI模型的输入,N为大于或等于1的整数,M为大于N的整数。这样,由于通信设备可以在第一域的M个元素中灵活选择N个元素作为AI模型的输入,因此,可以使得通信设备能够在不同的资源位置和/或资源数目的情况下,使用尽量少的AI模型,提升了AI模型的使用灵活性和泛化能力,降低了通信系统的开销。
本申请实施例提供的通信设备500还可以实现上述200方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
本申请实施例还提供一种通信设备,包括处理器和通信接口,处理器用于根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N个元素 作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。该通信设备实施例与上述第一通信设备方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该通信设备实施例中,且能达到相同的技术效果。
具体地,本申请实施例还提供了一种通信设备。如图6所示,该通信设备600包括:天线61、射频装置62、基带装置63、处理器64和存储器65。天线61与射频装置62连接。在上行方向上,射频装置62通过天线61接收信息,将接收的信息发送给基带装置63进行处理。在下行方向上,基带装置63对要发送的信息进行处理,并发送给射频装置62,射频装置62对收到的信息进行处理后经过天线61发送出去。
以上实施例中第一通信设备执行的方法可以在基带装置63中实现,该基带装置63包括基带处理器。
基带装置63例如可以包括至少一个基带板,该基带板上设置有多个芯片,如图6所示,其中一个芯片例如为基带处理器,通过总线接口与存储器65连接,以调用存储器65中的程序,执行以上方法实施例中所示的第一通信设备操作。
该通信设备还可以包括网络接口66,该接口例如为通用公共无线接口(common public radio interface,CPRI)。
具体地,本发明实施例的通信设备600还包括:存储在存储器65上并可在处理器64上运行的指令或程序,处理器64调用存储器65中的指令或程序执行图6所示各模块执行的方法,并达到相同的技术效果,为避免重复,故不在此赘述。
本申请实施例还提供一种可读存储介质,所述可读存储介质上存储有程序或指令,该程序或指令被处理器执行时实现上述模型输入的确定方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
其中,所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质,包括计算机可读存储介质,如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。
本申请实施例另提供了一种芯片,所述芯片包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现上述模型输入的确定方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
应理解,本申请实施例提到的芯片还可以称为系统级芯片,系统芯片,芯片系统或片上系统芯片等。
本申请实施例另提供了一种计算机程序/程序产品,所述计算机程序/程序产品被存储在存储介质中,所述计算机程序/程序产品被至少一个处理器执行以实现上述模型输入的确定方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
需要说明的是,在本文中,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者装置所 固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外,需要指出的是,本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能,还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能,例如,可以按不同于所描述的次序来执行所描述的方法,并且还可以添加、省去、或组合各种步骤。另外,参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以计算机软件产品的形式体现出来,该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中,包括若干指令用以使得一台终端(可以是手机,计算机,服务器,空调器,或者网络设备等)执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述,但是本申请并不局限于上述的具体实施方式,上述的具体实施方式仅仅是示意性的,而不是限制性的,本领域的普通技术人员在本申请的启示下,在不脱离本申请宗旨和权利要求所保护的范围情况下,还可做出很多形式,均属于本申请的保护之内。

Claims (26)

  1. 一种模型输入的确定方法,其中,包括:
    第一通信设备根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N个元素作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。
  2. 根据权利要求1所述的方法,其中,
    所述AI模型在所述第一通信设备中使用,所述AI模型和所述配置信息由所述第一通信设备配置得到,所述第一通信设备为终端或网络侧设备。
  3. 根据权利要求1所述的方法,其中,
    所述AI模型在所述第一通信设备中使用,所述AI模型和所述配置信息由第二通信设备配置给所述第一通信设备;
    其中,所述第一通信设备为终端,所述第二通信设备为网络侧设备;或,
    所述第一通信设备为网络侧设备,所述第二通信设备为终端;或,
    所述第一通信设备为第一终端,所述第二通信设备为第二终端;或,
    所述第一通信设备为第一网络侧设备,所述第二通信设备为第二网络侧设备。
  4. 根据权利要求3所述的方法,其中,第一通信设备在根据AI模型的配置信息确定所述AI模型的输入之前,所述方法还包括:
    接收所述第二通信设备发送的所述AI模型和所述配置信息。
  5. 根据权利要求1所述的方法,其中,所述第一域包括以下至少一项:
    频域;时域;空域;多普勒域;时延域;波束域。
  6. 根据权利要求1所述的方法,其中,所述第一通信设备支持使用所述第一域中的M个元素;
    其中,所述第一域中的M个元素由第二通信设备配置给所述第一通信设备,或由第二通信设备上报给所述第一通信设备,或由所述第一通信设备自定义,或由协议约定,或由第二域转换得到,所述第二域由所述第一通信设备中的指定模块提供或由第二通信设备中的指定模块提供。
  7. 根据权利要求1所述的方法,其中,所述N个元素包括以下任一项:
    所述N个元素为所述第一域中指定位置的N个元素,或指定位置的连续N个元素,或指定位置的连续等间隔的N个元素;
    所述N个元素为所述第一域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;
    所述N个元素为所述第一域的指定区域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;其中,所述指定区域中的元素总数目大于所述N且小于等于所述M。
  8. 根据权利要求7所述的方法,其中,包括以下任一项:
    在所述N个元素为所述指定位置的N个元素的情况下,所述配置信息中包括所述N个元素中每个元素在所述第一域中的位置信息;
    在所述N个元素为所述指定位置的连续N个元素的情况下,所述配置信息中包括所述N个元素在所述第一域中的起始位置信息、和/或中间位置信息和/或结束位置信息;
    在所述N个元素为所述指定位置的连续等间隔的N个元素的情况下,所述配置信息中包括所述N个元素的间隔信息以及所述N个元素在所述第一域中的起始位置信息、和/或中间位置信息和/或结束位置信息。
  9. 根据权利要求7所述的方法,其中,
    所述连续等间隔的N个元素之间的间隔为固定值,或为预设区间中的任意值。
  10. 根据权利要求7所述的方法,其中,所述N个元素满足以下至少一项:
    在所述AI模型的个数为1的情况下,所述N个元素为所述第一域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;
    在所述AI模型的个数为多个的情况下,多个所述N个元素的合集覆盖所述第一域,一个所述N个元素对应一个所述AI模型。
  11. 根据权利要求1所述的方法,其中,所述方法还包括:
    在确定所述N个元素后,在将所述N个元素输入所述AI模型时,对所述N个元素进行排序;
    其中,对所述N个元素进行排序,包括以下至少一项:
    根据所述N个元素在所述第一域中的位置或标识,按照所述位置或标识从大到小的顺序或从小到大的顺序对所述N个元素进行排序;
    根据所述N个元素的信道特征,按照所述信道特征从大到小的顺序或从小到大的顺序对所述N个元素进行排序,元素的信道特征包括以下至少一项:元素上信息的功率、幅度或相位;元素与其他元素的相关性,所述其他元素为所述N个元素中的其他元素或所述M个元素中的其他元素。
  12. 根据权利要求1所述的方法,其中,所述AI模型和所述AI模型的输入包括以下至少一项:
    所述AI模型用于信号处理,所述AI模型的输入包括以下至少一项:解调参考信号DMRS;探测参考信号SRS;同步信号和物理广播信道块SSB;跟踪参考信号TRS;相位跟踪参考信号PTRS;信道状态信息参考信号CSI-RS;
    所述AI模型用于信号传输、接收、解调或发送,所述AI模型的输入包括以下至少一项:物理下行控制信道PDCCH;物理下行共享信道PDSCH;物理上行控制信道PUCCH;物理上行共享信道PUSCH;物理随机接入信道PRACH;物理广播信道PBCH;
    所述AI模型用于获取信道状态信息,所述AI模型的输入包括以下至少一项:CSI;CSI-RS;SRS;
    所述AI模型用于波束管理,所述AI模型的输入包括以下至少一项:波束质量;波束信息;
    所述AI模型用于信道预测,所述AI模型的输入包括以下至少一项:历史时刻的信道信息;当前时刻的信道信息;
    所述AI模型用于干扰抑制,所述AI模型的输入包括以下至少一项:信号;干扰;
    所述AI模型用于定位,所述AI模型的输入包括以下至少一项:参考信号的信道信息;辅助位置估计或轨迹估计的信息;
    所述AI模型用于高层业务和/或参数的预测和管理,所述AI模型的输入包括以下至少一项:高层业务和/或参数;物理层的业务和/或参数;介质访问控制MAC层的业务和/或参数;
    所述AI模型用于解析控制信令,所述AI模型的输入包括以下至少一项:信令;控制信道的接收信息。
  13. 一种模型输入的确定装置,其中,包括:
    确定模块,用于根据AI模型的配置信息确定所述AI模型的输入,所述配置信息用于指示从第一域中选择N个元素作为所述AI模型的输入,N为大于或等于1的整数,所述第一域中包括M个元素,M为大于N的整数。
  14. 根据权利要求13所述的装置,其中,
    所述AI模型在第一通信设备中使用,所述AI模型和所述配置信息由所述第一通信设备配置得到,所述第一通信设备为终端或网络侧设备。
  15. 根据权利要求13所述的装置,其中,
    所述AI模型在第一通信设备中使用,所述AI模型和所述配置信息由第二通信设备配置给所述第一通信设备;
    其中,所述第一通信设备为终端,所述第二通信设备为网络侧设备;或,
    所述第一通信设备为网络侧设备,所述第二通信设备为终端;或,
    所述第一通信设备为第一终端,所述第二通信设备为第二终端;或,
    所述第一通信设备为第一网络侧设备,所述第二通信设备为第二网络侧设备。
  16. 根据权利要求15所述的装置,其中,所述确定模块,还用于:
    接收所述第二通信设备发送的所述AI模型和所述配置信息。
  17. 根据权利要求13所述的装置,其中,所述第一域包括以下至少一项:
    频域;时域;空域;多普勒域;时延域;波束域。
  18. 根据权利要求13所述的装置,其中,第一通信设备支持使用所述第一域中的M个元素;
    其中,所述第一域中的M个元素由第二通信设备配置给所述第一通信设备,或由第二通信设备上报给所述第一通信设备,或由所述第一通信设备自定义,或由协议约定,或由第二域转换得到,所述第二域由所述第一通信设备中的指定模块提供或由 第二通信设备中的指定模块提供。
  19. 根据权利要求13所述的装置,其中,所述N个元素包括以下任一项:
    所述N个元素为所述第一域中指定位置的N个元素,或指定位置的连续N个元素,或指定位置的连续等间隔的N个元素;
    所述N个元素为所述第一域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;
    所述N个元素为所述第一域的指定区域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;其中,所述指定区域中的元素总数目大于所述N且小于等于所述M。
  20. 根据权利要求19所述的装置,其中,包括以下任一项:
    在所述N个元素为所述指定位置的N个元素的情况下,所述配置信息中包括所述N个元素中每个元素在所述第一域中的位置信息;
    在所述N个元素为所述指定位置的连续N个元素的情况下,所述配置信息中包括所述N个元素在所述第一域中的起始位置信息、和/或中间位置信息和/或结束位置信息;
    在所述N个元素为所述指定位置的连续等间隔的N个元素的情况下,所述配置信息中包括所述N个元素的间隔信息以及所述N个元素在所述第一域中的起始位置信息、和/或中间位置信息和/或结束位置信息。
  21. 根据权利要求19所述的装置,其中,
    所述连续等间隔的N个元素之间的间隔为固定值,或为预设区间中的任意值。
  22. 根据权利要求19所述的装置,其中,所述N个元素满足以下至少一项:
    在所述AI模型的个数为1的情况下,所述N个元素为所述第一域中任意位置的N个元素,或任意位置的连续N个元素,或任意位置的连续等间隔的N个元素;
    在所述AI模型的个数为多个的情况下,多个所述N个元素的合集覆盖所述第一域,一个所述N个元素对应一个所述AI模型。
  23. 根据权利要求13所述的装置,其中,所述确定模块,还用于:
    在确定所述N个元素后,在将所述N个元素输入所述AI模型时,对所述N个元素进行排序;
    其中,对所述N个元素进行排序,包括以下至少一项:
    根据所述N个元素在所述第一域中的位置或标识,按照所述位置或标识从大到小的顺序或从小到大的顺序对所述N个元素进行排序;
    根据所述N个元素的信道特征,按照所述信道特征从大到小的顺序或从小到大的顺序对所述N个元素进行排序,元素的信道特征包括以下至少一项:元素上信息的功率、幅度或相位;元素与其他元素的相关性,所述其他元素为所述N个元素中的其他元素或所述M个元素中的其他元素。
  24. 根据权利要求13所述的装置,其中,所述AI模型和所述AI模型的输入包括以下至少一项:
    所述AI模型用于信号处理,所述AI模型的输入包括以下至少一项:解调参考信号DMRS、探测参考信号SRS、同步信号和物理广播信道块SSB、跟踪参考信号TRS、相位跟踪参考信号PTRS、信道状态信息参考信号CSI-RS;
    所述AI模型用于信号传输、接收、解调或发送,所述AI模型的输入包括以下至少一项:物理下行控制信道PDCCH、物理下行共享信道PDSCH、物理上行控制信道PUCCH、物理上行共享信道PUSCH、物理随机接入信道PRACH、物理广播信道PBCH;
    所述AI模型用于获取信道状态信息,所述AI模型的输入包括以下至少一项:CSI;CSI-RS;SRS;
    所述AI模型用于波束管理,所述AI模型的输入包括以下至少一项:波束质量;波束信息;
    所述AI模型用于信道预测,所述AI模型的输入包括以下至少一项:历史时刻的信道信息;当前时刻的信道信息;
    所述AI模型用于干扰抑制,所述AI模型的输入包括以下至少一项:信号;干扰;
    所述AI模型用于定位,所述AI模型的输入包括以下至少一项:参考信号的信道信息;辅助位置估计或轨迹估计的信息;
    所述AI模型用于高层业务和/或参数的预测和管理,所述AI模型的输入包括以下至少一项:高层业务和/或参数;物理层的业务和/或参数;介质访问控制MAC层的业务和/或参数;
    所述AI模型用于解析控制信令,所述AI模型的输入包括以下至少一项:信令;控制信道的接收信息。
  25. 一种通信设备,其中,包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求1至12任一项所述的模型输入的确定方法的步骤。
  26. 一种可读存储介质,其中,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如权利要求1至12任一项所述的模型输入的确定方法的步骤。
PCT/CN2023/087739 2022-04-14 2023-04-12 模型输入的确定方法及通信设备 WO2023198094A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210390701.5A CN116963099A (zh) 2022-04-14 2022-04-14 模型输入的确定方法及通信设备
CN202210390701.5 2022-04-14

Publications (1)

Publication Number Publication Date
WO2023198094A1 true WO2023198094A1 (zh) 2023-10-19

Family

ID=88328981

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/087739 WO2023198094A1 (zh) 2022-04-14 2023-04-12 模型输入的确定方法及通信设备

Country Status (2)

Country Link
CN (1) CN116963099A (zh)
WO (1) WO2023198094A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112052027A (zh) * 2019-06-06 2020-12-08 阿里巴巴集团控股有限公司 一种处理ai任务的方法及装置
CN112394950A (zh) * 2021-01-19 2021-02-23 共达地创新技术(深圳)有限公司 Ai模型部署方法、设备和存储介质
CN114070675A (zh) * 2020-08-05 2022-02-18 展讯半导体(南京)有限公司 Ai网络模型匹配方法及装置、存储介质、用户设备
CN114154641A (zh) * 2020-09-07 2022-03-08 华为云计算技术有限公司 Ai模型的训练方法、装置、计算设备和存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112052027A (zh) * 2019-06-06 2020-12-08 阿里巴巴集团控股有限公司 一种处理ai任务的方法及装置
CN114070675A (zh) * 2020-08-05 2022-02-18 展讯半导体(南京)有限公司 Ai网络模型匹配方法及装置、存储介质、用户设备
CN114154641A (zh) * 2020-09-07 2022-03-08 华为云计算技术有限公司 Ai模型的训练方法、装置、计算设备和存储介质
CN112394950A (zh) * 2021-01-19 2021-02-23 共达地创新技术(深圳)有限公司 Ai模型部署方法、设备和存储介质

Also Published As

Publication number Publication date
CN116963099A (zh) 2023-10-27

Similar Documents

Publication Publication Date Title
EP4231693A1 (en) Configuration method for ai network parameter and device
WO2022028373A1 (zh) 测量上报方法、装置及设备
EP4224733A1 (en) Beam processing method and apparatus, and related device
US20230291658A1 (en) Method for Processing Partial Input Missing of AI Network, and Device
WO2023040885A1 (zh) 参数选择方法、参数配置方法、终端及网络侧设备
WO2023040887A1 (zh) 信息上报方法、装置、终端及可读存储介质
WO2023284801A1 (zh) Tci状态确定方法、装置、终端及网络侧设备
WO2022105913A1 (zh) 通信方法、装置及通信设备
US20240137781A1 (en) Spatial relation indication method and device
US20240088970A1 (en) Method and apparatus for feeding back channel information of delay-doppler domain, and electronic device
WO2023208042A1 (zh) 预失真处理方法、装置及设备
US20230299910A1 (en) Communications data processing method and apparatus, and communications device
WO2023198058A1 (zh) 信息传输方法、装置、终端及网络侧设备
WO2023274369A1 (zh) 传输方法、装置、通信设备及存储介质
WO2023198094A1 (zh) 模型输入的确定方法及通信设备
CN114390554A (zh) 信道状态信息确定方法、上报设置确定方法、装置及相关设备
WO2023179753A1 (zh) 波束信息指示方法、装置、终端及网络侧设备
WO2024061287A1 (zh) 人工智能ai模型传输方法、装置、终端及介质
WO2023207842A1 (zh) 波束信息确定方法、终端及网络侧设备
WO2023198062A1 (zh) Csi测量和上报方法、装置、设备、系统及存储介质
WO2023160547A1 (zh) 信息响应方法、信息发送方法、终端及网络侧设备
WO2024027537A1 (zh) 感知方法、装置、通信设备及可读存储介质
WO2024093771A1 (zh) 参考信号确定方法、终端及网络侧设备
WO2023072239A1 (zh) 信道预测方法、装置、网络侧设备及终端
WO2023186018A1 (zh) 信息传输方法、装置、终端及可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23787737

Country of ref document: EP

Kind code of ref document: A1