WO2023179649A1 - 人工智能模型的输入处理方法、装置及设备 - Google Patents

人工智能模型的输入处理方法、装置及设备 Download PDF

Info

Publication number
WO2023179649A1
WO2023179649A1 PCT/CN2023/083041 CN2023083041W WO2023179649A1 WO 2023179649 A1 WO2023179649 A1 WO 2023179649A1 CN 2023083041 W CN2023083041 W CN 2023083041W WO 2023179649 A1 WO2023179649 A1 WO 2023179649A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
related information
artificial intelligence
intelligence model
optimal
Prior art date
Application number
PCT/CN2023/083041
Other languages
English (en)
French (fr)
Inventor
施源
孙鹏
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 WO2023179649A1 publication Critical patent/WO2023179649A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/24—Cell structures
    • H04W16/28—Cell structures using beam steering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/06—Testing, supervising or monitoring using simulated traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/10—Scheduling measurement reports ; Arrangements for measurement reports

Definitions

  • This application belongs to the field of communication technology, and specifically relates to an input processing method, device and equipment for an artificial intelligence model.
  • AI Artificial Intelligence
  • RSRP Reference Signal Received Power
  • the output of the AI model is the RSRP result of all beam pairs.
  • the beam pair is composed of a transmitting beam and a receiving beam.
  • the number of inputs to the AI model is related to the number of selected partial beam pairs, that is, the number of inputs to the AI model is fixed.
  • the number of outputs of the AI model is equal to the number of all beam pairs, and the number of all beam pairs is equal to the number of transmitting beams * the number of receiving beams. If the model is sent to the terminal from the network side, that is, the model is on the terminal side, the number of transmit beams corresponds to the number of transmit beams on the network side, and the number of receive beams corresponds to the number of terminal receive beams.
  • the embodiments of the present application provide an input processing method, device and equipment for an artificial intelligence model, which can It is enough to solve the problem that the AI model output quantity cannot be widely used.
  • an input processing method for an artificial intelligence model includes:
  • the first device determines the input information of the artificial intelligence model; wherein the artificial intelligence model is used for beam prediction related functions, and the input information includes desired information and at least one of the following:
  • an input processing device for an artificial intelligence model includes:
  • the first determination module is used to determine the input information of the artificial intelligence model through the first device; wherein the artificial intelligence model is used for beam-related functions, and the input information includes desired information and at least one of the following:
  • an input processing device for an artificial intelligence model includes a processor and a memory.
  • the memory stores programs or instructions that can be run on the processor.
  • a communication device including a processor and a communication interface, wherein the processor is used to determine input information of an artificial intelligence model; wherein the artificial intelligence model is used for beam related functions, and the input information Include expectation information and at least one of the following:
  • an input processing system for an artificial intelligence model including a communication device, and the communication device can be used to perform the steps of the method described in the first aspect.
  • 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 seventh aspect, includes a processor and a communication interface.
  • the communication interface is coupled to the processor, and 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 of input processing methods for artificial intelligence models.
  • the first device determines the input information of the artificial intelligence model; wherein the artificial intelligence model is used for beam prediction related functions, and the input information includes expected information and at least one of the following: beam quality related information; First beam related information; time related information.
  • the solution of the present application can obtain the desired output information through the AI model of the first device without being limited by the capability of the first device, and the number of outputs of the AI model is fixed and has nothing to do with the number or angle of the receiving beams of the terminal device. It solves the problem that the output quantity of the artificial intelligence model cannot be widely used due to the different capabilities of the terminal equipment, and increases the practicality of the artificial intelligence model.
  • Figure 1 is a schematic diagram of beam prediction using an artificial intelligence model in the prior art
  • Figure 2 is a block diagram of a wireless communication system to which the embodiment of the present application can be applied.
  • Figure 3 is a schematic diagram of the steps of the input processing method of the artificial intelligence model according to the embodiment of the present application.
  • Figure 4 is a schematic diagram of the input processing system of the artificial intelligence model according to the embodiment of the present application.
  • Figure 5 is a schematic structural diagram of the input processing device of the artificial intelligence model according to the embodiment of the present application.
  • Figure 6 is a schematic structural diagram of the input processing device of the artificial intelligence model according to the embodiment of the present application.
  • Figure 7 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 should be understood that the technique used in this way The terms “first” and “second” 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 the objects distinguished by “first” and “second” are generally of the same type.
  • the number of objects is not limited.
  • the first object may 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-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. 2 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 may 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 palmtop 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
  • MID mobile Internet Device
  • AR augmented reality
  • VR virtual reality
  • robots wearable devices
  • WUE Vehicle User Equipment
  • PUE Pedestrian User Equipment
  • smart home home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.
  • game consoles personal computers (personal computer, PC), teller machine or self-service machine and other terminal-side devices.
  • Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets) bracelets, smart anklets, etc.), Smart wristbands, smart clothing, etc.
  • the network side device 12 may include an access network device or a core network device, where the access network device 12 may also be called a radio access network device, a radio access network (Radio Access Network, RAN), a 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.
  • WLAN Wireless Local Area Network
  • the base station may be called a Node B, an Evolved Node B (eNB), an access point, Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), home B-node, home evolved B-node , Transmitting Receiving Point (TRP) or some other suitable term in the field, as long as the same technical effect is achieved, the base station is not limited to specific technical terms. It should be noted that in the embodiment of the present application This introduction only takes the base station in the NR system as an example, and does not limit the specific type of base station.
  • this embodiment of the present application provides an input processing method for an artificial intelligence model, which includes the following steps:
  • Step 301 The first device determines the input information of the artificial intelligence model; wherein the artificial intelligence model is used for beam-related functions, and the input information includes desired information and at least one of the following:
  • the beam-related functions include but are not limited to: beam prediction, beam indication, beam recovery, and beam training, that is, the artificial intelligence model has at least one function of beam prediction, beam indication, beam recovery, and beam training. .
  • the first device determines the input information of the artificial intelligence model; wherein the artificial intelligence model is used for beam prediction related functions, and the input information includes expected information and at least one of the following: beam quality related information; A beam associated information; time related information.
  • the solution of the present application can obtain the desired output information through the AI model of the first device without being limited by the capability of the first device, and the output quantity of the AI model is fixed, which is consistent with the receiving beam of the terminal device. It has nothing to do with the quantity or angle, which solves the problem that the output quantity of the artificial intelligence model cannot be widely applied due to the different capabilities of the terminal equipment, and increases the practicality of the artificial intelligence model.
  • the first device and the second device can be various combinations of the base station side, the terminal side, and the auxiliary network center unit; for example: the first device is the terminal side, and the second device is the network side; or, the first device is the network side, and the second device is the terminal side; or the first device and the second device are both the network side; or the first device and the second device are both the terminal side, or the first device is an auxiliary network center unit, and the first device is the auxiliary network center unit.
  • the second device is the network side, etc., or the first device and the second device interact through the auxiliary network center unit; wherein the auxiliary network center unit is a unit used for information interaction. That is, both the first device and the second device may be a sending device or a receiving device.
  • the first device is a receiving end device.
  • the desired information includes at least one of the following:
  • the first beam association information may include at least one of transmit beam association information, receive beam association information and beam pair association information.
  • the second beam association information may include at least one of transmit beam association information, receive beam association information and beam pair association information.
  • the first device receives the first reference signal sent by the second device
  • the first beam association information includes the beam association information sent by the second device, the beam association information received by the first device, and the beam sent by the second device and the beam received by the first device At least one of the beam pair associated information.
  • the second beam association information includes the beam association information expected by the first device to be transmitted by the second device, the association information of the beam expected by the first device to be received by the first device, and the beam association expected by the first device from the second device to be transmitted and received by the first device. At least one of the beam pair associated information of the beam.
  • the input processing method of the artificial intelligence model in the embodiment of the present application can determine, through the first device, the beam pair association information expected by the first device (receiving end device) and the receiving end device reception beam association information expected by the receiving end device. And the second device (the transmitting device) desired by the receiving device sends the beam association information. Beam correlation information of all beams can be obtained through the first device.
  • the beam quality related information includes at least one of the following: signal-to-noise ratio, received power, and received quality;
  • the first beam association information is used to characterize the input of the artificial intelligence model to select the first beam.
  • the associated information corresponding to the first beam wherein the first beam associated information includes at least one of the following: beam identification related information, beam angle related information, beam gain related information, and beam width related information;
  • the time-related information is used to characterize the time corresponding to the beam quality-related information, and/or the time relationship between multiple beam quality-related information.
  • beam quality-related information refers to information that can characterize beam quality, including but not limited to at least one of the following: Level 1 Signal to interference plus noise ratio (L1-SINR), Level 1 reference signal Received power (Layer 1 Reference signal received power, L1-RSRP), Layer 1 Reference Signal Received Quality (L1-RSRQ), Layer 3 signal-to-noise ratio (L3-SINR), Layer 3 Reference signal received power (L3-RSRP), reference signal received quality of layer 3 (L3-RSRQ), etc.
  • L1-SINR Level 1 Signal to interference plus noise ratio
  • L1-RSRP Level 1 Reference signal Received power
  • L1-RSRQ Layer 1 Reference Signal Received Quality
  • L3-SINR Layer 3 signal-to-noise ratio
  • L3-RSRP Layer 3 Reference signal received power
  • L3-RSRQ reference signal received quality of layer 3
  • the information related to the beam identity is also the information related to the beam identity, and is used to represent the information related to the identity of the beam, including but not limited to at least one of the following: sending a beam identifier (Identifier, ID) , receiving beam identifier, beam pair identifier, reference signal set identifier corresponding to the beam, reference signal resource identifier corresponding to the beam, random identifier of the unique identifier, coded value after processing by the additional AI model (or AI network), beam Angle related information;
  • the beam angle related information is used to characterize the angle related information corresponding to the beam, including but not limited to at least one of the following: beam angle related information, sending angle related information, receiving angle related information; wherein, the angle related information It is the relevant information used to characterize the angle, such as angle, radian, index code value, code value after additional AI network processing;
  • the angle-related information can be determined based on the global coordinate system (Global coordinate system, GLS) or the local coordinate system (LCS); optionally, if it is a local coordinate system, the origin of the coordinate system can be The first device or the second device is determined through network configuration, protocol agreement or reporting method;
  • GLS global coordinate system
  • LCS local coordinate system
  • the beam gain related information is used to characterize the gain related information of the beam and/or antenna, including but not limited to at least one of the following: antenna relative gain (unit dBi), equivalent isotropic radiation power beam power spectrum (Effective Isotropic Radiated Power, EIRP), beam angle gain, beam angle gain spectrum (that is, the gain of a beam relative to different angles, including complete or partial gain spectrum information), EIRP corresponding to each beam angle, main lobe angle, side lobe Angle, number of side lobes, side Lobe distribution, number of antennas, beam scanning horizontal coverage, beam scanning vertical coverage, 3dB width, 6dB width, etc.
  • antenna relative gain unit dBi
  • EIRP Equi isotropic radiation power beam power spectrum
  • beam angle gain that is, the gain of a beam relative to different angles, including complete or partial gain spectrum information
  • EIRP corresponding to each beam angle
  • main lobe angle side lobe Angle
  • number of side lobes side Lobe distribution
  • number of antennas
  • At least one of the beam angle related information and the beam identity related information may be described by two-dimensional components (which may be horizontal-vertical components), or described and determined by higher-dimensional component information.
  • the time-related information is used to characterize the time corresponding to the beam quality-related information, and/or the time relationship between multiple beam quality-related information. For example, input three sets of beam quality-related information, respectively corresponding to the information received at the 1st cycle, 2nd cycle and 3rd cycle time before the current time.
  • the input processing method of the artificial intelligence model in the embodiment of the present application can obtain, through the first device, beam quality-related information including signal-to-noise ratio, received power and reception quality, associated information corresponding to the first beam, and information characterizing the beam quality. The time corresponding to the relevant information, and/or the time-related information of the time relationship between multiple beam quality-related information, thereby determining the input information of the artificial intelligence model.
  • the second beam association information is used to characterize the association information corresponding to the desired second beam; wherein the second beam association information includes at least one of the following:
  • Beam identification related information beam angle related information, beam gain related information, beam width related information
  • the predicted time-related information is used to characterize the time-related information associated with the output information of the artificial intelligence model.
  • the prediction time-related information used to characterize the time-related information associated with the output information of the artificial intelligence model can be understood as: the predicted time-related information can indicate the time-related information associated with the results output by the AI model. information, and the prediction time-related information includes but is not limited to at least one of the following: reference time + z time units, system frame number, prediction period; wherein the reference time and z can be determined by the first device and the The interaction of the second device determines that the unit of the time unit may be a unit representing time, symbol, time slot, frame number, microsecond, millisecond or second.
  • the input processing method of the artificial intelligence model in the embodiment of the present application can determine the associated information corresponding to the desired second beam and the time-related information associated with the output information of the artificial intelligence model.
  • information corresponding to the same input type includes at least one time information on time.
  • the information corresponding to the same input type can be understood as:
  • the desired information and at least one of the following: beam quality related information, beam related information, the transmitting end sends the beam related information, and the receiving end receives the beam related information;
  • the expected information includes at least one of the following:
  • the receiving end device expects the receiving end device to receive beam association information
  • the receiving end device expects the transmitting end device to send beam association information.
  • the expected information includes at least one of the following: beam association information expected by the first device to be received by the first device, beam association information expected by the second device sent by the first device, beam association information expected by the first device Beam pair correlation information between the beam sent by the second device and the beam received by the first device and the desired prediction time correlation information.
  • the one time information may be information at a historical moment and/or information at a current moment.
  • the first beam association information and/or the time-related information are associated with the beam quality-related information; and/or
  • the quantity of the first beam-related information and/or the quantity of the time-related information is associated with the quantity of the beam quality-related information.
  • the relationship between the input information includes a correlation relationship between beam quality and a correlation relationship between beam quantity.
  • the first device is a receiving end device;
  • the correlation relationship between beam quality can be understood as: the second device sends beam correlation information and the first device receives beam correlation information and the beam quality correlation information is correlated ;
  • the time-related information is associated with the beam quality-related information; or, the second device sends the beam-related information, the first device receives the beam-related information, and the time-related information is associated with the beam quality-related information. ;and / or
  • the correlation between the number of beams can be understood as: the number of beam correlation information sent by the second device and the number of beam correlation information received by the first device are correlated with the quantity of the beam quality-related information; or, the time-related information
  • the quantity is associated with the quantity of the beam quality-related information; or, the quantity of the beam-related information sent by the second device, the quantity of the beam-related information received by the first device.
  • the quantity and the quantity of the time-related information are associated with the quantity of the beam quality-related information.
  • the input processing method of the artificial intelligence model in the embodiment of the present application can determine the correlation between beam quality and beam quantity between the input information, thereby obtaining the detailed content of the input information.
  • the first beam related information and/or the time related information are associated with the beam quality related information, including at least one of the following:
  • the beam corresponding to the beam quality related information corresponds to the beam represented by the first beam related information in a one-to-one correspondence
  • the beams corresponding to the beam quality related information correspond to the beams represented by the time related information
  • the beams corresponding to the beam quality related information correspond to the beams jointly represented by the first beam related information and the time related information.
  • the quantity of the first beam-related information and/or the quantity of the time-related information is associated with the quantity of the beam quality-related information, including at least one of the following:
  • the quantity of at least one of the beam pair-related correlation information, the transmit beam correlation information and the reception beam correlation information included in the first beam correlation information is equal to the quantity of the beam quality-related information
  • the amount of time-related information is equal to the amount of beam quality-related information
  • the first beam related information includes transmit beam related information and receive beam related information
  • the product of the quantity of the transmit beam related information and the quantity of the received beam related information is equal to the quantity of the beam quality related information
  • the product of the quantity of at least one of the beam pair-related correlation information, the transmit beam correlation information and the reception beam correlation information included in the first beam correlation information and the quantity of the time-related information is equal to the quantity of the beam quality-related information.
  • the first beam-related information includes transmit beam-related information and receive beam-related information
  • the product of the quantity of the transmit beam-related information and the quantity of the time-related information is equal to the quantity of the beam quality-related information
  • the product of the quantity of the reception beam correlation information and the quantity of the time correlation information is equal to The amount of information related to beam quality
  • the product of the quantity of the transmission beam correlation information, the quantity of the reception beam correlation information and the quantity of the time correlation information is equal to Describes the amount of information related to beam quality.
  • the amount of transmitting beam related information, the amount of receiving beam related information, and the amount of time-related information can be determined through interaction between the first device and the second device.
  • the beams corresponding to the beam quality related information correspond to the beams represented by the beam association information one-to-one, and the remaining information is determined through interaction;
  • y is beam quality related information or beam pair related information
  • a is the beam related information sent by the transmitter
  • b is the beam related information received by the receiving end
  • c is the beam pair related information expected by the receiving end
  • d is the reception expected by the receiving end.
  • the end receives the beam association information
  • e is the sending end that the receiving end expects to send the beam association information.
  • the beam corresponding to the beam quality related information corresponds to the beam represented by the beam association information one-to-one, and the remaining information is determined through interaction, and at least one input information contains multiple temporal information;
  • y*t is the beam quality related information
  • y*z*t or y*z is the beam pair related information
  • a*z*t or a*z is the beam related information sent by the transmitter
  • b*z*t or b *z is the receiving beam association information of the receiving end
  • c is the beam pair association information expected by the receiving end
  • d*z is the receiving end beam association information expected by the receiving end
  • e*z is the transmitting beam association information expected by the receiving end ;
  • t represents the number of times of input information.
  • the input processing method of the artificial intelligence model also includes:
  • the first device determines output information of the artificial intelligence model; wherein the output information includes at least one of the following:
  • the beam quality related information and/or the optimal beam related information of the corresponding beam pair Under the second receiving beam and the second transmitting beam, the beam quality related information and/or the optimal beam related information of the corresponding beam pair;
  • the first receiving beam and the second receiving beam are receiving beams represented by the second beam association information; the first transmitting beam and the second transmitting beam are transmitting beams represented by the second beam association information.
  • the output information includes at least one of the following:
  • the receiving end sends beam quality related information and/or optimal transmit beam related information to all transmitting ends under the receiving beam;
  • the receiving end When the receiving end expects the sending end to send beam related information, the receiving end receives beam quality related information and/or optimal receiving beam related information from all receiving ends of the transmitting beam;
  • the receiving end expects the beam quality related information and/or the optimal beam related information of the beam pair indicated by the transmitting beam set and the receiving beam set under the transmitting end beam association information and the receiving end receiving beam association information expected by the receiving end;
  • the transmission beam set is obtained based on the transmitting beam association information expected by the receiving end
  • the receiving beam set is obtained based on the receiving end receive beam association information desired by the receiving end.
  • the first device is a receiving end
  • the output information includes at least one of the following:
  • the first device Under the reception beam correlation information desired by the first device, the first device sends beam quality-related information and/or optimal transmission beam-related information to all second devices under the reception beam;
  • the first device When the first device expects the second device to send beam correlation information, the first device receives beam quality-related information and/or optimal reception beam-related information from all first devices that send the beam;
  • the first device expects the beam quality related information and/or optimal beam correlation of the beam pair indicated by the transmit beam set and the receive beam set under the second device transmit beam association information and the first device receive beam association information. information;
  • the receiving beam set is based on the first device receiving beam association expected by the first device.
  • Information is obtained, and the transmission beam set is obtained based on the transmission beam association information of the second device expected by the first device.
  • the optimal transmission related information, the optimal reception related information and the optimal beam related information refer to carrying information that can characterize the optimal beam, including but not limited to: beam identification related information and/or beam angle related information; in addition , there can be multiple optimal beams.
  • information corresponding to the same output type includes at least one piece of time information.
  • the one time information may be information at a future time and/or information at a current time.
  • the information at the historical moment and the information at the future moment are compared with the information at the current moment in time.
  • the period before the current moment is the historical moment, and the period after the current moment is the future moment.
  • the current moment is related to the moment when the quality-related information was last obtained in the AI model input beam quality-related information.
  • the moment of the last quality-related information obtained is marked as s moment, then s moment is the current moment, or s+x moment is the current moment.
  • Time, x is obtained interactively, which can be processing time.
  • the amount of output information is associated with the second beam correlation information and/or the prediction time-related information.
  • the amount of output information is associated with the second beam association information and/or the prediction time-related information, including:
  • the amount of the output information is at least one of the following: the optimal number of beams, the product of the optimal beam number and the number of predictions; or,
  • the amount of the output information is at least one of the following: the optimal number of beams, the product of the optimal beam number and the number of predictions; or,
  • the amount of the output information is at least one of the following: the number of at least some transmit beams, the number of at least some transmit beams and The product of prediction times;
  • the output information includes information related to the quality of all receiving beams under the first transmitting beam.
  • the amount of output information is at least one of the following: the number of at least partially received beams, the product of the number of at least partially received beams and the number of predictions;
  • the amount of the output information is as follows: At least one item: the number of beam pairs corresponding to the second receiving beam and the second transmitting beam, the product of the number of beam pairs and the number of predictions;
  • the number of predictions is indicated by the prediction time-related information, or is determined by the artificial intelligence model or obtained by interacting with the second device.
  • the amount of output information associated with the second beam associated information and/or the prediction time related information can be understood as: the amount of information output by the AI model is associated with the beam pair information expected by the receiving end, and the reception expected by the receiving end.
  • the receiving end receives the beam association information, the receiving end expects the transmitting end to send the beam association information, the expected prediction time related information and at least one of the respective corresponding quantities are related.
  • the output number of the AI model is the optimal beam number, or equal to the optimal beam number * the number of predictions indicated by the expected prediction time related information;
  • the optimal beam includes: an optimal beam, an optimal transmitting beam, and an optimal receiving beam; or,
  • the output number of the AI model is the optimal number of beams, or equal to the optimal number of beams * the number of predictions indicated by the expected prediction time related information;
  • the optimal beams include: Optimal beam, optimal transmit beam, and optimal receive beam; or,
  • the AI model output quantity is equal to the number of all transmitting end beams at the transmitting end or The number of all transmit beams at the end * the number of predictions indicated by the expected prediction time related information; or,
  • the AI model output quantity is equal to the number of all receiving end beams at the receiving end or the receiving end.
  • the number of all received beams * the number of predictions indicated by the desired prediction time-related information; or,
  • the AI model output is the desired transmitter transmit beam association information at the receiver end and B receive wave Under the beam correlation information, when the receiving end has beam quality related information and/or optimal beam related information of the beam pair indicated by the transmitting beam set and receiving beam set, the AI model output quantity is equal to the transmitting beam set and receiving beam set indicating The number of beam pairs or the number of beam pairs indicated by the transmit beam set and the receive beam set * the number of predictions indicated by the desired prediction time related information.
  • the optimal beam includes: an optimal beam, an optimal transmitting beam, and an optimal receiving beam;
  • Optimal beam-related information, optimal transmission-related information, and optimal reception-related information refer to information that can characterize the optimal beam, including but not limited to: beam identification-related information and/or beam angle-related information; in addition, the optimal beam Can be multiple.
  • the input processing of the artificial intelligence model also includes:
  • the first device determines the auxiliary information of the artificial intelligence model; wherein the auxiliary information is used to assist the first device in determining the input information and/or output information of the artificial intelligence model; the auxiliary information includes the following At least one:
  • the amount of information related to the beam identification is the amount of information related to the beam identification
  • the order between the input information is the order in which information is input into the AI model
  • the order between the output information is the order in which information is output from the AI model
  • the order between the input information and the order between the output information may be an order defined according to certain rules, or a specified order.
  • the input information includes desired information, beam quality related information, first beam related information and time related information
  • the input The order between the information is specified according to the information type, and the beam quality related information, the first beam related information, the time related information and the expected information are input in order.
  • the quantity of beam identification related information includes:
  • the amount of information related to the beam identification pairs of the transmitter and the receiver is the amount of information related to the beam identification pairs of the transmitter and the receiver
  • the amount of information related to the transmitting beam identification of the transmitting end is the amount of information related to the transmitting beam identification of the transmitting end
  • the amount of information related to the transmit beam identification of the receiving end is the amount of information related to the transmit beam identification of the receiving end
  • the amount of information related to the receiving beam identification of the transmitter is the amount of information related to the receiving beam identification of the transmitter
  • the amount of information related to the receiving beam identification of the receiving end is the amount of information related to the receiving beam identification of the receiving end.
  • the amount of beam identification related information includes:
  • the amount of information related to the beam pair identification of the second device and the first device is the amount of information related to the beam pair identification of the second device and the first device
  • the amount of information related to the transmit beam identification of the second device is the amount of information related to the transmit beam identification of the second device
  • the quantity of information related to the transmitting beam identification of the first device is the quantity of information related to the transmitting beam identification of the first device
  • the amount of information related to the receiving beam identification of the second device is the amount of information related to the receiving beam identification of the second device
  • the quantity of information related to the receiving beam identification of the first device is the quantity of information related to the receiving beam identification of the first device.
  • the amount of beam angle related information includes:
  • the amount of information related to the beam angle sent by the transmitter is the amount of information related to the beam angle sent by the transmitter.
  • the amount of information related to the beam angle sent by the receiving end is the amount of information related to the beam angle sent by the receiving end
  • the amount of information related to the beam angle received by the transmitter is the amount of information related to the beam angle received by the transmitter
  • the amount of information related to the beam angle received by the receiving end is the amount of information related to the beam angle received by the receiving end.
  • the amount of beam angle related information includes:
  • the number of beam angle related information sent by the second device is the number of beam angle related information sent by the second device.
  • the number of beam angle related information sent by the first device is the number of beam angle related information sent by the first device
  • the second device receives the amount of information related to the beam angle
  • the first device receives the amount of information related to the beam angle.
  • the maximum beam-related capabilities supported by the device include: the maximum capabilities of beam identification and beam angle supported by the device;
  • the maximum capability of beam identification supported by the device includes:
  • the maximum amount of information related to the beam identification of the transmitter and receiver is the maximum amount of information related to the beam identification of the transmitter and receiver
  • the maximum amount of information related to the transmitting beam identification of the transmitting end is the maximum amount of information related to the transmitting beam identification of the transmitting end
  • the maximum amount of information related to the receiving beam identification of the transmitting end is the maximum amount of information related to the receiving beam identification of the transmitting end
  • the maximum capabilities of the beam angles supported by the device include:
  • the maximum amount of beam angle related information sent by the receiving end is the maximum amount of beam angle related information sent by the receiving end
  • the maximum amount of information related to the beam angle received by the transmitter is the maximum amount of information related to the beam angle received by the transmitter
  • the maximum amount of information related to the beam angle received by the receiving end is the maximum amount of information related to the beam angle received by the receiving end.
  • the first device is a receiving end device
  • the maximum beam-related capabilities supported by the device include:
  • the maximum capability of beam identification supported by the device includes:
  • the maximum amount of beam identification related information of the second device and the first device is the maximum amount of beam identification related information of the second device and the first device
  • the maximum amount of information related to the transmit beam identification of the second device is the maximum amount of information related to the transmit beam identification of the second device
  • the maximum amount of information related to the transmit beam identification of the first device is the maximum amount of information related to the transmit beam identification of the first device
  • the maximum amount of information related to the receiving beam identification of the second device is the maximum amount of information related to the receiving beam identification of the second device
  • the maximum amount of information related to the receiving beam identification of the first device is the maximum amount of information related to the receiving beam identification of the first device
  • the maximum capabilities of the beam angles supported by the device include:
  • the maximum amount of beam angle related information sent by the second device is the maximum amount of beam angle related information sent by the second device.
  • the maximum amount of beam angle related information sent by the first device is the maximum amount of beam angle related information sent by the first device
  • the maximum amount of beam angle related information received by the second device is the maximum amount of beam angle related information received by the second device
  • the maximum amount of beam angle related information received by the first device is the maximum amount of beam angle related information received by the first device.
  • the number of repetitions of the transmission beam includes:
  • the number of repetitions of the beam sent by the sender can use different receiving beams under the same transmitting beam.
  • the number of cycles to output information includes:
  • the minimum number of cycles recommended by the sending end for the receiving end refers to the receiving end. Change the desired beam correlation information to obtain the number of times the AI model outputs under different desired conditions;
  • the minimum number of times cannot exceed the number or maximum ability of the interactive receiving end to receive beam identifiers or angle-related information.
  • the information related to the prediction time-related information may be prediction time-related information, or may be information used to derive the prediction time-related information, or may include the number of predicted cycles and the length of the cycle. For example, the predicted number of future cycles needs to be converted into the corresponding time by the receiving end.
  • the third beam association information includes: beam association information related to the AI model input information suggested by the transmitting end and the receiving end, including at least one of the following: information related to the recommended beam pair, and information related to the recommended beam identification sent by the transmitting end.
  • the suggested information is used by the sending end to inform the receiving end of the beam identifier and/or angle combination of the AI model input recommended by the sending end during AI model training.
  • the first device is a first device
  • the third beam association information includes: the second device recommends beam association information related to the AI model input information of the first device, including at least one of the following: 1. Recommended beam pair related information, the recommended second device sends beam identification related information, the recommended second device sends beam angle related information, the recommended first device receives beam identification related information, and the recommended first device receives beam angle Related Information.
  • the suggested information is used by the second device to inform the first device of the beam identifier and/or angle combination of the AI model input recommended by the second device during AI model training.
  • the first device determines the output information of the artificial intelligence model, it further includes:
  • the first device selects at least one of the following from the output information:
  • the optimal transmit beam related information, the optimal receive beam related information and the optimal beam related information may be one or more.
  • the optimal beam related information is the related information of the optimal beam pair.
  • a synchronization signal block (Synchronization Signal Block, SSB) related beam prediction, and when at least one SSB in the SSB burst set is not enabled, the optimal beam association information is selected from the beam of the first SSB;
  • SSB Synchronization Signal Block
  • the first SSB is an SSB that is associated with the output information and is enabled in the SSB burst set;
  • the optimal beam includes at least one of the following:
  • the beam corresponding to the optimal beam related information is the beam corresponding to the optimal beam related information.
  • the optimal beam for interaction can only be selected from the AI model output results associated with enabled SSBs in the SSB burst set.
  • At least one of the input information, output information and auxiliary information of the artificial intelligence model is determined by the first device through interaction with the second device;
  • the interaction method includes at least one of the following:
  • the first device is a receiving end and the second device is a transmitting end, then at least one of the input information, output information and auxiliary information of the artificial intelligence model can be the third device.
  • One device sends it to the second device; or,
  • the first device is the base station side and the second device is the terminal side, then at least one of the input information, output information and auxiliary information of the artificial intelligence model can be indicated by the first device to the third device.
  • the first device is configured to the second device, the second device reports to the first device, or the second device requests the first device; or,
  • the input processing method of the artificial intelligence model in the embodiment of the present application can be implemented through the input processing system of the artificial intelligence model as shown in Figure 4:
  • the reference signal received power (RSRP) indicates that the input is RSRP information
  • the angle information indicates that the input is the angle information of the beam corresponding to the RSRP
  • the expected information indicates that the input is the expected angle information of the beam corresponding to the RSRP.
  • the desired information includes: desired receiving angle information and/or desired transmitting angle information.
  • the expected information in the input information of the AI model, only includes the expected receiving angle information, then:
  • AI model input 8 RSRPs, 8 senders corresponding to 8 RSRPs send angle information, 8 receivers corresponding to 8 RSRPs receive angle information, and 1 desired receiving angle information;
  • AI model output RSRP corresponding to the 32 transmitting beams predicted under the expected receiving angle information
  • the receiving end cycles 8 times, that is, changes the expected receiving angle information, and can obtain a total of RSRP information corresponding to 32*8 beam pairs.
  • the expected information in the input information of the AI model, the expected information only contains the expected transmission angle information, then:
  • AI model input 8 RSRPs, 8 senders corresponding to 8 RSRPs send angle information, 8 receivers corresponding to 8 RSRPs receive angle information, and 1 desired sending angle information;
  • AI model output RSRP corresponding to the eight receiving end beams predicted under the expected transmission angle information
  • the receiving end cycles 32 times, that is, changes the desired transmission angle information, and can obtain a total of RSRP information corresponding to 8*32 beam pairs.
  • the expected information in the input information of the AI model, includes the expected sending angle information and the expected receiving angle information, then:
  • AI model input 8 RSRPs, 8 senders corresponding to 8 RSRPs to send angle information, 8 RSRPs corresponding to 8 receivers to receive angle information, 1 expected sending angle information and 1 expected receiving angle information
  • AI model output: a corresponding RSRP predicted under the expected sending angle information and the expected receiving angle information;
  • the receiving end cycles 256 times, that is, changes the desired transmission angle and desired reception angle information, and can obtain a total of RSRP information corresponding to 1*256 beam pairs.
  • the expected information in the input information of the AI model, includes the expected sending angle information and the expected receiving angle information, then:
  • AI model input 8 RSRPs, 8 senders corresponding to 8 RSRPs send angle information, 8 RSRPs correspond to 8 receiver receiving angle information, 1 expected transmit angle information and 1 expected receive angle information.
  • the 1 expected transmit angle information and 1 expected receive angle information contain 8 expected Transmitting angle and 2 desired receiving angles;
  • AI model output RSRP corresponding to the 16 beam pairs predicted under the expected transmit angle information and the expected receive angle information;
  • the receiving end cycles 16 times, that is, changes the desired transmission angle and desired reception angle information, and can obtain a total of RSRP information corresponding to 16*16 beam pairs.
  • the expected information in the input information of the AI model, includes the expected sending angle information and the expected receiving angle information, then:
  • AI model input 8 RSRPs, 8 RSRPs corresponding to 8 sender sending angle information, 8 RSRPs corresponding to 8 receiving end receiving angle information, 8 expected sending angle information and 2 expected receiving angle information,
  • Each desired transmission angle information and each desired reception angle information include one desired transmission angle and one desired reception angle.
  • AI model output RSRP corresponding to the 16 beam pairs predicted under the expected transmit angle information and the expected receive angle information;
  • the receiving end cycles 16 times, that is, changes the desired transmission angle and desired reception angle information, and can obtain a total of RSRP information corresponding to 16*16 beam pairs.
  • the expected information includes expected receiving angle information and expected prediction time related information, then:
  • AI model input 8 RSRPs, 8 senders corresponding to 8 RSRPs send angle information, 8 receivers corresponding to 8 RSRPs receive angle information, 1 expected receiving angle information, 1 expected prediction time related information Information indicating that the prediction is located 1 second after the reference time;
  • AI model output predict the RSRP corresponding to the 32 transmitting beams located 1 second after the reference time under the expected receiving angle information
  • the receiving end cycles 8 times, that is, changes the expected receiving angle information, and can obtain a total of RSRP information corresponding to 32*8 beam pairs located 1 second after the reference time.
  • this embodiment of the present application also provides an input processing device 500 for an artificial intelligence model, which includes:
  • the first determination module 501 is used to determine the input information of the artificial intelligence model through the first device; Wherein, the artificial intelligence model is used for beam prediction related functions, and the input information includes expected information and at least one of the following:
  • the input information of the artificial intelligence model is determined through the first determination module of the first device; wherein the artificial intelligence model is used for beam prediction related functions, and the input information includes desired information and at least one of the following : Beam quality related information; first beam related information; time related information.
  • the solution of the present application can obtain the desired output information through the AI model of the first device without being limited by the capability of the first device, and the number of outputs of the AI model is fixed and has nothing to do with the number or angle of the receiving beams of the terminal device. It solves the problem that the output quantity of the artificial intelligence model cannot be widely used due to the different capabilities of the terminal equipment, and increases the practicality of the artificial intelligence model.
  • the input processing device of the artificial intelligence model also includes:
  • a second determination module configured to determine the output information of the artificial intelligence model through the first device; wherein the output information includes at least one of the following:
  • the beam quality related information and/or the optimal beam related information of the corresponding beam pair Under the second receiving beam and the second transmitting beam, the beam quality related information and/or the optimal beam related information of the corresponding beam pair;
  • the first receiving beam and the second receiving beam are receiving beams represented by the second beam association information; the first transmitting beam and the second transmitting beam are transmitting beams represented by the second beam association information.
  • the output information includes at least one of the following:
  • the receiving end sends beam quality related information and/or optimal transmit beam related information to all transmitting ends under the receiving beam;
  • the receiving end receives beam quality related information and/or optimal receiving beam related information
  • the receiving end expects the beam quality related information and/or the optimal beam related information of the beam pair indicated by the transmitting beam set and the receiving beam set under the transmitting end beam association information and the receiving end receiving beam association information expected by the receiving end;
  • the first transmitting beam and the second transmitting beam are obtained according to the transmitting end beam association information expected by the receiving end, and the first receiving beam and the second receiving beam are obtained according to the receiving end receiving beam association information expected by the receiving end. get.
  • the input processing device of the artificial intelligence model also includes:
  • the third determination module is used to determine the auxiliary information of the artificial intelligence model through the first device; wherein the auxiliary information is used to assist the first device in determining the input information and/or output of the artificial intelligence model.
  • Information includes at least one of the following:
  • the amount of information related to the beam identification is the amount of information related to the beam identification
  • the order between the input information is the order in which information is input into the AI model
  • the order between the output information is the order in which information is input from the AI model
  • the order between the input information and the order between the output information may be an order defined according to certain rules, or a specified order.
  • the quantity of beam identification related information includes:
  • the amount of information related to the beam identification of the transmitter and receiver is the amount of information related to the beam identification of the transmitter and receiver
  • the amount of information related to the transmitting beam identification of the transmitting end is the amount of information related to the transmitting beam identification of the transmitting end
  • the amount of information related to the transmit beam identification of the receiving end is the amount of information related to the transmit beam identification of the receiving end
  • the amount of information related to the receiving beam identification of the transmitter is the amount of information related to the receiving beam identification of the transmitter
  • the amount of information related to the receiving beam identification of the receiving end is the amount of information related to the receiving beam identification of the receiving end.
  • the amount of beam angle related information includes:
  • the amount of information related to the beam angle sent by the transmitter is the amount of information related to the beam angle sent by the transmitter.
  • the amount of information related to the beam angle sent by the receiving end is the amount of information related to the beam angle sent by the receiving end
  • the amount of information related to the beam angle received by the transmitter is the amount of information related to the beam angle received by the transmitter
  • the amount of information related to the beam angle received by the receiving end is the amount of information related to the beam angle received by the receiving end.
  • the maximum beam-related capabilities supported by the device include: the maximum capabilities of beam identification and beam angle supported by the device;
  • the maximum capability of beam identification supported by the device includes:
  • the amount of beam identification related information or the maximum supported capabilities of the transmitter and receiver is the amount of beam identification related information or the maximum supported capabilities of the transmitter and receiver
  • the maximum capabilities of the beam angles supported by the device include:
  • the number of repetitions of the transmission beam includes:
  • the number of repetitions of the sender's transmit beam is used to indicate that the receiver can use different receive beams under the same transmit beam.
  • the number of cycles to output information includes:
  • the minimum number of loops recommended by the sending end for the receiving end refers to the number of times the receiving end changes the expected associated information to obtain the AI model output under different expected conditions
  • the minimum number of times cannot exceed the number or maximum ability of the interactive receiving end to receive beam identifiers or angle-related information.
  • the input processing device of the artificial intelligence model also includes:
  • a selection module configured to select at least one of the following from the output information through the first device:
  • the optimal transmit beam related information, the optimal receive beam related information and the optimal beam related information may be one or more.
  • the optimal beam related information is the related information of a beam pair.
  • the input processing device of the artificial intelligence model 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.
  • the input processing device of the artificial intelligence model provided by the embodiment of the present application can implement each process implemented by the method embodiment in Figure 3 and achieve the same technical effect. To avoid duplication, the details will not be described here.
  • this embodiment of the present application also provides an input processing device 600 for an artificial intelligence model, including a processor 601 and a memory 602.
  • the memory 602 stores data that can run on the processor 601.
  • Programs or instructions for example, when the input processing device 600 of the artificial intelligence model is a terminal, when the program or instructions are executed by the processor 601, each step of the input processing method embodiment of the artificial intelligence model is implemented, and the same technology can be achieved. Effect.
  • the input processing device 600 of the artificial intelligence model is a network-side device, when the program or instruction is executed by the processor 601, each step of the input processing method embodiment of the artificial intelligence model is implemented, and the same technical effect can be achieved. In order to avoid Repeat, I won’t go into details 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 information of the artificial intelligence model through the first device;
  • the artificial intelligence model is used for beam related functions, and the input information includes expected information and at least one of the following: beam quality related information;
  • FIG. 7 is a schematic diagram of the hardware structure of a communication device that implements an embodiment of the present application.
  • the communication device 700 includes but is not limited to: radio frequency unit 701, network module 702, audio output unit 703, input unit 704, sensor 705, display unit 706, user input unit 707, interface unit 708, memory 709, processor 710, etc. at least some parts of it.
  • the communication device 700 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 710 through a power management system, thereby managing charging, discharging, and Power consumption management and other functions.
  • the structure of the communication device shown in Figure 7 does not constitute a limitation of the communication device for input processing of the artificial intelligence model.
  • the communication device may include more or less components than shown in the figure, or combine certain components, or different The arrangement of components will not be described again here.
  • the input unit 704 may include a graphics processing unit (Graphics Processing Unit, GPU) 7041 and a microphone 7042.
  • the graphics processor 7041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras).
  • the display unit 706 may include a display panel 7061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
  • the user input unit 707 includes a touch panel 7071 and at least one of other input devices 7072 .
  • Touch panel 7071 also called touch screen.
  • the touch panel 7071 may include two parts: a touch detection device and a touch controller.
  • Other input devices 7072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
  • the radio frequency unit 701 after receiving downlink data from the network side device, can transmit it to the processor 710 for processing; in addition, the radio frequency unit 701 can send uplink data to the network side device.
  • the radio frequency unit 701 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low noise amplifier, duplexer, etc.
  • Memory 709 may be used to store software programs or instructions as well as various data.
  • the memory 709 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first The storage area can store an operating system, an application program or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc.
  • memory 709 may include volatile memory or non-volatile memory, or memory 709 may include both volatile and non-volatile memory.
  • the 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 710 may include one or more processing units; optionally, the processor 710 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-mentioned modem processor may not be integrated into the processor 710.
  • the processor 710 is configured to determine the input information of the artificial intelligence model through the first device; wherein the artificial intelligence model is used for beam-related functions, and the input information includes desired information and at least one of the following:
  • the processor 710 is also used by the first device to determine the output information of the artificial intelligence model; wherein the output information includes at least one of the following:
  • At least part of the receive beam quality related information and/or the optimal receive beam phase related information are included in the first transmit beam.
  • the beam quality related information and/or the optimal beam related information of the corresponding beam pair Under the second receiving beam and the second transmitting beam, the beam quality related information and/or the optimal beam related information of the corresponding beam pair;
  • the first receiving beam and the second receiving beam are receiving beams represented by the second beam association information; the first transmitting beam and the second transmitting beam are transmitting beams represented by the second beam association information.
  • the processor 710 is also used by the first device to determine the auxiliary information of the artificial intelligence model; wherein the auxiliary information is used to assist the first device in determining the input information of the artificial intelligence model and/or Output information; the auxiliary information includes at least one of the following:
  • the amount of information related to the beam identification is the amount of information related to the beam identification
  • the processor 710 is also configured for the first device to select at least one of the following from the output information:
  • Embodiments of the present application further provide a readable storage medium, which stores programs or instructions, and when the program or instructions are executed by a processor, each process of the above embodiment of the input processing method of the artificial intelligence model is implemented. , and can achieve the same technical effect, so to avoid repetition, they will not be described again here.
  • the processor is the processor in the communication device described in the above embodiment.
  • Said can Read storage media, including computer-readable storage media, such as computer-readable memory ROM, random access memory RAM, magnetic disks or optical disks, 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 input processing of the above-mentioned artificial intelligence model.
  • Each process of the method embodiment can achieve the same technical effect, so to avoid repetition, it will not be described again 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 input of the above artificial intelligence model.
  • Each process of the processing method embodiment can achieve the same technical effect, so to avoid repetition, it will not be described again here.
  • An embodiment of the present application further provides an input processing system for an artificial intelligence model, including: a communication device, which may be used to execute the input processing method for an artificial intelligence model as described above.
  • 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 essentially or the part that contributes to the existing technology can be used as a computer software product.
  • the computer software product is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes a number of instructions to enable a terminal (which can be a mobile phone, computer, server, air conditioner, or network equipment etc.) to perform the methods described in various embodiments of this application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种人工智能模型的输入处理方法、装置及设备，属于通信领域，本申请实施例的人工智能模型的输入处理方法包括：第一设备确定人工智能模型的输入信息；其中，所述人工智能模型用于波束相关功能，所述输入信息包括期望信息以及以下至少一项：波束质量相关信息；第一波束关联信息；时间相关信息。

Description

人工智能模型的输入处理方法、装置及设备
相关申请的交叉引用
本申请主张在2022年03月23日在中国提交的中国专利申请No.202210295984.5的优先权，其全部内容通过引用包含于此。
技术领域
本申请属于通信技术领域，具体涉及一种人工智能模型的输入处理方法、装置及设备。
背景技术
人工智能(Artificial Intelligence，AI)目前在各个领域获得了广泛的应用，所以，在通信技术领域也提出了使用AI进行波束预测的方案。如图1所示，使用部分波束对的参考信号接收功率(Reference Signal Received Power，RSRP)作为输入，AI模型的输出则是所有波束对的RSRP结果。其中，波束对是由发送波束和接收波束组成的。
其中，AI模型的输入数量与挑出来的部分波束对的数量相关，即AI模型的输入数量是固定的。而AI模型的输出数量等于所有波束对的数量，所有波束对的数量等于发送波束数量*接收波束数量。如果该模型是网络侧发送给终端的，也就是模型是在终端侧，发送波束数量对应于网络侧的发送波束数量，而接收波束数量对应于终端接收波束数量。
但是，考虑到泛化和实际可行性，一个小区内往往只有一个AI模型用于波束预测，也就是代表着该AI模型要适用于小区内的所有用户。然而，由于不同终端的能力是不同的，对应的终端接收波束的数量也是不同的，也就是总的波束对的数量对于不同终端的能力下是不同的。所以，就导致了AI模型无法广泛应用。
发明内容
本申请实施例提供一种人工智能模型的输入处理方法、装置及设备，能 够解决AI模型输出数量无法广泛应用的问题。
第一方面，提供了一种人工智能模型的输入处理方法，该方法包括：
第一设备确定人工智能模型的输入信息；其中，所述人工智能模型用于波束预测相关功能，所述输入信息包括期望信息以及以下至少一项：
波束质量相关信息；
第一波束关联信息；
时间相关信息。
第二方面，提供了一种人工智能模型的输入处理装置，该装置包括：
第一确定模块，用于通过第一设备确定人工智能模型的输入信息；其中，所述人工智能模型用于波束相关功能，所述输入信息包括期望信息以及以下至少一项：
波束质量相关信息；
第一波束关联信息；
时间相关信息。
第三方面，提供了一种人工智能模型的输入处理设备，该人工智能模型的输入处理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，所述处理器用于确定人工智能模型的输入信息；其中，所述人工智能模型用于波束相关功能，所述输入信息包括期望信息以及以下至少一项：
波束质量相关信息；
第一波束关联信息；
时间相关信息。
第五方面，提供了一种人工智能模型的输入处理系统，包括通信设备，所述通信设备可用于执行如第一方面所述的方法的步骤。
第六方面，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。
第七方面，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。
第八方面，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人工智能模型的输入处理方法的步骤。
在本申请实施例中，第一设备确定人工智能模型的输入信息；其中，所述人工智能模型用于波束预测相关功能，所述输入信息包括期望信息以及以下至少一项：波束质量相关信息；第一波束关联信息；时间相关信息。这样，本申请的方案能够不受限于第一设备的能力，通过第一设备的AI模型获得期望的输出信息，并且人工智能模型的输出数量固定，与终端设备的接收波束数量或角度无关，解决了因为终端设备的能力不同导致人工智能模型输出数量无法广泛应用的问题，增加了人工智能模型的实用性。
附图说明
图1是现有技术中人工智能模型进行波束预测的示意图；
图2是本申请实施例的本申请实施例可应用的一种无线通信系统的框图
图3是本申请实施例的人工智能模型的输入处理方法的步骤示意图；
图4是本申请实施例的人工智能模型的输入处理系统的示意图；
图5是本申请实施例的人工智能模型的输入处理装置的结构示意图；
图6是本申请实施例的人工智能模型的输入处理设备的结构示意图；
图7是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(6thGeneration，6G)通信系统。
图2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(Vehicle User Equipment，VUE)、行人终端(Pedestrian User Equipment，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(Evolved Node 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人工智能模型的输入处理进行详细地说明。
如图3所示，本申请实施例提供一种人工智能模型的输入处理方法，包括如下步骤：
步骤301，第一设备确定人工智能模型的输入信息；其中，所述人工智能模型用于波束相关功能，所述输入信息包括期望信息以及以下至少一项：
波束质量相关信息；
第一波束关联信息；
时间相关信息。
可选地，所述波束相关功能包括但不限于：波束预测，波束指示，波束恢复，波束训练，即所述人工智能模型具有波束预测，波束指示，波束恢复，波束训练中的至少一项功能。
本申请实施例中，第一设备确定人工智能模型的输入信息；其中，所述人工智能模型用于波束预测相关功能，所述输入信息包括期望信息以及以下至少一项：波束质量相关信息；第一波束关联信息；时间相关信息。这样，本申请的方案能够不受限于第一设备的能力，通过第一设备的AI模型获得期望的输出信息，并且人工智能模型的输出数量固定，与终端设备的接收波束 数量或角度无关，解决了因为终端设备的能力不同导致人工智能模型输出数量无法广泛应用的问题，增加了人工智能模型的实用性。
需要说明的是，第一设备和第二设备可以是基站侧、终端侧以及辅助网络中心单元的各种组合；例如：第一设备为终端侧，第二设备为网络侧；或，第一设备为网络侧，第二设备为终端侧；或，第一设备和第二设备均为网络侧；或，第一设备和第二设备均为终端侧，或第一设备是辅助网络中心单元，第二设备是网络侧等，或第一设备和第二设备通过辅助网络中心单元进行交互；其中，辅助网络中心单元是用于信息交互的单元。即，所述第一设备和所述第二设备均可以是发送端设备或者接收端设备。本申请一实施例中所述第一设备为接收端设备。
可选地，所述期望信息包括以下至少一项：
第二波束关联信息；
预测时间相关信息。
其中，第一波束关联信息可以包括发送波束关联信息，接收波束关联信息和波束对关联信息中的至少一项。同样的，第二波束关联信息可以包括发送波束关联信息，接收波束关联信息和波束对关联信息中的至少一项。例如，第一设备接收第二设备发送的第一参考信号，第一波束关联信息包括第二设备发送波束关联信息，第一设备接收波束关联信息，以及第二设备发送波束与第一设备接收波束的波束对关联信息中的至少一者。而第二波束关联信息则包括第一设备期望的第二设备发送波束关联信息，第一设备期望的第一设备接收波束关联信息，以及第一设备期望的第二设备发送波束与第一设备接收波束的波束对关联信息中的至少一者。
可选地，本申请实施例的人工智能模型的输入处理方法，通过第一设备能够确定第一设备(接收端设备)期望的波束对关联信息、接收端设备期望的接收端设备接收波束关联信息以及接收端设备期望的第二设备(发送端设备)发送波束关联信息。能够通过第一设备获得所有波束的波束关联信息。
可选地，所述波束质量相关信息包括以下至少一项：信噪比，接收功率，接收质量；
所述第一波束关联信息用于表征所述人工智能模型的输入选择第一波束 的情况下，所述第一波束对应的关联信息；其中，所述第一波束关联信息包括以下至少一项：波束标识相关信息，波束角度相关信息，波束增益相关信息，波束宽度相关信息；
所述时间相关信息用于表征所述波束质量相关信息对应的时间，和/或，多个所述波束质量相关信息之间的时间关系。
其中，波束质量相关信息是指能表征波束质量的信息，包括但不限于以下至少之一：层1的信噪比(Level 1 Signal to interference plus noise ratio，L1-SINR)，层1的参考信号接收功率(Layer 1 Reference signal received power，L1-RSRP)，层1的参考信号接收质量(Layer 1 Reference Signal Received Quality，L1-RSRQ)，层3的信噪比(L3-SINR)，层3的参考信号接收功率(L3-RSRP)，层3的参考信号接收质量(L3-RSRQ)等。
本申请一实施例中，所述波束标识相关信息，也是波束身份相关信息，用于表征所述波束的身份识别的相关信息，包含但不限于以下至少之一：发送波束标识(Identifier，ID)，接收波束标识，波束对标识，所述波束对应的参考信号集标识，所述波束对应的参考信号资源标识，唯一标识的随机标识，额外AI模型(或AI网络)处理后的编码值，波束角度相关信息；
所述波束角度相关信息用于表征所述波束对应的角度相关信息，包含但不限于以下至少之一：波束对角度相关信息，发送角度相关信息，接收角度相关信息；其中，所述角度相关信息是用于表征角度的相关信息，例如，角度，弧度，索引编码值，额外AI网络处理后的编码值；
可选地，角度相关信息，可以是基于全局坐标系(Global coordinate system，GLS)，或局部坐标系(local coordinate system，LCS)确定；可选地，若为局部坐标系，坐标系原点可以是第一设备或第二设备，通过网络配置，协议约定或上报方式确定；
所述波束增益相关信息用于表征所述波束和/或天线的增益相关信息，包括但不限于以下至少之一：天线相对增益(单位dBi)，等效全向辐射功率波束功率谱(Effective Isotropic Radiated Power，EIRP)，波束角度增益，波束角度增益谱(即一个波束相对于不同角度上的增益，包括完整的或部分增益谱信息)，每个波束角度对应的EIRP，主瓣角度，副瓣角度，副瓣数量，副 瓣分布，天线数量，波束扫描水平覆盖范围，波束扫描垂直覆盖范围，3dB宽度，6dB宽度等。
可选地，所述波束角度相关信息和波束标识相关信息中至少一项可以是通过二维分量(可以是水平-垂直分量)进行描述，或通过更高维度的分量信息进行描述确定。
本申请一实施例中，所述时间相关信息用于表征所述波束质量相关信息对应的时间，和/或，多个所述波束质量相关信息之间的时间关系。例如，输入3组波束质量相关信息，分别对应距离当前时刻之前的第1周期，第2周期和第3周期时刻接收的信息。
本申请实施例的人工智能模型的输入处理方法，能够通过第一设备获取包括信噪比，接收功率以及接收质量的波束质量相关信息、所述第一波束对应的关联信息以及表征所述波束质量相关信息对应的时间，和/或，多个所述波束质量相关信息之间的时间关系的时间相关信息，从而确定所述人工智能模型的输入信息。
可选地，所述第二波束关联信息用于表征期望的第二波束所对应的关联信息；其中，所述第二波束关联信息包括以下至少一项：
波束标识相关信息，波束角度相关信息，波束增益相关信息，波束宽度相关信息；
所述预测时间相关信息用于表征所述人工智能模型的输出信息关联的时间相关信息。
本申请一实施例中，所述预测时间相关信息用于表征所述人工智能模型的输出信息关联的时间相关信息可以理解为：所述预测时间相关信息能够指示AI模型输出的结果关联的时间相关信息，且所述预测时间相关信息包括但不限于以下至少之一：基准时刻+z个时间单元，系统帧号，预测所属周期；其中，基准时刻和z可以由所述第一设备与所述第二设备的交互确定，时间单元的单位可以是表征时间的单位，符号，时隙，帧号，微妙，毫秒或者秒。
本申请实施例的人工智能模型的输入处理方法，能够确定出期望的第二波束所对应的关联信息以及人工智能模型的输出信息关联的时间相关信息。
可选地，所述输入信息中，同一个输入类型对应的信息包括至少一个时 间上的信息。
可选地，所述同一个输入类型对应的信息可以理解为：
所述期望信息和以下至少一项：波束质量相关信息，波束关联信息，发送端发送波束关联信息，接收端接收波束关联信息；
所述期望信息包括以下至少一项：
接收端设备期望的波束关联信息；
接收端设备期望的接收端设备接收波束关联信息；
接收端设备期望的发送端设备发送波束关联信息。
在本申请一实施例中，所述期望信息包括以下至少一项：第一设备期望的第一设备接收波束关联信息、第一设备期望的第二设备发送波束关联信息、第一设备期望的第二设备发送波束与第一设备接收波束的波束对关联信息以及期望的预测时间相关信息。
可选地，对于输入信息，所述一个时间上的信息可以是历史时刻的信息和/或当前时刻的信息。
可选地，所述第一波束关联信息和/或所述时间相关信息，与所述波束质量相关信息相关联；和/或
所述第一波束关联信息的数量和/或所述时间相关信息的数量，与所述波束质量相关信息的数量相关联。
所述输入信息之间的关系包括波束质量的关联关系和波束数量的关联关系。
本申请一实施例中，所述第一设备为接收端设备；波束质量的关联关系可以理解为：第二设备发送波束关联信息和第一设备接收波束关联信息与所述波束质量相关信息相关联；或，所述时间相关信息与所述波束质量相关信息相关联；或，第二设备发送波束关联信息、第一设备接收波束关联信息和所述时间相关信息与所述波束质量相关信息相关联；和/或
波束数量的关联关系可以理解为：所述第二设备发送波束关联信息的数量和第一设备接收波束关联信息的数量与所述波束质量相关信息的数量相关联；或，所述时间相关信息的数量，与所述波束质量相关信息的数量相关联；或，所述第二设备发送波束关联信息的数量、第一设备接收波束关联信息的 数量和所述时间相关信息的数量与所述波束质量相关信息的数量相关联。
本申请实施例的人工智能模型的输入处理方法，能够确定所述输入信息之间波束质量的关联关系和波束数量的关联关系，从而得到所述输入信息的详细内容。
具体地，所述第一波束关联信息和/或所述时间相关信息，与所述波束质量相关信息相关联，包括以下至少一项：
所述波束质量相关信息对应的波束与所述第一波束关联信息表征的波束一一对应；
所述波束质量相关信息对应的波束与所述时间相关信息表征的波束一一对应；
所述波束质量相关信息对应的波束与所述第一波束关联信息和所述时间相关信息联合表征的波束一一对应。
具体地，所述第一波束关联信息的数量和/或所述时间相关信息的数量，与所述波束质量相关信息的数量相关联，包括以下至少一项：
所述第一波束关联信息包括的波束对相关的关联信息、发送波束关联信息以及接收波束关联信息中至少一者的数量，等于所述波束质量相关信息的数量；
所述时间相关信息的数量等于所述波束质量相关信息的数量；
所述第一波束关联信息包括发送波束关联信息和接收波束关联信息的情况下，所述发送波束关联信息的数量与所述接收波束关联信息的数量的乘积等于所述波束质量相关信息的数量；
所述第一波束关联信息包括的波束对相关的关联信息、发送波束关联信息以及接收波束关联信息中至少一者的数量与所述时间相关信息的数量的乘积等于所述波束质量相关信息的数量；
所述第一波束关联信息包括发送波束关联信息和接收波束关联信息的情况下，所述发送波束关联信息的数量与所述时间相关信息的数量的乘积等于所述波束质量相关信息的数量；
所述第一波束关联信息包括发送波束关联信息和接收波束关联信息的情况下，所述接收波束关联信息的数量与所述时间相关信息的数量的乘积等于 所述波束质量相关信息的数量；
所述第一波束关联信息包括发送波束关联信息和接收波束关联信息的情况下，所述发送波束关联信息的数量、所述接收波束关联信息的数量以及所述时间相关信息的数量的乘积等于所述波束质量相关信息的数量。
本申请一实施例中，所述发送波束关联信息的数量、所述接收波束关联信息的数量以及所述时间相关信息的数量可以通过第一设备与第二设备的交互确定。
本申请一实施例中：所述波束质量相关信息对应的波束与波束关联信息表征的波束一一对应，其余信息通过交互确定；
具体地，所述输入信息的数量之间满足以下至少之一：a*b＝y,a＝b＝y；
其中，y为波束质量相关信息或者波束对关联信息；a为发送端发送波束关联信息；b为接收端接收波束关联信息；c为接收端期望的波束对关联信息；d为接收端期望的接收端接收波束关联信息；e为接收端期望的发送端发送波束关联信息。
本申请另一实施例中：所述波束质量相关信息对应的波束与波束关联信息表征的波束一一对应，其余信息通过交互确定，至少一个输入信息中是包含多个时间上的信息；
具体地，所述输入信息的数量之间满足以下至少之一：a*b＝y,a＝b＝y；
其中，y*t为波束质量相关信息；y*z*t或y*z为波束对关联信息；a*z*t或a*z为发送端发送波束关联信息；b*z*t或b*z为接收端接收波束关联信息；c为接收端期望的波束对关联信息；d*z为接收端期望的接收端接收波束关联信息；e*z为接收端期望的发送端发送波束关联信息；
需要说明的是，t代表输入的信息的次数，一种可能性是t-1个历史信息+当前信息，z代表表征波束关联信息的分量信息的数量，若使用二维分量表示，对应仅使用波束关联信息时，z＝2,若使用发送波束关联信息和接收波束关联信息时，z＝1；c,d和e的数量由交互确定。
可选地，所述人工智能模型的输入处理方法，还包括：
所述第一设备确定所述人工智能模型的输出信息；其中，所述输出信息包括以下至少一项：
第一接收波束下，至少部分发送波束质量相关信息和/或最优发送波束相关信息；
第一发送波束下，至少部分接收波束质量相关信息和/或最优接收波束相关信息；
第二接收波束和第二发送波束下，对应的波束对的波束质量相关信息和/或最优波束相关信息；
其中，所述第一接收波束和第二接收波束为所述第二波束关联信息表征的接收波束；所述第一发送波束和第二发送波束为所述第二波束关联信息表征的发送波束。
本申请一实施例中，所述输出信息包括以下至少一项：
接收端期望的接收波束关联信息下，接收端在该接收波束下的所有发送端发送波束质量相关信息和/或最优发送波束相关信息；
接收端期望的发送端发送波束关联信息下，接收端在该发送波束的所有接收端接收波束质量相关信息和/或最优接收波束相关信息；
接收端期望的发送端发送波束关联信息和接收端接收波束关联信息下，接收端在发送波束集合和接收波束集合指示的波束对的波束质量相关信息和/或最优波束相关信息；
其中，所述发送波束集合是根据接收端期望的发送端发送波束关联信息获得，所述接收波束集合是根据接收端期望的接收端接收波束关联信息获得。
本申请一实施例中，所述第一设备为接收端，所述输出信息包括以下至少一项：
第一设备期望的接收波束关联信息下，第一设备在该接收波束下的所有第二设备发送波束质量相关信息和/或最优发送波束相关信息；
第一设备期望的第二设备发送波束关联信息下，第一设备在该发送波束的所有第一设备接收波束质量相关信息和/或最优接收波束相关信息；
第一设备期望的第二设备发送波束关联信息和第一设备接收波束关联信息下，第一设备在该发送波束集合和接收波束集合指示的波束对的波束质量相关信息和/或最优波束相关信息；
其中，所述接收波束集合是根据第一设备期望的第一设备接收波束关联 信息获得，所述发送波束集合是根据第一设备期望的第二设备发送波束关联信息获得。
可选地，最优发送相关信息、最优接收相关信息以及最优波束相关信息是指携带能表征最优波束的信息，包括但不限于：波束标识相关信息和/或波束角度相关信息；此外，最优波束可以是多个。
可选地，所述输出信息中，同一个输出类型对应的信息包括至少一个时间上的信息。
可选地，所述一个时间上的信息可以是未来时刻的信息和/或当前时刻的信息。
需要说明的是，历史时刻的信息以及未来时刻的信息都是与当前时刻的信息在时间进行对比的，在当前时刻之前的即为历史时刻，在当前时刻之后的即为未来时刻。当前时刻与AI模型输入波束质量相关信息中最后获得的质量相关信息的时刻有关，最后获得的质量相关信息的时刻标记为s时刻，则s时刻即为当前时刻，或s+x时刻即为当前时刻，x为交互获得，可以是处理时间。
可选地，所述输出信息的数量与所述第二波束关联信息和/或所述预测时间相关信息相关联。
可选地，所述输出信息的数量与所述第二波束关联信息和/或所述预测时间相关信息相关联，包括：
在所述输出信息包括所述最优发送波束相关信息的情况下，所述输出信息的数量为以下至少一项：最优波束数量，最优波束数量与预测次数的乘积；或者，
在所述输出信息包括所述最优接收波束相关信息的情况下，所述输出信息的数量为以下至少一项：最优波束数量，最优波束数量与预测次数的乘积；或者，
在所述输出信息包括所述第一接收波束下的所有发送波束质量相关信息的情况下，所述输出信息的数量为以下至少一项：至少部分发送波束的数量，至少部分发送波束的数量与预测次数的乘积；
在所述输出信息包括所述第一发送波束下的所有接收波束质量相关信息 的情况下，所述输出信息的数量为以下至少一项：至少部分接收波束的数量，至少部分接收波束的数量与预测次数的乘积；
在所述输出信息包括所述第二接收波束和所述第二发送波束下，对应的波束对的波束质量相关信息和/或最优波束相关信息的情况下，所述输出信息的数量为以下至少一项：所述第二接收波束和所述第二发送波束对应的波束对的数量，所述波束对的数量与预测次数的乘积；
其中，所述预测次数是由所述预测时间相关信息指示的，或者所述人工智能模型确定的或者与第二设备交互获得的。
所述输出信息的数量与所述第二波束关联信息和/或所述预测时间相关信息相关联可以理解为：AI模型输出的信息数量与接收端期望的波束对关联信息，接收端期望的接收端接收波束关联信息，接收端期望的发送端发送波束关联信息，期望的预测时间相关信息以及分别对应的数量中的至少之一有关联。
本申请一实施例中，当AI模型输出为最优发送波束相关信息时，AI模型的输出数量即为最优波束数量，或等于最优波束数量*期望的预测时间相关信息指示的预测次数；所述最优波束包括：最优波束、最优发送波束以及最优接收波束；或者，
当AI模型输出为最优接收波束相关信息时，AI模型的输出数量即为最优波束数量，或等于最优波束数量*期望的预测时间相关信息指示的预测次数；所述最优波束包括：最优波束、最优发送波束以及最优接收波束；或者，
当AI模型输出为在接收端期望的接收端接收波束关联信息下，接收端在该接收波束下的所有发送端发送波束质量相关信息时，AI模型输出数量等于发送端所有发送波束的数量或发送端所有发送波束的数量*期望的预测时间相关信息指示的预测次数；或者，
当AI模型输出为在接收端期望的发送端发送波束关联信息下，接收端在该发送波束的所有接收端接收波束质量相关信息时，AI模型输出数量等于接收端所有接收波束的数量或接收端所有接收波束的数量*期望的预测时间相关信息指示的预测次数；或者，
当AI模型输出为在接收端期望的发送端发送波束关联信息和B接收波 束关联信息下，接收端在该发送波束集合和接收波束集合指示的波束对的波束质量相关信息和/或最优波束相关信息时，AI模型输出数量等于该发送波束集合和接收波束集合指示的波束对的数量或该发送波束集合和接收波束集合指示的波束对的数量*期望的预测时间相关信息指示的预测次数。
可选地，所述最优波束包括：最优波束、最优发送波束以及最优接收波束；
最优波束相关信息、最优发送相关信息以及最优接收相关信息是指携带能表征最优波束的信息，包括但不限于：波束标识相关信息和/或波束角度相关信息；此外，最优波束可以是多个。
可选地，所述人工智能模型的输入处理，还包括：
所述第一设备确定所述人工智能模型的辅助信息；其中，所述辅助信息用于辅助所述第一设备确定所述人工智能模型的输入信息和/或输出信息；所述辅助信息包括以下至少一项：
所述输入信息间的顺序；
所述输出信息间的顺序；
所述人工智能模型对输入信息的处理方式；
所述人工智能模型对输出信息的处理方式；
波束标识相关信息的数量；
波束角度相关信息的数量；
设备支持的波束相关最大能力；
发送波束的重复次数；
输出信息的循环次数；
与所述预测时间相关信息有关的信息；
第三波束关联信息。
可选地，所述输入信息间的顺序为信息输入AI模型的顺序；所述输出信息间的顺序为信息从AI模型输出的顺序；
本申请一实施例中，所述输入信息间的顺序和所述输出信息间的顺序，可以是按照一定规则定义的顺序，或指定的顺序。例如，输入信息包括期望信息、波束质量相关信息、第一波束关联信息和时间相关信息时，所述输入 信息间的顺序是按照信息类型指定的，依次输入波束质量相关信息、第一波束关联信息、时间相关信息和期望信息。
可选地，所述波束标识相关信息的数量，包括：
发送端和接收端的波束标识对相关信息数量；
发送端的发送波束标识相关信息数量；
接收端的发送波束标识相关信息数量；
发送端的接收波束标识相关信息数量；
接收端的接收波束标识相关信息数量。
本申请一实施例中，所述第一设备为接收端设备，则所述波束标识相关信息的数量，包括：
第二设备和第一设备的波束对标识相关信息数量；
第二设备的发送波束标识相关信息数量；
第一设备的发送波束标识相关信息数量；
第二设备的接收波束标识相关信息数量；
第一设备的接收波束标识相关信息数量。
可选地，所述波束角度相关信息的数量，包括：
发送端发送波束角度相关信息数量；
接收端发送波束角度相关信息数量；
发送端接收波束角度相关信息数量；
接收端接收波束角度相关信息数量。
本申请一实施例中，所述第一设备为接收端设备，则所述波束角度相关信息的数量，包括：
第二设备发送波束角度相关信息数量；
第一设备发送波束角度相关信息数量；
第二设备接收波束角度相关信息数量；
第一设备接收波束角度相关信息数量。
可选地，所述设备支持的波束相关最大能力，包括：所述设备支持的波束标识的最大能力和波束角度的最大能力；
其中，所述设备支持的波束标识的最大能力，包括：
发送端和接收端的波束标识相关信息的最大数量；
发送端的发送波束标识相关信息的最大数量；
接收端的发送波束标识相关信息的最大数量；
发送端的接收波束标识相关信息的最大；
接收端的接收波束标识相关信息的最大；
所述设备支持的波束角度的最大能力，包括：
发送端发送波束角度相关信息的最大数量；
接收端发送波束角度相关信息的最大数量；
发送端接收波束角度相关信息的最大数量；
接收端接收波束角度相关信息的最大数量。
本申请一实施例中，所述第一设备为接收端设备，则所述设备支持的波束相关最大能力，包括：
所述设备支持的波束标识的最大能力和波束角度的最大能力；
其中，所述设备支持的波束标识的最大能力，包括：
第二设备和第一设备的波束标识相关信息的最大数量；
第二设备的发送波束标识相关信息的最大数量；
第一设备的发送波束标识相关信息的最大数量；
第二设备的接收波束标识相关信息的最大数量；
第一设备的接收波束标识相关信息的最大数量；
所述设备支持的波束角度的最大能力，包括：
第二设备发送波束角度相关信息的最大数量；
第一设备发送波束角度相关信息的最大数量；
第二设备接收波束角度相关信息的最大数量；
第一设备接收波束角度相关信息的最大数量。
可选地，所述发送波束的重复次数，包括：
发送端发送波束的重复次数。其中，在重复过程中，接收端可以在相同的发送波束下使用不同的接收波束。
可选地，输出信息的循环次数，包括：
发送端建议的接收端循环的最低次数，所述循环的最低次数是指接收端 改变期望的波束关联信息获取在不同期望条件下的AI模型输出的次数；
需要说明的是，该最低次数不能超过交互的接收端接收波束标识或角度相关信息数量或最大能力。
可选地，所述与所述预测时间相关信息有关的信息可以是预测时间相关信息，也可以是用于推算出预测时间相关信息有关的信息，也可以包含预测的周期次数，周期长度。例如预测的未来周期次数，则需要接收端换算成对应的时刻。
可选地，所述第三波束关联信息，包括：发送端建议接收端AI模型输入信息有关的波束关联信息，包括以下至少之一，建议的波束对相关信息，建议的发送端发送波束标识相关信息，建议的发送端发送波束角度相关信息，建议的接收端接收波束标识相关信息，建议的接收端接收波束角度相关信息。所述建议的信息用于发送端告知接收端AI模型训练时发送端推荐的AI模型输入的波束标识和/或角度组合。
本申请一实施例中，所述第一设备为第一设备设备，则所述第三波束关联信息，包括：第二设备建议第一设备AI模型输入信息有关的波束关联信息，包括以下至少之一，建议的波束对相关信息，建议的第二设备发送波束标识相关信息，建议的第二设备发送波束角度相关信息，建议的第一设备接收波束标识相关信息，建议的第一设备接收波束角度相关信息。所述建议的信息用于第二设备告知第一设备AI模型训练时第二设备推荐的AI模型输入的波束标识和/或角度组合。
可选地，所述第一设备确定所述人工智能模型的输出信息之后，还包括：
所述第一设备在所述输出信息中选出以下至少一项：
最优发送波束相关信息；
最优接收波束相关信息；
最优波束相关信息。
需要说明的是，所述最优发送波束相关信息，所述最优接收波束相关信息以及所述最优波束相关信息均可以为一个或者多个。所述最优波束相关信息为最优波束对的相关信息。
可选地，在所述人工智能模型用于同步信号块(Synchronization Signal  Block，SSB)相关的波束预测，且SSB突发集中至少有一个SSB未使能时，最优波束关联信息是从第一SSB的波束中选择的；
其中，所述第一SSB是所述输出信息关联到的、所述SSB突发集中使能的SSB；
所述最优波束包括以下至少一项：
所述最优发送波束相关信息对应的波束；
所示最优接收波束相关信息对应的波束；
所述最优波束相关信息对应的波束。
可选地，在所述SSB突发集中，SSB是否使能通过位图指示确定；交互的最优波束仅可以从AI模型输出结果中关联到SSB突发集中使能的SSB中选择。
可选地，所述人工智能模型的输入信息、输出信息以及辅助信息中的至少一项，是所述第一设备通过与第二设备的交互确定的；
其中，所述交互的方式包括以下至少一项：
发送；上报；指示；配置；请求；预先约定。
本申请一实施例中，所述第一设备为接收端，所述第二设备为发送端，则所述人工智能模型的输入信息、输出信息以及辅助信息中的至少一项可以为所述第一设备发送给所述第二设备；或者，
所述第一设备为基站侧，所述第二设备为终端侧，则所述人工智能模型的输入信息、输出信息以及辅助信息中的至少一项可以为所述第一设备指示给所述第二设备、所述第一设备配置给所述第二设备、所述第二设备向所述第一设备上报或者所述第二设备向所述第一设备请求；或者，
所述第一设备与所述第二设备之间的预先约定。
本申请实施例的人工智能模型的输入处理方法可以通过如图4所示的人工智能模型的输入处理系统实现：
参考信号接收功率(Reference Signal Received Power，RSRP)表示输入的是RSRP信息，角度信息表示输入的是与RSRP对应的波束的角度信息，期望信息表示输入的是期望的与RSRP对应的波束的角度信息，其中，所述期望信息包括：期望的接收角度信息和/或期望的发送角度信息。
假设发送端有32个发送波束，接收端有8个接收波束：
本申请的一实施例中：AI模型的输入信息中，期望信息仅包含期望的接收角度信息，则：
AI模型输入：8个RSRP，8个RSRP对应的8个发送端发送角度信息，8个RSRP对应的8个接收端接收角度信息，1个期望的接收角度信息；
AI模型输出：在该期望的接收角度信息下预测出来的32个发送端发送波束对应的RSRP；
此时接收端循环8次，即改变期望的接收角度信息，可一共获得32*8个波束对对应的RSRP信息。
本申请的另一实施例中：AI模型的输入信息中，期望信息仅包含期望的发送角度信息，则：
AI模型输入：8个RSRP，8个RSRP对应的8个发送端发送角度信息，8个RSRP对应的8个接收端接收角度信息，1个期望的发送角度信息；
AI模型输出：在该期望的发送角度信息下预测出来的8个接收端接收波束对应的RSRP；
此时，接收端循环32次，即改变期望的发送角度信息，可一共获得8*32个波束对对应的RSRP信息。
本申请的又一实施例中，AI模型的输入信息中，期望信息包含期望的发送角度信息以及期望的接收角度信息，则：
AI模型输入：8个RSRP，8个RSRP对应的8个发送端发送角度信息，8个RSRP对应的8个接收端接收角度信息，1个期望的发送角度信息和1个期望的接收角度信息
AI模型输出：在该期望的发送角度信息和期望的接收角度信息下预测出来的1个对应的RSRP；
此时，接收端循环256次，即改变期望的发送角度和期望的接收角度信息，可一共获得1*256个波束对对应的RSRP信息。
本申请的又一实施例中，AI模型的输入信息中，期望信息包含期望的发送角度信息以及期望的接收角度信息，则：
AI模型输入：8个RSRP，8个RSRP对应的8个发送端发送角度信息， 8个RSRP对应的8个接收端接收角度信息，1个期望的发送角度信息和1个期望的接收角度信息，该1个期望的发送角度信息和1个期望的接收角度信息包含8个期望的发送角度和2个期望的接收角度；
AI模型输出：在该期望的发送角度信息和期望的接收角度信息下预测出来的16个波束对对应的RSRP；
此时，接收端循环16次，即改变期望的发送角度和期望的接收角度信息，可一共获得16*16个波束对对应的RSRP信息。
本申请的又一实施例中，AI模型的输入信息中，期望信息包含期望的发送角度信息以及期望的接收角度信息，则：
AI模型输入：8个RSRP，8个RSRP对应的8个发送端发送角度信息，8个RSRP对应的8个接收端接收角度信息，8个期望的发送角度信息和2个期望的接收角度信息，每个期望的发送角度信息和每个期望的接收角度信息包含1个期望的发送角度和1个期望的接收角度。
AI模型输出：在该期望的发送角度信息和期望的接收角度信息下预测出来的16个波束对对应的RSRP；
此时接收端循环16次，即改变期望的发送角度和期望的接收角度信息，可一共获得16*16个波束对对应的RSRP信息。
本申请的又一实施例中，AI模型输入信息中，期望信息包含期望的接收角度信息和期望预测时间相关信息，则：
AI模型输入：8个RSRP，8个RSRP对应的8个发送端发送角度信息，8个RSRP对应的8个接收端接收角度信息，1个期望的接收角度信息，1个期望预测时间相关信息用于指示预测位于基准时刻之后1秒的信息；
AI模型输出：在该期望的接收角度信息下预测位于基准时刻之后1秒的32个发送端发送波束对应的RSRP；
此时接收端循环8次，即改变期望的接收角度信息，可一共获得位于基准时刻之后1秒的32*8个波束对对应的RSRP信息。
如图5所示，本申请实施例还提供一种人工智能模型的输入处理装置500，包括：
第一确定模块501，用于通过第一设备确定人工智能模型的输入信息； 其中，所述人工智能模型用于波束预测相关功能，所述输入信息包括期望信息以及以下至少一项：
波束质量相关信息；
第一波束关联信息；
时间相关信息。
在本申请实施例中，通过第一设备的第一确定模块确定人工智能模型的输入信息；其中，所述人工智能模型用于波束预测相关功能，所述输入信息包括期望信息以及以下至少一项：波束质量相关信息；第一波束关联信息；时间相关信息。这样，本申请的方案能够不受限于第一设备的能力，通过第一设备的AI模型获得期望的输出信息，并且人工智能模型的输出数量固定，与终端设备的接收波束数量或角度无关，解决了因为终端设备的能力不同导致人工智能模型输出数量无法广泛应用的问题，增加了人工智能模型的实用性。
可选地，所述人工智能模型的输入处理装置还包括：
第二确定模块，用于通过所述第一设备确定所述人工智能模型的输出信息；其中，所述输出信息包括以下至少一项：
第一接收波束下，至少部分发送波束质量相关信息和/或最优发送波束相关信息；
第一发送波束下，至少部分接收波束质量相关信息和/或最优接收波束相关信息；
第二接收波束和第二发送波束下，对应的波束对的波束质量相关信息和/或最优波束相关信息；
其中，所述第一接收波束和第二接收波束为所述第二波束关联信息表征的接收波束；所述第一发送波束和第二发送波束为所述第二波束关联信息表征的发送波束。
本申请一实施例中，所述输出信息包括以下至少一项：
接收端期望的接收波束关联信息下，接收端在该接收波束下的所有发送端发送波束质量相关信息和/或最优发送波束相关信息；
接收端期望的发送端发送波束关联信息下，接收端在该发送波束的所有 接收端接收波束质量相关信息和/或最优接收波束相关信息；
接收端期望的发送端发送波束关联信息和接收端接收波束关联信息下，接收端在该发送波束集合和接收波束集合指示的波束对的波束质量相关信息和/或最优波束相关信息；
其中，所述第一发送波束和第二发送波束是根据接收端期望的发送端发送波束关联信息获得，所述第一接收波束和第二接收波束是根据接收端期望的接收端接收波束关联信息获得。
可选地，所述人工智能模型的输入处理装置还包括：
第三确定模块，用于通过所述第一设备确定所述人工智能模型的辅助信息；其中，所述辅助信息用于辅助所述第一设备确定所述人工智能模型的输入信息和/或输出信息；所述辅助信息包括以下至少一项：
所述输入信息间的顺序；
所述输出信息间的顺序；
所述人工智能模型对输入信息的处理方式；
所述人工智能模型对输出信息的处理方式；
波束标识相关信息的数量；
波束角度相关信息的数量；
设备支持的波束相关最大能力；
发送波束的重复次数；
输出信息的循环次数；
与所述预测时间相关信息有关的信息；
第三波束关联信息。
可选地，所述输入信息间的顺序为信息输入AI模型的顺序；所述输出信息间的顺序为信息从AI模型输入的顺序；
本申请一实施例中，所述输入信息间的顺序和所述输出信息间的顺序，可以是按照一定规则定义的顺序，或指定的顺序。
可选地，所述波束标识相关信息的数量，包括：
发送端和接收端的波束标识相关信息数量；
发送端的发送波束标识相关信息数量；
接收端的发送波束标识相关信息数量；
发送端的接收波束标识相关信息数量；
接收端的接收波束标识相关信息数量。
可选地，所述波束角度相关信息的数量，包括：
发送端发送波束角度相关信息数量；
接收端发送波束角度相关信息数量；
发送端接收波束角度相关信息数量；
接收端接收波束角度相关信息数量。
可选地，所述设备支持的波束相关最大能力，包括：所述设备支持的波束标识的最大能力和波束角度的最大能力；
其中，所述设备支持的波束标识的最大能力，包括：
发送端和接收端的波束标识相关信息数量或支持的最大能力；
发送端的发送波束标识相关信息数量或支持的最大能力；
接收端的发送波束标识相关信息数量或支持的最大能力；
发送端的接收波束标识相关信息数量或支持的最大能力；
接收端的接收波束标识相关信息数量或支持的最大能力；
所述设备支持的波束角度的最大能力，包括：
发送端发送波束角度相关信息数量或支持的最大能力；
接收端发送波束角度相关信息数量或支持的最大能力；
发送端接收波束角度相关信息数量或支持的最大能力；
接收端接收波束角度相关信息数量或支持的最大能力。
可选地，所述发送波束的重复次数，包括：
发送端发送波束的重复次数，用于指示接收端可以在相同的发送波束下使用不同的接收波束。
可选地，输出信息的循环次数，包括：
发送端建议的接收端循环的最低次数，所述循环的最低次数是指接收端改变期望的关联信息获取在不同期望条件下的AI模型输出的次数；
需要说明的是，该最低次数不能超过交互的接收端接收波束标识或角度相关信息数量或最大能力。
可选地，所述人工智能模型的输入处理装置还包括：
选择模块，用于通过所述第一设备在输出信息中选出以下至少一项：
最优发送波束相关信息；
最优接收波束相关信息；
最优波束相关信息。
需要说明的是，所述最优发送波束相关信息，所述最优接收波束相关信息以及所述最优波束相关信息均可以为一个或者多个。所述最优波束相关信息为波束对的相关信息。
本申请实施例中的人工智能模型的输入处理设备可以是电子设备，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人工智能模型的输入处理设备能够实现图3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。
可选地，如图6所示，本申请实施例还提供一种人工智能模型的输入处理设备600，包括处理器601和存储器602，存储器602上存储有可在所述处理器601上运行的程序或指令，例如，该人工智能模型的输入处理设备600为终端时，该程序或指令被处理器601执行时实现上述人工智能模型的输入处理方法实施例的各个步骤，且能达到相同的技术效果。该人工智能模型的输入处理设备600为网络侧设备时，该程序或指令被处理器601执行时实现上述人工智能模型的输入处理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，处理器用于通过第一设备确定人工智能模型的输入信息；
其中，所述人工智能模型用于波束相关功能，所述输入信息包括期望信息以及以下至少一项：波束质量相关信息；
第一波束关联信息；
时间相关信息。
该通信设备实施例与上述第一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终端实施例中，且能达到相同的技术效果。具体地，图7为实现本申请实施例的通信设备的硬件结构示意图。
该通信设备700包括但不限于：射频单元701、网络模块702、音频输出单元703、输入单元704、传感器705、显示单元706、用户输入单元707、接口单元708、存储器709以及处理器710等中的至少部分部件。
本领域技术人员可以理解，该通信设备700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710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7中示出的该通信设备结构并不构成对人工智能模型的输入处理该通信设备的限定，该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704可以包括图形处理单元(Graphics Processing Unit，GPU)7041和麦克风7042，图形处理器7041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706可包括显示面板7061，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7061。用户输入单元707包括触控面板7071以及其他输入设备7072中的至少一种。触控面板7071，也称为触摸屏。触控面板7071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7072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701接收来自网络侧设备的下行数据后，可以传输给处理器710进行处理；另外，射频单元701可以向网络侧设备发送上行数据。通常，射频单元701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709可用于存储软件程序或指令以及各种数据。存储器709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709可以包括易失性存储器或非易失性存储器，或者，存储器709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709包括但不限于这些和任意其它适合类型的存储器。
处理器710可包括一个或多个处理单元；可选地，处理器710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710中。
其中，处理器710，用于通过第一设备确定人工智能模型的输入信息；其中，所述人工智能模型用于波束相关功能，所述输入信息包括期望信息以及以下至少一项：
波束质量相关信息；
第一波束关联信息；
时间相关信息。
所述处理器710，还用于所述第一设备确定所述人工智能模型的输出信息；其中，所述输出信息包括以下至少一项：
第一接收波束下，至少部分发送波束质量相关信息和/或最优发送波束相关信息；
第一发送波束下，至少部分接收波束质量相关信息和/或最优接收波束相 关信息；
第二接收波束和第二发送波束下，对应的波束对的波束质量相关信息和/或最优波束相关信息；
其中，所述第一接收波束和第二接收波束为所述第二波束关联信息表征的接收波束；所述第一发送波束和第二发送波束为所述第二波束关联信息表征的发送波束。
所述处理器710，还用于所述第一设备确定所述人工智能模型的辅助信息；其中，所述辅助信息用于辅助所述第一设备确定所述人工智能模型的输入信息和/或输出信息；所述辅助信息包括以下至少一项：
所述输入信息间的顺序；
所述输出信息间的顺序；
所述人工智能模型对输入信息的处理方式；
所述人工智能模型对输出信息的处理方式；
波束标识相关信息的数量；
波束角度相关信息的数量；
设备支持的波束相关最大能力；
发送波束的重复次数；
输出信息的循环次数；
与所述预测时间相关信息有关的信息；
第三波束关联信息。
所述处理器710，还用于所述第一设备在输出信息中选出以下至少一项：
最优发送波束相关信息；
最优接收波束相关信息；
最优波束相关信息。
本申请实施例另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，还程序或者指令被处理器执行时实现如上人工智能模型的输入处理方法的实施例的各个过程，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通信设备中的处理器。所述可 读存储介质，包括计算机可读存储介质，如计算机可读存储器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，实现上述人工智能模型的输入处理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人工智能模型的输入处理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。
本申请实施例另提供了一种人工智能模型的输入处理系统，包括：通信设备，所述通信设备可用于执行如上所述的人工智能模型的输入处理方法。
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 (22)

  1. 一种人工智能模型的输入处理方法，包括：
    第一设备确定人工智能模型的输入信息；其中，所述人工智能模型用于波束相关功能，所述输入信息包括期望信息以及以下至少一项：
    波束质量相关信息；
    第一波束关联信息；
    时间相关信息。
  2. 根据权利要求1所述的方法，其中，所述期望信息包括以下至少一项：
    第二波束关联信息；
    预测时间相关信息。
  3. 根据权利要求1所述的方法，其中，所述波束质量相关信息包括以下至少一项：信噪比，接收功率，接收质量；
    所述第一波束关联信息用于表征所述人工智能模型的输入选择第一波束的情况下，所述第一波束对应的关联信息；其中，所述第一波束关联信息包括以下至少一项：波束标识相关信息，波束角度相关信息，波束增益相关信息，波束宽度相关信息；
    所述时间相关信息用于表征所述波束质量相关信息对应的时间，和/或，多个所述波束质量相关信息之间的时间关系。
  4. 根据权利要求2所述的方法，其中，所述第二波束关联信息用于表征期望的第二波束所对应的关联信息；其中，所述第二波束关联信息包括以下至少一项：
    波束标识相关信息，波束角度相关信息，波束增益相关信息，波束宽度相关信息；
    所述预测时间相关信息用于表征所述人工智能模型的输出信息关联的时间相关信息。
  5. 根据权利要求1所述的方法，其中，所述输入信息中，同一个输入类型对应的信息包括至少一个时间上的信息。
  6. 根据权利要求1所述的方法，其中，所述第一波束关联信息和/或所述 时间相关信息，与所述波束质量相关信息相关联；和/或
    所述第一波束关联信息的数量和/或所述时间相关信息的数量，与所述波束质量相关信息的数量相关联。
  7. 根据权利要求6所述的方法，其中，所述第一波束关联信息和/或所述时间相关信息，与所述波束质量相关信息相关联，包括以下至少一项：
    所述波束质量相关信息对应的波束与所述第一波束关联信息表征的波束一一对应；
    所述波束质量相关信息对应的波束与所述时间相关信息表征的波束一一对应；
    所述波束质量相关信息对应的波束与所述第一波束关联信息和所述时间相关信息联合表征的波束一一对应。
  8. 根据权利要求6所述的方法，其中，所述第一波束关联信息的数量和/或所述时间相关信息的数量，与所述波束质量相关信息的数量相关联，包括以下至少一项：
    所述第一波束关联信息包括的波束对相关的关联信息、发送波束关联信息以及接收波束关联信息中至少一者的数量，等于所述波束质量相关信息的数量；
    所述时间相关信息的数量等于所述波束质量相关信息的数量；
    所述第一波束关联信息包括发送波束关联信息和接收波束关联信息的情况下，所述发送波束关联信息的数量与所述接收波束关联信息的数量的乘积等于所述波束质量相关信息的数量；
    所述第一波束关联信息包括的波束对相关的关联信息、发送波束关联信息以及接收波束关联信息中至少一者的数量与所述时间相关信息的数量的乘积等于所述波束质量相关信息的数量；
    所述第一波束关联信息包括发送波束关联信息和接收波束关联信息的情况下，所述发送波束关联信息的数量与所述时间相关信息的数量的乘积等于所述波束质量相关信息的数量；
    所述第一波束关联信息包括发送波束关联信息和接收波束关联信息的情况下，所述接收波束关联信息的数量与所述时间相关信息的数量的乘积等于 所述波束质量相关信息的数量；
    所述第一波束关联信息包括发送波束关联信息和接收波束关联信息的情况下，所述发送波束关联信息的数量、所述接收波束关联信息的数量以及所述时间相关信息的数量的乘积等于所述波束质量相关信息的数量。
  9. 根据权利要求2所述的方法，其中，还包括：
    所述第一设备确定所述人工智能模型的输出信息；其中，所述输出信息包括以下至少一项：
    第一接收波束下，至少部分发送波束质量相关信息和/或最优发送波束相关信息；
    第一发送波束下，至少部分接收波束质量相关信息和/或最优接收波束相关信息；
    第二接收波束和第二发送波束下，对应的波束对的波束质量相关信息和/或最优波束相关信息；
    其中，所述第一接收波束和第二接收波束为所述第二波束关联信息表征的接收波束；所述第一发送波束和第二发送波束为所述第二波束关联信息表征的发送波束。
  10. 根据权利要求9所述的方法，其中，所述输出信息中，同一个输出类型对应的信息包括至少一个时间上的信息。
  11. 根据权利要求9所述的方法，其中，所述输出信息的数量与所述第二波束关联信息和/或所述预测时间相关信息相关联。
  12. 根据权利要求11所述的方法，其中，所述输出信息的数量与所述第二波束关联信息和/或所述预测时间相关信息相关联，包括：
    在所述输出信息包括所述最优发送波束相关信息的情况下，所述输出信息的数量为以下至少一项：最优波束数量，最优波束数量与预测次数的乘积；或者，
    在所述输出信息包括所述最优接收波束相关信息的情况下，所述输出信息的数量为以下至少一项：最优波束数量，最优波束数量与预测次数的乘积；或者，
    在所述输出信息包括所述第一接收波束下的所有发送波束质量相关信息 的情况下，所述输出信息的数量为以下至少一项：至少部分发送波束的数量，至少部分发送波束的数量与预测次数的乘积；
    在所述输出信息包括所述第一发送波束下的所有接收波束质量相关信息的情况下，所述输出信息的数量为以下至少一项：至少部分接收波束的数量，至少部分接收波束的数量与预测次数的乘积；
    在所述输出信息包括所述第二接收波束和所述第二发送波束下，对应的波束对的波束质量相关信息和/或最优波束相关信息的情况下，所述输出信息的数量为以下至少一项：所述第二接收波束和所述第二发送波束对应的波束对的数量，所述波束对的数量与预测次数的乘积；
    其中，所述预测次数是由所述预测时间相关信息指示的，或者所述人工智能模型确定的或者与第二设备交互获得的。
  13. 根据权利要求2所述的方法，其中，还包括：
    所述第一设备确定所述人工智能模型的辅助信息；其中，所述辅助信息用于辅助所述第一设备确定所述人工智能模型的输入信息和/或输出信息；所述辅助信息包括以下至少一项：
    所述输入信息间的顺序；
    所述输出信息间的顺序；
    所述人工智能模型对输入信息的处理方式；
    所述人工智能模型对输出信息的处理方式；
    波束标识相关信息的数量；
    波束角度相关信息的数量；
    设备支持的波束相关最大能力；
    发送波束的重复次数；
    输出信息的循环次数；
    与所述预测时间相关信息有关的信息；
    第三波束关联信息。
  14. 根据权利要求9所述的方法，其中，所述第一设备确定所述人工智能模型的输出信息之后，还包括：
    所述第一设备在所述输出信息中选出以下至少一项：
    最优发送波束相关信息；
    最优接收波束相关信息；
    最优波束相关信息。
  15. 根据权利要求9或14所述的方法，其中，在所述人工智能模型用于同步信号块SSB相关的波束预测，且SSB突发集中至少有一个SSB未使能时，最优波束关联信息是从第一SSB的波束中选择的；
    其中，所述第一SSB是所述输出信息关联到的、所述SSB突发集中使能的SSB；
    所述最优波束包括以下至少一项：
    所述最优发送波束相关信息对应的波束；
    所示最优接收波束相关信息对应的波束；
    所述最优波束相关信息对应的波束。
  16. 根据权利要求1所述的方法，其中，所述人工智能模型的输入信息、输出信息以及辅助信息中的至少一项，是所述第一设备通过与第二设备的交互确定的；
    其中，所述交互的方式包括以下至少一项：
    发送；上报；指示；配置；请求；预先约定。
  17. 一种人工智能模型的输入处理装置，包括：
    第一确定模块，用于通过第一设备确定人工智能模型的输入信息；其中，所述人工智能模型用于波束预测相关功能，所述输入信息包括期望信息以及以下至少一项：
    波束质量相关信息；
    第一波束关联信息；
    时间相关信息。
  18. 根据权利要求17所述的装置，其中，还包括：
    第二确定模块，用于通过所述第一设备确定所述人工智能模型的输出信息；其中，所述输出信息包括以下至少一项：
    第一接收波束下，至少部分发送波束质量相关信息和/或最优发送波束相关信息；
    第一发送波束下，至少部分接收波束质量相关信息和/或最优接收波束相关信息；
    第二接收波束和第二发送波束下，对应的波束对的波束质量相关信息和/或最优波束相关信息；
    其中，所述第一接收波束和第二接收波束为第二波束关联信息表征的接收波束；所述第一发送波束和第二发送波束为所述第二波束关联信息表征的发送波束。
  19. 根据权利要求17所述的装置，其中，还包括：
    第三确定模块，用于通过所述第一设备确定所述人工智能模型的辅助信息；其中，所述辅助信息用于辅助所述第一设备确定所述人工智能模型的输入信息和/或输出信息；所述辅助信息包括以下至少一项：
    所述输入信息间的顺序；
    所述输出信息间的顺序；
    所述人工智能模型对输入信息的处理方式；
    所述人工智能模型对输出信息的处理方式；
    波束标识相关信息的数量；
    波束角度相关信息的数量；
    设备支持的波束相关最大能力；
    发送波束的重复次数；
    输出信息的循环次数；
    与所述预测时间相关信息有关的信息；
    第三波束关联信息。
  20. 根据权利要求17所述的装置，其中，还包括：
    选择模块，用于通过所述第一设备在输出信息中选出以下至少一项：
    最优发送波束相关信息；
    最优接收波束相关信息；
    最优波束相关信息。
  21. 一种人工智能模型的输入处理设备，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 执行时实现如权利要求1至16任一项所述的人工智能模型的输入处理方法的步骤。
  22. 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权利要求1至16任一项所述的人工智能模型的输入处理方法的步骤。
PCT/CN2023/083041 2022-03-23 2023-03-22 人工智能模型的输入处理方法、装置及设备 WO2023179649A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210295984.5 2022-03-23
CN202210295984.5A CN116847368A (zh) 2022-03-23 2022-03-23 人工智能模型的输入处理方法、装置及设备

Publications (1)

Publication Number Publication Date
WO2023179649A1 true WO2023179649A1 (zh) 2023-09-28

Family

ID=88100025

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/083041 WO2023179649A1 (zh) 2022-03-23 2023-03-22 人工智能模型的输入处理方法、装置及设备

Country Status (2)

Country Link
CN (1) CN116847368A (zh)
WO (1) WO2023179649A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180262918A1 (en) * 2017-03-08 2018-09-13 Sony Corporation Electronic apparatus and method in wireless communication system
CN111954228A (zh) * 2019-05-16 2020-11-17 北京三星通信技术研究有限公司 波束管理方法、装置、电子设备及计算机可读存储介质
US10938464B1 (en) * 2019-08-31 2021-03-02 Lg Electronics Inc. Intelligent beamforming method, apparatus and intelligent computing device
CN113365352A (zh) * 2020-03-06 2021-09-07 大唐移动通信设备有限公司 波束推荐方法、装置、网络侧设备、终端及存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180262918A1 (en) * 2017-03-08 2018-09-13 Sony Corporation Electronic apparatus and method in wireless communication system
CN111954228A (zh) * 2019-05-16 2020-11-17 北京三星通信技术研究有限公司 波束管理方法、装置、电子设备及计算机可读存储介质
US10938464B1 (en) * 2019-08-31 2021-03-02 Lg Electronics Inc. Intelligent beamforming method, apparatus and intelligent computing device
CN113365352A (zh) * 2020-03-06 2021-09-07 大唐移动通信设备有限公司 波束推荐方法、装置、网络侧设备、终端及存储介质

Also Published As

Publication number Publication date
CN116847368A (zh) 2023-10-03

Similar Documents

Publication Publication Date Title
CN116266216A (zh) 客户端筛选方法及装置、客户端及中心设备
WO2023246757A1 (zh) 算力服务方法、装置及终端
WO2023179649A1 (zh) 人工智能模型的输入处理方法、装置及设备
WO2023179651A1 (zh) 波束处理方法、装置及设备
WO2023179653A1 (zh) 波束处理方法、装置及设备
WO2023186014A1 (zh) 信号发送方法、信号接收方法及通信设备
WO2023198093A1 (zh) 模型有效性的确定方法、装置及通信设备
WO2023198167A1 (zh) 模型微调的方法、装置及设备
WO2024061111A1 (zh) 资源处理方法、装置及通信设备
WO2023241448A1 (zh) 测量处理方法、终端及网络侧设备
WO2023151650A1 (zh) 信息激活方法、终端及网络侧设备
WO2023109828A1 (zh) 数据收集方法及装置、第一设备、第二设备
WO2024022166A1 (zh) Ra方法、装置、ue、网络侧设备、通信系统及可读存储介质
WO2024007950A1 (zh) Sl定位方法、设备及可读存储介质
WO2023227022A1 (zh) 信道状态信息csi的传输方法、装置及终端
WO2024022168A1 (zh) 协商方法、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
WO2023246756A1 (zh) 算力服务方法、装置、终端及核心网设备
WO2023186018A1 (zh) 信息传输方法、装置、终端及可读存储介质
WO2024001981A1 (zh) 预编码矩阵的指示方法、终端及网络侧设备
WO2024012303A1 (zh) 一种ai网络模型交互方法、装置和通信设备
WO2024037514A1 (zh) 信息传输方法、装置及通信设备
WO2024017239A1 (zh) 数据采集方法及装置、通信设备
WO2023241449A1 (zh) 测量处理方法、装置及设备
WO2023169430A1 (zh) Pusch传输方法、终端及网络侧设备

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: 23773896

Country of ref document: EP

Kind code of ref document: A1