CN114189889A - Wireless communication artificial intelligence processing method and device - Google Patents
Wireless communication artificial intelligence processing method and device Download PDFInfo
- Publication number
- CN114189889A CN114189889A CN202111471065.0A CN202111471065A CN114189889A CN 114189889 A CN114189889 A CN 114189889A CN 202111471065 A CN202111471065 A CN 202111471065A CN 114189889 A CN114189889 A CN 114189889A
- Authority
- CN
- China
- Prior art keywords
- information
- artificial intelligence
- feedback
- downlink
- model
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 154
- 238000004891 communication Methods 0.000 title claims abstract description 23
- 238000003672 processing method Methods 0.000 title claims abstract description 12
- 238000000034 method Methods 0.000 claims abstract description 54
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000013528 artificial neural network Methods 0.000 claims abstract description 7
- 238000004590 computer program Methods 0.000 claims description 14
- 238000010295 mobile communication Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 9
- 230000004044 response Effects 0.000 claims description 8
- 230000011664 signaling Effects 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 abstract description 10
- 238000013480 data collection Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 14
- 238000007726 management method Methods 0.000 description 14
- 230000000694 effects Effects 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 7
- 238000010276 construction Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 101000741965 Homo sapiens Inactive tyrosine-protein kinase PRAG1 Proteins 0.000 description 1
- 102100038659 Inactive tyrosine-protein kinase PRAG1 Human genes 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The application discloses a wireless communication artificial intelligence processing method, which comprises the following steps: the downlink information comprises first information, the first information is used for indicating N artificial intelligence models to download services for selection, and each artificial intelligence model comprises a neural network structural feature and parameters; each artificial intelligence model is used for corresponding network side and mobile side performance processing. The application also includes an apparatus for implementing the method. The method and the device provided by the invention can effectively carry out wireless data collection, processing and model circulation, and support the application of various actual wireless communication systems based on AI technology.
Description
Technical Field
The present application relates to the field of wireless communications technologies, and in particular, to an artificial intelligence processing method and device.
Background
The problems faced in the mobile communication system are complex and various, and researches show that the performance of a mobile communication network side and a wireless side can be effectively improved by utilizing an Artificial Intelligence (AI) technology. The mobile communication network contains a large amount of data resources, and the reasonable utilization and development of the 5G data resources by utilizing the AI technology can effectively improve the mobile communication system.
When designing a communication system using the AI technique, a series of processes such as data collection and model transfer are required. The data set and model construction of a mobile communication system directly affect the effect of solving related problems by using the AI technique. The actual system construction needs to combine the basic theory of the wireless communication system and the AI basic theory and consider the constraints of various practical situations.
For the design of a wireless physical layer, the good process design can effectively utilize an AI technology to solve the practical problem and improve the performance of a wireless system. Poor design can make it difficult for AI techniques to take real value and even degrade system performance. AI-based air interface techniques are not explicitly supported in the current 5G standard. The functions of data collection, model transmission and the like related to the AI model between the network device and the terminal are not embodied in the 5G standard.
Disclosure of Invention
The invention provides a wireless communication artificial intelligence processing method and equipment, and provides a solution scheme combining data set construction and AI model use.
In a first aspect, the present application provides a method for processing artificial intelligence in wireless communication, including the following steps:
the downlink information comprises first information, the first information is used for indicating N artificial intelligence models to download services for selection, and each artificial intelligence model comprises a neural network structural feature and parameters;
each artificial intelligence model is used for corresponding network side and mobile side performance processing.
Further, the first information is a high-level signaling or a downlink control signaling DCI; the first information comprises N bits, and each 1 bit identifies 1 artificial intelligence model downloading service.
Preferably, the first information is further used for indicating a first feedback time for the first information.
Further, the uplink information includes third information as feedback to the first information;
the third information comprises selection indications of the N artificial intelligence model downloading services.
Further, the downlink information further includes second information, and the second information is used for indicating a second feedback time and a type of feedback data; and at the second feedback time, the uplink information comprises the feedback data.
Further, according to the feedback of the first information in the third information, the downlink information comprises fourth information, and the fourth information comprises 1 or more of the N artificial intelligence models.
The method according to any one of the embodiments of the first aspect of the present application, for a network device, includes the following steps:
sending downlink information, wherein the downlink information comprises the first information;
receiving uplink information, wherein the uplink information comprises a selection instruction of the downloading service of the N artificial intelligence models;
in response to the selection indication, the downstream information includes 1 or more of the N artificial intelligence models.
The method according to any one of the embodiments of the first aspect of the present application, applied to a terminal device, includes the following steps:
receiving downlink information, wherein the downlink information comprises the first information;
and sending uplink information, wherein the uplink information comprises selection instructions of the downloading services of the N artificial intelligence models.
Further, receiving downlink information, wherein the downlink information comprises 1 or more artificial intelligence models corresponding to the selection indication;
and executing the artificial intelligence model to generate feedback data.
Further, receiving downlink information, wherein the downlink information also comprises the type of feedback data and an indication of feedback time;
and sending the feedback data at the feedback time indicated by the downlink information.
In a second aspect, the present application further provides a network device, configured to implement the method in any embodiment of the first aspect of the present application, where at least one module in the network device is configured to perform at least one of the following functions: sending downlink information, wherein the downlink information comprises the first information; receiving uplink information, wherein the uplink information comprises a selection instruction of the downloading service of the N artificial intelligence models; in response to the selection indication, the downstream information includes 1 or more of the N artificial intelligence models.
In a third aspect, the present application further provides a terminal device, configured to implement the method in any embodiment of the first aspect of the present application, where at least one module in the terminal device is configured to: receiving downlink information, wherein the downlink information comprises the first information; sending uplink information, wherein the uplink information comprises a selection instruction of the downloading service of the N artificial intelligence models; receiving downlink information, wherein the downlink information comprises 1 or more artificial intelligence models corresponding to the selection indication; executing the artificial intelligence model to generate feedback data; receiving downlink information, wherein the downlink information also comprises the type of feedback data and an indication of feedback time; and sending the feedback data at the feedback time indicated by the downlink information.
In a fourth aspect, the present application further provides a communication device, including: memory, a processor and a computer program stored on the memory and executable on the processor, the computer program, when executed by the processor, implementing the steps of the method according to any one of the embodiments of the first aspect of the application.
In a fifth aspect, the present application also proposes a computer-readable medium on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the method according to any one of the embodiments of the first aspect of the present application.
In a sixth aspect, the present application further provides a mobile communication system, which includes at least one network device according to any embodiment of the present application and/or at least one terminal device according to any embodiment of the present application.
The embodiment of the application adopts at least one technical scheme which can achieve the following beneficial effects:
the method, the equipment and the communication system provided by the invention can enable the network equipment and the terminal equipment to realize the construction, the AI model training, the AI model deployment and the use of data sets of different applications through necessary information exchange, thereby achieving the effect of solving a plurality of key problems in the wireless mobile communication system through the artificial intelligence technology. This information exchange mechanism is very critical to solve the mobile communication problem using artificial intelligence technology. The method provided by the invention ensures that the terminal equipment and the network equipment determine the scene using the AI model and how to construct the data set through information exchange, thereby realizing efficient and accurate data set establishment and AI model exchange and improving the efficiency of solving key problems by using the AI technology in the communication system.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a flow chart of an embodiment of the method of the present application;
FIG. 2 is a schematic diagram of an embodiment in which multiple artificial intelligence models are simultaneously delivered with second information and fourth information;
FIG. 3 is a schematic diagram of an embodiment in which multiple artificial intelligence models are downloaded independently;
FIG. 4 is a schematic diagram of an embodiment of an artificial intelligence model download of a pre-upload data requirement;
FIG. 5 is a flow chart of an embodiment of a method of the present application for a network device;
FIG. 6 is a flowchart of an embodiment of a method of the present application for a terminal device;
FIG. 7 is a schematic diagram of an embodiment of a network device;
FIG. 8 is a schematic diagram of an embodiment of a terminal device;
fig. 9 is a schematic structural diagram of a network device according to another embodiment of the present invention;
fig. 10 is a block diagram of a terminal device of another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Consider a communication system comprised of network devices and terminal devices. One network device can simultaneously transmit and receive data to a plurality of terminal devices. The network equipment and the terminal equipment transmit data through a downlink data shared channel (PDSCH) and an uplink data shared channel (PUSCH); and the control information exchange is carried out through a downlink control channel (PDCCH) and an uplink access channel (PRACH) and a control channel (PUCCH) of a synchronization and broadcast channel (SS/PBCH). The SS/PBCH sends synchronization signals and broadcast information, and the terminal control unit receives the SS/PBCH to carry out synchronization and acquire basic system information. The PDCCH transmits Downlink Control Information (DCI) and performs specific transmission format-related contents of the PDSCH, PUSCH, and PUCCH. The terminal initiates access based on PRACH to the network equipment according to the control information sent by the network equipment and the receiving condition of the terminal data, or feeds back whether the data correctly receives ACK/NACK information. The basic time transmission unit in the system is a symbol, and 14 symbols form a time slot. A time slot of length 1/2kAnd ms, where k is a positive integer and corresponds to different subcarrier intervals, respectively, and when k is 0,1,2,3,4,5, and 6, the subcarrier intervals correspond to 15kHz, 30kHz, 60kHz, 120kHz, 240kHz, 480kHz, and 960 kHz.
The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is a flow chart of an embodiment of the method of the present application.
In a first aspect, the present application provides a wireless communication artificial intelligence processing method, including the following steps 101-104:
Each artificial intelligence model comprises a neural network structural feature and a parameter; each artificial intelligence model is used for corresponding network side and mobile side performance processing.
Further, the first information is a high-level signaling or a downlink control signaling DCI; the first information comprises N bits, and each 1 bit identifies 1 artificial intelligence model downloading service.
Preferably, the first information is further used for indicating a first feedback time for the first information.
102, the uplink information comprises third information serving as feedback of the first information;
the third information comprises selection indications of the N artificial intelligence model downloading services.
Step 103, further, the downlink information further includes second information, where the second information is used to indicate a second feedback time and a type of feedback data; and at the second feedback time, the uplink information comprises the feedback data.
It should be noted that the feedback data includes at least one of the following data:
generating and training data required by the artificial intelligence model;
operating an artificial intelligence model to obtain result data;
and (3) evaluating the operation effect of the artificial intelligent model.
And 104, further, according to the feedback of the first information, the downlink information comprises fourth information, and the fourth information comprises 1 or more of the N artificial intelligence models.
The fourth information may be carried by the PDSCH, or jointly indicated by DCI information carried by the PDCCH, and the artificial intelligence model is included in the PDSCH indicated by the DCI.
Through the steps, the network device provides the AI service which can be provided by the network device to the terminal device through the first information and the second information, and provides information, data and corresponding feedback positions which need to be fed back. And the terminal equipment feeds back the third information according to the first information and the second information, determines the used AI model and starts corresponding data feedback. And the network equipment issues the fourth information according to the third information and data feedback of the terminal.
In order to execute step 104, according to the feedback of the first information and the second information, the data set construction and the AI model training of different applications can be realized, and then the AI model deployment is provided. And after the terminal equipment receives the fourth information, generating a processing result by using the AI model.
Fig. 2 is a schematic diagram of an embodiment in which multiple artificial intelligence models are delivered simultaneously with second information and fourth information.
In this embodiment, the standard specifies that the 5G network can explicitly support 5 AI model services, such as a model for "channel estimation," a model for "channel information feedback," a "positioning" model, "a" beam management "model," a "mobility management" model, and N is 5. The network device may provide 3 AI model services, such as channel estimation, channel information feedback, and beam management, where 5 bits of the corresponding first information M indicate 11010, where each 1 bit represents an artificial intelligence model downloading service, a 1 st bit is 1 to indicate that "channel estimation" artificial intelligence model downloading service can be provided, a 2 nd bit is 1 to indicate that "channel information feedback" artificial intelligence model downloading service can be provided, a 3 rd bit is 0 to indicate that "positioning" artificial intelligence model downloading service cannot be provided, a 4 th bit is 1 to indicate that "beam management" artificial intelligence model downloading service can be provided, and a 5 th bit is 0 to indicate that "mobility management" artificial intelligence model downloading service cannot be provided. The first information is sent by DCI information carried by the PDCCH.
The first information indicates a first feedback time at the same time, for example, the third information is fed back to the 10 th time slot after the terminal carries the first information PDCCH.
The AI model to be used is determined by the selection indication in the third information. For example, the terminal may wish to use AI-based "channel estimation" and "channel information feedback", where after the first information indicates a location, i.e. after the first information PDCCH is carried, F is 10 slots, the third information is fed back as: 11000.
and after receiving the third information sent by the terminal equipment, the network equipment sends the second information to the terminal equipment, wherein the second information indicates second feedback time and the type and the quantity of feedback data. For example, the terminal device supports two types of AI model feedback data, the amount of each type of data, and a feedback period. Specifically, the data supporting the "channel estimation" and needing to be fed back is feedback of 1 bit fed back to the model effect every 100 time slots, the feedback supporting the "channel information feedback" is feedback of 10 CSI-to-be-compressed original data every 100 time slots, the period is P-100 time slots, and the feedback time is S-80 time slots after the second information PDCCH is sent. The "1-bit model 1 feedback" in fig. 2 contains evaluation feedback for model 1, such as indicating "good" or "bad".
In an embodiment, the network device sends the fourth information to the terminal device while sending the second information, where the fourth information includes 1 each of AI models used for "channel estimation" and "channel information feedback", and is sent on the PDSCH where the second information is located.
FIG. 3 is a schematic diagram of an embodiment in which multiple artificial intelligence models are downloaded independently.
In this embodiment, the standard specifies that the 5G network can explicitly support 5 AI model services, such as a model for "channel estimation," a model for "channel information feedback," a "positioning" model, "a" beam management "model," a "mobility management" model, and N is 5. The network device may provide 3 AI model services, such as channel estimation, channel information feedback, and beam management, and the corresponding first information M ═ 5 bits are indicated as 11010, which means the same as the previous embodiment. The first information is sent by DCI information carried by the PDCCH.
The first information indicates a first feedback time at the same time, for example, the third information is fed back to the 10 th time slot after the terminal carries the first information PDCCH, and the AI model used in the third information is determined by the selection indication. For example, the terminal may wish to use an AI-based "channel estimation" and "positioning" model to feed back the third information at the 10 th time slot after the first information indicates a location, i.e. carries the first information PDCCH: 10100.
after receiving the third information sent by the terminal device, the network device first sends the second information related to channel estimation to the terminal device, and indicates the type, the amount and the second feedback time of the feedback data. For example, the data that needs to be fed back to support channel estimation is P1-100 time slots, and 1 bit of data is fed back to evaluate the model effect, and the feedback starting time is S1-80 time slots after the distance from the channel model transmission.
In this embodiment, in the PDSCH in the same time slot as the second information, the network device sends "channel estimation" to the terminal device, where the fourth information includes an AI model used for channel estimation.
And in a second time slot after the channel estimation model is sent, the network equipment sends second information to the terminal again, wherein the second information is indication information related to positioning and indicates the type, the quantity and the second feedback time of feedback data. For example, the data supporting positioning and needing to be fed back is P2-50 slots, the terminal feeds back the estimation of the channel impulse response, and the feedback starting time is S2-10 slots after the second information is transmitted.
Further, the network device sends the fourth information corresponding to positioning to the terminal device in the PDSCH of the same time slot as the second information related to "positioning", where the fourth information includes an AI model used for positioning. The channel impulse response in fig. 3 is a general term for the influence of the channel on the amplitude and phase information of the input signal, etc.
FIG. 4 is a schematic diagram of an artificial intelligence model downloading embodiment of a pre-upload data requirement.
In this embodiment, the standard specifies that the 5G network can explicitly support 5 AI model services, such as a model for "channel estimation," a model for "channel information feedback," a "positioning" model, "a" beam management "model," a "mobility management" model, and N is 5. The network device may provide 3 AI model services, such as "channel estimation", "channel information feedback", and "beam management", where the corresponding first information M ═ 5 bits indicate 11010, and is transmitted by DCI information carried by a PDCCH.
The first information indicates a first feedback time at the same time, for example, the terminal is instructed to feed back the third information in the 10 th time slot after carrying the first information PDCCH, and the AI model to be used is determined. The terminal hopes to adopt an AI-based beam management model, and feeds back the third information at the 10 th time slot after the first information indicates the position, that is, the PDCCH carrying the first information: 00010. after receiving the third information sent by the terminal device, the network device first sends the second information related to beam management to the terminal device, and indicates a second feedback time and a feedback data type. For example, the data indicating that feedback is required to support beam management is P1 ═ 100 time slots for feeding back mobility-related data. The feedback start time is 80 time slots from the second information transmission end S1.
And the fourth information is sent after the terminal feeds back the first-time mobility-related data and D is 10 time slots.
Fig. 5 is a flowchart of an embodiment of a method of the present application for a network device.
The method according to any one embodiment of the first aspect of the present application, applied to a network device, includes the following steps 201-205:
Each artificial intelligence model comprises a neural network structural feature and a parameter; each artificial intelligence model is used for corresponding network side and mobile side performance processing.
And the network equipment informs the AI model types supported by the terminal equipment through first information. Specifically, the network device supports at least N AI model types, where N is a positive integer greater than or equal to 1, and each AI model may serve different functions. Each AI model function is agreed by the network device and the terminal device.
The first information may be DCI information carried by a PDCCH, or may be higher layer signaling. And N bits are contained to represent the supporting conditions of the network equipment to the N types of AI models. The first information may include confirmation request information indicating whether the terminal device feeds back the first information indication model, and a first feedback time.
202, receiving uplink information, wherein the uplink information comprises a selection instruction of the downloading service of the N artificial intelligence models;
specifically, the uplink information includes third information as feedback to the first information; the third information comprises selection indications of the N artificial intelligence model downloading services.
And the network equipment receives the fed back third information at the corresponding time point according to the first information. The third information feeds back whether the first information is used for providing indication information of an AI model. The model indicated by the first information is selected using N bits. The corresponding time point is a first feedback time indicated by the first information.
for example, the network device sends data feedback information required by each model in the third information to the terminal device through the second information. The data feedback information comprises the type of feedback data, the quantity of each type of data, the feedback data period, the feedback starting point and the like.
And step 204, responding to the selection indication, wherein the downlink information comprises 1 or more of the N artificial intelligence models.
Specifically, according to the feedback of the first information, the downlink information comprises fourth information, and the fourth information comprises 1 or more of the N artificial intelligence models.
For example, after receiving the third information fed back by the terminal, the network device sends the fourth information to the terminal. The fourth information comprises model feedback of the model which needs to be used and is fed back in the third information.
It should be noted that step 204 may be executed multiple times, that is, the fourth information may be sent multiple times, so as to implement model updating.
Further, the number of the fourth information for the same type of AI model may be greater than or equal to 1.
Further, the fourth information may be fed back in the same time slot as the second information, and indicated by the DCI where the second information is located.
In order to execute step 204, the network device can implement data set construction and AI model training for different applications according to the feedback of the terminal device on the first information and the second information, thereby providing AI model deployment.
It should be noted that, the sequence of step 204 and step 205 is optional, and when the fed-back data includes data required for generating or training the artificial intelligence model, the feedback data in the uplink information should be received at a second feedback time before step 204. When the feedback data includes result data obtained by operating the artificial intelligence model or data for evaluating the operation effect of the artificial intelligence model, the feedback data in the uplink information should be received at a second feedback time after step 204.
Fig. 6 is a flowchart of an embodiment of a method of the present application for a terminal device.
The method according to any one of the embodiments of the first aspect of the present application, applied to a terminal device, includes the following steps:
Each artificial intelligence model comprises a neural network structural feature and a parameter; each artificial intelligence model is used for corresponding network side and mobile side performance processing.
The terminal equipment identifies the first information and determines an artificial intelligence model and first feedback time for selection.
And 302, sending uplink information, wherein the uplink information comprises selection instructions of the downloading services of the N artificial intelligence models.
Specifically, the uplink information includes third information as feedback to the first information; the third information comprises selection indications of the N artificial intelligence model downloading services.
And the terminal equipment feeds back third information at a corresponding time point according to the first information. The third information feeds back whether the first information is used for providing indication information of an AI model. The model indicated by the first information is selected using N bits. The corresponding time point is a first feedback time indicated by the first information.
for example, downlink information is received, where the downlink information includes second information, and the second information is used to indicate a second feedback time and a type and a quantity of feedback data.
It should be noted that the type of the feedback data includes at least one of the following data:
generating and training data required by the artificial intelligence model;
operating an artificial intelligence model to obtain result data;
and (3) evaluating the operation effect of the artificial intelligent model.
specifically, according to the selection indication of the third information, fourth information is included in the downstream information, and the fourth information includes 1 or more kinds of the N artificial intelligence models.
It should be noted that step 304 is executed multiple times, that is, the fourth information may be received multiple times, so as to implement model updating.
And 305, executing the artificial intelligence model to generate feedback data.
After the artificial intelligence model is executed, the generated feedback data comprises: operating an artificial intelligence model to obtain result data; and (3) evaluating the operation effect of the artificial intelligent model.
And step 306, sending the feedback data at the feedback time indicated by the downlink information.
For example, at the second feedback time, uplink information is received, where the uplink information includes the feedback data.
It should be noted that step 306 should be performed before step 304 when the type of feedback data is the data required to generate and train the artificial intelligence model.
And the terminal equipment feeds back the data information according to the feedback time and the feedback data type indicated by the second information. The feedback data required by multiple models can be sent simultaneously, and when the types and the periods of the feedback data required by the multiple models are the same, only one piece of data can be fed back.
Fig. 7 is a schematic diagram of an embodiment of a network device.
An embodiment of the present application further provides a network device, where, using the method according to any one of the embodiments of the present application, the network device is configured to: sending downlink information, wherein the downlink information comprises at least one of the first information, the second information and the fourth information; and receiving uplink information, wherein the uplink information comprises the third information. Responding to the first information, receiving the uplink information at a first feedback time, wherein the uplink information comprises a selection instruction of the downloading service of the N artificial intelligence models; in response to the selection indication, downstream information comprises 1 or more of the N artificial intelligence models; the feedback data is received at a second feedback time in response to the second information and/or the fourth information.
In order to implement the foregoing technical solution, the network device 400 provided in the present application includes a network sending module 401, a network determining module 402, and a network receiving module 403.
The network sending module is configured to send downlink information including at least one of the first information, the second information, and the fourth information.
The network determination module is used for determining first feedback time, second feedback time and N types of artificial intelligence model downloading services for selection, wherein each type of artificial intelligence model comprises a neural network structure characteristic and parameter and an artificial intelligence model type selected by uplink information indication. Further, the network determining module is further configured to train the artificial intelligence model according to the feedback data.
The network receiving module is configured to receive uplink information, which includes third information and also includes the feedback data.
The specific method for implementing the functions of the network sending module, the network determining module, and the network receiving module is described in the embodiments of the methods of the present application, and is not described herein again.
Fig. 8 is a schematic diagram of an embodiment of a terminal device.
The present application further provides a terminal device, which uses the method of any one of the embodiments of the present application, and is configured to: receiving downlink information, wherein the downlink information comprises at least one of the first information, the second information and the third information; and sending uplink information, wherein the uplink information comprises the third information. Responding to the first information, and sending the uplink information at a first feedback time, wherein the uplink information comprises a selection instruction of the downloading service of the N artificial intelligence models; receiving downlink information, wherein the downlink information comprises 1 or more artificial intelligence models corresponding to the selection indication; executing the artificial intelligence model to generate feedback data; receiving downlink information, wherein the downlink information also comprises the type of feedback data and an indication of feedback time; and sending the feedback data at a second feedback time indicated by the downlink information.
In order to implement the foregoing technical solution, the terminal device 500 provided in the present application includes a terminal sending module 501, a terminal determining module 502, and a terminal receiving module 503.
The terminal receiving module is used for receiving downlink information, and the downlink information comprises at least one of first information, second information and fourth information.
And the terminal determining module is used for determining N types of artificial intelligence model downloading services for selection, determining selection instructions of the artificial intelligence models, operating the artificial intelligence models and determining feedback data. Further, the terminal determining module is further configured to train an artificial intelligence model according to the feedback data.
And the terminal sending module is used for sending uplink information and third information, and also comprises the feedback data.
The specific method for implementing the functions of the terminal sending module, the terminal determining module and the terminal receiving module is as described in the method embodiments of the present application, and is not described herein again.
The terminal equipment can be mobile terminal equipment.
Fig. 9 is a schematic structural diagram of a network device according to another embodiment of the present invention. As shown, the network device 600 includes a processor 601, a wireless interface 602, and a memory 603. Wherein the wireless interface may be a plurality of components, i.e. including a transmitter and a receiver, providing means for communicating with various other apparatus over a transmission medium. The wireless interface implements a communication function with the terminal device, and processes wireless signals through the receiving and transmitting devices, and data carried by the signals are communicated with the memory or the processor through the internal bus structure. The memory 603 contains a computer program that executes any of the embodiments of the present application, running or changed on the processor 601. When the memory, processor, wireless interface circuit are connected through a bus system. The bus system includes a data bus, a power bus, a control bus, and a status signal bus, which are not described herein.
Fig. 10 is a block diagram of a terminal device of another embodiment of the present invention. The terminal device 700 comprises at least one processor 701, a memory 702, a user interface 703 and at least one network interface 704. The various components in the terminal device 700 are coupled together by a bus system. A bus system is used to enable connection communication between these components. The bus system includes a data bus, a power bus, a control bus, and a status signal bus.
The user interface 703 may include a display, a keyboard, or a pointing device, such as a mouse, a trackball, a touch pad, or a touch screen, among others.
The memory 702 stores executable modules or data structures. The memory may have stored therein an operating system and an application program. The operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, and the like, and is used for implementing various basic services and processing hardware-based tasks. The application programs include various application programs such as a media player, a browser, and the like for implementing various application services.
In the embodiment of the present invention, the memory 702 contains a computer program for executing any of the embodiments of the present application, and the computer program runs or changes on the processor 701.
The memory 702 contains a computer readable storage medium, and the processor 701 reads the information in the memory 702 and combines the hardware to complete the steps of the above-described method. In particular, the computer-readable storage medium has stored thereon a computer program which, when being executed by the processor 701, carries out the steps of the method embodiments as described above with reference to any of the embodiments.
The processor 701 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the method of the present application may be implemented by hardware integrated logic circuits in the processor 701 or by instructions in the form of software. The processor 701 may be a general purpose processor, a digital signal processor, an application specific integrated circuit, an off-the-shelf programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components. The various methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. In a typical configuration, the device of the present application includes one or more processors (CPUs), an input/output user interface, a network interface, and a memory.
Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application therefore also proposes a computer-readable medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of the embodiments of the present application. For example, the memory 603, 702 of the present invention may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM).
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
Based on the embodiments of fig. 7 to 10, the present application further provides a mobile communication system including at least 1 embodiment of any terminal device in the present application and/or at least 1 embodiment of any network device in the present application.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In the present application, "first", "second", "third", and "fourth" are used for distinguishing a plurality of objects having the same name, and do not mean a size or an order, and have no other special meaning unless otherwise specified.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (15)
1. A wireless communication artificial intelligence processing method is characterized by comprising the following steps:
the downlink information comprises first information, the first information is used for indicating N artificial intelligence models to download services for selection, and each artificial intelligence model comprises a neural network structural feature and parameters;
each artificial intelligence model is used for corresponding network side and mobile side performance processing.
2. The wireless communication artificial intelligence processing method of claim 1,
the first information is high-level signaling or downlink control signaling DCI;
the first information comprises N bits, and each 1 bit identifies 1 artificial intelligence model downloading service.
3. The wireless communication artificial intelligence processing method of claim 1,
the first information is further used for indicating a first feedback time of the first information.
4. The wireless communication artificial intelligence processing method of claim 1,
the uplink information comprises third information serving as feedback to the first information;
the third information comprises selection indications of the N artificial intelligence model downloading services.
5. The wireless communication artificial intelligence processing method of claim 1,
the downlink information also comprises second information, and the second information is used for indicating second feedback time and the type of feedback data;
and at the second feedback time, the uplink information comprises the feedback data.
6. The wireless communication artificial intelligence processing method of claim 4,
and according to the feedback of the first information, downlink information comprises fourth information, and the fourth information comprises 1 or more of the N artificial intelligence models.
7. The method according to any of claims 1 to 6, for a network device,
sending downlink information, wherein the downlink information comprises the first information;
receiving uplink information, wherein the uplink information comprises a selection instruction of the downloading service of the N artificial intelligence models;
in response to the selection indication, the downstream information includes 1 or more of the N artificial intelligence models.
8. The method according to any of claims 1 to 6, for a terminal device,
receiving downlink information, wherein the downlink information comprises the first information;
and sending uplink information, wherein the uplink information comprises selection instructions of the downloading services of the N artificial intelligence models.
9. The method of claim 8,
receiving downlink information, wherein the downlink information comprises 1 or more artificial intelligence models corresponding to the selection indication;
and executing the artificial intelligence model to generate feedback data.
10. The method of claim 8,
receiving downlink information, wherein the downlink information also comprises the type of feedback data and an indication of feedback time;
and sending the feedback data at the feedback time indicated by the downlink information.
11. A network device for implementing the method of any one of claims 1 to 7,
at least one module in the network device for at least one of the following functions: sending downlink information, wherein the downlink information comprises the first information; receiving uplink information, wherein the uplink information comprises a selection instruction of the downloading service of the N artificial intelligence models; in response to the selection indication, the downstream information includes 1 or more of the N artificial intelligence models.
12. A terminal device for implementing the method of any one of claims 1 to 6 and 8 to 10,
at least one module in the terminal device is used for at least one of the following functions: receiving downlink information, wherein the downlink information comprises the first information; sending uplink information, wherein the uplink information comprises a selection instruction of the downloading service of the N artificial intelligence models; receiving downlink information, wherein the downlink information comprises 1 or more artificial intelligence models corresponding to the selection indication; executing the artificial intelligence model to generate feedback data; receiving downlink information, wherein the downlink information also comprises the type of feedback data and an indication of feedback time; and sending the feedback data at the feedback time indicated by the downlink information.
13. A communication device, comprising: memory, processor and computer program stored on the memory and executable on the processor, which computer program, when executed by the processor, carries out the steps of the method according to any one of claims 1 to 10.
14. A computer-readable medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 8.
15. A mobile communication system comprising at least 1 network device according to claim 11 and/or at least 1 terminal device according to claim 12.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111471065.0A CN114189889B (en) | 2021-12-03 | 2021-12-03 | Wireless communication artificial intelligence processing method and device |
| PCT/CN2022/070730 WO2023097870A1 (en) | 2021-12-03 | 2022-01-07 | Wireless communication artificial intelligence processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111471065.0A CN114189889B (en) | 2021-12-03 | 2021-12-03 | Wireless communication artificial intelligence processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114189889A true CN114189889A (en) | 2022-03-15 |
| CN114189889B CN114189889B (en) | 2024-07-05 |
Family
ID=80542248
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111471065.0A Active CN114189889B (en) | 2021-12-03 | 2021-12-03 | Wireless communication artificial intelligence processing method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114189889B (en) |
| WO (1) | WO2023097870A1 (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023198186A1 (en) * | 2022-04-15 | 2023-10-19 | 维沃移动通信有限公司 | Model determining method and apparatus, information transmission method and apparatus, and related device |
| WO2023245589A1 (en) * | 2022-06-23 | 2023-12-28 | 北京小米移动软件有限公司 | Positioning model determination method and apparatus |
| WO2024016363A1 (en) * | 2022-07-22 | 2024-01-25 | 北京小米移动软件有限公司 | Model interaction method, apparatus and system for heterogeneous artificial intelligence (ai) framework |
| WO2024031582A1 (en) * | 2022-08-11 | 2024-02-15 | 北京小米移动软件有限公司 | Communication method and apparatus, storage medium, and chip |
| WO2024040378A1 (en) * | 2022-08-22 | 2024-02-29 | Apple Inc. | Ai/ml model management and activation/deactivation |
| WO2024073969A1 (en) * | 2023-01-12 | 2024-04-11 | Lenovo (Beijing) Limited | Methods and apparatuses for ai model management |
| WO2024082195A1 (en) * | 2022-10-19 | 2024-04-25 | 北京小米移动软件有限公司 | Ai model-based terminal positioning method and apparatus |
| WO2024092660A1 (en) * | 2022-11-03 | 2024-05-10 | 北京小米移动软件有限公司 | Model selection method and apparatus |
| WO2024092812A1 (en) * | 2022-11-04 | 2024-05-10 | 北京小米移动软件有限公司 | Ai or ml model monitoring method and apparatus, and communication device and storage medium |
| WO2024103543A1 (en) * | 2023-02-10 | 2024-05-23 | Zte Corporation | Transfer of artificial intelligence network management models in wireless communication systems |
| WO2024130518A1 (en) * | 2022-12-19 | 2024-06-27 | 北京小米移动软件有限公司 | Model management methods and apparatuses |
| WO2024140442A1 (en) * | 2022-12-29 | 2024-07-04 | 维沃移动通信有限公司 | Model updating method and apparatus, and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180005134A1 (en) * | 2016-06-30 | 2018-01-04 | Cogniac, Corp. | Operating Machine-Learning Models on Different Platforms |
| US20190356516A1 (en) * | 2018-05-18 | 2019-11-21 | Parallel Wireless, Inc. | Machine Learning for Channel Estimation |
| CN111510848A (en) * | 2019-01-11 | 2020-08-07 | 中国移动通信有限公司研究院 | Terminal position prediction method, device, medium and equipment |
| CN111819872A (en) * | 2020-06-03 | 2020-10-23 | 北京小米移动软件有限公司 | Information transmission method, device, communication equipment and storage medium |
| CN113543305A (en) * | 2020-04-22 | 2021-10-22 | 维沃移动通信有限公司 | Positioning method, communication equipment and network equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20210053036A (en) * | 2019-11-01 | 2021-05-11 | 엘지전자 주식회사 | Artificial intelligence server |
| CN111222046B (en) * | 2020-01-03 | 2022-09-20 | 腾讯科技（深圳）有限公司 | Service configuration method, client for service configuration, equipment and electronic equipment |
| CN115380594A (en) * | 2020-04-10 | 2022-11-22 | Oppo广东移动通信有限公司 | Resource allocation method, device, equipment and storage medium |
-
2021
- 2021-12-03 CN CN202111471065.0A patent/CN114189889B/en active Active
-
2022
- 2022-01-07 WO PCT/CN2022/070730 patent/WO2023097870A1/en unknown
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180005134A1 (en) * | 2016-06-30 | 2018-01-04 | Cogniac, Corp. | Operating Machine-Learning Models on Different Platforms |
| US20190356516A1 (en) * | 2018-05-18 | 2019-11-21 | Parallel Wireless, Inc. | Machine Learning for Channel Estimation |
| CN111510848A (en) * | 2019-01-11 | 2020-08-07 | 中国移动通信有限公司研究院 | Terminal position prediction method, device, medium and equipment |
| CN113543305A (en) * | 2020-04-22 | 2021-10-22 | 维沃移动通信有限公司 | Positioning method, communication equipment and network equipment |
| CN111819872A (en) * | 2020-06-03 | 2020-10-23 | 北京小米移动软件有限公司 | Information transmission method, device, communication equipment and storage medium |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023198186A1 (en) * | 2022-04-15 | 2023-10-19 | 维沃移动通信有限公司 | Model determining method and apparatus, information transmission method and apparatus, and related device |
| WO2023245589A1 (en) * | 2022-06-23 | 2023-12-28 | 北京小米移动软件有限公司 | Positioning model determination method and apparatus |
| WO2024016363A1 (en) * | 2022-07-22 | 2024-01-25 | 北京小米移动软件有限公司 | Model interaction method, apparatus and system for heterogeneous artificial intelligence (ai) framework |
| WO2024031582A1 (en) * | 2022-08-11 | 2024-02-15 | 北京小米移动软件有限公司 | Communication method and apparatus, storage medium, and chip |
| WO2024040378A1 (en) * | 2022-08-22 | 2024-02-29 | Apple Inc. | Ai/ml model management and activation/deactivation |
| WO2024082195A1 (en) * | 2022-10-19 | 2024-04-25 | 北京小米移动软件有限公司 | Ai model-based terminal positioning method and apparatus |
| WO2024092660A1 (en) * | 2022-11-03 | 2024-05-10 | 北京小米移动软件有限公司 | Model selection method and apparatus |
| WO2024092812A1 (en) * | 2022-11-04 | 2024-05-10 | 北京小米移动软件有限公司 | Ai or ml model monitoring method and apparatus, and communication device and storage medium |
| WO2024130518A1 (en) * | 2022-12-19 | 2024-06-27 | 北京小米移动软件有限公司 | Model management methods and apparatuses |
| WO2024140442A1 (en) * | 2022-12-29 | 2024-07-04 | 维沃移动通信有限公司 | Model updating method and apparatus, and device |
| WO2024073969A1 (en) * | 2023-01-12 | 2024-04-11 | Lenovo (Beijing) Limited | Methods and apparatuses for ai model management |
| WO2024103543A1 (en) * | 2023-02-10 | 2024-05-23 | Zte Corporation | Transfer of artificial intelligence network management models in wireless communication systems |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114189889B (en) | 2024-07-05 |
| WO2023097870A1 (en) | 2023-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114189889B (en) | Wireless communication artificial intelligence processing method and device | |
| CN113922936B (en) | AI technology channel state information feedback method and equipment | |
| CN114172765B (en) | Wireless communication artificial intelligent channel estimation method and device | |
| RU2768371C1 (en) | Method, device and data transmission system for direct communication | |
| KR102255949B1 (en) | HARQ feedback information transmission method, UE, base station and system | |
| EP3627753B1 (en) | Method for transmitting uplink control information, terminal device, and network device | |
| CN111970094B (en) | Method and equipment for determining side link feedback resource | |
| CN110915157B (en) | Method, device, terminal and storage medium for data transmission and reception | |
| CN113727446A (en) | Method and equipment for dynamically sending sensing signal | |
| CN103891386A (en) | Apparatus and method for transmitting and receiving persistent scheduling change information in wireless communication system | |
| CN113726491B (en) | Method and equipment for sending sensing signal according to needs | |
| CN112039644A (en) | Aperiodic positioning pilot frequency transmission method and device | |
| US20220217696A1 (en) | Feedback information transmission method and apparatus, and device and storage medium | |
| CN115580877A (en) | Method and equipment for jointly deploying mobile communication network data and model | |
| JP2024523521A (en) | Method and device for determining beam usage time | |
| JP2020523884A (en) | Uplink control channel transmission method, reception method, device, terminal and base station | |
| TWI770962B (en) | A method, device and computer storage medium for determining a feedback codebook | |
| CN112449425A (en) | Communication method and device | |
| CN111953458B (en) | PUCCH resource determination method and communication equipment | |
| CN115767570A (en) | AI model registration method and equipment for wireless communication system | |
| CN115767605A (en) | AI model deployment method and equipment for mobile communication network | |
| CN113261235B (en) | Feedback method and device for confirming hybrid automatic repeat request | |
| CN112737750B (en) | Uplink SRS pilot frequency transmission method and device | |
| WO2022077514A1 (en) | Communication method and apparatus | |
| CN115413052A (en) | Uplink information scheduling method, device, electronic equipment, storage medium and base station |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |