CN116418880A

CN116418880A - AI network information transmission method and device and communication equipment

Info

Publication number: CN116418880A
Application number: CN202111666710.4A
Authority: CN
Inventors: 任千尧; 孙鹏
Original assignee: Vivo Mobile Communication Co Ltd
Current assignee: Vivo Mobile Communication Co Ltd
Priority date: 2021-12-31
Filing date: 2021-12-31
Publication date: 2023-07-11
Also published as: WO2023125934A1

Abstract

The application discloses an AI network information transmission method, an AI network information transmission device and communication equipment, which belong to the technical field of communication, and the AI network information transmission method in the embodiment of the application comprises the following steps: the first end compresses AI network information, wherein the AI network information comprises at least one of a network structure and a network parameter; and the first end sends the compressed AI network information to a second end.

Description

AI network information transmission method and device and communication equipment

Technical Field

The application belongs to the technical field of communication, and particularly relates to an AI network information transmission method, an AI network information transmission device and communication equipment.

Background

Artificial intelligence (Artificial Intelligence, AI) is a new technical science for researching and developing theories, methods, techniques and application systems for simulating, extending and expanding human intelligence, and is receiving a great deal of attention from people, and the application of AI is becoming more and more widespread. At present, research into the use of AI networks in communication systems, for example, communication data may be transmitted between a network-side device and a terminal through the AI network, has been begun. In a communication system, the entire AI network is generally transferred together, resulting in a large overhead.

Disclosure of Invention

The embodiment of the application provides an AI network information transmission method, an AI network information transmission device and communication equipment, which can solve the problem of high AI network transmission overhead of communication equipment in the related technology.

In a first aspect, an AI network information transmission method is provided, including:

the first end compresses AI network information, wherein the AI network information comprises at least one of a network structure and a network parameter;

and the first end sends the compressed AI network information to a second end.

In a second aspect, there is provided an AI network information transfer method, including:

the second end receives the compressed AI network information sent by the first end, where the AI network information includes at least one of a network structure and a network parameter.

In a third aspect, there is provided an AI network information transfer apparatus including:

the compression module is used for compressing AI network information, wherein the AI network information comprises at least one of a network structure and a network parameter;

and the sending module is used for sending the compressed AI network information to the second end.

In a fourth aspect, there is provided an AI network information transfer apparatus including:

and the receiving module is used for receiving the compressed AI network information sent by the first end, wherein the AI network information comprises at least one of a network structure and a network parameter.

In a fifth aspect, there is provided a communication device comprising a processor and a memory storing a program or instructions executable on the processor, which when executed by the processor, implement the steps of the AI-network-information-transmission method of the first aspect, or implement the steps of the AI-network-information-transmission method of the second aspect.

In a sixth aspect, there is provided a readable storage medium storing thereon a program or instructions which, when executed by a processor, implement the steps of the AI-network-information-transmission method as set forth in the first aspect, or implement the steps of the AI-network-information-transmission method as set forth in the second aspect.

In a seventh aspect, a chip is provided, where the chip includes a processor and a communication interface, where the communication interface is coupled to the processor, and the processor is configured to execute a program or instructions to implement the AI-network information transmission method according to the first aspect, or implement the AI-network information transmission method according to the second aspect.

In an eighth aspect, there is provided a computer program/program product stored in a storage medium, the computer program/program product being executed by at least one processor to implement the steps of the AI-network-information-transmission method as set forth in the first aspect, or to implement the steps of the AI-network-information-transmission method as set forth in the second aspect.

In this embodiment of the present application, the first end may send compressed AI network information to the second end, where the AI network information includes at least one of a network structure and a network parameter, and further in a communication process, it is not necessary to transmit the entire AI network including all network structures and network parameters together, so that the network structure and the network parameter of the AI network may be sent separately, and further transmission overhead in the communication process may be effectively reduced.

Drawings

Fig. 1 is a block diagram of a wireless communication system to which embodiments of the present application are applicable;

fig. 2 is a flowchart of an AI network information transmission method provided in an embodiment of the present application;

fig. 3 is a flowchart of another AI-network information transmission method provided by an embodiment of the application;

fig. 4 is a block diagram of an AI network information transmission apparatus provided in an embodiment of the application;

fig. 5 is a block diagram of another AI-network information transfer apparatus provided by an embodiment of the application;

fig. 6 is a block diagram of a communication device according to an embodiment of the present application;

fig. 7 is a block diagram of a terminal according to an embodiment of the present application;

fig. 8 is a block diagram of a network side device according to an embodiment of the present application;

fig. 9 is a block diagram of another network side device according to an embodiment of the present application.

Detailed Description

Technical solutions in the embodiments of the present application will be clearly described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application are within the scope of the protection of the present application.

The terms first, second and the like in the description and in the claims, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application are capable of operation in sequences other than those illustrated or otherwise described herein, and that the terms "first" and "second" are generally intended to be used in a generic sense and not to limit the number of objects, for example, the first object may be one or more. Furthermore, in the description and claims, "and/or" means at least one of the connected objects, and the character "/" generally means a relationship in which the associated object is an "or" before and after.

It is noted that the techniques described in embodiments of the present application are not limited to long term evolution (Long Term Evolution, LTE)/LTE evolution (LTE-Advanced, LTE-a) systems, but may also be used in other wireless communication systems, such as code division multiple access (Code Division Multiple Access, CDMA), time division multiple access (Time Division Multiple Access, TDMA), frequency division multiple access (Frequency Division Multiple Access, FDMA), orthogonal frequency division multiple access (Orthogonal Frequency Division Multiple Access, OFDMA), single carrier frequency division multiple access (Single-carrier Frequency Division Multiple Access, SC-FDMA), and other systems. The terms "system" and "network" in embodiments of the present application are often used interchangeably, and the techniques described may be used for both the above-mentioned systems and radio technologies, as well as other systems and radio technologies. The following description describes a New air interface (NR) system for purposes of example and uses NR terminology in much of the description that follows, but these techniques are also applicable to applications other than NR system applications, such as generation 6 (6) ^th Generation, 6G) communication system.

Fig. 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable. The wireless communication system includes a terminal 11 and a network device 12. The terminal 11 may be a mobile phone, a tablet (Tablet Personal Computer), a Laptop (Laptop Computer) or a terminal-side Device called a notebook, a personal digital assistant (Personal Digital Assistant, PDA), a palm top, a netbook, an ultra-mobile personal Computer (ultra-mobile personal Computer, UMPC), a mobile internet appliance (Mobile Internet Device, MID), an augmented reality (augmented reality, AR)/Virtual Reality (VR) Device, a robot, a Wearable Device (weather Device), a vehicle-mounted Device (VUE), a pedestrian terminal (PUE), a smart home (home Device with a wireless communication function, such as a refrigerator, a television, a washing machine, or a furniture), a game machine, a personal Computer (personal Computer, PC), a teller machine, or a self-service machine, and the Wearable Device includes: intelligent wrist-watch, intelligent bracelet, intelligent earphone, intelligent glasses, intelligent ornament (intelligent bracelet, intelligent ring, intelligent necklace, intelligent anklet, intelligent foot chain etc.), intelligent wrist strap, intelligent clothing etc.. Note that, the specific type of the terminal 11 is not limited in the embodiment of the present application. The network-side device 12 may comprise an access network device or core network device, wherein the access network device may also be referred to as a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function or a radio access network element. The access network device may include a base station, a WLAN access point, a WiFi node, or the like, where the base station may be referred to as a node B, an evolved node B (eNB), an access point, a base transceiver station (Base Transceiver Station, BTS), a radio base station, a radio transceiver, a basic service set (Basic Service Set, BSS), an extended service set (Extended Service Set, ESS), a home node B, a home evolved node B, a transmission receiving point (Transmitting Receiving Point, TRP), or some other suitable terminology in the field, and the base station is not limited to a specific technical vocabulary so long as the same technical effect is achieved, and it should be noted that in the embodiment of the present application, only the base station in the NR system is described by way of example, and the specific type of the base station is not limited. The core network device may include, but is not limited to, at least one of: core network nodes, core network functions, mobility management entities (Mobility Management Entity, MME), access mobility management functions (Access and Mobility Management Function, AMF), session management functions (Session Management Function, SMF), user plane functions (User Plane Function, UPF), policy control functions (Policy Control Function, PCF), policy and charging rules function units (Policy and Charging Rules Function, PCRF), edge application service discovery functions (Edge Application Server Discovery Function, EASDF), unified data management (Unified Data Management, UDM), unified data repository (Unified Data Repository, UDR), home subscriber server (Home Subscriber Server, HSS), centralized network configuration (Centralized network configuration, CNC), network storage functions (Network Repository Function, NRF), network opening functions (Network Exposure Function, NEF), local NEF (or L-NEF), binding support functions (Binding Support Function, BSF), application functions (Application Function, AF), and the like. In the embodiment of the present application, only the core network device in the NR system is described as an example, and the specific type of the core network device is not limited.

The AI network information transmission method provided by the embodiments of the present application is described in detail below with reference to the accompanying drawings through some embodiments and application scenarios thereof.

Referring to fig. 2, fig. 2 is a flowchart of an AI network information transmission method according to an embodiment of the disclosure, and as shown in fig. 2, the method includes the following steps:

step 201, a first end compresses AI network information, wherein the AI network information comprises at least one of a network structure and a network parameter;

step 202, the first end sends the compressed AI network information to a second end.

In this embodiment of the present application, the first end and the second end are communication devices having a transmitting and receiving function.

Optionally, the first end is one of a network side device and a terminal, and the second end is the other of the network side device and the terminal; or the first end and the second end are different nodes of the terminal; or the first end and the second end are different nodes of the network side equipment.

The network-side device may include an access network device (e.g., a base station) and a core network device. Optionally, the first end may be an access network device, and the second end is a core network device; or the first end is a terminal, and the second end is core network equipment or access network equipment; or the first end and the second end are different nodes of access network equipment; or, the first end and the second end are different nodes of the core network device, and the embodiments of the present application are not specifically listed.

In this embodiment of the present application, the AI network information includes at least one of a network structure and a network parameter. Alternatively, the AI network information may be a network structure and/or a network parameter of a certain AI network, or a network structure and/or a network parameter of a plurality of AI networks. In some embodiments, the AI network may also be referred to as an AI neural network, AI model, or the like. Wherein, the network parameters comprise weight parameters, super parameters and the like of the AI network.

The compressing the AI network information may refer to compressing the AI network information into a file corresponding to a preset model expression mode according to a preset model expression mode, where the model expression mode is a data structure, and describes information such as an AI network structure and network parameters according to a certain rule.

Optionally, the AI network information includes a network structure and/or a network parameter, and the compression of the AI network information by the first end includes compressing the network structure and/or the weight parameter, and transmitting the compressed network structure and/or the compressed weight parameter to the second end. For example, if the AI network information includes only a network structure, the first end compresses and transmits only the network structure; alternatively, the AI network information may include only the network parameter, and the first end may compress and transmit only the network parameter; alternatively, the AI network information includes a partial network structure and a partial network parameter, and the first end compresses and transmits the partial network structure and the partial network parameter. Of course, the specific information content included in the AI network information may be other situations, which are not described in detail herein.

In the embodiment of the application, the AI network information comprises at least one of a network structure and network parameters, and the whole AI network comprising all the network structures and the network parameters is not required to be transmitted together in the communication process, so that the network structures and the network parameters of the AI network can be transmitted separately, and the transmission cost in the communication process can be effectively reduced.

Optionally, the AI network information may include a network structure and a network parameter, and the step 201 may include any one of:

the first end performs merging compression on the network structure and the network parameters;

the first end compresses the network structure and the network parameters respectively.

For example, the first end may combine and compress the network structure and the network parameters of the AI network into one transmission file based on a preset model expression mode, or may compress the network structure and the weight parameters into two independent transmission files based on a preset model expression mode.

Optionally, in the case where the first end compresses the network structure and the network parameter separately, the step 202 may include any one of the following:

The first end merges and sends the compressed network structure and the compressed network parameters to the second end;

the first end sends the compressed network structure and the compressed network parameters to the second end respectively.

In an exemplary case where the AI network information includes a network structure and a network parameter, the first end compresses the network structure and the network parameter into two independent transmission files, and sends the two compressed transmission files to the second end together, or may send the two compressed transmission files separately, or may send only one of the compressed transmission files, so that the first end is more flexible for the transmission manner of the AI network information after compression.

Optionally, the step 201 may further include any one of the following:

the first end converts AI network information into a corresponding transmission file based on a preset model expression mode, and compresses the transmission file;

the first end compresses the AI network information based on a preset data format;

the first end obtains AI network information to be sent and the AI network information existing in the second end, and compresses an AI network information difference value between the AI network information to be sent and the AI network information existing in the second end;

The first end obtains the AI network information to be sent and the AI network information of a preset AI network, and compresses the AI network information difference value between the AI network information to be sent and the AI network information of the preset AI network.

The preset model expression mode may be an AI network expression mode common to both the first end and the second end, for example, an open neural network interaction (open neural network exchange, ONNX), a TensorFlow, or the like. The first end may convert the AI network information into a corresponding transmission file based on ONNX, compress the transmission file, and send the compressed transmission file to the second end, where the second end decompresses the compressed transmission file, and may also convert the decompressed transmission file into a network structure and network parameters applicable to itself based on ONNX.

The file structures of the AI network information storage under two different neural network frameworks are different, the AI network information can not be directly read, and the AI network information is converted into a corresponding transmission file through a preset model expression mode and then is compressed and transmitted, so that the two communication devices using the different neural network frameworks can realize the reading and the application of the AI network information.

Optionally, the first end may compress the AI network information based on a preset data format, for example, a protobuf data format used by ONNX, and the like.

Or, the first end may compress the AI network information difference between the AI network information to be sent and the AI network information existing in the second end, and the first end only needs to send the compressed AI network information difference to the second end, so that the same AI network information between the AI network information to be sent and the AI network information existing in the second end does not need to be compressed and sent, thereby effectively saving the transmission overhead of the first end.

Or, the first end may compress the AI network information difference between the AI network information to be sent and the AI network information of the preset AI network, where the first end only needs to send the compressed AI network information difference to the second end, so as to save transmission overhead of the first end. The preset AI network may be a protocol reservation or a higher-level configuration, such as some fixed AI network templates, or may be an AI network template common to the communication devices. The preset AI network includes an initial value of a network structure and an initial value of a network parameter.

Optionally, the AI network information difference value includes at least one of:

a specified network parameter;

indexing network parameters;

modified network parameters;

modified parameter values in the modified network parameters;

the location of the modified parameter value in the modified network parameter;

the location of a modified reference value in a modified network parameter, the reference value being the maximum value in the network parameter;

a non-zero value in the modified network parameter;

the location of non-zero values in the modified network parameters;

a newly added network structure;

a deleted network structure;

modified network architecture.

In this embodiment of the present application, the preset model expression manner includes any one of the following: the protocol appoints a model expression mode and a self-defined model expression mode. The custom model expression mode may refer to a data structure custom-defined by the communication device, and is used for expressing a network structure and network parameters of the AI network.

Optionally, the content of the customized model expression mode includes at least one of the following: network structure of AI network, attribute of network parameter of AI network, value of network parameter of AI network. The attribute of the network parameter comprises information such as name, identification, dimension and the like of the network parameter; the values of the parameters of the AI network may be one or more.

Optionally, the representation of the network structure in the preset model representation includes at least one of the following:

association relationship between network structures of AI network;

attributes of network parameters of the AI network;

the location of non-zero values in network parameters of the AI network;

updated numerical positions in network parameters of the AI network.

The association relationship between the network structures may refer to a connection relationship between input and output of each network structure (or also referred to as a node), for example, an output of a first node is connected to an input of a second node, an output of the second node is connected to an input of a third node, and so on. The attribute of the network parameter comprises information such as name, identification, dimension and the like of the network parameter.

In this embodiment of the present application, the first end converts AI network information into a corresponding transmission file based on a preset model expression mode, compresses the transmission file, and includes:

the first end converts AI network information into at least one transmission file based on at least one preset model expression mode, and one preset model expression mode corresponds to the at least one transmission file;

and the first end performs merging and compression on the at least one transmission file, or the first end performs merging after compressing the at least one transmission file respectively.

It should be noted that there may be a plurality of preset model expression modes, and one preset model expression mode may correspond to at least one transmission file, for example, the network structure and the network parameters are respectively converted into corresponding transmission files based on one preset model expression mode. Of course, the network structure and the network parameters may be converted into a transmission file based on a preset model expression mode.

The first end converts the AI network information into at least one transmission file based on at least one preset model expression mode, merges the at least one transmission file together for compression, and sends the compressed transmission file, or respectively compresses the at least one transmission file independently, merges the at least one transmission file together for sending, or respectively sends each compressed transmission file.

Optionally, the AI network information includes a network structure and network parameters, the first end converts the AI network information into a corresponding transmission file based on a preset model expression mode, and compresses the transmission file, including:

the first end converts the network structure into a first transmission file based on a preset model expression mode, converts the network parameters into a second transmission file based on the preset model expression mode, and compresses the first transmission file and the second transmission file respectively.

That is, the first end may convert the network structure and the network parameters into corresponding transmission files based on a preset model expression mode, so as to obtain two transmission files, compress the two transmission files respectively, and may send the compressed two transmission files respectively, or may be combined together for sending.

For example, the first end is a base station, the second end is a terminal, and when the terminal is initially accessed, the base station may store the trained network structure of the AI network into a transmission file in a corresponding format based on a preset model expression mode, compress the transmission file, and send the transmission file to the terminal; if network parameters are needed, the base station saves the network parameters into a transmission file based on the preset model expression mode for compression and sends the transmission file to the terminal. Alternatively, the transmission file corresponding to the network structure and the transmission file corresponding to the network parameter may be different file types. In addition, the base station may transmit the compressed transmission file through a data channel.

Optionally, in the case that the AI network information includes a network parameter, the compressed AI network information also includes the compressed network parameter, and the first end sends the compressed AI network information to the second end, including:

And the first end sends the compressed network parameters to the second end according to the priority order based on the priority order of the compressed network parameters.

For example, when the first end is taken as a base station and the second end is taken as a terminal, the base station may divide the compressed network parameters into N groups according to a preset priority order, and issue the groups with high priority first, issue the groups with low priority later, or not issue the groups with limited transmission resources.

Optionally, the first end sends the compressed network parameters to the second end according to the priority order based on the priority order of the compressed network parameters, including:

the first end groups the compressed network parameters based on the priority order of the compressed network parameters;

in some specific scenarios, for example, in the case that the transmission resource is smaller than a preset threshold, the first end discards the network parameters after the packets according to a preset sequence, and sends the remaining network parameters, where the preset sequence is the sequence of the priority of the network parameters after the packets from low to high.

For example, the first end is a base station, the second end is a terminal, when the base station sends the compressed network parameters, the base station may divide the compressed network parameters into N groups according to a preset priority order, and when the transmission resources of the base station are smaller than a preset threshold, that is, when the transmission resources of the base station are limited, or when there is a burst of high priority service, the base station discards the network parameters according to a priority order from low to high, that is, the network parameters of the group with the lowest priority are discarded first until the transmission resources of the base station are enough to send the network parameters of the remaining groups, that is, it can be ensured that the network parameters sent by the base station are network parameters with higher priority.

After receiving the network structure and the network parameters sent by the base station, the terminal may use default values agreed by the protocol or 0 for the network parameters not received.

In this embodiment of the present application, before the first end compresses the AI network information, the method further includes:

the first end receives first request information sent by the second end, wherein the first request information is used for requesting acquisition of target AI network information;

in this case, the first end compresses AI network information, including:

and the first end compresses the target AI network information.

That is, the second end can acquire the designated target AI network information based on the first request information, and the first end compresses the target AI network information based on the first request information and then sends the compressed target AI network information to the second end.

Optionally, before the first end compresses the target AI network information, the method further includes:

the first end judges whether the target AI network information needs to be updated or not;

updating the target AI network information under the condition that the target AI network information is judged to need to be updated;

In this case, the first end compresses the target AI network information, including:

and the first end compresses the updated target AI network information.

After receiving a first update request sent by a second end, the first end can determine which target AI network information the second end wants to acquire based on the first update request, and before the first end compresses the target AI network information, the first end can determine whether the target AI network information needs to be updated, if so, the target AI network information is updated, and the updated target AI network information is compressed and then sent to the second end; if the target AI network information does not need to be updated, the first end may directly compress the target AI network information and then send the target AI network information to the second end.

For example, the second end sends a weight parameter update request (Request of Weight Updating) to the first end, designates that a certain or some specific weight parameters need to be acquired, and when the first end determines that the updating is required based on the weight parameter update request, updates the designated weight parameters, compresses the updated weight parameters based on a preset model expression mode and sends the compressed weight parameters to the second end.

Optionally, the first request information includes at least one of:

the name of the requested network parameter;

identification of the requested network parameters;

a network structure update request;

a network parameter update request;

network effect metric value of AI network.

The network effect metric value may refer to that the second end calculates the effect of the AI network according to a certain agreed method, reports the calculated value, and the first end judges whether the AI network information needs to be updated based on the calculated value carried in the first request information. For example, for channel state information (Channel State Information, CSI) prediction, the terminal (second end) calculates the correlation between the predicted result and the measured result and reports the correlation to the base station (first end), where the correlation is the network effect metric value, and the base station determines whether to update the AI network information according to the comparison between the correlation reported by the terminal and the correlation calculated by the AI network. If the correlation calculated by the AI network at the base station side is obviously better than the correlation reported by the terminal, the parameters of the AI network of the terminal are outdated, the parameters cannot be matched with the current channel, and the base station needs to resend the parameters of the AI network.

Optionally, the target AI network information includes a first target network parameter, and the first end compresses the target AI network information, including:

the first end converts the attribute and the parameter value of the first target network parameter into a preset format based on a preset model expression mode and compresses the converted attribute and parameter value;

wherein the attribute of the first target network parameter includes at least one of a name, a dimension, and a length.

The first end is taken as a base station, the second end is taken as a terminal, and when the terminal is initially accessed, one AI network can be selected from preset AI networks, the identification of the AI network is sent to the terminal, and the attribute and the parameter value of the network parameter of the AI network can be converted into a preset format and then compressed, for example, compressed according to a preset model expression mode and then sent to the terminal.

Optionally, in a case that the first end is a network side device, the second end is a terminal, the AI network information includes a network parameter, and the second end is switched from the first cell to the second cell, before the first end compresses the AI network information, the method further includes:

the first end calculates the correlation between the network parameters of the first cell and the network parameters of the second cell, and obtains second target network parameters, wherein the second target network parameters comprise at least one of the following: the method comprises the steps of (1) network parameters with the correlation smaller than a preset threshold value and the first N network parameters in a preset sequence, wherein the preset sequence is a sequence in which the correlations of the network parameters are arranged in a sequence from small to large;

In this case, the first end compresses AI network information, including:

the first end compresses the second target network parameters;

the first end sends the compressed AI network information to a second end, including:

and the first end sends the compressed second target network parameters to a second end.

For example, in the case of a terminal switching cells, the network-side device (e.g., a base station) may send a complete network parameter of a certain AI network to the terminal, or may send a partial network parameter to the terminal. For example, the terminal switches from the first cell to the second cell, the second cell may acquire network parameters corresponding to the first cell from the first cell, calculate the correlation of each network parameter, compress the network parameters with the correlation smaller than a preset threshold or the first N network parameters with the correlation smaller, and send the compressed network parameters to the terminal, and the terminal may only update the received network parameters.

In this embodiment, for updating and transferring network parameters, the first end and the second end may interact with attributes of the network parameters to be transferred, including names, dimensions, lengths, etc. in advance, and then convert the network parameters to be transferred into corresponding files through a preset model expression mode, and then compress and transfer the files. When the network structure is known, the sequence of all network parameters can be considered to be known, and the positions of the network parameters needing to be updated in the whole list can be interacted through a bitmap (bitmap) or a combination number and the like.

Optionally, when describing the network parameters based on the preset model expression, the first end may compress and transmit the attribute and the parameter value of the network parameters together, and the second end determines the length and the corresponding weight of the network parameters based on the received attribute of the network parameters.

The network parameter may be a number or a list of a plurality of groups, and when updating the network parameter, if only a part of values in the list are changed, the first end can indicate the position of the changed values through the value position, the first end only transmits the changed values, and the second end only updates the received values. Alternatively, when the first end transmits the complete network parameter, the position of the non-zero value in the network parameter can be indicated by the value position, the first end only transmits the non-zero value, and the second end regards the value which is not received as 0 or a certain preset value. Wherein the numerical position indication may be an additional indication independent of network parameters, either a radio resource control (Radio Resource Control, RRC) configuration or a medium access control element (Medium Access Control Control Element, MACCE) configuration.

For a better understanding of the technical solutions provided by the embodiments of the present application, the following description is given by way of example only with reference to several specific embodiments.

Example 1

The terminal and the base station use a combined AI network to perform CSI feedback, namely, the terminal converts channel information into CSI feedback information of a plurality of bits (bits) through the AI network and reports the CSI feedback information to the base station, the base station receives the bit information fed back by the terminal, and the channel information is recovered through the AI network at the base station side.

Because the network of the base station and the terminal need to perform joint training, different cell channel conditions are different, and new network parameters may also be needed, when the terminal accesses the network, the base station needs to send the network parameters used by the terminal to the terminal.

The network for CSI feedback can be divided into two parts, a terminal coding part and a base station decoding part, and typically, the base station only needs to transmit the AI network of the terminal coding part to the terminal. The base station may save the network structure and network parameters of the AI network to be transmitted as a file corresponding to a preset model expression, for example, an ONNX file or a pth file of a pytorch, and then, send the whole file to the terminal after the whole file is laminated by a packet data convergence protocol (Packet Data Convergence Protocol, PDCP). After receiving the file, the terminal loads the file into the AI frame according to the file format to realize the inference by using the AI network or continue training.

Or the base station stores the AI network structure into a file corresponding to a preset model expression mode, for example, a meta file of TensorFlow, and then sends the meta file to the terminal through ZIP compression. After receiving the file, the terminal builds a corresponding AI network under its own AI framework, and the weight parameter which is not received defaults to 0 or a certain fixed initial value, and the terminal can train by using the initial value or can directly infer by using the default value.

Or, according to a predefined AI network template (i.e. the above-mentioned preset AI network), the base station sends a certain index value (index) to indicate the corresponding AI network template, and then stores the network parameters into a file corresponding to a preset model expression mode, for example, a data file of a TensorFlow, or stores the index of the network template and the network parameters together into a file corresponding to a certain self-defined network expression mode, and sends the file to the terminal. After receiving the corresponding file, the terminal loads the corresponding network structure in the AI frame according to the index of the AI network template defined in advance, initializes the network parameter according to the network parameter of the AI network template, then updates the network parameter sent by the base station to the AI network, and if some network parameters are not received, uses the initial value in the AI template.

Example two

In the process of measuring the downlink channel by the terminal, the terminal may use the AI network to perform channel measurement, including channel estimation such as CSI reference signal (CSI Reference Signal, CSI-RS), demodulation reference signal (Demodulation Reference Signal, DMRS), and radio resource management (Radio resource management, RRM) measurement. Also, due to the channel differences of different cells, when a terminal switches cells, it is often necessary to switch the network used because the network of the old cell cannot support the channel change situation under the new cell, and the new cell needs to send new network parameters to the terminal.

The network structure of the AI network used by the terminal in the new cell can be generally considered to be consistent with the AI network structure of the old cell, but because of the change of the channel quality, the new cell needs to be retrained and the waste of time is difficult to meet the real-time requirement, so that the new cell can send the network parameters which are trained by the terminal and have the same network structure as the terminal to the terminal, the terminal can continue training and deducing on the basis of the parameters, the real-time efficiency can be improved, the training times can be reduced, and the network can be quickly converged.

The new cell may obtain the network parameters last sent to the terminal from the old cell, compare its own network parameters with the network parameters last sent to the user, and calculate the average correlation of the weight values, for example:

Wherein W is ₁ ⁽ⁱ⁾ An ith value, W, representing the last network parameter sent to the terminal ₂ ⁽ⁱ⁾ The i-th value representing the network parameter this time sent to the terminal, N being the number of values of this network parameter, the smaller the calculated C the more relevant. The specific correlation calculation formulas can be various, and the base station judges which network parameters need to be updated according to the correlation of the weight parameters of two times, and selects M2 network parameters needing to be updated from the total M1 network parameters to update.

The base station can inform the terminal of which network parameters need to be updated firstly, because the terminal knows the network structure, the base station and the terminal naturally know a total of M1 network parameters, the base station and the terminal adopt the same ordering mode for the M1 network parameters, the ordering mode corresponds to a preset model expression mode, the base station informs the terminal of the positions of the M2 network parameters needing to be updated in the ordering mode by using a bitmap (combination number) or the like, after the terminal receives the notification of the base station, the terminal determines which M2 network parameters need to be updated, and then receives specific parameter values.

Because which network parameters need to be updated in advance is determined, the base station directly compresses the values of the network parameters needing to be updated into the data file in the preset model expression mode, the values are directly transmitted without containing information such as weight names, dimensions and the like, and after the terminal receives the values, the terminal analyzes the parameter values of the network parameters at corresponding positions according to the dimensions of each network parameter known by the terminal, and updates the network parameters in the AI frame of the terminal.

Or the base station can directly compress the network parameter name and the numerical value to be updated into a file of a preset model expression mode, and the terminal firstly analyzes the network parameter name and then acquires the corresponding numerical value according to the name.

Optionally, for each network parameter to be updated, the base station calculates the corresponding correlation among all K values of the network parameter, if the correlation is greater than a certain fixed threshold, the value is considered to be not updated, then the positions of all the values to be updated are compressed to a file corresponding to a preset model expression mode together with the values to be transmitted by a bitmap or combination number method, after receiving the compressed file, the terminal firstly decodes out which network parameters need to be updated, and then judges which values of the network parameter need to be updated, and only the values of the corresponding positions are updated.

Specifically, the network parameters to be updated, the value position to be updated in each network parameter may be indicated or configured to the terminal in advance, and then the corresponding value information is directly compressed and then sent to the terminal.

Example III

When two nodes using the AI network interact with the AI network information, the AI network information needs to be compressed into a file corresponding to a certain preset model expression mode, wherein the preset model expression mode is a data structure, and describes the AI network according to a certain rule, and describes information such as AI network parameters.

For example, the preset model expression mode includes TensorFlow, pyTorch, ONNX, which describes the AI network description according to a fixed rule, for example, describes the network as a combination of a plurality of independent nodes, or describes the network as a combination of a plurality of layers, each layer has independent functions, the specific functions are expressed by means of some weight parameters, activation functions and the like, different frameworks have own definition modes for the network, and in order to be uniformly transferred in the communication system, some network description modes meeting the communication requirements can be defined.

The network structure and network parameters of the AI network should be separated to support the delivery of independent network structures and the delivery of independent network parameters. The network structure may be based on nodes, defining the basic functions of the nodes, mapping all weights to inputs and outputs, and representing connections by using the same numbers through the inputs and outputs, so that the entire network is described by the numbers of the nodes and the inputs and outputs.

For network parameters, at least one of the following is included:

1. the number of the parameter is consistent with the number described in the network structure;

2. the dimension of the parameter needs to be matched with other parameters connected with the parameter;

3. The numerical value of the parameter is the specific parameter content and is matched with the dimension;

4. in order to reduce the overhead of parameter transmission, the effective value position information in the parameters is not changed or the parameters approaching 0 can not be transmitted, and the corresponding position information is required to indicate the position of the effective value.

These may be combined arbitrarily into separate files, with the complete network parameters being represented as a result after differencing, i.e. with part of the parameters not passed, or with the complete network being represented as a result of part weights, i.e. corresponding compression operations.

Example IV

In some scenarios, some functions only need the terminal to use the AI network, and do not need to train with the network side equipment, for example, channel prediction, positioning and the like at the terminal side, and the AI networks are affected by the channel environment, and along with the movement of the terminal, the AI networks need to train continuously and update the networks. When the terminal B enters the area of the terminal a, the terminal a can transmit the trained AI network to the terminal B, and the terminal B trains based on the AI network of the terminal a.

The training complexity can be reduced by transferring the trained AI network between any two associated nodes, and even the AI network trained by other nodes can be directly used.

The terminal a and the terminal B are the same for the AI network structure of the same function, the main network parameters of the two needs to be interacted, the terminal a can divide the network parameters into N parts in consideration of the transmission limit of a sidelink (sidelink), the network parameters are respectively transmitted to the terminal B in different time slots (slots), and when each transmission is performed, the information of network parameter names, dimensions and the like can be uniformly compressed parameter values or can be directly transmitted values after independent configuration. After receiving part of the network parameters of the terminal a, the terminal B may directly update the network parameters of its corresponding location and use the updated network to directly infer, or the terminal B may wait until all the network parameters are received completely and update its corresponding network parameters together, which mainly depends on whether the part of the network parameters can operate independently, and this capability may be notified to the terminal B by the terminal a, or through a common base station configuration, if the capability can operate independently, the terminal B may update its network after each time of receiving a new network parameter, otherwise wait until all the network parameters are received completely and update together.

The terminal B may also designate a certain network parameter to be updated, send the information such as the name and/or dimension of the network parameter to be updated to the terminal a, or may include the information such as the resource location where the network parameter is expected to be received at the same time, where the terminal a directly compresses the value of the corresponding network parameter into a file corresponding to a certain preset model expression mode according to the sequence of the network parameter required by the terminal a according to the signaling of the terminal B, and send the file to the terminal B, and after the terminal B receives the file, update the network parameter required by itself.

Referring to fig. 3, fig. 3 is another AI-network information transmission method provided in an embodiment of the disclosure, where, as shown in fig. 3, the AI-network information transmission method includes the following steps:

step 301, the second end receives the compressed AI network information sent by the first end, where the AI network information includes at least one of a network structure and a network parameter.

It can be appreciated that the first end compresses the AI network information, and sends the compressed AI network information to the second end. The compressing the AI network information may refer to compressing the AI network information into a file corresponding to a preset model expression mode according to a preset model expression mode, where the model expression mode is a data structure, and describes information such as an AI network structure, network parameters, and the like according to a certain rule. The implementation process of the compression of the AI network information by the first end may refer to the specific description in the method embodiment described in fig. 2, which is not repeated herein.

The first end compresses the AI network information, namely, compresses the network structure and/or the weight parameter, and sends the compressed network structure and/or the compressed weight parameter to the second end. For example, if the AI network information includes only a network structure, the first end compresses and transmits only the network structure; alternatively, the AI network information may include only the network parameter, and the first end may compress and transmit only the network parameter; alternatively, the AI network information includes a partial network structure and a partial network parameter, and the first end compresses and transmits the partial network structure and the partial network parameter. Of course, the specific information content included in the AI network information may be other situations, which are not described in detail herein.

Optionally, before the step 301, the method further includes:

the second end sends first request information to the first end, wherein the first request information is used for requesting to acquire target AI network information;

in this case, the step 301 includes:

and the second end receives the compressed target AI network information sent by the first end.

Optionally, the first request information includes at least one of:

the name of the requested network parameter;

identification of the requested network parameters;

a network structure update request;

a network parameter update request;

network effect metric value of AI network.

It should be noted that, the AI network information transmission method provided in the embodiment of the present application is applied to the second end, and corresponds to the AI network information transmission method provided in the embodiment of fig. 2 and applied to the first end, and the specific implementation process of the relevant steps in the embodiment of the present application may refer to the description in the embodiment of the method described in fig. 2, so that repetition is avoided.

In this embodiment of the present application, the second end receives compressed AI network information sent by the first end, where the AI network information includes at least one of a network structure and a network parameter, and further in a communication process, it is not necessary to perform compression transmission on the entire AI network including all network structures and network parameters together, so that the network structure and the network parameter of the AI network may be sent separately, and further transmission overhead in the communication process may be effectively reduced.

According to the AI network information transmission method provided by the embodiment of the application, the execution main body can be an AI network information transmission device. In the embodiment of the present application, an AI network information transmission device executes an AI network information transmission method by taking an AI network information transmission device as an example, and the AI network information transmission device provided in the embodiment of the present application is described.

Referring to fig. 4, fig. 4 is a block diagram of an AI-network information transmission apparatus according to an embodiment of the disclosure, and as shown in fig. 4, an AI-network information transmission apparatus 400 includes:

a compression module 401, configured to compress AI network information, where the AI network information includes at least one of a network structure and a network parameter;

a sending module 402, configured to send the compressed AI network information to a second end.

Optionally, the AI network information includes a network structure and network parameters, and the compression module 401 is configured to perform any one of the following:

combining and compressing the network structure and the network parameters;

and respectively compressing the network structure and the network parameters.

Optionally, in the case that the compression module 401 compresses the network structure and the network parameter separately, the sending module 402 is configured to perform any one of the following:

The compressed network structure and the compressed network parameters are combined and sent to the second end;

and respectively sending the compressed network structure and the compressed network parameters to the second end.

Optionally, the compression module 401 is configured to perform any one of the following:

converting AI network information into corresponding transmission files based on a preset model expression mode, and compressing the transmission files;

compressing the AI network information based on a preset data format;

acquiring AI network information to be sent and the AI network information existing at the second end, and compressing an AI network information difference value between the AI network information to be sent and the AI network information existing at the second end;

and acquiring the AI network information to be transmitted and the AI network information of a preset AI network, and compressing an AI network information difference value between the AI network information to be transmitted and the AI network information of the preset AI network.

a specified network parameter;

indexing network parameters;

modified network parameters;

modified parameter values in the modified network parameters;

the location of the modified parameter value in the modified network parameter;

a non-zero value in the modified network parameter;

the location of non-zero values in the modified network parameters;

a newly added network structure;

a deleted network structure;

modified network architecture.

Optionally, the preset model expression mode includes any one of the following: the protocol appoints a model expression mode and a self-defined model expression mode.

Optionally, the content of the customized model expression mode includes at least one of the following: network structure of AI network, attribute of network parameter of AI network, value of network parameter of AI network.

association relationship between network structures of AI network;

attributes of network parameters of the AI network;

the location of non-zero values in network parameters of the AI network;

updated numerical positions in network parameters of the AI network.

Optionally, the compression module 401 is further configured to:

converting AI network information into at least one transmission file based on at least one preset model expression mode, wherein one preset model expression mode corresponds to the at least one transmission file;

And combining and compressing the at least one transmission file, or respectively compressing and then combining the at least one transmission file.

Optionally, the AI network information includes a network structure and network parameters, and the compression module 401 is further configured to:

and converting the network structure into a first transmission file based on a preset model expression mode, converting the network parameters into a second transmission file based on the preset model expression mode, and respectively compressing the first transmission file and the second transmission file.

Optionally, the compressed AI network information includes compressed network parameters, and the sending module 402 is further configured to:

and transmitting the compressed network parameters to a second end according to the priority order based on the priority order of the compressed network parameters.

Optionally, the sending module 402 is further configured to:

grouping the compressed network parameters based on the priority order of the compressed network parameters;

and under the condition that the transmission resources are smaller than a preset threshold value, discarding the grouped network parameters according to a preset sequence and transmitting the rest network parameters, wherein the preset sequence is the sequence of the priority of the grouped network parameters from low to high.

Optionally, the apparatus further comprises:

the receiving module is used for receiving first request information sent by the second end, wherein the first request information is used for requesting acquisition of target AI network information;

the compression module 401 is further configured to: and compressing the target AI network information.

Optionally, the first request information includes at least one of:

the name of the requested network parameter;

identification of the requested network parameters;

a network structure update request;

a network parameter update request;

network effect metric value of AI network.

Optionally, the apparatus further comprises:

the judging module is used for judging whether the target AI network information needs to be updated or not;

the compression module 401 is further configured to:

and compressing the updated target AI network information.

Optionally, the target AI network information includes a first target network parameter, and the compression module 401 is further configured to:

converting the attribute and parameter value of the first target network parameter into a preset format based on a preset model expression mode, and compressing;

Optionally, the apparatus is one of a network side device and a terminal, and the second end is the other of the network side device and the terminal; or,

the device and the second end are different nodes of the terminal; or,

the device and the second end are different nodes of the network side equipment.

Optionally, the apparatus is a network side device, the second end is a terminal, the AI network information includes a network parameter, and before the first end compresses the AI network information when the second end is handed over from the first cell to the second cell, the method further includes:

the compression module 401 is further configured to:

Compressing the second target network parameter;

the sending module 402 is further configured to:

and sending the compressed second target network parameters to a second end.

In this embodiment of the present application, the device may send compressed AI network information to the second end, where the AI network information includes at least one of a network structure and a network parameter, and further in a communication process, it is not necessary to transmit the entire AI network including all network structures and network parameters together, so that the network structure and the network parameter of the AI network may be sent separately, and further transmission overhead in the communication process may be effectively reduced.

The AI-network information transmission apparatus 400 in this embodiment of the application may be an electronic device, for example, an electronic device with an operating system, or may be a component in an electronic device, for example, an integrated circuit or a chip. The electronic device may be a terminal, or may be other devices than a terminal. By way of example, terminals may include, but are not limited to, the types of terminals 11 listed above, other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., and embodiments of the application are not specifically limited.

The AI network information transmission apparatus 400 provided in this embodiment of the present application can implement each process implemented by the method embodiment described in fig. 2, and achieve the same technical effects, and for avoiding repetition, a detailed description is omitted herein.

Referring to fig. 5, fig. 5 is a block diagram of another AI-network information transmission apparatus provided in an embodiment of the disclosure, and as shown in fig. 5, the AI-network information transmission apparatus 500 includes:

a receiving module 501, configured to receive compressed AI network information sent by a first end, where the AI network information includes at least one of a network structure and a network parameter.

Optionally, the apparatus further comprises:

the sending module is used for sending first request information to the first end, wherein the first request information is used for requesting to acquire target AI network information;

the receiving module 501 is further configured to:

and receiving the compressed target AI network information sent by the first end.

Optionally, the first request information includes at least one of:

the name of the requested network parameter;

identification of the requested network parameters;

a network structure update request;

a network parameter update request;

network effect metric value of AI network.

Optionally, the first end is one of a network side device and a terminal, and the apparatus is the other of the network side device and the terminal; or,

The first end and the device are different nodes of a terminal; or,

the first end and the device are different nodes of network side equipment.

In this embodiment of the present application, the device receives compressed AI network information sent by the first end, where the AI network information includes at least one of a network structure and a network parameter, and further in a communication process, it is not necessary to perform compression transmission on an entire AI network including all network structures and network parameters together, so that the network structure and the network parameter of the AI network may be sent separately, and further transmission overhead in the communication process may be effectively reduced.

The AI network information transmission apparatus 500 provided in this embodiment of the present application can implement each process implemented by the method embodiment illustrated in fig. 3, and achieve the same technical effects, and for avoiding repetition, a detailed description is omitted herein.

Optionally, as shown in fig. 6, the embodiment of the present application further provides a communication device 600, including a processor 601 and a memory 602, where the memory 602 stores a program or an instruction that can be executed on the processor 601, and the program or the instruction implements each step of the embodiment of the AI network information transmission method described in fig. 2 or fig. 3 when executed by the processor 601, and achieves the same technical effects. In order to avoid repetition, a description thereof is omitted.

The embodiment of the application further provides a terminal, and each implementation process and implementation manner of the embodiment of the method of fig. 2 or fig. 3 are applicable to the embodiment of the terminal, and the same technical effects can be achieved. Specifically, fig. 7 is a schematic hardware structure of a terminal for implementing an embodiment of the present application.

The terminal 700 includes, but is not limited to: at least some of the components of the radio frequency unit 701, the network module 702, the audio output unit 703, the input unit 704, the sensor 705, the display unit 706, the user input unit 707, the interface unit 708, the memory 709, and the processor 710.

Those skilled in the art will appreciate that the terminal 700 may further include a power source (e.g., a battery) for powering the various components, and that the power source may be logically coupled to the processor 710 via a power management system so as to perform functions such as managing charging, discharging, and power consumption via the power management system. The terminal structure shown in fig. 7 does not constitute a limitation of the terminal, and the terminal may include more or less components than shown, or may combine certain components, or may be arranged in different components, which will not be described in detail herein.

It should be appreciated that in embodiments of the present application, the input unit 704 may include a graphics processing unit (Graphics Processing Unit, GPU) 7041 and a microphone 7042, with the graphics processor 7041 processing image data of still pictures or video obtained by an image capturing device (e.g., a camera) in a video capturing mode or an image capturing mode. The display unit 706 may include a display panel 7061, and the display panel 7061 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 at least one of a touch panel 7071 and other input devices 7072. The touch panel 7071 is also referred to as a 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, a physical keyboard, function keys (e.g., volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and so forth, which are not described in detail herein.

In this embodiment, after receiving downlink data from the network side device, the radio frequency unit 701 may transmit the downlink data to the processor 710 for processing; in addition, the radio frequency unit 701 may send uplink data to the network side device. Typically, the radio unit 701 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, and the like.

The memory 709 may be used to store software programs or instructions and various data. The memory 709 may mainly include a first storage area storing programs or instructions and a second storage area storing data, wherein the first storage area may store an operating system, application programs or instructions (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like. Further, the memory 709 may include volatile memory or nonvolatile memory, or the memory 709 may include both volatile and nonvolatile memory. The nonvolatile Memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable EPROM (EEPROM), or a flash Memory. The volatile memory may be random access memory (Random Access Memory, RAM), static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (ddr SDRAM), enhanced SDRAM (Enhanced SDRAM), synchronous DRAM (SLDRAM), and Direct RAM (DRRAM). Memory 709 in embodiments of the present application includes, but is not limited to, these and any other suitable types of memory.

Processor 710 may include one or more processing units; optionally, processor 710 integrates an application processor that primarily processes operations involving an operating system, user interface, application programs, and the like, and a modem processor that primarily processes wireless communication signals, such as a baseband processor. It will be appreciated that the modem processor described above may not be integrated into the processor 710.

In one implementation of the embodiments of the present application, the terminal 700 is the first end. Wherein the processor 710 is configured to: compressing AI network information, the AI network information including at least one of a network structure and a network parameter;

the radio frequency unit 701 is configured to: and sending the compressed AI network information to a second end.

Optionally, the AI network information includes a network structure and network parameters, and the processor 710 is configured to perform any one of:

combining and compressing the network structure and the network parameters;

and respectively compressing the network structure and the network parameters.

Optionally, in the case that the processor 710 is configured to compress the network structure and the network parameters respectively, the radio frequency unit 701 is further configured to perform any one of the following:

Optionally, the processor 710 is configured to perform any one of the following:

compressing the AI network information based on a preset data format;

a specified network parameter;

indexing network parameters;

modified network parameters;

modified parameter values in the modified network parameters;

the location of the modified parameter value in the modified network parameter;

a non-zero value in the modified network parameter;

the location of non-zero values in the modified network parameters;

a newly added network structure;

a deleted network structure;

modified network architecture.

association relationship between network structures of AI network;

attributes of network parameters of the AI network;

the location of non-zero values in network parameters of the AI network;

updated numerical positions in network parameters of the AI network.

Optionally, the processor 710 is further configured to:

Optionally, the AI network information includes a network structure and network parameters, and the processor 710 is further configured to:

Optionally, the compressed AI network information includes compressed network parameters, and the radio frequency unit 701 is further configured to:

Optionally, the radio frequency unit 701 is further configured to:

receiving first request information sent by the second end, wherein the first request information is used for requesting acquisition of target AI network information;

the processor 710 is further configured to:

and compressing the target AI network information.

Optionally, the first request information includes at least one of:

the name of the requested network parameter;

identification of the requested network parameters;

a network structure update request;

a network parameter update request;

network effect metric value of AI network.

Optionally, the processor 710 is further configured to:

judging whether the target AI network information needs to be updated or not;

and compressing the updated target AI network information.

Optionally, the target AI network information includes a first target network parameter, and the processor 710 is further configured to:

In another implementation manner of the embodiment of the present application, the terminal 700 is a second terminal. Wherein, the radio frequency unit 701 is further configured to: and receiving the compressed AI network information sent by the first end, wherein the AI network information comprises at least one of a network structure and a network parameter.

Optionally, the radio frequency unit 701 is further configured to:

sending first request information to the first end, wherein the first request information is used for requesting acquisition of target AI network information;

Optionally, the first request information includes at least one of:

the name of the requested network parameter;

identification of the requested network parameters;

a network structure update request;

a network parameter update request;

network effect metric value of AI network.

According to the technical scheme, the network structure and the network parameters of the AI network can be sent separately, so that the transmission overhead in the communication process can be effectively reduced.

The embodiment of the application further provides a network side device, and each implementation process and implementation manner of the embodiments of the methods described in fig. 2 and fig. 3 are applicable to the embodiment of the network side device, and the same technical effects can be achieved.

Specifically, the embodiment of the application also provides network side equipment. As shown in fig. 8, the network side device 800 includes: an antenna 81, a radio frequency device 82, a baseband device 83, a processor 84 and a memory 85. The antenna 81 is connected to a radio frequency device 82. In the uplink direction, the radio frequency device 82 receives information via the antenna 81, and transmits the received information to the baseband device 83 for processing. In the downlink direction, the baseband device 83 processes information to be transmitted, and transmits the processed information to the radio frequency device 82, and the radio frequency device 82 processes the received information and transmits the processed information through the antenna 81.

The method performed by the network side device in the above embodiment may be implemented in the baseband apparatus 83, and the baseband apparatus 83 includes a baseband processor.

The baseband device 83 may, for example, include at least one baseband board, where a plurality of chips are disposed, as shown in fig. 8, where one chip, for example, a baseband processor, is connected to the memory 85 through a bus interface, so as to call a program in the memory 85 to perform the network device operation shown in the above method embodiment.

The network-side device may also include a network interface 86, such as a common public radio interface (common public radio interface, CPRI).

Specifically, the network side device 800 of the embodiment of the present invention further includes: instructions or programs stored in the memory 85 and executable on the processor 84, the processor 84 invokes the instructions or programs in the memory 85 to perform the methods performed by the modules shown in fig. 4 or fig. 5, and achieve the same technical effects, and are not repeated here.

Specifically, the embodiment of the application also provides another network side device. As shown in fig. 9, the network side device 900 includes: a processor 901, a network interface 902, and a memory 903. The network interface 902 is, for example, a common public radio interface (common public radio interface, CPRI).

Specifically, the network side device 900 of the embodiment of the present invention further includes: instructions or programs stored in the memory 903 and executable on the processor 901, the processor 901 invokes the instructions or programs in the memory 903 to perform the methods performed by the modules shown in fig. 4 or fig. 5, and achieve the same technical effects, so that repetition is avoided and thus they are not described herein.

The embodiment of the present application further provides a readable storage medium, where a program or an instruction is stored, and when the program or the instruction is executed by a processor, the processes of the embodiment of the method described in fig. 2 or fig. 3 are implemented, and the same technical effects can be achieved, so that repetition is avoided, and no further description is given here.

Wherein the processor is a processor in the terminal described in the above embodiment. The readable storage medium includes computer readable storage medium such as computer readable memory ROM, random access memory RAM, magnetic or optical disk, etc.

The embodiment of the present application further provides a chip, where the chip includes a processor and a communication interface, where the communication interface is coupled to the processor, and the processor is configured to run a program or an instruction, implement each process of the method embodiment described in fig. 2 or fig. 3, and achieve the same technical effect, so that repetition is avoided, and no further description is given here.

It should be understood that the chips referred to in the embodiments of the present application may also be referred to as system-on-chip chips, or the like.

The embodiments of the present application further provide a computer program/program product, which is stored in a storage medium, and executed by at least one processor to implement the respective processes of the embodiments of the methods described in fig. 2 or fig. 3, and achieve the same technical effects, and are not repeated herein.

The embodiment of the application also provides a communication system, which comprises: a terminal and a network side device, wherein the terminal is used for executing the steps of the method as shown in fig. 2, and the network side device is used for executing the steps of the method as shown in fig. 3; alternatively, the terminal may be configured to perform the steps of the method as described in fig. 3 above, and the network-side device may be configured to perform the steps of the method as described in fig. 2.

It should be noted that, in this document, 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 one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. Furthermore, it should be noted that the scope of the methods and apparatus in the embodiments of the present application is not limited to performing the functions in the order shown or discussed, but may also include performing the functions in a substantially simultaneous manner or in an opposite order depending on the functions involved, e.g., the described methods may be performed in an order different from that described, and various steps may also be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solutions of the present application may be embodied essentially or in a part contributing to the prior art in the form of a computer software product stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), comprising several instructions for causing a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method described in the embodiments of the present application.

The embodiments of the present application have been described above with reference to the accompanying drawings, but the present application is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those of ordinary skill in the art without departing from the spirit of the present application and the scope of the claims, which are also within the protection of the present application.

Claims

1. An artificial intelligence AI network information transmission method, comprising:

and the first end sends the compressed AI network information to a second end.

2. The method of claim 1, wherein the AI network information comprises a network structure and network parameters, and wherein the first end compresses the AI network information, including any one of:

3. The method of claim 2, wherein, in the case where the first end compresses the network structure and the network parameters, respectively, the first end sends the compressed AI network information to a second end, including any one of:

4. The method of claim 1, wherein the first end compresses AI network information, comprising any one of:

5. The method of claim 4, wherein the AI network information difference comprises at least one of:

A specified network parameter;

indexing network parameters;

modified network parameters;

modified parameter values in the modified network parameters;

the location of the modified parameter value in the modified network parameter;

a non-zero value in the modified network parameter;

the location of non-zero values in the modified network parameters;

a newly added network structure;

a deleted network structure;

modified network architecture.

6. The method according to claim 4, wherein the preset model expression includes any one of the following: the protocol appoints a model expression mode and a self-defined model expression mode.

7. The method of claim 6, wherein the content of the custom model representation comprises at least one of: network structure of AI network, attribute of network parameter of AI network, value of network parameter of AI network.

8. The method of claim 4, wherein the representation of the network structure in the preset model representation includes at least one of:

association relationship between network structures of AI network;

Attributes of network parameters of the AI network;

the location of non-zero values in network parameters of the AI network;

updated numerical positions in network parameters of the AI network.

9. The method of claim 4, wherein the first end converts AI network information into a corresponding transmission file based on a preset model expression, and compresses the transmission file, comprising:

10. The method of claim 4, wherein the AI network information includes a network structure and network parameters, wherein the first end converts the AI network information into a corresponding transmission file based on a preset model expression, and wherein compressing the transmission file includes:

11. The method of any of claims 1-10, wherein the compressed AI network information comprises a compressed network parameter, and wherein the first end sends the compressed AI network information to a second end, comprising:

12. The method of claim 11, wherein the first end sends the compressed network parameters to a second end in a priority order based on the priority order of the compressed network parameters, comprising:

and under the condition that the transmission resource is smaller than a preset threshold value, the first end discards the grouped network parameters according to a preset sequence and transmits the rest network parameters, wherein the preset sequence is the sequence of the priority of the grouped network parameters from low to high.

13. The method of any of claims 1-10, wherein prior to the first end compressing AI network information, the method further comprises:

the first end compresses AI network information, including:

and the first end compresses the target AI network information.

14. The method of claim 13, wherein the first request information comprises at least one of:

the name of the requested network parameter;

identification of the requested network parameters;

a network structure update request;

a network parameter update request;

network effect metric value of AI network.

15. The method of claim 13, wherein prior to the first end compressing the target AI network information, the method further comprises:

the first end compresses the target AI network information, including:

and the first end compresses the updated target AI network information.

16. The method of claim 13, wherein the target AI network information comprises a first target network parameter, the first end compressing the target AI network information, comprising:

17. The method according to any of claims 1-10, wherein the first end is one of a network side device and a terminal and the second end is the other of a network side device and a terminal; or,

the first end and the second end are different nodes of a terminal; or,

the first end and the second end are different nodes of network side equipment.

18. The method according to any one of claims 1-10, wherein the first end is a network-side device, the second end is a terminal, the AI network information includes a network parameter, and the method further comprises, before the first end compresses the AI network information in case the second end is handed over from the first cell to the second cell:

The first end compresses AI network information, including:

the first end compresses the second target network parameters;

19. An AI network information transmission method, comprising:

20. The method of claim 19, wherein before the second end receives the compressed AI network information sent by the first end, the method further comprises:

the second terminal receives the compressed AI network information sent by the first terminal, including:

21. The method of claim 20, wherein the first request information comprises at least one of:

The name of the requested network parameter;

identification of the requested network parameters;

a network structure update request;

a network parameter update request;

network effect metric value of AI network.

22. The method of claim 19, wherein the first end is one of a network side device and a terminal, and the second end is the other of the network side device and the terminal; or,

the first end and the second end are different nodes of a terminal; or,

the first end and the second end are different nodes of network side equipment.

23. An AI network information transmission apparatus, characterized by comprising:

24. An AI network information transmission apparatus, characterized by comprising:

25. A communication device comprising a processor and a memory storing a program or instructions executable on the processor, which when executed by the processor, implement the steps of the AI network information transfer method of any of claims 1-18, or implement the steps of the AI network information transfer method of any of claims 19-22.

26. A readable storage medium, wherein a program or instructions is stored on the readable storage medium, which when executed by a processor, implements the steps of the AI network information transfer method of any of claims 1-18, or implements the steps of the AI network information transfer method of any of claims 19-22.