Interaction method and device for AI model parameters

Abstract

The embodiment of the invention provides an interaction method and device of AI model parameters, the method is applied to a WIFI sending end, and the method comprises the following steps: after an auto-encoder of the WIFI sending end is embedded with the AI model parameters, sending a data frame with a special sequence to a WIFI receiving end through the auto-encoder of the WIFI sending end, and sending the special sequence to a WIFI receiving end, so that the WIFI receiving end identifies the data frame according to the special sequence, and decoding the data frame through an auto-encoder of the WIFI receiving end under the condition of successful identification. The self-encoder improves the coding efficiency and the decoding accuracy of a traditional WIFI transmitter and a traditional WIFI receiver system, so that the WIFI network performance and the user experience are enhanced, and the problem that the effect of enhancing the network performance and the user experience through the AI / ML cannot be achieved due to the fact that the specific implementation mode that the AI / ML is used for the WLAN does not exist in the related technology is solved.

Description

Technical Field

[0001] This invention relates to the field of communications, and more specifically, to a method and apparatus for interacting with AI model parameters. Background Technology

[0002] Traditional artificial intelligence / machine learning (AI / ML) technologies are based on centralized models, requiring the exchange of large amounts of data between data sources and centralized servers. Recently, distributed AI / ML algorithms such as federated learning have been developed, which will allow for more analysis at the source and reduce the amount of data that needs to be exchanged.

[0003] Research shows that AI / ML algorithms can help improve the performance of wireless communication networks by providing better resource utilization, lower energy consumption, higher reliability, and robustness to changing environments. As these algorithms become more mature and cost-effective, Wireless Local Area Networks (WLANs) may leverage AI / ML to enhance network performance and user experience.

[0004] However, there is currently no concrete implementation of AI / ML in WLAN, so it is impossible to achieve the effect of enhancing network performance and user experience through AI / ML. Summary of the Invention

[0005] This invention provides an interactive method and apparatus for AI model parameters, which at least solves the problem that related technologies have not yet applied AI / ML to specific implementations of WLAN, thus failing to achieve the effect of enhancing network performance and user experience through AI / ML.

[0006] According to an embodiment of the present invention, an interactive method for AI model parameters is provided, applied to a WIFI transmitter, wherein the WIFI transmitter includes an autoencoder, and the method includes:

[0007] After the autoencoder at the WIFI transmitter is fed with AI model parameters, it sends a data frame with a special sequence to the WIFI receiver so that the WIFI receiver can identify the data frame according to the special sequence. If the identification is successful, the autoencoder at the WIFI receiver decodes the data frame.

[0008] In one exemplary embodiment, after transmitting data frames with a special sequence to the WIFI receiver via the self-encoder of the WIFI transmitter, the method further includes:

[0009] Listen to the feedback messages from the WIFI receiver;

[0010] If no feedback message is received from the WIFI receiver within a preset time, it is determined that the AI ​​model parameters of the autoencoder of the WIFI transmitter need to be updated.

[0011] In one exemplary embodiment, it further includes:

[0012] When the autoencoder at the WIFI transmitter needs to update the AI ​​model parameters, it sends a trigger frame with a special sequence to the WIFI receiver within an inter-frame interval to trigger the autoencoder at the WIFI receiver to feed back new AI model parameters.

[0013] In one exemplary embodiment, it further includes:

[0014] The system receives new AI model parameters from the autoencoder of the WIFI receiver and sends an acknowledgment message to the WIFI receiver.

[0015] In one exemplary embodiment, the data frame with a special sequence carries feedback Channel State Information (CSI) content.

[0016] According to another embodiment of the present invention, another method for interacting AI model parameters is provided, applied to a WIFI receiver, wherein the WIFI receiver includes an autoencoder and a classifier of the WIFI receiver, and the method includes:

[0017] After the autoencoder at the WIFI transmitter is fed with AI model parameters, the classifier receives data frames with special sequences sent by the autoencoder at the WIFI transmitter and identifies the data frames according to the special sequences.

[0018] If identification is successful, the data frame is decoded by the self-encoder of the WIFI receiver.

[0019] In one exemplary embodiment, after decoding the data frame using the self-encoder of the WIFI receiver, the method further includes:

[0020] When the autoencoder at the WIFI transmitter needs to update the AI ​​model parameters, it receives a trigger frame with a special sequence sent by the WIFI transmitter.

[0021] The trigger frame is identified by the classifier based on the special sequence;

[0022] If the identification is successful, the autoencoder of the WIFI receiver is trained to obtain new AI model parameters;

[0023] The new AI model parameters are fed back to the autoencoder of the WIFI transmitter.

[0024] In an exemplary embodiment, the step of identifying the trigger frame based on the special sequence using the classifier includes:

[0025] Match the special sequence of the trigger frame with the special sequence stored locally;

[0026] If a match is found, the identification is considered successful.

[0027] According to another embodiment of the present invention, an interactive device for AI model parameters is provided, applied to a WIFI transmitter, wherein the WIFI transmitter includes a self-encoder, and the device includes:

[0028] The first transmitting module is configured to transmit a data frame with a special sequence to the WIFI receiver through the autoencoder of the WIFI transmitter after the AI ​​model parameters are loaded into the autoencoder of the WIFI transmitter, so that the WIFI receiver can identify the data frame according to the special sequence, and decode the data frame through the autoencoder of the WIFI receiver if the identification is successful.

[0029] According to another embodiment of the present invention, a different interactive device for AI model parameters is provided, applied to a WIFI receiver, wherein the WIFI receiver includes a self-decoder and a classifier, and the device includes:

[0030] The first receiving module is used to receive data frames with a special sequence sent by the encoder of the WIFI transmitter through the classifier after the AI ​​model parameters are put into the autoencoder of the WIFI transmitter, and to identify the data frames according to the special sequence.

[0031] The decoding module is used to decode the data frame using the autoencoder of the WIFI receiver when the identification is successful.

[0032] According to yet another embodiment of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to perform the steps in any of the above method embodiments when executed.

[0033] According to another embodiment of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any one of the above method embodiments.

[0034] This invention employs autoencoders at both the Wi-Fi transmitter and receiver. After AI model parameters are input into the autoencoder at the Wi-Fi transmitter, data frames with a specific sequence are sent to the Wi-Fi receiver via the autoencoder. The Wi-Fi receiver identifies the data frames based on the specific sequence, and if identification is successful, decodes the data frames using its own autoencoder. This autoencoder improves the encoding efficiency and decoding accuracy of traditional Wi-Fi transmitter and receiver systems, thereby enhancing Wi-Fi network performance and user experience. It also addresses the problem that related technologies have not utilized AI / ML in specific WLAN implementations, thus failing to achieve the desired effect of enhancing network performance and user experience through AI / ML. Attached Figure Description

[0035] Figure 1 This is a hardware structure block diagram of a mobile terminal for an AI model parameter interaction method according to an embodiment of the present invention.

[0036] Figure 2 This is a network architecture diagram of a WIFI system according to an embodiment of the present invention;

[0037] Figure 3 This is a flowchart illustrating the interaction of AI model parameters according to an embodiment of the present invention;

[0038] Figure 4 This is a schematic diagram of AI parameter interaction during the DIFS frame interval according to an embodiment of the present invention;

[0039] Figure 5 This is a schematic diagram of AI parameter interaction during the EIFS frame interval time according to an embodiment of the present invention;

[0040] Figure 6 This is a flowchart illustrating the interaction of AI model parameters according to another embodiment of the present invention;

[0041] Figure 7 This is a structural block diagram of an interactive device for AI model parameters according to an embodiment of the present invention;

[0042] Figure 8 This is a structural block diagram of an interactive device for AI model parameters according to another embodiment of the present invention. Detailed Implementation

[0043] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples.

[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0045] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for an AI model parameter interaction method according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0046] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the AI ​​model parameter interaction method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0047] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0048] Figure 2 This is a network architecture diagram of a WIFI system according to an embodiment of the present invention. This embodiment can be run on... Figure 2In the network architecture shown, such as Figure 2 As shown, the network architecture of the WIFI system in this embodiment of the invention mainly includes a WIFI transmitter and a WIFI receiver. The WIFI transmitter includes an autoencoder (AE) and a traditional WIFI transmitter, and the WIFI receiver includes an autoencoder (AE), a classifier, and a traditional WIFI receiver. The functions of each module are as follows:

[0049] The autoencoder can be set at both the WIFI transmitter and the WIFI receiver, including an AE encoder and an AE decoder. The AE encoder takes the encoded vector as its output, and the AE decoder takes the encoded vector as its input. The space between the encoder's input and the encoded vector, and between the decoder's output and the decoded vector, is called a hidden layer. The structure of the hidden layer can be a convolutional neural network (CNN) structure, a fully connected structure, or a recurrent neural network (RNN) structure.

[0050] Among them, the AE encoder is used to encode the transmitted data, redefine the function coding and modulation, and perform coding and modulation similar to those in traditional WIFI transmitters, or compress and encode the transmitted content such as channel estimation data as a transmitter.

[0051] An AE decoder is used to decode the Channel State Information (CSI) content compressed and fed back by the encoder, or as a receiver to detect and demodulate the transmitted content.

[0052] Traditional Wi-Fi transmitters: Traditional Wi-Fi transmitters consist of an encoded modulation radio frequency front-end, used to transmit encoded data.

[0053] Classifier: The classifier is connected to the autoencoder of the WIFI transmitter and the traditional WIFI transmitter respectively. It receives the data sent by the WIFI transmitter and is one of the most important hardware devices in this invention. Its main functions are twofold: to identify the trigger frame sent by the AE of the WIFI transmitter and to identify the encoded vector encoded by the AE of the WIFI transmitter.

[0054] The trigger frame's data field is a special sequence. Identifying the trigger signal involves matching this special sequence with locally stored feature sequences. If a match is successful, identification is confirmed, and AE parameter interaction can continue. The AE-encoded vector from the Wi-Fi transmitter can be scrambled with a special pseudo-noise (PN) sequence. The classifier can use a locally cached PN sequence for detection and identification. If identification is successful, the AE-encoded vector from the Wi-Fi transmitter is sent to the AE decoder at the Wi-Fi receiver for decoding. If identification fails, the traditional Wi-Fi receiver detection process begins.

[0055] Traditional WIFI receiver: used for demodulating and decoding the received data.

[0056] In addition, it should be noted that the feedback link is not a hardware component, but a frame interaction sequence adapted to this hardware structure. When the autoencoder (AE) is used for signal modulation and demodulation, the feedback link is used to feed back the parameters in the AE encoder trained by the WIFI receiver to the WIFI transmitter and put them into the AE encoder of the WIFI transmitter.

[0057] This embodiment provides a method for interacting with AI model parameters of a network architecture running on the aforementioned WIFI system. Figure 3 This is a flowchart illustrating the interaction of AI model parameters according to an embodiment of the present invention, applied to a WIFI transmitter, wherein the WIFI transmitter includes a self-encoder, such as... Figure 3 As shown, the process includes the following steps:

[0058] Step S301: After the AI ​​model parameters are loaded into the autoencoder of the WIFI transmitter, a data frame with a special sequence is sent to the WIFI receiver through the autoencoder of the WIFI transmitter, so that the WIFI receiver can identify the data frame according to the special sequence, and if the identification is successful, the data frame is decoded by the autoencoder of the WIFI receiver.

[0059] As an example, when the autoencoder of the WIFI transmitter is fed with AI model parameters, the autoencoder AE of the WIFI transmitter can be used first for frame transmission. During transmission, special fields can be added to the data frames for frame recognition.

[0060] As an example, a WIFI receiver can identify data frames based on a specific sequence, and if the identification is successful, decode the data frames using the WIFI receiver's autoencoder.

[0061] In one alternative embodiment, the data frame with a special sequence carries feedback Channel State Information (CSI) content.

[0062] In one optional embodiment, after sending data frames with a special sequence to the WIFI receiver via the autoencoder of the WIFI transmitter, the method further includes: listening to the feedback message of the WIFI receiver; if no feedback message is received from the WIFI receiver within a preset time, it is determined that the AI ​​model parameters of the autoencoder of the WIFI transmitter need to be updated.

[0063] As an example, after the WIFI transmitter sends a data frame with a special sequence to the WIFI receiver, it enters the listening phase and can listen for feedback messages from the WIFI receiver. If no feedback message is received from the WIFI receiver within a preset time, it is determined that the AI ​​model parameters of the autoencoder of the WIFI transmitter need to be updated.

[0064] In one optional embodiment, when the autoencoder at the WIFI transmitter needs to update the AI ​​model parameters, a trigger frame with a special sequence is sent to the WIFI receiver within an inter-frame interval to trigger the autoencoder at the WIFI receiver to feed back new AI model parameters.

[0065] The definition of inter-frame spacing specified in 802.11 is as follows:

[0066] aSIFSTime=16μs(OFDM System)

[0067] Slottime = 9μs

[0068] aDIFSTime=aSIFSTime+2*Slottime

[0069] aEIFSTime=aSIFSTime+ACKTransmissionTIme+aDIFSTime

[0070] SIFS stands for Short Interframe Space, which is equal to the time required for a node to switch from a transmitting state to a receiving state and to decode correctly, or the time required to switch from a receiving state to a transmitting state.

[0071] OFDM stands for Orthogonal Frequency-Division Multiplexing.

[0072] DIFS stands for Distributed-Coordination-Function Interframe Space. In the DCF protocol, a node needs to check if the channel is idle before it can start sending data. If the channel is idle, the node must wait for the DIFS period before sending data; if the channel is detected as busy at any time during the DIFS period, the node must postpone its data transmission.

[0073] EIFS stands for Extended Interframe Space, which means that if an error occurs in the previous frame, the transmitting node has to delay for the EIFS period instead of the DIFS period before transmitting the next frame.

[0074] As an example, when the autoencoder at the WIFI transmitter needs to update the AI ​​model parameters, it sends a trigger frame with a special sequence to the WIFI receiver within a DIFS inter-frame interval or an EIFS inter-frame interval to trigger the autoencoder at the WIFI receiver to feed back the new AI model parameters.

[0075] It should be noted that the above interaction process can be achieved using traditional WIFI transmitters and traditional WIFI receivers, with the aim of enabling the self-encoder AE and other hardware added to the WIFI transmitter and receiver to function properly.

[0076] In one optional embodiment, new AI model parameters fed back by the autoencoder of the WIFI receiver are received, and an acknowledgment message is sent to the WIFI receiver.

[0077] As an example, the WIFI transmitter can receive new AI model parameters fed back by the autoencoder of the WIFI receiver within a SIFS inter-frame interval, and send an acknowledgment (ACK) message to the WIFI receiver to notify the WIFI receiver whether the reception was successful.

[0078] To further facilitate understanding by those skilled in the art, the following two embodiments illustrate the interaction process during the DIFS frame interval and the interaction process during the EIFS frame interval:

[0079] Example 1:

[0080] Figure 4 This is a schematic diagram of AI parameter interaction during the DIFS frame interval according to an embodiment of the present invention, such as... Figure 4As shown, after completing a transmission or reception, the WIFI device can enter the listening phase. The listening time is the DIFS time shown in the figure. If the AP needs to update the AI ​​model parameters, the WIFI transmitter can send a trigger frame with a special field to trigger the WIFI receiver to send new AI model parameters.

[0081] After receiving the trigger frame, the WIFI receiver can first match the special fields on the trigger frame with the special fields stored locally. If a match is found, it can start constructing the new AI model parameters required by the WIFI transmitter based on the content of the trigger frame.

[0082] After one SIFS time, the WIFI receiver can return the new AI model parameters to the WIFI transmitter via the feedback link and put the new AI model parameters into the neural network of the WIFI transmitter, thus completing the interaction process.

[0083] Example 2:

[0084] Figure 5 This is a schematic diagram of AI parameter interaction during the EIFS frame interval according to an embodiment of the present invention, such as... Figure 5 As shown, during the nth transmission, the WIFI transmitter needs to first send an error frame that fails the verification before entering the listening phase. The listening time is the EIFS time shown in the figure. If the WIFI transmitter needs to update the model parameters, it can send a trigger frame with a special field to trigger the WIFI receiver to send the AI ​​model parameters.

[0085] After receiving the trigger frame, the WIFI receiver can first match the special fields on the trigger frame with the special fields stored locally. If a match is found, it can start constructing the new AI model parameters required by the WIFI transmitter based on the content of the trigger frame.

[0086] When a Wi-Fi receiver sends new AI model parameters back to a Wi-Fi transmitter, the use of EIFS intervals allows for continuous transmission of multiple frames, compared to the transmission duration of DIFS. For example, the AI ​​model parameters can be divided into multiple parts and fed back to the Wi-Fi transmitter multiple times; alternatively, complete AI model parameters can be sent back to the Wi-Fi transmitter multiple times, and the Wi-Fi transmitter can then calculate the average value based on these multiple complete AI model parameters.

[0087] After one SIFS time, the WIFI receiver can return the new AI model parameters to the WIFI transmitter via the feedback link and put the new AI model parameters into the neural network of the WIFI transmitter, thus completing the interaction process.

[0088] After completing the reception, the WIFI transmitter can send an ACK to the WIFI receiver to notify the receiver that the reception was successful.

[0089] This invention provides an autoencoder at both the Wi-Fi transmitter and receiver. After AI model parameters are input into the autoencoder at the Wi-Fi transmitter, a data frame with a special sequence is sent to the Wi-Fi receiver via the autoencoder. The Wi-Fi receiver identifies the data frame based on the special sequence, and if identification is successful, decodes the data frame using the autoencoder at the Wi-Fi receiver. This autoencoder improves the encoding efficiency and decoding accuracy of traditional Wi-Fi transmitters and receivers, thereby enhancing Wi-Fi network performance and user experience. It also solves the problem that related technologies do not utilize AI / ML in specific WLAN implementations, thus failing to achieve the effect of enhancing network performance and user experience through AI / ML.

[0090] This embodiment provides another method for interacting with AI model parameters on the network architecture of the aforementioned WIFI system. Figure 6 This is a flowchart illustrating the interaction of AI model parameters according to another embodiment of the present invention, applied to a WIFI receiver, wherein the WIFI receiver includes an autoencoder and a classifier, such as... Figure 5 As shown, the process includes the following steps:

[0091] Step S601: After the AI ​​model parameters are loaded into the autoencoder of the WIFI transmitter, the classifier receives the data frame with a special sequence sent by the autoencoder of the WIFI transmitter, and identifies the data frame according to the special sequence.

[0092] As an example, after the autoencoder at the WIFI transmitter is fed with AI model parameters, a classifier can receive data frames with special sequences sent by the autoencoder at the WIFI transmitter and identify the data frames based on the special sequences.

[0093] As an example, a data frame can be an AE-encoded vector from a WIFI transmitter, which can be scrambled by a special pseudo noise (PN) sequence. The classifier can use a locally cached PN sequence for detection and identification.

[0094] Step S602: If the identification is successful, the data frame is decoded by the self-encoder of the WIFI receiver.

[0095] As an example, the data frame can be an AE-encoded vector from the WIFI transmitter. If the recognition is successful, the AE-encoded vector from the WIFI transmitter can be sent to the AE decoder at the WIFI receiver. The AE decoder at the WIFI receiver can then perform decoding, which improves the accuracy of the autoencoder.

[0096] In an optional embodiment, after the data frame is decoded by the autoencoder of the WIFI receiver, the method further includes: receiving a trigger frame with a special sequence sent by the WIFI transmitter when the autoencoder of the WIFI transmitter needs to update the AI ​​model parameters; identifying the trigger frame according to the special sequence using the classifier; training the autoencoder of the WIFI receiver to obtain new AI model parameters if the identification is successful; and feeding the new AI model parameters back to the autoencoder of the WIFI transmitter.

[0097] As an example, when the autoencoder on the WIFI transmitter needs to update the AI ​​model parameters, the WIFI transmitter can send a trigger frame with a special sequence to the classifier, which can then identify the trigger frame based on the special sequence on the trigger frame.

[0098] As an example, if recognition is successful, the WIFI receiver can train its autoencoder to obtain new AI model parameters, which can then be fed back to the WIFI transmitter's autoencoder via a feedback link.

[0099] As an example, in this embodiment of the invention, the WIFI receiver undertakes the main computational work, namely the training work, and the steps are as follows:

[0100] 1. Set the data to be sent as s = [s1, s2, ..., s...] n Normalization, the normalization method is as follows:

[0101]

[0102] 2. Normalize the data s using an encoder. norm Mapped to encoding vector x k

[0103] x k =f i (f i-1 (...f1(s norm |θ1)|θ i-i |θ i )

[0104] 3. The encoded vector x is decoded by a decoder. k Estimation mapped to the original data

[0105]

[0106] in, θ represents an estimate of the original data. i This represents the trainable network parameters, i.e., w. i and b i ;

[0107] Basic network unit function f i The general expression for (x) is as follows:

[0108] f i (x)=f a (w i ⊙x+b i )

[0109] Where ⊙ represents the network operator defined by the neural network type, commonly used networks include fully connected networks (DNN, Dense Neural Network, DNN), convolutional neural networks (CNN), and recurrent neural networks (RNN). a (x) represents the activation function, which can include ReLU activation function, Tanh activation function, Sigmoid activation function, etc.

[0110] 4. Train network parameters using a loss function. The loss function characterizes the difference between the deep learning evaluation module and the actual performance. The mean squared error loss function is used and is defined as follows:

[0111]

[0112] Since the basic network units use normalized evaluation parameters, cross-entropy can also be used as the cost function:

[0113]

[0114] The Wi-Fi receiver periodically calls the trainer for training. The Wi-Fi receiver can use new data collected in real time during the current period to train the existing network. The training object is the trainable parameters w. i b i The training method is gradient descent, which can include algorithms such as RMSProp and Adam. After training, the trained network parameters, i.e., w, can be set through a preset interactive mechanism. i and b iFeedback is sent to the WIFI transmitter. The data exchanged during the feedback process consists of trainable parameters of the encoder or decoder in the AE.

[0115] In one optional embodiment, the step of identifying the trigger frame based on the special sequence by the classifier includes: matching the special sequence of the trigger frame with a locally stored special sequence; and determining that the identification is successful if the match is successful.

[0116] As an example, the data field of the trigger frame can be a special sequence. The trigger signal identification can match the special sequence on the trigger frame with the feature sequence stored locally. If the match is successful, the identification can be confirmed as successful, and the AE parameter interaction can continue.

[0117] After the autoencoder at the Wi-Fi transmitter is fed with AI model parameters, a classifier receives data frames with special sequences sent by the autoencoder at the Wi-Fi transmitter. The classifier then identifies the data frames based on these special sequences. If identification is successful, the autoencoder at the Wi-Fi receiver decodes the data frames. This autoencoder improves the encoding efficiency and decoding accuracy of traditional Wi-Fi transmitter and receiver systems, thereby enhancing Wi-Fi network performance and user experience. It addresses the issue that related technologies have not yet incorporated AI / ML into specific WLAN implementations, thus failing to achieve the desired effect of enhancing network performance and user experience through AI / ML.

[0118] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0119] This embodiment also provides an interactive device for AI model parameters, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0120] Figure 7This is a structural block diagram of an interactive device for AI model parameters according to an embodiment of the present invention, applied to a WIFI transmitter. The WIFI transmitter includes a self-encoder for the WIFI transmitter, such as... Figure 7 As shown, the device includes:

[0121] The first transmitting module 701 is used to transmit a data frame with a special sequence to the WIFI receiver through the autoencoder of the WIFI transmitter after the AI ​​model parameters are put into the autoencoder of the WIFI transmitter, so that the WIFI receiver can identify the data frame according to the special sequence, and decode the data frame through the autoencoder of the WIFI receiver if the identification is successful.

[0122] In an optional embodiment, it further includes:

[0123] The monitoring module is used to monitor the feedback messages of the WIFI receiver after the WIFI transmitter sends a data frame with a special sequence to the WIFI receiver through the autoencoder of the WIFI transmitter.

[0124] The determination module is used to determine that if no feedback message is received from the WIFI receiver within a preset time, the AI ​​model parameters of the autoencoder of the WIFI transmitter need to be updated.

[0125] In an optional embodiment, it further includes:

[0126] The second sending module is used to send a trigger frame with a special sequence to the WIFI receiver within an inter-frame interval when the autoencoder at the WIFI transmitting end needs to update the AI ​​model parameters, so as to trigger the autoencoder at the WIFI receiver to feed back new AI model parameters.

[0127] In an optional embodiment, it further includes:

[0128] The third sending module is used to receive new AI model parameters fed back by the autoencoder of the WIFI receiver and send an acknowledgment message to the WIFI receiver.

[0129] In one alternative embodiment, the data frame with a special sequence carries feedback Channel State Information (CSI) content.

[0130] Figure 8 This is a structural block diagram of an interactive device for AI model parameters according to another embodiment of the present invention, applied to a WIFI receiver, wherein the WIFI receiver includes an autoencoder and a classifier, such as... Figure 8 As shown, the device includes:

[0131] The first receiving module 801 is used to receive data frames with a special sequence sent by the encoder of the WIFI transmitter through the classifier after the AI ​​model parameters are put into the autoencoder of the WIFI transmitter, and to identify the data frames according to the special sequence.

[0132] The decoding module 802 is used to decode the data frame through the autoencoder of the WIFI receiver when the identification is successful.

[0133] In an optional embodiment, it further includes:

[0134] The second receiving module is used to receive a trigger frame with a special sequence sent by the WIFI transmitter when the WIFI transmitter needs to update the AI ​​model parameters after the data frame is decoded by the autoencoder of the WIFI receiver.

[0135] The identification module is used to identify the trigger frame based on the special sequence through the classifier;

[0136] The training module is used to train the autoencoder of the WIFI receiver to obtain new AI model parameters when the recognition is successful.

[0137] The feedback module is used to feed back the new AI model parameters to the autoencoder of the WIFI transmitter.

[0138] In one optional embodiment, the identification module includes:

[0139] The matching submodule is used to match the special sequence of the trigger frame with the special sequence stored locally;

[0140] The determination submodule is used to confirm successful recognition if a match is found.

[0141] It should be noted that the above modules can be implemented through software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0142] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program, wherein the computer program is configured to perform the steps in any of the above method embodiments when executed.

[0143] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0144] Embodiments of the present invention also provide an electronic device including a memory and a processor, the memory storing a computer program and the processor being configured to run the computer program to perform the steps in any of the above method embodiments.

[0145] In one exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0146] Specific examples in this embodiment can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0147] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0148] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An interactive method for AI model parameters, characterized in that: Applied to a WIFI transmitter, wherein the WIFI transmitter includes an autoencoder of the WIFI transmitter, and the method includes: After the autoencoder of the WIFI transmitting end is placed into the AI ​​model parameters, a data frame with a special sequence is sent to the WIFI receiving end through the autoencoder of the WIFI transmitting end, so that the WIFI receiving end identifies the data frame according to the special sequence, and if the identification is successful, the data frame is decoded by the autoencoder of the WIFI receiving end.

2. The method according to claim 1, characterized in that After the data frame with the special sequence is sent to the WIFI receiving end through the autoencoder of the WIFI transmitting end, the method further includes: Monitoring the feedback message of the WIFI receiving end; If no feedback message sent by the WIFI receiving end is monitored within the preset time, it is determined that the AI ​​model parameters of the autoencoder of the WIFI sending end need to be updated.

3. The method according to claim 2, characterized in that Also includes: When the autoencoder of the WIFI transmitting end needs to update the AI ​​model parameters, a trigger frame with a special sequence is sent to the WIFI receiving end within an inter-frame interval to trigger the autoencoder of the WIFI receiving end to feed back new AI model parameters.

4. The method according to claim 3, characterized in that Also includes: Receive new AI model parameters fed back by the autoencoder of the WIFI receiving end, and send a confirmation message to the WIFI receiving end.

5. The method according to claim 1, characterized in that The data frame with the special sequence carries the feedback channel state information CSI content.

6. An interactive method for AI model parameters, characterized in that: Applied to a WIFI receiving end, wherein the WIFI receiving end includes an autoencoder and a classifier of the WIFI receiving end, and the method includes: After the autoencoder of the WIFI transmitter is placed into the AI ​​model parameters, the data frame with the special sequence sent by the autoencoder of the WIFI transmitter is received by the classifier, and the data frame is identified according to the special sequence; When the recognition is successful, the data frame is decoded by the autoencoder of the WIFI receiving end.

7. The method according to claim 6, characterized in that After the data frame is decoded by the autoencoder of the WIFI receiving end, the method further includes: When the autoencoder of the WIFI transmitter needs to update the AI ​​model parameters, receiving a trigger frame with a special sequence sent by the WIFI transmitter; identifying the trigger frame according to the special sequence by the classifier; If the recognition is successful, the autoencoder of the WIFI receiving end is trained to obtain new AI model parameters; The new AI model parameters are fed back to the autoencoder of the WIFI sending end.

8. The method according to claim 7, characterized in that The identifying the trigger frame according to the special sequence by the classifier includes: Matching the special sequence of the trigger frame with a special sequence stored locally; In case of successful matching, it is determined that the recognition is successful.

9. An interactive device for AI model parameters, characterized in that: Applied to a WIFI transmitter, the WIFI transmitter includes an autoencoder of the WIFI transmitter, and the device includes: The first sending module is used to send a data frame with a special sequence to a WIFI receiving end through the autoencoder of the WIFI sending end after the AI ​​model parameters are placed in the autoencoder of the WIFI sending end, so that the WIFI receiving end identifies the data frame according to the special sequence, and decodes the data frame through the autoencoder of the WIFI receiving end if the identification is successful.

10. An interactive device for AI model parameters, characterized in that: Applied to a WIFI receiving end, the WIFI receiving end includes a self-decoder and a classifier of the WIFI receiving end, and the device includes: A first receiving module, configured to receive, through the classifier, a data frame with a special sequence sent by the encoder of the WIFI transmitting end after the autoencoder of the WIFI transmitting end is placed into the AI ​​model parameters, and identify the data frame according to the special sequence; The decoding module is used to decode the data frame through the autoencoder of the WIFI receiving end when the recognition is successful.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 5 are implemented, or the steps of the method described in any one of claims 6 to 8 are implemented.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the processor implements the steps of the method described in any one of claims 1 to 5, or implements the steps of the method described in any one of claims 6 to 8.

Images