WO2023202499A1

WO2023202499A1 - Data transmission method and related device

Info

Publication number: WO2023202499A1
Application number: PCT/CN2023/088550
Authority: WO
Inventors: 杜雨晴; 马梦瑶; 杨讯
Original assignee: 华为技术有限公司
Priority date: 2022-04-22
Filing date: 2023-04-15
Publication date: 2023-10-26
Also published as: CN116980972A

Abstract

The present application relates to a remote identification or classification scene, and in particular, to a data transmission method and a related device. In the method, a receiver determines a distance between each piece of data in a database and first data, selects second data with the shortest distance from the first data from the database, and determines whether to send an identification result represented by the second data according to the number of the second data and an absolute difference between a first distance and a second distance, wherein the first distance is the shortest distance in the distance between each piece of data in the database and the first data, and the second distance is the shortest distance except the first distance. The data transmission method is based on a feedback principle of classification accuracy, and can improve the accuracy of the identification result. The application field of the present application includes but is not limited to systems supporting next generation WIFI protocol of 802.11ax, e.g. 802.11be, Wi-Fi 7 or EHT, or next generation of 802.11be, i.e., Wi-Fi 8, etc.

Description

Data transmission method and related device

This application claims priority to the Chinese patent application filed with the China Patent Office on April 22, 2022, with the application number 202210427681.4 and the application title "Data transmission method and related devices", the entire content of which is incorporated into this application by reference.

Technical field

The present application relates to the field of communication technology, and in particular, to a data transmission method and related devices.

Background technique

In the scenario of remote identification, the data collected by the second device often needs to be transmitted to the first device, and the first device uses the data for classification and identification. The data transmitted by the second device to the first device is high-dimensional information extracted after performing principal component analysis on the collected data. The high-dimensional information may be a high-dimensional matrix. The number of rows of the high-dimensional matrix may range from hundreds of thousands or even millions, and the transmission of the high-dimensional matrix has considerable communication overhead. In order to reduce communication overhead, a low-precision quantizer can be used to quantize and transmit the high-dimensional matrix. However, the errors before and after quantization will affect the accuracy of the recognition results.

Therefore, how to improve the accuracy of recognition results is an urgent problem to be solved.

Contents of the invention

This application provides a data transmission method and related devices, which can improve the accuracy of identification results.

The present application is introduced from different aspects below, and it should be understood that the implementation modes and beneficial effects of different aspects can be referred to each other.

In a first aspect, the present application provides a data transmission method, which can be executed by a first communication device. The first communication device can be a communication device or a communication device, such as a chip, that can support the communication device to implement the functions required by the method. For example, the first communication device is a receiver, or a chip provided in the cloud for implementing the functions of the receiver, or other components for implementing the functions of the receiver. In the following introduction process, the receiver is used as the execution subject as an example.

In this method, the receiver determines the distance between each data in the database and the first data, which is the received data to be identified or a matrix composed of received rows or columns in the data to be identified. Among them, the shortest distance between each data and the first data is the first distance, and the shortest distance other than the first distance is the second distance; the distance between the judgment database and the first data is the first distance. Whether the number of the second data of the distance is one or more; if the number of the second data is one and the absolute difference between the first distance and the second distance is greater than the preset value, send the recognition result represented by the second data .

It can be seen that this application uses the shortest distance principle to not only select the second data with the shortest distance from the first data from the database, but also determine based on the number of second data and the absolute difference between the first distance and the second distance. The accuracy of the identification result represented by the second data is used to determine whether to send the identification result represented by the second data. This data transmission method can also be called the feedback principle based on classification accuracy, which can improve the accuracy of the recognition results.

In an optional implementation, before the receiver determines the distance between each data in the database and the first data, the method further includes: the receiver sends a transmission instruction, the transmission instruction is used to indicate the currently requested transmission to be identified. The column index of the column or the row index of the row in the data; the receiver receives the element of the column corresponding to the column index or the element of the row corresponding to the row index; and determines the first data using the received column elements or row elements. It can be seen that in this embodiment, the receiver can request the sender to send elements of partial columns or partial rows through transmission instructions for matching identification, thereby reducing the data that needs to be transmitted as much as possible, which is beneficial to reducing communication overhead and reducing communication time. extension.

In an optional implementation, if the receiver determines that the number of second data in the database is multiple, or the absolute difference between the first distance and the second distance is not greater than a preset value, the receiver can Perform the step of sending a transmission instruction again until the number of second data is one and the absolute difference between the first distance and the second distance is greater than the preset value, or until all columns or all rows in the data to be identified All have been transmitted, and the recognition result represented by the second data is sent. It can be seen that in this embodiment, when making a matching judgment based on the rows or columns that have been transmitted and the data in the database, if it is judged that the matching conditions are not met, the transmission instruction can be used again to request the transmitter to send elements of other rows or columns, using the updated The first data is matched again. Therefore, it is beneficial to reduce the data that needs to be transmitted as much as possible, reduce the communication overhead, and reduce the communication delay.

In an optional implementation, the transmission instruction is also used to indicate the transmission parameters used for compressed transmission of the column or row currently requested to be transmitted; the transmission parameters include the length of each block into which the column or row is divided, and the length of each block into which the column or row is divided. The codebook information used for quantization; the element of the column corresponding to the received column index is obtained by decompressing the element of the column corresponding to the column index of the compressed transmission based on the transmission parameters, or the element of the row corresponding to the received row index is It is obtained by decompressing the elements of the row corresponding to the row index of the compressed transmission based on the transmission parameters. Therefore, it is beneficial to reduce the data that needs to be transmitted, thereby reducing communication overhead.

In another optional implementation, before the receiver determines the distance between each data in the database and the first data, the method further includes: the receiver sends a transmission instruction, the transmission instruction is used to request transmission of the data to be identified. Data; receiving the data to be identified as the first data.

Optionally, the transmission instruction is also used to indicate the transmission parameters used for compressed transmission of the data to be identified that are requested to be transmitted; the transmission parameters include the length of each block divided into each column or row of the data to be identified, and the Each block is quantized with codebook information used; the received data to be identified is obtained by decompressing the compressed and transmitted data to be identified based on the transmission parameters. Therefore, it is beneficial to reduce the data that needs to be transmitted, thereby reducing communication overhead.

In this implementation, if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than the preset value, the receiver can send a retransmission instruction, and the retransmission instruction is used to Indicate the row index or column index of the row currently requested to be retransmitted in the data to be identified; receive the element of the row corresponding to the row index or the element of the column corresponding to the column index; update the first data using the elements of the row or the elements of the column, And perform the step of determining the distance between each data in the database and the first data again until the number of second data is one and the absolute difference between the first distance and the second distance is greater than the preset value , or all columns or all rows in the data to be identified have been retransmitted, and the identification results represented by the second data are sent. It can be seen that in this embodiment, when making a matching judgment based on the transmitted data to be identified, if it is judged that the matching conditions are not met, a retransmission instruction can be sent to request the transmitter to retransmit elements of some rows or columns, using the updated first The data is matched again. That is to say, this implementation method can solve the problem of large communication data volume in the existing method and reduce the communication delay to a great extent through partial row/column compression and retransmission based on the classification accuracy feedback criterion.

Optionally, the retransmission instruction is also used to indicate the transmission parameters used for compressed retransmission of the row or column currently requested to be retransmitted; the transmission parameters include the length of each block into which the column or row is divided, and the quantization of each block. The codebook information used; the element of the column corresponding to the received column index is obtained by decompressing the element of the column corresponding to the column index of the compressed transmission based on the transmission parameter, or the element of the row corresponding to the received row index is based on The transmission parameters are obtained by decompressing the elements of the row corresponding to the row index of the compressed retransmission. Therefore, it is beneficial to reduce the amount of data transmitted by retransmission.

Optionally, the quantization precision used for compressed retransmission of the row or column currently requested by the receiver to be retransmitted is higher than the quantization precision used for compressed transmission of the data to be identified that the receiver requests transmission. Therefore, the communication overhead and communication delay caused by using a high-precision quantizer for the first transmission of the data to be identified is avoided.

In another optional implementation, when the receiver uses the received data to be identified to perform matching judgment, if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not is greater than the preset value, send a plurality of fourth data, the plurality of fourth data are a plurality of second data in the database, or the second data in the database and the distance between the database and the first data is the second The third data of distance. It can be seen that this implementation method improves the recognition accuracy on the premise of By feeding back part of the data in the database in the downlink, the problem of large uplink transmission volume in the identification scenario can be solved. Compared with the current retransmission mechanism, the communication delay is also reduced.

In a second aspect, the present application provides a data transmission method, which can be executed by a second communication device. The second communication device can be a communication device or a communication device, such as a chip, that can support the communication device to implement the functions required by the method. For example, the second communication device is a transmitter, or a chip provided in the sensor to implement the function of the transmitter, or other components used to implement the function of the transmitter. In the following introduction process, the sender is used as the execution subject as an example.

In this method, the transmitter sends a task request, which is used to request an identification task. The task request includes the number N of rows and the number M of columns of data to be identified in the identification task; the transmitter receives a transmission instruction, which is used to Indicates the column index of the column or the row index of the row currently requested to be transmitted in the data to be identified; the sender sends the element of the column corresponding to the column index or the element of the row corresponding to the row index.

It can be seen that after the task request is sent to the receiver in the embodiment of the present application, it can be transmitted in rows or columns according to the transmission instructions sent by the receiver for identification by the receiver, thereby helping to reduce communication overhead.

In an optional implementation, the transmitter receives the recognition result represented by the second data. The second data is data whose distance from the first data in the database of the device responding to the recognition task is a first distance. A distance is the shortest distance between each data in the database and the first data, which is composed of elements of rows or columns that have been sent in the data to be identified. It can be seen that compared with the method of uploading all the data to be identified using a high-precision quantizer, this data transmission method can reduce communication overhead while ensuring the accuracy of the identification results.

In an optional implementation, the transmission instruction is also used to indicate the transmission parameters used for compressed transmission of the column or row currently requested to be transmitted; the transmission parameters include the length of each block into which the column or row is divided, and the length of each block into which the column or row is divided. The codebook information used by the block to be quantized; the elements of the column corresponding to the sent column index are obtained by compressing the elements of the column corresponding to the column index based on the transmission parameters, or the elements of the row corresponding to the sent row index are obtained based on the transmission The parameter is obtained by compressing the elements of the row corresponding to the row index. It can be seen that this implementation method is beneficial to further reducing communication overhead.

In a third aspect, the present application also provides a data transmission method, which can be executed by a second communication device. The second communication device can be a communication device or a communication device, such as a chip, that can support the communication device to implement the functions required by the method. For example, the second communication device is a transmitter, or a chip provided in the sensor to implement the function of the transmitter, or other components used to implement the function of the transmitter. In the following introduction process, the sender is used as the execution subject as an example.

In this method, the transmitter receives a transmission instruction, which is used to request the transmission of data to be identified in the identification task; sends the data to be identified; in addition, the transmitter can also receive a retransmission instruction, which is used to indicate that the data to be identified is to be transmitted. The row index of the row or the column index of the column in the data that is currently requested to be retransmitted. Then, the sender can retransmit the element of the column corresponding to the column index or the element of the row corresponding to the row index.

It can be seen that in this method, after the transmitter sends the data to be identified, it can also retransmit the elements of some rows or columns based on the retransmission instruction to update the data to be identified, which is conducive to using a compression transmission scheme with lower quantization accuracy for transmission. The data is to be identified, which in turn helps reduce communication overhead.

In an optional implementation, the method further includes: receiving a recognition result represented by second data, where the distance between the second data and the first data in the database in the device responding to the recognition task is a first distance. data, the first distance is the shortest distance between each data in the database and the first data. The first data is updated data based on the elements of the row or column that have been retransmitted. It can be seen that in this embodiment, the transmitter can enable the receiver to determine the recognition result based on the updated data to be recognized, thereby reducing communication overhead while ensuring the accuracy of the recognition result.

In an optional implementation, the transmission instruction is also used to indicate the transmission parameters to be used for compressed transmission of the data to be identified that is requested to be transmitted; the transmission parameters include the data of each block divided into each column or row in the data to be identified. length, and the codebook information used to quantize each block; accordingly, the data to be identified sent above is obtained by compressing the data to be identified based on the transmission parameters.

In an optional implementation, the retransmission instruction is also used to indicate the transmission parameters used for compressed retransmission of the row or column currently requested to be retransmitted; the transmission parameters indicated by the retransmission instruction include each of the columns or rows divided into The length of the block, and the codebook information used to quantize each block; accordingly, the elements of the column corresponding to the column index sent above are obtained by compressing the elements of the column corresponding to the column index based on the transmission parameters, or, The elements of the row corresponding to the sent row index are obtained by compressing the elements of the row corresponding to the row index based on the transmission parameters indicated by the retransmission instruction.

In an optional implementation, the quantization precision used for compressed retransmission of the row or column currently requested to be retransmitted is higher than the quantization precision used for compressed transmission of the data to be identified that is requested to be transmitted. In this way, communication overhead can be reduced as much as possible while ensuring the accuracy of the recognition results.

In a fourth aspect, the present application provides a data transmission method, which can be executed by a second communication device. The second communication device can be a communication device or a communication device, such as a chip, that can support the communication device to implement the functions required by the method. For example, the second communication device is a transmitter, or a chip provided in the sensor to implement the function of the transmitter, or other components used to implement the function of the transmitter. In the following introduction process, the sender is used as the execution subject as an example.

In this method, the transmitter receives a transmission instruction, which is used to request the transmission of the data to be identified in the identification task; sends the data to be identified; receives a plurality of fourth data; and determines the difference between each received fourth data and the data to be identified. distance between them; determine whether the number of fifth data among the plurality of fourth data that is the third distance from the data to be identified is one or more; if the number of fifth data is one and the third The absolute difference between the distance and the fourth distance is greater than the preset value, and the recognition result represented by the fifth data is output.

Wherein, the plurality of fourth data are the second data in the database, or the second data in the database and the third data in the database. The second data is data in the database of the device that identifies the task, and the distance between it and the first data is the first distance, and the third data is the data in the database of the device that identifies the task, and the distance between it and the first data is the second distance. distance data. Wherein, the shortest distance between each data in the database and the first data is the first distance, and the shortest distance other than the first distance is the second distance. The first data is data to be identified received by the device; wherein the shortest distance between each fourth data in the plurality of fourth data and the data to be identified is the third distance, and the shortest distance other than the third distance is The distance is the fourth distance.

It can be seen that in this method, after the transmitter sends the data to be identified, if the identification result cannot be obtained, part of the data in the database can be received, and the transmitter performs the above matching judgment to obtain the identification result. This avoids sending the data to be identified again and greatly reduces the amount of data for uplink transmission.

In an optional implementation, if the number of fifth data is multiple, or the absolute difference between the third distance and the fourth distance is not greater than a preset value, prompt information is output. Used to indicate that the data to be identified failed to match.

In a fifth aspect, the present application provides a communication device. The communication device may be a receiver or a chip in the receiver. The communication device includes a processor and a transceiver. The processor is used to determine the distance between each data in the database and the first data distance, the first data is the received data to be identified or a matrix composed of received rows or columns in the data to be identified; the shortest distance between each data and the first data is the first distance, except for the first The shortest distance outside the distance is the second distance; and it is judged whether the number of second data in the database that is the first distance from the first data is one or more; the transceiver is used to determine if the first data is the second data. If the number of two pieces of data is one, and the absolute difference between the first distance and the second distance is greater than a preset value, the identification result represented by the second data is sent.

Optionally, the communication device further includes a memory, the memory is used to store a computer program, and the computer program includes program instructions.

In a sixth aspect, the present application provides a communication device. The communication device may be a transmitter or a chip in the transmitter. The communication device includes a transceiver.

In an optional implementation, the transceiver is used to send a task request. The task request is used to request an identification task. The task request includes the number N of rows and the number M of columns of data to be identified in the identification task; receiving the transmission instruction, The transfer instruction is used to indicate the column index of the column or the row index of the row currently requested to be transferred in the data to be identified; send the element of the column corresponding to the column index or the element of the row corresponding to the row index.

In another optional implementation, the transceiver is used to receive a transmission instruction, which is used to request the transmission of data to be identified in the identification task; to send the data to be identified; and to receive a retransmission instruction, which is used to indicate The row index of the row or column index of the column currently requested to be retransmitted in the data to be identified; the element of the column corresponding to the retransmitted column index or the element of the row corresponding to the row index.

In yet another optional implementation, the communication device further includes a processor, the receiver is used to receive a transmission instruction, the transmission instruction is used to request the transmission of data to be identified in the identification task; to send the data to be identified; and to receive a plurality of third Four data, the plurality of fourth data is the second data in the database, or the second data and the third data in the database; the second data is the same as the first data in the database of the device responding to the recognition task. The distance between the data is the data of the first distance, and the third data is the data in the database whose distance from the first data is the second distance. Wherein, the shortest distance between each data in the database and the first data is the first distance, and the shortest distance other than the first distance is the second distance. The first data is data received by the device to be identified. The processor is used to determine the distance between each received fourth data and the data to be identified; the shortest distance between each fourth data in the plurality of fourth data and the data to be identified is the third distance. , the shortest distance other than the third distance is the fourth distance; and determining whether the number of fifth data among the plurality of fourth data that is the third distance from the data to be identified is one or more; The receiver is configured to output a recognition result represented by the fifth data if the number of the fifth data is one and the absolute difference between the third distance and the fourth distance is greater than a preset value.

In the seventh aspect, embodiments of the present application provide a communication device, which is implemented in the form of a chip product and includes a processor and an interface circuit; the interface circuit is used to receive code instructions and transmit them to the processor; the processor uses The code instructions are executed to execute the data transmission method described in the first aspect or any possible implementation of the first aspect. Optionally, the communication device further includes a memory, and the memory is connected to the processor through a circuit.

Optionally, the processor and the memory may be physically independent units, or the memory may be integrated with the processor.

In an eighth aspect, embodiments of the present application provide a communication device, which is implemented in the form of a chip product and includes a processor and an interface circuit; the interface circuit is used to receive code instructions and transmit them to the processor; the processor uses Running the code instructions to execute the above second aspect or any possible implementation of the above second aspect, or the above third aspect or any possible implementation of the above third aspect, or the fourth aspect or the third aspect Any possible realization of the four aspects The data transmission method described in the method. Optionally, the communication device further includes a memory, and the memory is connected to the processor through a circuit.

In a ninth aspect, embodiments of the present application provide a computer-readable storage medium. A computer program is stored in the computer-readable storage medium. The computer program includes program instructions. When the program instructions are run on a computer, they cause the computer to execute the above. The data transmission method described in any one of the first to fourth aspects.

In a tenth aspect, embodiments of the present application provide a computer program product that, when run on a computer, causes the computer to execute the data transmission method described in any one of the above-mentioned first to fourth aspects.

In an eleventh aspect, embodiments of the present application provide a communication system, which includes the receiver described in the above first aspect or any possible implementation manner of the above first aspect, and the above second to third aspects. The transmitter described in any one of the four aspects or any possible implementation manner thereof; or, the communication system includes the communication device described in the fifth aspect above, and any possible implementation manner described in the sixth aspect above. Described communication device.

Description of the drawings

Figure 1 is a schematic architectural diagram of a data transmission system provided by an embodiment of the present application;

Figure 2 is an architectural schematic diagram of another data transmission system provided by an embodiment of the present application;

Figure 3 is a schematic diagram of a "cutting" operation provided by an embodiment of the present application;

Figure 4 is a schematic flowchart of the data transmission method 100 provided by the embodiment of the present application;

Figure 5 is a schematic flowchart of the data transmission method 200 provided by the embodiment of the present application;

Figure 6 is a schematic flowchart of the data transmission method 300 provided by the embodiment of the present application;

Figure 7 is a schematic flowchart of the data transmission method 400 provided by the embodiment of the present application;

Figure 8 is a schematic structural diagram of a communication device provided by an embodiment of the present application;

Figure 9 is a schematic structural diagram of another communication device provided by an embodiment of the present application.

Detailed ways

In order for the receiver to use the data received from the transmitter to better reconstruct the data to be identified. When the transmitter compresses and transmits the data to be identified, it needs to make the minimum mean square error of the signals before and after quantization as small as possible to compress the original data to be identified. However, in order to achieve the purpose of reconstruction, a high-precision quantizer is needed, which will result in a large amount of compressed data. In addition, based on the feedback retransmission mechanism, after all the compressed data to be identified is uploaded to the cloud, if an error at the bit or symbol level is detected, the transmitter will be required to retransmit the data. This results in huge communication overhead.

Therefore, in the face of the demand for accuracy of recognition results and the need to reduce communication overhead, how to handle the data transmission method is an issue that needs to be solved urgently. This application provides a data transmission method that can reduce communication overhead as much as possible while ensuring the accuracy of recognition results. Figures 1 and 2 are respectively schematic architectural diagrams of a data transmission system provided by embodiments of the present application. It should be understood that the system architecture described in this application is to more clearly illustrate the technical solutions of the embodiments of this application, and does not constitute a limitation on the technical solutions provided by the embodiments of this application.

The technical solution provided by this application can be applied in remote identification or classification data transmission scenarios. As shown in Figure 1, the data transmission system includes a transmitter and a receiver. Taking the data transmission scenario of remote identification as an example, the transmitter can send part or all of the data to be identified to the receiver. After the receiver receives all or part of the data to be identified, it can use the data in the database to judge and match. If the recognition result is obtained, the recognition result can be sent to the transmitter. Among them, data can be transmitted between the transmitter and the receiver through a wireless link.

As shown in Figure 2, for the transmitter, the data to be identified is transmitted through compression, which involves mapping, cutting, and blocking. Encoding and entropy coding operations; for the receiver, the data to be identified is obtained through decompression, which involves entropy decoding, block decoding, and inverse mapping operations. Further, the receiver can perform relevant operations based on the classification accuracy feedback principle through judgment operations, and feed back the recognition results to the transmitter when the recognition result feedback conditions are met. When the recognition result feedback conditions are not met, the receiver can send the recognition results to the transmitter. Transfer instructions, or retransmit instructions, or second data in the database. For further explanation of the data transmission method, please refer to the data transmission method 100 to the data transmission method 400 described below.

In the embodiment of the present application, the transmitter may be a wireless communication chip, a wireless sensor or a wireless communication terminal that has the ability to collect or extract data but does not have the function of data identification or classification. For example, support narrowband communication functions (including but not limited to 802.11ax next-generation wireless fidelity (Wi-Fi) protocol, such as 802.11be, Wi-Fi 7 or extremely high throughput (EHT), and then Such as 802.11be next generation, Wi-Fi 8 and other systems, Bluetooth, Zigbee) and/or user terminals and user devices that support cellular communications (including but not limited to 2G, 3G, 4G, 5G and future new generation communications), access Device, subscriber station, subscriber unit, mobile station, user agent, user equipment, wherein the user terminal can include various handheld devices, vehicle-mounted devices, wearable devices, and Internet of Things (IoT) devices with wireless communication functions , computing equipment or other processing equipment connected to a wireless modem, and various forms of user equipment (UE), mobile station (MS), terminal, terminal equipment (terminal equipment), portable communications device, handset, portable computing device, entertainment device, gaming device or system, global positioning system device or any other suitable device configured for network communications via a wireless medium, etc.

In this embodiment of the present application, the receiver may be a cloud with data identification or classification capabilities, a server in the cloud, or other devices that are remote from the sender. The database in the receiver may store data for matching with the data to be identified, so as to obtain the identification result represented by the data to be identified. The receiver may include but is not limited to: evolved Node B (eNB), radio network controller (RNC), Node B (Node B, NB), base station controller (BSC) ), base transceiver station (BTS), home base station (e.g., home evolved NodeB, or homeNode B, HNB), baseband unit (BBU), wireless fidelity (WiFi) system Access point (AP), wireless relay node, wireless backhaul node, transmission point (TP) or transmission and reception point (TRP), etc., can also be 5G, such as NR, A gNB in the system, or a transmission point (TRP or TP), one or a group (including multiple antenna panels) of antenna panels of a base station in a 5G system, or it can also be a network node that constitutes a gNB or transmission point, such as Baseband unit (BBU), or distributed unit (DU), etc., or base stations in future communication systems, etc.

The following is a brief introduction to some relevant content, terms or nouns involved in this application.

1. Data to be identified

The data to be identified can also be called the data to be classified, which is the data transmitted from the transmitter to the receiver for matching judgment. The data described in this article are data that have been stored in the database of the receiver and are used to make matching judgments in conjunction with the data to be identified.

In an optional implementation, the data to be identified can be one or more of the following data collected or extracted by various sensors used to collect biometric characteristics, such as fingerprint image sensors, face image sensors, iris image sensors, etc. : Such as face, fingerprint, iris and other biometric features; or, it can be various sensors used to collect environmental features, such as wireless ranging sensors, radars, cameras, etc., to collect or extract information about the surrounding environment and vehicles; or, It can be gestures, actions and other motor behavior characteristics.

In another optional implementation, the data to be identified is data in which principal components are extracted from the original data using principal component analysis (PCA). The raw data can be data collected by the various sensors mentioned above.

Optionally, the data to be identified can also be called information to be classified or source signals. Correspondingly, the data in the database of the receiver can be called signals in the library.

In this application, the first data may be the data to be identified received by the receiver, data that has been compressed and suffered compression loss on the basis of the data to be identified on the transmitter side, or it may be the part of the data to be identified that is received by the receiver. Row or column data, optionally, there is a compression loss in the data to be identified received by the receiver compared with the data in the row or column of the data to be identified on the transmitter side. The first data may be the current data to be identified by the receiver, such as the data to be identified received after the receiver sends a transmission instruction, or the data to be identified updated using the received row or column elements after the receiver sends a retransmission instruction, Or, the elements of some rows or columns in the data to be identified are received after the receiver sends the transmission instruction.

2. Compression

The transmitter sends the data to be identified to the receiver. In order to save transmission overhead, the data to be identified needs to be compressed and transmitted. After the receiver receives the compressed and transmitted data to be identified, it decompresses and reconstructs the data to be identified. Among them, as shown in Figure 2, for the transmitter, compression involves the operations of mapping, cutting, block coding, and entropy coding; for the receiver, decompression involves the operations of entropy decoding, block decoding, and inverse mapping. The steps involved in compressed transmission are described below:

(1)Mapping

The main function of mapping is dimensionality reduction, which can also be called spatial mapping. The data to be identified is a matrix with a high number of rows and/or columns, referred to as a high-dimensional matrix. Therefore, the high-dimensional matrix needs to be dimensionally reduced. Alternatively, a dimensionality reduction method is compressed sensing, that is, reducing the dimensionality of a high-dimensional matrix by compressing the matrix F. For example, assuming a high-dimensional matrix X, then the dimensionally reduced matrix X′=FX.

Optionally, the dimensionally reduced matrix X′ has the same number of rows as the high-dimensional matrix X, but the number of columns is reduced; or, the dimensionally reduced matrix Compared with the high-dimensional matrix X, the dimensionally reduced matrix .

(2) Cutting

Cutting refers to cutting each column of the matrix into small segments, called "chunks". As shown in Figure 3, the M-column matrix assumes that each column has 50 rows, that is, each column is a 50×1 column vector. The column vector can be cut into 5 "blocks", and each "block" is 10× subvector of 1. In this way, if the dimensionality of the reduced matrix X′ is still relatively large, that is, the quantization space is still relatively large, if the matrix There are also high issues. Therefore, through the "cutting" operation, each column vector can be cut into multiple sub-vectors, and the complexity of designing a codebook and performing quantized encoding for each sub-vector will be reduced.

where the length of each "chunk" is the number of rows of that subvector.

(3) Block coding

For each "block", a vector codebook (quantizer) with specified precision can be used to quantize the "block" according to the length of the "block" to obtain a codeword number sequence. Specifically, based on the distance between the "block" and each codeword in the vector codebook, the codeword corresponding to the shortest distance is found. Optionally, one is to use the codeword corresponding to the shortest distance as the quantized signal of the "block", and the other is to use the number of the codeword in the vector codebook as the quantized signal of the "block".

(4)Entropy coding

Entropy coding can compress the number of the codeword corresponding to each "block" in the vector codebook. For example, the entropy coding method can be: Huffman, arithmetic coding, etc., which is not limited in this application.

(5)Entropy decoding

The receiver can use the decoder to perform entropy decoding on the received data to obtain the number of the codeword corresponding to each "block" in the vector codebook.

(6) Block decoding

For each "block", find the codeword corresponding to the "block" in the corresponding vector codebook according to the number, and then estimate the dimensionally reduced matrix X′.

(7)Inverse mapping

Inverse mapping is used to estimate the received block after decoding Perform estimation recovery to obtain a high-dimensional matrix. For example, using the compression matrix F for estimation recovery is as follows:

i.e. estimate Left multiply the inverse matrix of the compression matrix F to obtain the estimated high-dimensional matrix

Since there is a loss in data compression, the transmitter compresses the high-dimensional matrix X and transmits it to the receiver. The high-dimensional matrix obtained by the receiver through the above decompression operation There are differences from the high-dimensional matrix X.

For ease of explanation, the data to be identified described in this application is, for the transmitter, the above-mentioned original data to be identified, for example, the high-dimensional matrix Identify data, e.g., high-dimensional matrices

3. Distance

In this application, distance refers to the distance between different matrices. The distance between two matrices is the sum of the angles between the hyperplanes represented by the two matrices. For example, the angle between matrix U and matrix S can be calculated as follows:

Perform singular value decomposition on U ^T S, recorded as [U, Σ, V] = svd (U ^T S), where the diagonal elements of the diagonal matrix Σ are between the hyperplanes represented by the two matrices. The cosine of the included angle. For example, consider the diagonal matrix Σ:

Then, the distance between matrix U and matrix S:
d=arcos(0.8611)+arcos(0.6389)+arcos(0.3163)+arcos(0.2128)=4.0168.

Correspondingly, in this application, "the shortest distance" refers to calculating the distance between each data separately based on the above distance calculation method, where the distance with the minimum value is the shortest distance or the shortest distance.

This application provides a data transmission method 100, which can use the shortest distance principle to select the second data with the shortest distance from the first data from the database, and according to the number of second data and the distance between the first distance and the second distance The absolute difference determines the accuracy of the recognition result of the second data representation, and determines whether to send the recognition result of the second data representation. The data transmission method 100 can also be called a feedback mechanism based on classification accuracy, thereby reducing communication overhead as much as possible while ensuring recognition accuracy.

This application also provides a data transmission method 200, which is an incremental transmission mechanism based on classification accuracy. The receiver makes a matching judgment based on the rows or columns that have been transmitted and the data in the database. If the judgment does not meet the recognition result feedback conditions, it can be processed again. Through the transmission instruction, the transmitter is requested to send elements of other rows or columns, and the updated first data is used to perform matching judgment again. Therefore, on the premise of ensuring recognition accuracy, the data to be transmitted can be reduced as much as possible and the communication overhead can be reduced.

This application also provides a data transmission method 300, which is a row-by-row or column-by-column retransmission mechanism based on classification accuracy. When the receiver makes a matching judgment based on the transmitted data to be identified, if the judgment does not meet the recognition result feedback conditions, it can Send a retransmission command to request the sender to retransmit elements of some rows or columns, and use the updated first data to make matching judgments again. Therefore, on the premise of ensuring recognition accuracy, the data to be transmitted can be reduced as much as possible and the communication overhead can be reduced.

This application also provides a data transmission method 400, which is a library signal feedback mechanism based on the principle of classification accuracy feedback. When the receiver makes a matching judgment based on the transmitted data to be identified, if it is judged that the identification result feedback conditions are not met, the database can be Part of the data is sent to the transmitter. It can be seen that on the premise of improving the recognition accuracy, this implementation method can reduce the amount of data transmitted in the uplink and reduce the communication delay by feeding back part of the data in the database downlink.

The technical solutions provided by this application will be described in detail below with reference to more drawings.

Referring to Figure 4, Figure 4 is a schematic flow chart of the data transmission method 100 provided by the embodiment of the present application. As shown in Figure 4, the data transmission method 100 includes but is not limited to the following steps:

S101. The receiver determines the distance between each data in the database and the first data, which is the received data to be identified or a matrix composed of received rows or columns in the data to be identified;

Wherein, the shortest distance between each data and the first data is the first distance, and the shortest distance other than the first distance is the second distance. For example, assume that the database includes data 1, data 2, and data 3. The distances between the three data and the first data are d1, d2, and d3 respectively; among them, d1 is the shortest distance among the three distances, and d2 is the shortest distance among the three distances. The shortest distance among the three distances except the shortest distance d1 can also be called the second shortest distance among the three distances. In this way, the first distance is d1 and the second distance is d2.

It should be noted that this application mentions that each data, the first data, the second data, the third data to be described below, etc. are all data in matrix form, so the distance between the data can be based on the distance described above. The calculation method is calculated and will not be described in detail here.

Optionally, the first data may be part or all elements of the data to be identified. The receiver may obtain the first data by: receiving it from a remote device, or by continuously transmitting it in rows or columns as described in the data transmission method 200, or by transmitting it for the first time as described in the data transmission method 300. Updates are obtained by retransmitting rows or columns after transmission.

S102. The receiver determines whether the number of second data in the database that is a first distance away from the first data is one or more;

S103. If the number of second data is one and the absolute difference between the first distance and the second distance is greater than the preset value, the receiver sends the recognition result represented by the second data.

Optionally, the preset value may be set in advance, or may be an empirical value that meets certain accuracy requirements through offline verification.

Optionally, the recognition result of the second data representation may be user information or permission granting information corresponding to the data to be recognized, which is not limited in this application.

In an optional implementation, if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than the preset value, the receiver can output prompt information. The prompt information Used to indicate that the data to be identified failed to match.

In another optional implementation, if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than the preset value, the receiver can perform the data transmission method 200 as shown in The sending of the transmission instruction instructs the transmitter to continue transmission in rows or columns, or it can also be the sending of a retransmission instruction as described in the data transmission method 300 to retransmit in rows or columns to update the first data, etc. The operation may also be an operation such as sending part of the data in the database to the sender as described in the data transmission method 400. Various possible operations mentioned in this implementation mode will be described one by one in the corresponding embodiments, and will not be described in detail here.

The data transmission method 100 performs feedback transmission based on the principle of classification accuracy. That is, after using the first data to make a matching judgment with the data in the database, the second data that meets the first distance is classified as a certain type of "confidence" signal, and then, when the number of second data is one and only one And when the absolute difference between the first distance and the second distance is greater than the preset value, the signal accuracy of this type of "confidence level" is determined to be high, and the recognition result can be fed back to the transmitter; otherwise, the "confidence level" of this type is determined. The signal accuracy is low, the recognition result is not fed back, but the transmission instruction (such as data transmission method 200) is continued to be sent to the transmitter, or the Send instructions (such as data transmission method 300), or issue library signals (such as data transmission method 400). Therefore, it is possible to reduce communication overhead as much as possible while ensuring the accuracy of the recognition results.

Referring to Figure 5, Figure 5 is a schematic flowchart of the data transmission method 200 provided by the embodiment of the present application. The difference between the data transmission method 200 and the data transmission method 100 is that in the data transmission method 200, the first data is sent by the transmitter in rows or columns. Therefore, if the elements of the sent rows or columns are constituted During the matching judgment of the first data, if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than the preset value, then the receiver can send the transmission to the transmitter again Instruction to obtain elements of other rows or columns and then perform matching judgment after updating the first data. Detailed description will be given below. As shown in Figure 5, the data transmission method 200 includes but is not limited to the following steps:

S201. The transmitter sends a task request, and the receiver receives the task request. The task request is used to request a recognition task. The task request includes the number of rows N and the number of columns M of the data to be recognized in the recognition task;

In an optional implementation, if the receiver cannot process the task request in time after receiving the task request, it can send a waiting instruction to the sender; if the receiver can process the task request in time after receiving the task request, Then step S202 can be executed.

Optionally, in addition to information about the data to be identified, such as the number of rows N and the number of columns M, the task request may also include other information.

S202. The receiver sends a transmission instruction, and accordingly, the transmitter receives the transmission instruction; the transmission instruction is used to indicate the column index of the column or the row index of the row in the data to be identified that is currently requested to be transmitted;

S203. The transmitter sends the element of the column corresponding to the column index or the element of the row corresponding to the row index; accordingly, the receiver receives the element of the column corresponding to the column index or the element of the row corresponding to the row index, and uses the received column or row The elements construct the first data;

In an optional implementation, the transmission instruction is also used to indicate the transmission parameters used for compressed transmission of the column or row currently requested to be transmitted; the transmission parameters include the length of each block into which the column or row is divided, and the length of each block. The codebook information used for quantization. Correspondingly, the elements of the column corresponding to the column index received by the receiver are obtained by decompressing the elements of the column corresponding to the column index of compressed transmission based on the transmission parameters, or the elements of the row corresponding to the row index received by the receiver are obtained based on the transmission parameters. The parameter is obtained by decompressing the element of the row corresponding to the row index of the compressed transmission.

That is to say, in this implementation, the transmission parameters at least include but are not limited to the following information:

Information (11) The column index of the column or the row index of the row currently requested to be transferred;

Information (12) block division configuration, used to indicate the length of each block into which the column or row currently requested to be transmitted is divided;

Information (13) The codebook information used by each block to be quantized.

Optionally, the receiver determines that there may be multiple block lengths indicated by the block division configuration in the transmission parameters. For example, the lengths of the blocks into which the columns indicated by the block division configuration are L ₀ and L 1 are L 0 and L ₁ , where L ₁ =mod(N,L ₀ ) and L ₁ is the length of the last “block” of each column. In this way, for the case where the dimension N of each column is not divisible by L ₀ , the length of the last "block" of each column is set to the remainder of N divided by L ₀ , that is, L ₁ . Correspondingly, "blocks" of the same length can be quantized using codebooks with the same quantization accuracy, and "blocks" of different lengths can be quantized using different codebooks to meet the need to adjust the compression rate.

Among them, the specific operation of the transmitter using the information (12) to block the column or row currently requested for transmission can be found in the above description and will not be described in detail here. Correspondingly, the transmitter uses information (13) to quantize the divided "blocks". Please refer to the above description of the compression operation, which will not be described in detail here.

In another optional implementation, the transmission parameters for the first transmission may also include:

Information (14) mapping reference information configuration, used to indicate mapping information; optionally, the mapping information may include the compression matrix as described above;

Information (15) entropy coding configuration is used to indicate the entropy coding method; optionally, the entropy coding method may be arithmetic coding or Huffman coding as mentioned above.

That is to say, the transmission parameters of other sub-transmissions may not include information (14) and information (15). In this way, the elements of the rows or columns of other transmissions can use the mapping information and entropy coding method indicated by the mapping reference information configuration in the transmission parameters of the first transmission, and the block length indicated by the block division configuration in the transmission parameters of this transmission. and codebook information for compressed transmission. Among them, the elements of the column corresponding to the column index or the elements of the row corresponding to the row index sent by the transmitter each time are based on the information (11) to information (13) indicated by the corresponding transmission parameters and the information in the transmission parameters of the first transmission. (14) and information (15) are compressed and transmitted. Correspondingly, the elements of the column corresponding to the column index received by the receiver are obtained by decompressing the elements of the column corresponding to the column index of compressed transmission based on the transmission parameters, or the elements of the row corresponding to the row index received by the receiver are obtained based on the transmission parameters. The parameter is obtained by decompressing the element of the row corresponding to the row index of the compressed transmission.

In another optional implementation, if the data transmission method is applied in a data transmission system that supports cellular communication, that is, the transmitter sends the data to be identified to the receiver through cellular communication, then the transmission parameters may also include : Information (16) resource configuration, used to indicate the resources used by the sender to transmit data. In this way, the transmitter can send the compressed element of the column corresponding to the column index or the element of the row corresponding to the row index on the indicated resource. Correspondingly, the receiver can receive the element of the column or row corresponding to the column index on the indicated resource. The element of the row corresponding to the index.

S204. The receiver determines the distance between each data in the database and the first data;

Correspondingly, the receiver also needs to determine the above-mentioned first distance and second distance from the distance between each data in the database and the first data.

S205. The receiver determines whether the number of second data in the database that is a first distance from the first data is one or more; if the number of second data is one and the first distance is the same as the second distance, If the absolute difference between them is greater than the preset value, step S206 is executed; if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than the preset value, execution continues. Steps S202 to S205, until if the number of second data is one and the absolute difference between the first distance and the second distance is greater than the preset value, or all columns in the data to be identified have been transmitted, then step S206 is executed. .

S206. The receiver sends the identification result represented by the second data. Correspondingly, the transmitter can receive the identification result represented by the second data.

Optionally, after receiving the identification result of the second data representation, the transmitter can also output the identification result of the second data representation.

Among them, after the receiver receives the element of the column corresponding to the column index or the element of the row corresponding to the row index in the first loop, the receiver can perform step S204; the receiver receives the element of the column corresponding to the column index in a non-first loop. After the element of the row corresponding to the row index or the row index is retrieved, the first data needs to be updated, and then step S204 is executed.

In an optional implementation, the column index of the column currently requested to be transferred or the row index of the row indicated by each transfer instruction may be continuous. For example, the column index indicated by the transmission instruction sent for the first time is the first column from the left in the data to be transmitted; if the number of second data is multiple, or the distance between the first distance and the second distance is If the absolute difference is greater than the preset value, step S202 is continued, that is, the column index indicated by the second transmission instruction is the second column from the left in the data to be transmitted, and so on, until if the If the number of the second data is one, and the absolute difference between the first distance and the second distance is greater than the preset value, or all columns in the data to be recognized have been transmitted, the recognition result represented by the second data is sent.

In another optional implementation, the column index of the column currently requested to be transferred or the row index of the row indicated by each transfer instruction may be discontinuous. For example, the column index indicated by the transmission instruction sent for the first time is the first column from the left in the data to be transmitted; if the number of second data is multiple, or the distance between the first distance and the second distance is If the absolute difference is greater than the preset value, step S202 is continued, that is, the column index indicated by the second transmission instruction is the third column from the left in the data to be transmitted, and so on, until the second The number of data is one, and the distance between the first distance and the second distance The absolute difference is greater than the preset value, or all columns in the data to be recognized have been transmitted, and the recognition result represented by the second data is sent.

The data transmission method 200 is an incremental transmission mechanism based on the classification accuracy feedback principle. That is, the data to be identified is partially transmitted by rows or columns. Once the classification accuracy feedback principle is met, the transmission is stopped and the identification results are fed back. Therefore, the elements of columns or rows in the data to be identified need to be transmitted as much as possible, which greatly reduces the communication delay caused by the large amount of communication data. The feedback principle of classification accuracy is also used to ensure the feedback results. accuracy.

Referring to Figure 6, Figure 6 is a schematic flowchart of a data transmission method 300 provided by an embodiment of the present application. The difference between the data transmission method 300 and the data transmission method 100 is that in the data transmission method 300, the first data is sent by the transmitter for the first time, or is updated by using retransmitted row or column elements. Therefore, If the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than the preset value, then the receiver can send a retransmission instruction to obtain the updated element of the row or column. After receiving the data, the matching judgment will be made. Detailed description will be given below. As shown in Figure 6, the data transmission method 300 includes but is not limited to the following steps:

S301. The transmitter sends a task request, and the receiver receives the task request. The task request is used to request a recognition task. The task request includes the number of rows N and the number of columns M of the data to be recognized in the recognition task;

For the relevant description of step S301, please refer to the above description of S201, and will not be described in detail here.

S302. The receiver sends a transmission instruction, and accordingly, the transmitter receives the transmission instruction; the transmission instruction is used to request the transmission of the data to be identified in the identification task;

S303. The transmitter sends the data to be identified, and accordingly, the receiver receives the data to be identified and obtains the first data;

In an optional implementation, the transmission instruction is also used to indicate the transmission parameters to be used for compressed transmission of the data to be identified that is requested to be transmitted; the transmission parameters include the length of each block divided into each column or row of the data to be identified. , and the codebook information used by each block to be quantized. Correspondingly, the data to be identified sent by the transmitter is obtained by compressing the data to be identified based on the transmission parameters.

Information (21) The data to be identified that is currently requested to be transmitted;

Information (22) block division configuration, used to indicate the length of each block into which each column or row in the data to be identified is divided;

Information (23) The codebook information used by each block to be quantized.

Optionally, the block partitioning configuration may indicate the length of blocks into which all columns or all rows are divided. Among them, the transmitter uses the information (22) to perform the specific operation of block division of all columns or rows, and the transmitter uses the information (23) to quantize the divided "blocks". Please refer to the explanation of the above compression operation, here No more details.

In another optional implementation, the transmission parameters may also include:

Information (24) mapping reference information configuration, used to indicate mapping information; optionally, the mapping information can be a compression matrix as described above;

Information (25) entropy coding configuration is used to indicate the entropy coding method; optionally, the entropy coding method may be arithmetic coding or Huffman coding as mentioned above.

In another optional implementation, if the data transmission method is applied in a data transmission system that supports cellular communication, that is, the transmitter sends the data to be identified to the receiver through cellular communication, then the transmission parameters may also include : Information (26) resource configuration, used to indicate the resources used by the sender to transmit data. In this way, the transmitter can send the compressed data to be identified on the indicated resource, and accordingly, the receiver can receive the compressed data to be identified on the indicated resource.

S304. The receiver determines the distance between each data in the database and the first data;

Correspondingly, the receiver also needs to determine the first distance and the second distance based on the distance between each data in the database and the first data. Among them, the shortest distance between each data and the first data is the first distance, except for the first distance The shortest distance is the second distance.

S305. The receiver determines whether the number of second data in the database that is a first distance from the first data is one or more; if the number of second data is one and the first distance is the same as the second distance, If the absolute difference between them is greater than the preset value, step S306 is executed; if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than the preset value, execution continues. Step S307;

S306. The receiver sends the identification result represented by the second data. Correspondingly, the transmitter can receive the identification result represented by the second data.

S307. The receiver sends a retransmission instruction. Correspondingly, the sender can receive the retransmission instruction. The retransmission instruction is used to indicate the row index or column index of the row currently requested to be retransmitted in the data to be identified;

S308. The transmitter retransmits the element of the column corresponding to the column index or the element of the row corresponding to the row index. Correspondingly, the receiver receives the element of the column corresponding to the column index or the element of the row corresponding to the row index, and uses The received column or row elements update the first data, and perform steps S304 to S305 again until if the number of second data is one and the absolute difference between the first distance and the second distance is greater than the preset value, Or if all columns or rows in the data to be identified have been retransmitted, step S306 is executed.

In an optional implementation, the retransmission instruction is also used to indicate the transmission parameters used for compressed retransmission of the row or column currently requested to be retransmitted; the transmission parameters include the length of each block into which the column or row is divided, and the length of each block. The codebook information used for each block to be quantized; accordingly, the elements of the column corresponding to the column index retransmitted by the transmitter are obtained by compressing the elements of the column corresponding to the column index based on the transmission parameters, or the elements of the column retransmitted by the transmitter The elements of the row corresponding to the row index are obtained by compressing the elements of the row corresponding to the row index based on the transmission parameters.

That is to say, in this implementation, the transmission parameters indicated by the retransmission instruction at least include but are not limited to the following information:

Information (31) The column index of the column or the row index of the row currently requested to be retransmitted;

Information (32) block division configuration, used to indicate the length of each block into which the column or row currently requested to be retransmitted is divided;

Information (33) The codebook information used by each block to be quantized.

Optionally, the length of the blocks indicated by the block division configuration in the information (32) may be multiple. Among them, the transmitter uses the information (32) to perform the specific operation of dividing the column or row currently requested to be transmitted into blocks, and the transmitter uses the information (33) to quantify the divided "blocks". Please refer to the explanation of the above compression operation. No further details will be given here.

In another optional implementation, if the data transmission method is applied in a data transmission system that supports cellular communication, that is, the transmitter sends the data to be identified to the receiver through cellular communication, then the transmission indicated by the retransmission instruction The parameters may also include: information (34) resource configuration, used to instruct the transmitter to retransmit the resources of the row or column. In this way, the transmitter can send the compressed element of the column corresponding to the column index or the element of the row corresponding to the row index on the indicated resource. Correspondingly, the receiver can receive the element of the column or row corresponding to the column index on the indicated resource. The element of the row corresponding to the index.

That is to say, the retransmission instruction includes two parts, one part is used to indicate the position of the retransmitted row or column, that is, information (31), and the other part is used to indicate the quantization accuracy of the retransmitted row or column, that is, information (32) , information (33); or, the retransmission instruction includes three parts, that is, the retransmission instruction also includes a part used to indicate the resources of the row or column to be retransmitted.

In an optional implementation, the quantization precision used for compressed retransmission of the row or column currently requested to be retransmitted is higher than the quantization precision used for compressed transmission of the data to be identified that is requested to be transmitted. That is to say, the codebook information of information (33) has higher quantization accuracy than the codebook information of information (23).

The data transmission method 300 is a row-by-row or column-by-column retransmission mechanism based on classification accuracy. That is, after the identification data is transmitted, once the classification accuracy feedback principle is not met, partial retransmission will be performed by row or column. In this way, the transmission of the data to be identified can be quantized using low-precision quantization parameters. When the classification accuracy feedback principle is not met, high-precision quantization parameters can be used to retransmit rows or columns to correct the data to be identified. thereby reducing the amount of data transmission required as much as possible, The communication delay is reduced to a great extent, and the accuracy of the feedback results is also ensured by using the classification accuracy feedback principle.

Referring to Figure 7, Figure 7 is a schematic flowchart of the data transmission method 400 provided by the embodiment of the present application. The difference between the data transmission method 400 and the data transmission method 100 is that in the data transmission method 400, if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than Default value, then the receiver can send part of the data in the database, and the sender itself makes a matching judgment. Detailed description will be given below. As shown in Figure 7, the data transmission method 400 includes but is not limited to the following steps:

S401. The transmitter sends a task request, and the receiver receives the task request. The task request is used to request a recognition task. The task request includes the number of rows N and the number of columns M of the data to be recognized in the recognition task;

For the relevant description of step S401, please refer to the above description of S201, which will not be described in detail here.

S402. The receiver sends a transmission instruction, and accordingly, the transmitter receives the transmission instruction; the transmission instruction is used to request the transmission of the data to be identified in the identification task;

S403. The transmitter sends the data to be identified, and accordingly, the receiver receives the data to be identified and obtains the first data;

For further description of steps S402 and S403, please refer to the relevant description of the above steps S302 and S303, which will not be described in detail here.

S404. The receiver determines the distance between each data in the database and the first data;

Correspondingly, the receiver also needs to determine the first distance and the second distance based on the distance between each data in the database and the first data. Wherein, the shortest distance between each data and the first data is the first distance, and the shortest distance other than the first distance is the second distance.

S405. The receiver determines whether the number of second data in the database that is a first distance from the first data is one or more; if the number of second data is one and the first distance is the same as the second distance, If the absolute difference between them is greater than the preset value, step S406 is executed; if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than the preset value, execution continues. Step S407;

S406. The receiver sends the identification result represented by the second data. Correspondingly, the transmitter can receive the identification result represented by the second data.

S407. The receiver sends a plurality of fourth data, and accordingly, the transmitter can receive the plurality of fourth data;

Wherein, the plurality of fourth data are a plurality of second data in the database, or the second data in the database and the third data in the database whose distance from the first data is a second distance. data.

In this embodiment, if the number of second data is multiple, the receiver may send all the second data in the database that are at a first distance from the first data to the transmitter. If the absolute difference between the first distance and the second distance is not greater than the preset value, that is, the second data corresponding to the first distance is one but the absolute difference is not greater than the preset value, then the receiver can store the corresponding data in the database. Data at the shortest two distances are sent to the transmitter, that is, the second data corresponding to the first distance and the third data corresponding to the second distance.

S408. The transmitter determines the distance between each received fourth data and the data to be identified;

Wherein, the shortest distance between the one or more second data and the data to be identified is the third distance, and the shortest distance other than the third distance is the fourth distance.

It should be noted that the distance determined by the transmitter is the distance between each fifth data and the data to be identified, and the distance determined by the receiver in step S404 is the distance between each data in the database and the first data. That is to say, the first data is the data to be identified received by the receiver, which is compressed and transmitted. Compared with the data to be identified on the sender side, there is a partial loss. Therefore, the receiver can send part of the data in the database. , the transmitter uses the uncompressed and transmitted data to be identified to determine the matching operation.

S409. The transmitter determines whether there is one or more fifth data among the plurality of fourth data whose distance from the data to be identified is the third distance; if the number of fifth data is one and the third distance If the absolute difference from the fourth distance is greater than the preset value, then step S410 is executed; otherwise, step S411 is executed;

That is to say, if the number of fifth data is multiple, or the absolute difference between the third distance and the fourth distance is not greater than the preset value, step S411 is performed.

It should be noted that the fifth data is data among the plurality of fourth data received by the transmitter. That is to say, step S409 is for the transmitter to perform an operation based on the feedback principle of accurate classification based on the source signal (ie, the data to be identified) and the library signal (ie, a plurality of fourth data) sent by the receiver.

S410. The transmitter outputs the recognition result represented by the fifth data.

S411. The transmitter outputs prompt information, which is used to indicate that the data to be identified fails to match.

The data transmission method 400 is a library signal feedback mechanism based on the classification accuracy feedback principle. That is, after the identification data is transmitted, once the classification accuracy feedback principle is not satisfied, the receiver will send the second data in the library, or the second data and the third data, to the transmitter, and the transmitter will based on the classification accuracy feedback principle. Make matching judgments. In this way, the transmission of the data to be identified can be quantized using low-precision quantization parameters, and library signals are sent when the classification accuracy feedback principle is not met, thereby reducing the amount of data transmission required for uplink transmission as much as possible, and greatly reducing Communication delay, the use of classification accuracy feedback principle also ensures the accuracy of feedback results.

In the description of this application, unless otherwise stated, "/" means "or". For example, A/B can mean A or B. "And/or" in this article is just an association relationship that describes related objects, indicating that there can be three relationships, for example, A and/or B, which can mean: A exists alone, A and B exist simultaneously, and B exists alone these three situations. In addition, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions thereof refers to any combination of these items, including any combination of a single item (items) or a plurality of items (items). For example, at least one of a, b, or c can represent: a, b, c; a and b; a and c; b and c; or a, b, and c. Among them, a, b, c can be single or multiple.

In the description of this application, words such as "first" and "second" are only used to distinguish different objects and do not limit the number and execution order, and words such as "first" and "second" are not limiting. It must be different. Furthermore, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes unlisted steps or units, or optional It also includes other steps or units inherent to these processes, methods, products or equipment.

In this application, the words "exemplary" or "such as" are used to mean examples, illustrations or explanations. Any embodiment or design described herein as "exemplary," "for example," or "such as" is not intended to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the words "exemplary," "for example," or "such as" is intended to present the concept in a concrete manner.

It should be understood that in this application, "when", "if" and "if" all mean that the device will perform corresponding processing under certain objective circumstances. They do not limit the time, nor do they require the device to be implemented. There must be an act of judgment, and it does not mean that there are other limitations.

Elements referred to in the singular in this application are intended to mean "one or more" and not "one and only one" unless otherwise specified.

It can be understood that in various embodiments of the present application, "A corresponds to B" means that A and B have a corresponding relationship, and B can be determined based on A. But it should also be understood that determining (or generating) B based on (or based on) A does not mean solely based on (or based on) A To determine (or generate) B, B may also be determined (or generated) according to (or based on) A and/or other information.

In this application, unless otherwise specified, the same or similar parts between various embodiments or implementations may be referred to each other. In the various embodiments of this application and the various implementation methods/implementation methods/implementation methods in each embodiment, if there are no special instructions or logical conflicts, the differences between different embodiments and the various implementation methods/implementation methods in each embodiment will be different. The terminology and/or descriptions between implementation methods/implementation methods are consistent and can be referenced to each other. Different embodiments, as well as the technical features in each implementation method/implementation method/implementation method in each embodiment are based on their inherent Logical relationships can be combined to form new embodiments, implementations, implementation methods, or implementation methods. The embodiments of the present application described below do not constitute a limitation on the protection scope of the present application.

In order to realize the functions of any of the communication methods provided by the embodiments of the present application, the transmitter and the receiver may respectively include a hardware structure and a software module, and may be implemented in the form of a hardware structure, a software module, or a hardware structure plus a software module. Each of the above functions. A certain function among the above functions can be executed by a hardware structure, a software module, or a hardware structure plus a software module. Figures 8 and 9 are schematic structural diagrams of possible communication devices provided by embodiments of the present application. These communication devices can be used to implement the functions of the receiver or transmitter in the above method embodiments, and therefore can also achieve the beneficial effects of the above method embodiments.

The communication device shown in FIG. 8 may include a communication unit 801. The communication unit 801 may include a sending unit and a receiving unit. The sending unit is used to implement the sending function, and the receiving unit is used to implement the receiving function. The communication unit 801 may implement the sending function and/or the receiving function. The communication unit can also be described as a transceiver unit.

The communication device shown in Figure 8 may be a receiver or a device in the receiver.

In an optional implementation, the communication device can implement the related operations of the receiver in the above data transmission method 100. For example, the communication unit 801 is used to receive data to be identified or elements of rows or columns in the data to be identified; the processing unit 802 is used to determine the distance between each data in the database and the first data, and the first data is The received data to be identified may be a matrix composed of received rows or columns in the data to be identified; the shortest distance between each data in the database and the first data is the first distance, except for the first distance. The shortest distance outside is the second distance; and determine whether the number of second data in the database that is the first distance from the first data is one or more; if the number of second data is one, and the If the absolute difference between the first distance and the second distance is greater than the preset value, the recognition result represented by the second data is sent.

For more detailed descriptions about the processing unit 802 and the communication unit 801, please refer to the relevant descriptions in the above method embodiments.

In another optional implementation, the communication device may implement the related operations of the receiver in the above data transmission method 200. For example, the communication unit 801 is used to send a transmission instruction; the transmission instruction is used to indicate the column index of the column or the row index of the row currently requested to be transmitted in the data to be identified; and receive the element of the column corresponding to the column index or the row index corresponding to elements of the row, and uses the received elements of the column or row to construct the first data; the processing unit 802 is used to determine the distance between each data in the database and the first data; the receiver determines the distance between the database and the first data Is the number of second data whose distance is the first distance one or more? If the number of second data is one and the absolute difference between the first distance and the second distance is greater than the preset value, trigger The communication unit 801 sends the recognition result represented by the second data; if the number of second data is multiple, or the absolute difference between the first distance and the second distance is greater than the preset value, the communication unit 801 can be triggered to execute the process again. The above-mentioned operation of sending a transmission instruction is performed until if the number of second data is one and the absolute difference between the first distance and the second distance is greater than the preset value, or all columns in the data to be identified have been transmitted.

In yet another optional implementation, the communication device can implement the related operations of the receiver in the above data transmission method 300. For example, the communication unit 801 is used to send a transmission instruction, which is used to request the transmission of data to be identified in the identification task; and to receive the data to be identified and obtain the first data; the processing unit 802 is used to determine the relationship between each data in the database and The distance between the first data, and whether the number of second data in the database that is the first distance is one or more; if the number of second data is one, and the first The absolute difference between the distance and the second distance is greater than the preset value, triggering the communication unit 801 to send the recognition result represented by the second data; if the number of second data is multiple, or the sum of the first distance and the second distance is If the absolute difference between the two is greater than the preset value, the communication unit 801 is triggered to send a retransmission instruction. The retransmission instruction is used to indicate the row index or column index of the row currently requested to be retransmitted in the data to be identified; and receive the column index. The element of the corresponding column or row indexes the element of the corresponding row, and updates the first data using the received element of the column or row; furthermore, the processing unit 802 can execute again to determine the distance between each data in the database and the first data. The operation is performed until if the number of second data is one and the absolute difference between the first distance and the second distance is greater than the preset value, or all columns or rows in the data to be identified have been retransmitted.

In yet another optional implementation, the communication device can implement the related operations of the receiver in the above data transmission method 400. For example, the communication unit 801 is used to send a transmission instruction, which is used to request the transmission of data to be identified in the identification task; and to receive the data to be identified and obtain the first data; the processing unit 802 is used to determine each data in the database distance from the first data, and determine whether the number of second data in the database that is the first distance from the first data is one or more; if the number of second data is one, and the If the absolute difference between the first distance and the second distance is greater than the preset value, the communication unit 801 is triggered to send the identification result represented by the second data; if the number of second data is multiple, or the first distance and the second distance If the absolute difference between them is greater than the preset value, the communication unit 801 is triggered to send a plurality of fourth data. The plurality of fourth data are a plurality of second data in the database, or the second data in the database and the said The distance between the third data and the first data in the database is the second distance.

The communication device shown in FIG. 8 may be a transmitter, or may be a device in the transmitter.

In an optional implementation, the communication device can perform related operations of the transmitter in the above data transmission method 200. For example, the communication unit 801 is used to send a task request, which is used to request an identification task. The task request includes the number N of rows and the number M of columns of data to be identified in the identification task; and to receive a transmission instruction, which is used Indicates the column index of the column or the row index of the row currently requested to be transmitted in the data to be identified; and sends the element of the column corresponding to the column index or the element of the row corresponding to the row index.

For more detailed descriptions about the communication unit 801, please refer to the relevant descriptions in the above method embodiments.

In another optional implementation, the communication device can perform related operations of the transmitter in the above data transmission method 300. For example, the communication unit 801 is used to receive a transmission instruction, which is used to request the transmission of the data to be identified in the identification task; to send the data to be identified; and to receive a retransmission instruction, which is used to indicate the current data in the data to be identified. The row index of the row or the column index of the column that is requested to be retransmitted, and the element of the column corresponding to the column index or the element of the row corresponding to the row index is retransmitted.

In yet another optional implementation, the communication device can perform related operations of the transmitter in the above data transmission method 400. For example, the communication unit 801 is used to receive a transmission instruction, which is used to request the transmission of data to be identified in the identification task; to send the data to be identified; and to receive a plurality of fourth data; the processing unit 802 is used to determine the received The distance between each fourth data and the data to be identified; determine the distance between the plurality of fourth data and the data to be identified as the third distance The number of the fifth data is one or more; if the number of the fifth data is one and the absolute difference between the third distance and the fourth distance is greater than the preset value, the communication unit 801 is triggered to output the third distance. Five data representation identification results.

The plurality of fourth data are the second data in the database of the device that identifies the task, or the second data in the database and the third data in the database. The second data is the data in the database whose distance from the first data is the first distance, and the third data is the data in the database whose distance from the first data is the second distance. Wherein, the shortest distance between each data in the database and the first data is the first distance, and the shortest distance other than the first distance is the second distance. The first data is data to be identified received by the device; wherein the shortest distance between each fourth data in the plurality of fourth data and the data to be identified is the third distance, and the shortest distance other than the third distance is The distance is the fourth distance.

The communication device shown in FIG. 9 may include a processor 901 and an interface circuit 902. The processor 901 and the interface circuit 902 are coupled to each other. It can be understood that the interface circuit 902 may be an interface circuit or an input-output interface. Optionally, the communication device shown in FIG. 9 may also include a memory 903 for storing instructions executed by the processor 901 or input data required for the processor 901 to run the instructions or data generated after the processor 901 executes the instructions.

The communication device is a receiver or a device in a receiver: the processor 901 can execute S101 to S103 in Figure 4; or the interface circuit 902 can execute S201 to S203 and S206 in Figure 5, and the processor 901 can execute S204 in Figure 5 , S205; or, the interface circuit 902 executes S301 to S303, S306 to S308 in Fig. 6, and the processor 901 can execute S304, S305 in Fig. 6; or, the interface circuit 902 executes S401 to S403, S406 to S407 in Fig. 7, and the processing The processor 901 can execute S404 and S405 in Figure 7.

The communication device is a transmitter or a device in the transmitter: the interface circuit 902 executes S201 to S203 and S206 in Figure 5; or the interface circuit 902 executes S301 to S303 and S306 to S308 in Figure 6; or the interface circuit 902 executes In S401 to S403 and S406 to S407 in Figure 7 , the processor 901 can execute S408 to S411 in Figure 7 .

It should be understood that the processor mentioned in the embodiments of this application can be a CPU, or other general-purpose processor, digital signal processor (digital signal processor, DSP), application specific integrated circuit (ASIC), off-the-shelf processor Field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

It should also be understood that the memory mentioned in the embodiments of the present application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Among them, non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically removable memory. Erase electrically programmable read-only memory (EPROM, EEPROM) or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of illustration, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (synchlink DRAM, SLDRAM) ) and direct memory bus random access memory (direct rambus RAM, DR RAM).

It should be noted that when the processor is a general-purpose processor, DSP, ASIC, FPGA or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, the memory (storage module) is integrated in the processor.

It should be noted that the memories described herein are intended to include, but are not limited to, these and any other suitable types of memories.

It should be understood that in the various embodiments of the present application, the size of the sequence numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its functions and internal logic, and should not be used in the embodiments of the present application. The implementation process constitutes any limitation.

Those of ordinary skill in the art can appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each specific application, but such implementations should not be considered beyond the scope of this application.

Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be described again here.

In the several embodiments provided in this application, it should be understood that the disclosed communication methods and devices can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components may be combined or may be Integrated into another system, or some features can be ignored, or not implemented. On the other hand, the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical, mechanical or other forms.

The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional module in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

If this function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present application or the part that contributes or the part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions to A computer device (which may be a personal computer, a server, or a network device, etc.) is caused to execute all or part of the steps of the method in each embodiment of the present application. The aforementioned computer-readable storage medium can be any available medium that can be accessed by a computer. Taking this as an example but not limited to: computer-readable media may include random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (electrically erasable programmable) read only memory (EEPROM), compact disc read-only memory (CD-ROM), universal serial bus flash disk (universal serial bus flash disk), mobile hard disk, or other optical disk storage, magnetic disk storage media or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures that can be accessed by a computer.

What is shown above is only a specific implementation mode of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any person familiar with the technical field can easily think of changes within the technical scope disclosed in the embodiments of the present application. or replacement, all should be covered by the protection scope of the embodiments of this application. Therefore, the protection scope of the embodiments of the present application should be subject to the protection scope of the claims.

Claims

A data transmission method, characterized in that the method includes:

Determine the distance between each data in the database and the first data, which is the received data to be identified or a matrix composed of received rows or columns in the data to be identified; each data in the database is The shortest distance among the distances between the first data is the first distance, and the shortest distance other than the first distance is the second distance;

Determine whether the number of second data in the database that is a first distance from the first data is one or more;

If the number of the second data is one and the absolute difference between the first distance and the second distance is greater than a preset value, the recognition result represented by the second data is sent.
The method according to claim 1, characterized in that before determining the distance between each data in the database and the first data, the method further includes:

Send a transmission instruction, the transmission instruction being used to indicate the column index of the column or the row index of the row in the data to be identified that is currently requested to be transmitted;

Receive elements of the column corresponding to the column index or elements of the row corresponding to the row index;

The first data is determined using the received column elements or row elements.
The method of claim 2, further comprising:

If the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than the preset value, perform the step of sending a transmission instruction again until the The number of the second data is one and the absolute difference between the first distance and the second distance is greater than a preset value, or all columns or all rows in the data to be identified have been transmitted and sent The second data represents the identification result.
The method according to claim 1, characterized in that before determining the distance between each data in the database and the first data, the method further includes:

Send a transmission instruction, which is used to request the transmission of the data to be identified;

The data to be identified is received as first data.
The method of claim 4, further comprising:

If the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than a preset value, a retransmission instruction is sent, and the retransmission instruction is used to indicate The row index or column index of the row currently requested to be retransmitted in the data to be identified;

Receive elements of the row corresponding to the row index or elements of the column corresponding to the column index;

Update the first data using elements of the row or elements of the column, and perform the step of determining the distance between each data in the database and the first data again until the individual data of the second data The number is one and the absolute difference between the first distance and the second distance is greater than the preset value, or all columns or all rows in the data to be identified have been retransmitted, and the second data is sent Recognition results of representations.
The method of claim 4, further comprising:

If the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than a predetermined Set value, send a plurality of fourth data, the plurality of fourth data are the second data in the database, or the second data in the database and the data in the database that are the same as the third data. The distance between one data is the second distance and the third data.
The method according to claim 2 or 3, characterized in that the transmission instruction is also used to indicate the transmission parameters used for compressed transmission of the column or row currently requested to be transmitted;

The transmission parameters include the length of each block into which the column or row is divided, and the codebook information used by which each block is quantized;

The received elements of the column corresponding to the column index are obtained by decompressing the elements of the column corresponding to the compressed and transmitted column index based on the transmission parameters; or

The received row elements corresponding to the row index are obtained by decompressing the compressed and transmitted row elements corresponding to the row index based on the transmission parameters.
The method according to any one of claims 4 to 6, characterized in that the transmission instruction is also used to indicate the transmission parameters to be used for compressed transmission of the data to be identified that is requested to be transmitted;

The transmission parameters include the length of each block divided into each column or row in the data to be identified, and the codebook information used by each block to be quantized;

The received data to be identified is obtained by decompressing the compressed and transmitted data to be identified based on the transmission parameters.
The method according to claim 8, characterized in that the retransmission instruction is also used to indicate the transmission parameters used for compressed retransmission of the row or column currently requested to be retransmitted;

The transmission parameters include the length of each block into which the column or row is divided, and the codebook information used by which each block is quantized;

The received elements of the column corresponding to the column index are obtained by decompressing the elements of the column corresponding to the compressed and retransmitted column index based on the transmission parameters; or

The received row elements corresponding to the row index are obtained by decompressing the compressed and retransmitted row elements corresponding to the row index based on the transmission parameters.
The method according to claim 9, characterized in that the quantization precision used for compressed retransmission of the row or column currently requested to be retransmitted is higher than the quantization precision used for compressed transmission of the data to be identified that is requested to be transmitted.
A data transmission method, characterized in that the method includes:

Send a task request, the task request is used to request a recognition task, the task request includes the number of rows N and the number of columns M of the data to be recognized in the recognition task;

Receive a transmission instruction, the transmission instruction being used to indicate the column index of the column or the row index of the row currently requested to be transmitted in the data to be identified;

Send the element of the column corresponding to the column index or the element of the row corresponding to the row index.
The method according to claim 11, characterized in that, the method further includes:

Receive the recognition result represented by the second data, the second data is the data in the database in the device that responds to the recognition task, and the distance between the first data and the first data is a first distance, the first distance is the database Each data in the The shortest distance among the distances between data. The first data is composed of elements of rows or columns that have been sent in the data to be identified.
The method according to claim 11 or 12, characterized in that the transmission instruction is also used to indicate the transmission parameters used for compressed transmission of the column or row currently requested to be transmitted;

The transmission parameters include the length of each block into which the column or row is divided, and the codebook information used by which each block is quantized;

The sent elements of the column corresponding to the column index are obtained by compressing the elements of the column corresponding to the column index based on the transmission parameters; or

The sent elements of the row corresponding to the row index are obtained by compressing the elements of the row corresponding to the row index based on the transmission parameters.
A data transmission method, characterized in that the method includes:

Receive a transmission instruction, which is used to request the transmission of data to be identified in the identification task;

Send the data to be identified;

Receive a retransmission instruction, the retransmission instruction being used to indicate the row index or the column index of the row currently requested to be retransmitted in the data to be identified;

Retransmit the element of the column corresponding to the column index or the element of the row corresponding to the row index.
The method of claim 14, further comprising:

Receive the recognition result represented by the second data, the second data is the data in the database in the device that responds to the recognition task, and the distance between the first data and the first data is a first distance, the first distance is the database The shortest distance between each data in the data and the first data, the first data is the data that is updated based on the elements of the row or column that have been retransmitted to be identified.
The method according to claim 14 or 15, characterized in that the transmission instruction is also used to instruct the transmission parameters to be used for compressed transmission of the data to be identified that is requested to be transmitted; the transmission parameters include the data to be identified. The length of each block into which each column or row is divided, and the codebook information used by said each block to be quantized;

The sent data to be identified is obtained by compressing the data to be identified based on the transmission parameters.
The method according to claim 16, characterized in that the retransmission instruction is also used to indicate the transmission parameters used for compressed retransmission of the row or column currently requested to be retransmitted;

The transmission parameters include the length of each block into which the column or row is divided, and the codebook information used by which each block is quantized;

The sent elements of the column corresponding to the column index are obtained by compressing the elements of the column corresponding to the column index based on the transmission parameters; or

The sent elements of the row corresponding to the row index are obtained by compressing the elements of the row corresponding to the row index based on the transmission parameters.
The method according to claim 17, characterized in that the quantization precision used for compressed retransmission of the row or column currently requested to be retransmitted is higher than the quantization precision used for compressed transmission of the data to be identified that is requested to be transmitted.
A data transmission method, characterized in that the method includes:

Receive a transmission instruction, which is used to request the transmission of data to be identified in the identification task;

Send the data to be identified;

Receive a plurality of fourth data, the plurality of fourth data being second data in the database of the device responding to the identification task, or second data and third data in the database, the second data being is the data in the database whose distance from the first data is the first distance, the third data is the data in the database whose distance from the first data is the second distance, the database The shortest distance between each data and the first data is the first distance, the shortest distance other than the first distance is the second distance, and the first data is the The data to be identified received by the above device;

Determine the distance between each received fourth data and the data to be identified; the shortest distance among the distances between each fourth data in the plurality of fourth data and the data to be identified is the third distance , the shortest distance other than the third distance is the fourth distance;

Determine whether the number of fifth data among the plurality of fourth data that is a third distance from the data to be identified is one or more;

If the number of the fifth data is one and the absolute difference between the third distance and the fourth distance is greater than the preset value, the recognition result represented by the fifth data is output.
The method according to claim 19, characterized in that the transmission instruction is also used to instruct the transmission parameters to be used for compressed transmission of the data to be identified that is requested to be transmitted; the transmission parameters include each column in the data to be identified. Or the length of each block into which each row is divided, and the codebook information used by each block to be quantized;

The sent data to be identified is obtained by compressing the data to be identified based on the transmission parameters.
The method according to claim 19 or 20, characterized in that if the number of the fifth data is multiple, or the absolute difference between the third distance and the fourth distance is not greater than a preset value, then output prompt information, the prompt information is used to indicate that the data to be identified fails to match.
A communication device, characterized in that it includes a processor and a transceiver, wherein the transceiver is used to send and receive signals, and the processor is used to execute program instructions, so that the communication device executes any of claims 1-10. method described in one item.
A communication device, characterized in that it includes a processor and a transceiver, wherein the transceiver is used to send and receive information, and the processor is used to execute program instructions, so that the communication device executes any of claims 11-13. The method described in any one of claims 14-18, or the method described in any one of claims 19-21.
A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by a computer, the computer is caused to execute the instructions as claimed in claim 1 -The method described in any one of 10.
A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by a computer, the computer is caused to execute the claims. The method according to any one of claims 11-13, or the method according to any one of claims 14-18, or the method according to any one of claims 19-21.
A computer program product, characterized in that, when the computer program product is run on a computer, it causes the computer to execute the method according to any one of claims 1-10.
A computer program product, characterized in that, when the computer program product is run on a computer, it causes the computer to execute the method as claimed in any one of claims 11-13, or as claimed in claims 14-18. The method of any one of claims 19-21.
A communication device, characterized in that the device includes:

A processing unit configured to determine the distance between each data in the database and the first data, which is the received data to be identified or a matrix composed of received rows or columns in the data to be identified; the database The shortest distance among the distances between each data and the first data is the first distance, and the shortest distance other than the first distance is the second distance;

The processing unit is also configured to determine whether the number of second data in the database that is a first distance away from the first data is one or more;

A communication unit configured to send the second data if the processing unit determines that the number of second data is one and the absolute difference between the first distance and the second distance is greater than a preset value. Recognition results of data representation.
The device according to claim 28, characterized in that:

The communication unit is also configured to send a transmission instruction before determining the distance between each data in the database and the first data. The transmission instruction is used to indicate the column index of the column in the data to be identified that is currently requested to be transmitted. or the row index of the row;

The communication unit is further configured to receive elements of the column corresponding to the column index or elements of the row corresponding to the row index;

The processing unit is also configured to determine the first data using the received column elements or row elements.
The device according to claim 29, characterized in that:

The communication unit is also configured to execute the step again if the number of second data is multiple or the absolute difference between the first distance and the second distance is not greater than a preset value. The step of sending a transmission instruction is until the number of the second data is one and the absolute difference between the first distance and the second distance is greater than a preset value, or until all columns in the data to be identified are Or all rows have been transmitted, and the identification result represented by the second data is sent.
The device according to claim 28, characterized in that:

The communication unit is also used to send a transmission instruction before determining the distance between each data in the database and the first data. The transmission instruction is used to request the transmission of the data to be identified; and to receive the data to be identified as the third data. One data.
The device according to claim 31, characterized in that:

The communication unit is also configured to send a retransmission instruction if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than a preset value, The retransmission instruction is used to indicate the row index or column index of the row currently requested to be retransmitted in the data to be identified; and to receive the element of the row corresponding to the row index or the element of the column corresponding to the column index. ;

The processing unit is also configured to update the first data using elements of the row or elements of the column, and execute again Perform the step of determining the distance between each data in the database and the first data until the number of the second data is one and the absolute difference between the first distance and the second distance is is greater than the preset value, or all columns or rows in the data to be identified have been retransmitted, and the identification result represented by the second data is sent.
The device according to claim 31, characterized in that:

The communication unit is also configured to send multiple fourth data if the number of second data is multiple, or the absolute difference between the first distance and the second distance is not greater than a preset value, The plurality of fourth data are the second data in the database, or the distance between the second data in the database and the first data in the database is a second distance the third data.
The device according to claim 29 or 30, characterized in that the transmission instruction is also used to indicate the transmission parameters used for compressed transmission of the column or row currently requested to be transmitted;

The transmission parameters include the length of each block into which the column or row is divided, and the codebook information used by which each block is quantized;

The received elements of the column corresponding to the column index are obtained by decompressing the elements of the column corresponding to the compressed and transmitted column index based on the transmission parameters; or

The received row elements corresponding to the row index are obtained by decompressing the compressed and transmitted row elements corresponding to the row index based on the transmission parameters.
The device according to any one of claims 31 to 33, wherein the transmission instruction is also used to indicate the transmission parameters to be used for compressed transmission of the data to be identified that is requested to be transmitted;

The transmission parameters include the length of each block divided into each column or row in the data to be identified, and the codebook information used by each block to be quantized;

The received data to be identified is obtained by decompressing the compressed and transmitted data to be identified based on the transmission parameters.
The device according to claim 35, characterized in that the retransmission instruction is also used to indicate the transmission parameters used for compressed retransmission of the row or column currently requested to be retransmitted;

The transmission parameters include the length of each block into which the column or row is divided, and the codebook information used by which each block is quantized;

The received elements of the column corresponding to the column index are obtained by decompressing the elements of the column corresponding to the compressed and retransmitted column index based on the transmission parameters; or

The received row elements corresponding to the row index are obtained by decompressing the compressed and retransmitted row elements corresponding to the row index based on the transmission parameters.
The device according to claim 36, wherein the quantization precision used for compressed retransmission of the row or column currently requested to be retransmitted is higher than the quantization precision used for compressed transmission of the data to be identified that is requested to be transmitted.
A data transmission device, characterized in that the device includes:

A communication unit configured to send a task request, the task request being used to request a recognition task, the task request including the number N of rows and the number M of columns of data to be recognized in the recognition task;

The communication unit is further configured to receive a transmission instruction, the transmission instruction being used to indicate the column index of the column or the row index of the row currently requested to be transmitted in the data to be identified;

The communication unit is further configured to send elements of the column corresponding to the column index or elements of the row corresponding to the row index.
The device according to claim 38, characterized in that:

The communication unit is also configured to receive the recognition result represented by the second data. The second data is the data whose distance from the first data in the database in the device responding to the recognition task is the first distance, so The first distance is the shortest distance between each data in the database and the first data, which is composed of elements of rows or columns that have been sent in the data to be identified.
The device according to claim 38 or 39, characterized in that the transmission instruction is also used to indicate the transmission parameters used for compressed transmission of the column or row currently requested to be transmitted;

The transmission parameters include the length of each block into which the column or row is divided, and the codebook information used by which each block is quantized;

The sent elements of the column corresponding to the column index are obtained by compressing the elements of the column corresponding to the column index based on the transmission parameters; or

The sent elements of the row corresponding to the row index are obtained by compressing the elements of the row corresponding to the row index based on the transmission parameters.
A data transmission device, characterized in that the device includes:

A communication unit configured to receive a transmission instruction, which is used to request the transmission of data to be identified in the identification task;

The communication unit is also used to send the data to be identified;

The communication unit is further configured to receive a retransmission instruction, where the retransmission instruction is used to indicate the row index or column index of the row currently requested to be retransmitted in the data to be identified;

The communication unit is further configured to retransmit elements of the column corresponding to the column index or elements of the row corresponding to the row index.
The device according to claim 41, characterized in that:

The communication unit is also configured to receive the recognition result represented by the second data. The second data is the data whose distance from the first data in the database in the device responding to the recognition task is the first distance, so The first distance is the shortest distance between each data in the database and the first data. The first data is the sent data to be identified after updating based on the elements of the row or column that have been retransmitted. The data.
The device according to claim 41 or 42, characterized in that the transmission instruction is also used to instruct the transmission parameters to be used for compressed transmission of the data to be identified that is requested to be transmitted; the transmission parameters include the data to be identified. The length of each block into which each column or row is divided, and the codebook information used by said each block to be quantized;

The sent data to be identified is obtained by compressing the data to be identified based on the transmission parameters.
The device according to claim 43, characterized in that the retransmission instruction is also used to indicate the transmission parameters used for compressed retransmission of the row or column currently requested to be retransmitted;

The transmission parameters include the length of each block into which the column or row is divided, and the codebook information used by which each block is quantized;

The sent elements of the column corresponding to the column index are obtained by compressing the elements of the column corresponding to the column index based on the transmission parameters; or

The sent elements of the row corresponding to the row index are obtained by compressing the elements of the row corresponding to the row index based on the transmission parameters.
The device according to claim 44, wherein the quantization precision used for compressed retransmission of the row or column currently requested to be retransmitted is higher than the quantization precision used for compressed transmission of the data to be identified that is requested to be transmitted.
A data transmission device, characterized in that the device includes:

A communication unit configured to receive a transmission instruction, which is used to request the transmission of data to be identified in the identification task;

The communication unit is also used to send the data to be identified;

The communication unit is also configured to receive a plurality of fourth data, the plurality of fourth data being second data in the database of the device responding to the identification task, or the second data in the database and the third data. Three data, the second data is data whose distance from the first data in the database is a first distance, and the third data is data whose distance from the first data in the database is a third distance. For data of two distances, the shortest distance between each data in the database and the first data is the first distance, and the shortest distance other than the first distance is the second distance. , the first data is the data to be identified received by the device;

a processing unit, configured to determine the distance between each received fourth data and the data to be identified; the shortest distance between each fourth data in the plurality of fourth data and the data to be identified; The distance is the third distance, and the shortest distance other than the third distance is the fourth distance;

The processing unit is further configured to determine whether the number of fifth data among the plurality of fourth data that is a third distance from the data to be identified is one or more;

The communication unit is also configured to, if the processing unit determines that the number of the fifth data is one and the absolute difference between the third distance and the fourth distance is greater than a preset value, output the Describe the identification results of the fifth data representation.
The device according to claim 46, characterized in that the transmission instruction is also used to instruct the transmission parameters to be used for compressed transmission of the data to be identified that is requested to be transmitted; the transmission parameters include each column in the data to be identified. Or the length of each block into which each row is divided, and the codebook information used by each block to be quantized;

The sent data to be identified is obtained by compressing the data to be identified based on the transmission parameters.
The device according to claim 46 or 47, characterized in that if the number of the fifth data is multiple, or the absolute difference between the third distance and the fourth distance is not greater than a preset value, then output prompt information, the prompt information is used to indicate that the data to be identified fails to match.