Apparatus and method for detecting and correcting semantic error in wireless communication system

Abstract

The present disclosure relates to an apparatus and a method for detecting a semantic error in a wireless communication system, and this method by which a first device operates in a wireless communication system comprises the steps of: receiving, from a second device, control information for transmission; generating transmission data on the basis of the control information; generating redundancy data for error detection and / or error correction of the transmission data; and transmitting the transmission data and the redundancy data, wherein the transmission data includes a semantic representation (SR) for semantic communication, and the redundancy data can be generated by a generation function selected from among a plurality of generation functions on the basis of a task to be performed by the second device.

Description

Device and method for detecting and correcting semantic errors in a wireless communication system

[0001] The following description relates to a wireless communication system, and to a device and method for detecting and correcting semantic errors in a wireless communication system.

[0002] Wireless access systems are widely deployed to provide various types of communication services, such as voice and data. Typically, wireless access systems are multiple access systems that support communications with multiple users by sharing available system resources (e.g., bandwidth, transmission power). Examples of multiple access systems include code division multiple access (CDMA), frequency division multiple access (FDMA), time division multiple access (TDMA), orthogonal frequency division multiple access (OFDMA), and single-carrier frequency division multiple access (SC-FDMA).

[0003] In particular, as numerous communication devices demand greater communication capacity, enhanced mobile broadband (eMBB) communication technologies are being proposed, improving upon existing radio access technology (RAT). Furthermore, massive machine type communications (mMTC), which connects numerous devices and objects to provide diverse services anytime and anywhere, as well as communication systems that consider reliability and latency-sensitive services / user equipment (UE), are being proposed. Various technological configurations are being proposed for these purposes.

[0004] The present disclosure can provide a device and method for effectively transmitting and receiving semantic information of a message in a wireless communication system.

[0005] The present disclosure may provide a device and method for transmitting and receiving a semantic representation (SR) and semantic redundancy in a wireless communication system.

[0006] The present disclosure may provide a device and method for generating a semantic expression based on task information related to semantic communication in a wireless communication system.

[0007] The present disclosure may provide a device and method for generating semantic redundancy based on task information related to semantic communication in a wireless communication system.

[0008] The present disclosure may provide a device and method for selecting a semantic redundancy generation function based on task information to be performed in a receiving device in a wireless communication system.

[0009] The present disclosure may provide a device and method for checking errors in semantic expressions based on semantic redundancy in a wireless communication system.

[0010] The present disclosure may provide a device and method for detecting semantic errors based on a difference between received semantic redundancy and generated semantic redundancy in a wireless communication system.

[0011] The present disclosure may provide a device and method for correcting semantic errors in a wireless communication system.

[0012] The present disclosure may provide a device and method for correcting a semantic expression using an optimization technique in a wireless communication system.

[0013] The present disclosure may provide a device and method for correcting a semantic representation using gradient descent in a wireless communication system.

[0014] The present disclosure may provide a device and method for detecting errors in a corrected semantic representation in a wireless communication system.

[0015] The present disclosure can provide a device and method for transmitting and receiving control information necessary for error detection and correction of semantic expressions in a wireless communication system.

[0016] The technical objectives to be achieved in the present disclosure are not limited to those mentioned above, and other technical tasks not mentioned can be considered by a person having ordinary skill in the technical field to which the technical configuration of the present disclosure is applied from the embodiments of the present disclosure described below.

[0017] As an example of the present disclosure, a method of operating a first device in a wireless communication system includes the steps of receiving control information for transmission from a second device, generating transmission data based on the control information, generating redundancy data for at least one of error checking and error correction for the transmission data, and transmitting the transmission data and the redundancy data, wherein the transmission data includes a semantic representation (SR) for semantic communication, and the redundancy data can be generated by a generation function selected from among a plurality of generation functions based on a task to be performed by the second device.

[0018] As an example of the present disclosure, a method of operating a second device in a wireless communication system includes the steps of transmitting control information to a first device, receiving transmission data and redundancy data for the transmission data from the first device, generating redundancy data based on the received transmission data, and checking an error in the transmission data based on a difference between the received redundancy data and the generated redundancy data, wherein the transmission data includes a semantic representation (SR) for semantic communication, and the generated redundancy data can be generated by a generation function selected from among a plurality of generation functions based on a task to be performed by the second device.

[0019] As an example of the present disclosure, in a wireless communication system, a first device includes a transceiver and a processor connected to the transceiver, wherein the processor receives control information for transmission from a second device, generates transmission data based on the control information, generates redundancy data for at least one of error checking and error correction for the transmission data, and controls transmission of the transmission data and the redundancy data, wherein the transmission data includes a semantic representation (SR) for semantic communication, and the redundancy data can be generated by a generation function selected from among a plurality of generation functions based on a task to be performed by the second device.

[0020] As an example of the present disclosure, in a wireless communication system, a second device includes a transceiver and a processor connected to the transceiver, wherein the processor transmits control information to a first device, receives transmission data and redundancy data for the transmission data from the first device, generates redundancy data based on the received transmission data, and checks for an error in the transmission data based on a difference between the received redundancy data and the generated redundancy data, wherein the transmission data includes a semantic representation (SR) for semantic communication, and the generated redundancy data can be generated by a generation function selected from among a plurality of generation functions based on a task to be performed by the second device.

[0021] As an example of the present disclosure, a communication device includes at least one processor, and at least one computer memory coupled to the at least one processor and storing instructions that direct operations when executed by the at least one processor, the operations including: receiving control information for transmission from another communication device; generating transmission data based on the control information; generating redundancy data for at least one of error checking and error correction for the transmission data; and transmitting the transmission data and the redundancy data, wherein the transmission data includes a semantic representation (SR) for semantic communication, and the redundancy data can be generated by a generation function selected from among a plurality of generation functions based on a task to be performed in the another communication device.

[0022] As an example of the present disclosure, a non-transitory computer-readable medium storing at least one instruction includes at least one instruction executable by a processor, wherein the at least one instruction controls a device to receive control information for transmission from a second device, generate transmission data based on the control information, generate redundancy data for at least one of error checking and error correction for the transmission data, and transmit the transmission data and the redundancy data, wherein the transmission data includes a semantic representation (SR) for semantic communication, and the redundancy data can be generated by a generation function selected from among a plurality of generation functions based on a task to be performed by the second device.

[0023] The above-described aspects of the present disclosure are only some of the preferred embodiments of the present disclosure, and various embodiments reflecting the technical features of the present disclosure can be derived and understood by a person having ordinary skill in the art based on the detailed description of the present disclosure to be described below.

[0024] The following effects may be achieved by embodiments based on the present disclosure.

[0025] According to the present disclosure, the reliability of semantic communication can be improved.

[0026] The effects that can be obtained from the embodiments of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned can be clearly derived and understood by those skilled in the art to which the technical configuration of the present disclosure is applied, from the description of the embodiments of the present disclosure below. In other words, unintended effects resulting from implementing the configuration described in the present disclosure can also be derived from the embodiments of the present disclosure by those skilled in the art.

[0027] The accompanying drawings are intended to aid understanding of the present disclosure and, together with detailed descriptions, may provide embodiments of the present disclosure. However, the technical features of the present disclosure are not limited to specific drawings, and the features disclosed in each drawing may be combined with each other to form new embodiments. Reference numerals in each drawing may indicate structural elements.

[0028] Figure 1 illustrates an example of a communication system applicable to the present disclosure.

[0029] FIG. 2 illustrates an example of a wireless device applicable to the present disclosure.

[0030] FIG. 3 illustrates a method for processing a transmission signal applicable to the present disclosure.

[0031] Figure 4 illustrates a communication procedure between a terminal and a base station applicable to the present disclosure.

[0032] FIG. 5 illustrates an example of a communication structure that can be provided in a 6G (6th generation) system applicable to the present disclosure.

[0033] Figure 6 illustrates an electromagnetic spectrum applicable to the present disclosure.

[0034] FIG. 7 illustrates a THz wireless communication transceiver applicable to the present disclosure.

[0035] Figure 8 illustrates a THz signal generation method applicable to the present disclosure.

[0036] FIG. 9 illustrates a wireless communication transceiver applicable to the present disclosure.

[0037] Figure 10 illustrates a transmitter structure applicable to the present disclosure.

[0038] Figure 11 illustrates a modulator structure applicable to the present disclosure.

[0039] Figure 12 illustrates the structure of a perceptron included in an artificial neural network applicable to the present disclosure.

[0040] Figure 13 illustrates an artificial neural network structure applicable to the present disclosure.

[0041] Figure 14 illustrates an example of a functional framework for application of artificial intelligence technology applicable to the present disclosure.

[0042] Figure 15 illustrates an example of a procedure for utilizing an artificial intelligence model applicable to the present disclosure.

[0043] Figure 16 illustrates another example of a procedure for utilizing an artificial intelligence model applicable to the present disclosure.

[0044] Figure 17 illustrates another example of a procedure for utilizing an artificial intelligence model applicable to the present disclosure.

[0045] Figure 18 illustrates an AI technology-based communication procedure applicable to the present disclosure.

[0046] Figure 19 illustrates an applicable communication model of the present disclosure.

[0047] Figure 20 illustrates an example framework of semantic communication applicable to the present disclosure.

[0048] Figure 21 illustrates an example of a semantic communication error.

[0049] Figures 22a and 22b illustrate examples of a semantic error checking structure based on a deep neural network (DNN).

[0050] Figure 23 illustrates an example of a semantic error correction technique based on a generative adversarial network (GAN).

[0051] Figure 24 shows an example of stochastic neighbor embedding (SNE) for the MNIST data set.

[0052] Figure 25 shows an example of SNE for the MSTAR data set.

[0053] FIG. 26 illustrates an example of semantic error detection and correction according to one embodiment of the present disclosure.

[0054] FIG. 27 illustrates an example of a structure for transmitting and receiving a semantic expression according to one embodiment of the present disclosure.

[0055] FIG. 28a illustrates an example of a structure for detecting errors in semantic expressions according to one embodiment of the present disclosure.

[0056] FIG. 28b illustrates an example of a structure for correcting errors in semantic expressions according to one embodiment of the present disclosure.

[0057] FIG. 29 illustrates an example of a procedure for transmitting a semantic expression according to one embodiment of the present disclosure.

[0058] FIG. 30 illustrates an example of a procedure for receiving a semantic representation according to one embodiment of the present disclosure.

[0059] FIG. 31 illustrates an example of a procedure for detecting and correcting errors in semantic expressions according to one embodiment of the present disclosure.

[0060] FIG. 32 illustrates an example of error correction of a semantic expression according to one embodiment of the present disclosure.

[0061] FIG. 33 illustrates an example of a signal exchange for detecting and correcting semantic errors according to one embodiment of the present disclosure.

[0062] The following embodiments combine the components and features of the present disclosure in a predetermined form. Each component or feature may be considered optional unless explicitly stated otherwise. Each component or feature may be implemented without being combined with other components or features. Furthermore, some components and / or features may be combined to form embodiments of the present disclosure. The order of operations described in the embodiments of the present disclosure may be changed. Some components or features of one embodiment may be included in another embodiment or may be replaced with corresponding components or features of another embodiment.

[0063] In the description of the drawings, procedures or steps that may obscure the gist of the present disclosure are not described, and procedures or steps that can be understood by a person skilled in the art are also not described.

[0064] Throughout the specification, when a part is said to "comprising" or "including" a component, this does not mean that other components may be included, but rather that other components may be excluded, unless otherwise specifically stated. In addition, terms such as "...part," "...unit," and "module" described in the specification mean a unit that processes at least one function or operation, which may be implemented by hardware, software, or a combination of hardware and software. In addition, the words "a" or "an," "one," "the," and similar related words may be used in the context of describing the present disclosure (especially in the context of the claims below) to include both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context.

[0065] Embodiments of the present disclosure described herein focus on the data transmission and reception relationship between a base station and a mobile station. Here, the base station is understood as a terminal node of a network that directly communicates with the mobile station. Certain operations described herein as being performed by the base station may, in some cases, be performed by an upper node of the base station.

[0066] That is, in a network consisting of multiple network nodes including a base station, various operations performed for communication with a mobile station may be performed by the base station or other network nodes other than the base station. In this case, the term 'base station' may be replaced by terms such as fixed station, Node B, eNB (eNode B), gNB (gNode B), ng-eNB, advanced base station (ABS), or access point.

[0067] Additionally, in the embodiments of the present disclosure, the term terminal may be replaced with terms such as user equipment (UE), mobile station (MS), subscriber station (SS), mobile subscriber station (MSS), mobile terminal, or advanced mobile station (AMS).

[0068] Additionally, a transmitter refers to a fixed and / or mobile node that provides data or voice services, and a receiver refers to a fixed and / or mobile node that receives data or voice services. Therefore, for uplink, a mobile station can be the transmitter, and a base station can be the receiver. Similarly, for downlink, a mobile station can be the receiver, and a base station can be the transmitter.

[0069] Embodiments of the present disclosure may be supported by standard documents disclosed in at least one of wireless access systems, such as IEEE 802.xx system, 3rd Generation Partnership Project (3GPP) system, 3GPP Long Term Evolution (LTE) system, 3GPP 5th generation (5G) NR (New Radio) system and 3GPP2 system, and in particular, embodiments of the present disclosure may be supported by 3GPP TS (technical specification) 38.211, 3GPP TS 38.212, 3GPP TS 38.213, 3GPP TS 38.321 and 3GPP TS 38.331 documents.

[0070] Furthermore, the embodiments of the present disclosure can be applied to other wireless access systems and are not limited to the systems described above. For example, they can be applied to systems implemented after the 3GPP 5G NR system and are not limited to a specific system.

[0071] That is, obvious steps or parts not described in the embodiments of the present disclosure can be explained by referring to the above documents. In addition, all terms disclosed in this document can be explained by the above standard documents.

[0072] Hereinafter, preferred embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. The detailed description set forth below, together with the accompanying drawings, is intended to illustrate exemplary embodiments of the present disclosure and is not intended to represent the only embodiments in which the technical configurations of the present disclosure may be implemented.

[0073] Additionally, specific terms used in the embodiments of the present disclosure are provided to aid in understanding of the present disclosure, and the use of such specific terms may be changed to other forms without departing from the technical spirit of the present disclosure.

[0074] The following technology can be applied to various wireless access systems such as CDMA (code division multiple access), FDMA (frequency division multiple access), TDMA (time division multiple access), OFDMA (orthogonal frequency division multiple access), and SC-FDMA (single carrier frequency division multiple access).

[0075] For clarity, the following description is based on 3GPP communication systems (e.g., LTE, NR, etc.), but the technical spirit of the present disclosure is not limited thereto. LTE may refer to technology after 3GPP TS 36.xxx Release 8. Specifically, LTE technology after 3GPP TS 36.xxx Release 10 may be referred to as LTE-A, and LTE technology after 3GPP TS 36.xxx Release 13 may be referred to as LTE-A pro. 3GPP NR may refer to technology after TS 38.xxx Release 15. 3GPP 6G may refer to technology after TS Release 17 and / or Release 18. "xxx" refers to a standard document detail number. LTE / NR / 6G may be collectively referred to as a 3GPP system.

[0076] For background information, terms, abbreviations, etc. used in this disclosure, reference may be made to standard documents published prior to this disclosure. For example, reference may be made to standard documents 36.xxx and 38.xxx.

[0077] Communication system applicable to the present disclosure

[0078] Although not limited thereto, the various descriptions, functions, procedures, proposals, methods and / or operational flowcharts of the present disclosure disclosed in this document may be applied to various fields requiring wireless communication / connectivity (e.g., 5G) between devices.

[0079] Hereinafter, more specific examples will be provided with reference to the drawings. In the drawings / descriptions below, the same drawing reference numerals may represent identical or corresponding hardware blocks, software blocks, or functional blocks, unless otherwise described.

[0080] Figure 1 illustrates an example of a communication system applied to the present disclosure.

[0081] Referring to FIG. 1, a communication system (100) applied to the present disclosure includes a wireless device, a base station, and a network. Here, the wireless device refers to a device that performs communication using a wireless access technology (e.g., LTE, LTE-A, LTE-A pro, NR, 5G, 5G-A, 6G) and may be referred to as a communication / wireless / 5G device. Although not limited thereto, the wireless device may include a robot (100a), a vehicle (100b-1, 100b-2), an XR (extended reality) device (100c), a hand-held device (100d), a home appliance (100e), an IoT (Internet of Things) device (100f), and an AI (artificial intelligence) device / server (100g). For example, the vehicle may include a vehicle equipped with a wireless communication function, an autonomous vehicle, a vehicle capable of performing vehicle-to-vehicle communication, etc. Here, the vehicles (100b-1, 100b-2) may include unmanned aerial vehicles (UAVs) (e.g., drones). The XR devices (100c) include augmented reality (AR) / virtual reality (VR) / mixed reality (MR) devices, and may be implemented in the form of head-mounted devices (HMDs), head-up displays (HUDs) installed in vehicles, televisions, smartphones, computers, wearable devices, home appliances, digital signage, vehicles, robots, etc. The portable devices (100d) may include smartphones, smart pads, wearable devices (e.g., smartwatches, smart glasses), computers (e.g., laptops, etc.), etc. The home appliances (100e) may include TVs, refrigerators, washing machines, etc. The IoT devices (100f) may include sensors, smart meters, etc.For example, the base station (120) and the network (130) may also be implemented as wireless devices, and a specific wireless device (120a) may act as a base station / network node to other wireless devices.

[0082] Wireless devices (100a to 100f) can be connected to a network (130) via a base station (120). AI technology can be applied to the wireless devices (100a to 100f), and the wireless devices (100a to 100f) can be connected to an AI server (100g) via a network (130). The network (130) can be configured using a 3G network, a 4G (e.g., LTE) network, a 5G (e.g., NR), or a 6G network. The wireless devices (100a to 100f) can communicate with each other via the base station (120) / network (130), but can also communicate directly (e.g., sidelink communication) without going through the base station (120) / network (130). For example, vehicles (100b-1, 100b-2) can communicate directly (e.g., V2V (vehicle to vehicle) / V2X (vehicle to everything) communication). Additionally, an IoT device (100f) (e.g., a sensor) can communicate directly with another IoT device (e.g., a sensor) or another wireless device (100a to 100f).

[0083] Wireless communication / connection (150a, 150b, 150c) can be established between wireless devices (100a to 100f) / base stations (120), and base stations (120) / base stations (120). Here, the wireless communication / connection can be established through various wireless access technologies such as uplink / downlink communication (150a), sidelink communication (150b) (or D2D communication), and base station-to-base station communication (150c) (e.g., relay, IAB (integrated access backhaul)). Through the wireless communication / connection (150a, 150b, 150c), the wireless device and base station / wireless device, and base stations and base stations can transmit / receive wireless signals to / from each other. For example, the wireless communication / connection (150a, 150b, 150c) can transmit / receive signals through various physical channels. To this end, based on various proposals of the present disclosure, at least some of various configuration information setting processes for transmitting / receiving wireless signals, various signal processing processes (e.g., channel encoding / decoding, modulation / demodulation, resource mapping / demapping, etc.), resource allocation processes, etc. may be performed.

[0084] Devices applicable to the present disclosure

[0085] FIG. 2 illustrates an example of a wireless device applicable to the present disclosure.

[0086] Referring to FIG. 2, the wireless device (200) can transmit and receive wireless signals via various wireless access technologies (e.g., LTE, LTE-A, LTE-A pro, NR, 5G, 5G-A, 6G). The wireless device (200) includes at least one processor (202) and at least one memory (204), and may additionally include at least one transceiver (206) and / or at least one antenna (208).

[0087] The processor (202) controls the memory (204) and / or the transceiver (206), and may be configured to implement the descriptions, functions, procedures, proposals, methods, and / or operational flowcharts disclosed in this document. For example, the processor (202) may process information in the memory (204) to generate first information / signal, and then transmit a wireless signal including the first information / signal via the transceiver (206). In addition, the processor (202) may receive a wireless signal including second information / signal via the transceiver (206), and then store information obtained from signal processing of the second information / signal in the memory (204). The memory (204) may be connected to the processor (202) and may store various information related to the operation of the processor (202). For example, the memory (204) may store software code including instructions for performing some or all of the processes controlled by the processor (202), or for performing the descriptions, functions, procedures, proposals, methods, and / or operational flowcharts disclosed herein. Here, the processor (202) and the memory (204) may be part of a communication modem / circuit / chip designed to implement wireless communication technology. The transceiver (206) may be connected to the processor (202) and may transmit and / or receive wireless signals via at least one antenna (208). The transceiver (206) may include a transmitter and / or a receiver. The transceiver (206) may be used interchangeably with an RF (radio frequency) unit. In the present disclosure, a wireless device may also mean a communication modem / circuit / chip.

[0088] Hereinafter, the hardware elements of the wireless device (200) will be described in more detail. Although not limited thereto, at least one protocol layer may be implemented by at least one processor (202). For example, at least one processor (202) may implement at least one layer (e.g., a functional layer such as physical (PHY), media access control (MAC), radio link control (RLC), packet data convergence protocol (PDCP), radio resource control (RRC), and service data adaptation protocol (SDAP)). At least one processor (202) may generate at least one Protocol Data Unit (PDU) and / or at least one Service Data Unit (SDU) according to the descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this document. At least one processor (202) may generate a message, control information, data, or information according to the descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this document. At least one processor (202) can generate a signal (e.g., a baseband signal) including a PDU, an SDU, a message, control information, data or information according to the functions, procedures, proposals and / or methods disclosed in this document, and provide the signal to at least one transceiver (206). At least one processor (202) can receive a signal (e.g., a baseband signal) from at least one transceiver (206) and obtain the PDU, SDU, message, control information, data or information according to the descriptions, functions, procedures, proposals, methods and / or operational flowcharts disclosed in this document.

[0089] At least one processor (202) may be referred to as a controller, a microcontroller, a microprocessor, or a microcomputer. The at least one processor (202) may be implemented by hardware, firmware, software, or a combination thereof. For example, at least one application specific integrated circuit (ASIC), at least one digital signal processor (DSP), at least one digital signal processing device (DSPD), at least one programmable logic device (PLD), or at least one field programmable gate array (FPGA) may be included in the at least one processor (202). The descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this document may be implemented using firmware or software, and the firmware or software may be implemented to include modules, procedures, functions, etc. The descriptions, functions, procedures, proposals, methods, and / or operation flowcharts disclosed in this document may be included in the at least one processor (202), or may be stored in at least one memory (204) and driven by the at least one processor (202). The descriptions, functions, procedures, suggestions, methods and / or flowcharts disclosed in this document may be implemented using firmware or software in the form of code, instructions and / or sets of instructions.

[0090] At least one memory (204) can be connected to at least one processor (202) and can store various forms of data, signals, messages, information, programs, codes, instructions and / or commands. The at least one memory (204) can be configured as a read only memory (ROM), a random access memory (RAM), an erasable programmable read only memory (EPROM), a flash memory, a hard drive, a register, a cache memory, a computer readable storage medium and / or a combination thereof. The at least one memory (204) can be located internally and / or externally to the at least one processor (202). In addition, the at least one memory (204) can be connected to the at least one processor (202) via various technologies such as a wired or wireless connection.

[0091] At least one transceiver (206) can transmit user data, control information, wireless signals / channels, etc., mentioned in the methods and / or flowcharts of this document to at least one other device. At least one transceiver (206) can receive user data, control information, wireless signals / channels, etc. mentioned in the descriptions, functions, procedures, proposals, methods and / or flowcharts disclosed in this document from at least one other device. For example, at least one transceiver (206) can be connected to at least one processor (202) and can transmit and receive wireless signals. For example, at least one processor (202) can control at least one transceiver (206) to transmit user data, control information, or wireless signals to at least one other device. Furthermore, at least one processor (202) can control at least one transceiver (206) to receive user data, control information, or wireless signals from at least one other device. In addition, at least one transceiver (206) may be connected to at least one antenna (208), and at least one transceiver (206) may be configured to transmit and receive user data, control information, wireless signals / channels, etc. mentioned in the descriptions, functions, procedures, proposals, methods and / or operation flowcharts disclosed in this document through at least one antenna (208). In this document, at least one antenna may be a plurality of physical antennas or a plurality of logical antennas (e.g., antenna ports). At least one transceiver (206) may convert the received wireless signals / channels, etc. from RF band signals to baseband signals in order to process the received user data, control information, wireless signals / channels, etc. using at least one processor (202). At least one transceiver (206) may convert the processed user data, control information, wireless signals / channels, etc. from baseband signals to RF band signals using at least one processor (202).For this purpose, at least one transceiver (206) may include an (analog) oscillator and / or filter.

[0092] The components of the wireless device described with reference to FIG. 2 may be referred to by different terms in terms of functionality. For example, the processor (202) may be referred to as a control unit, the transceiver (206) as a communication unit, and the memory (204) as a storage unit. In some cases, the communication unit may be used to mean at least a portion of the processor (202) and the transceiver (206).

[0093] The structure of the wireless device described with reference to FIG. 2 can be understood as the structure of at least a portion of various devices. For example, the structure of the wireless device illustrated in FIG. 2 can be at least a portion of various devices described with reference to FIG. 1 (e.g., a robot (100a), a vehicle (100b-1, 100b-2), an XR device (100c), a portable device (100d), a home appliance (100e), an IoT device (100f), an AI device / server (100g)). Furthermore, according to various embodiments, in addition to the components illustrated in FIG. 2, the device may further include other components.

[0094] For example, the device may be a portable device such as a smartphone, a smart pad, a wearable device (e.g., a smart watch, smart glasses), or a portable computer (e.g., a laptop, etc.). In this case, the device may further include at least one of a power supply unit that supplies power and includes a wired / wireless charging circuit, a battery, etc., an interface unit that includes at least one port for connection with another device (e.g., an audio input / output port, a video input / output port), and an input / output unit for inputting and outputting image information / signals, audio information / signals, data, and / or information input from a user.

[0095] For example, the device may be a mobile device such as a mobile robot, a vehicle, a train, an aerial vehicle (AV), a ship, etc. In this case, the device may further include at least one of a driving unit including at least one of an engine, a motor, a power train, wheels, brakes, and a steering unit of the device, a power supply unit including a wired / wireless charging circuit, a battery, etc. that supplies power, a sensor unit that senses status information, environmental information, and user information of the device or its surroundings, an autonomous driving unit that performs functions such as path maintenance, speed control, and destination setting, and a position measurement unit that obtains location information of the mobile device through a global positioning system (GPS) and various sensors.

[0096] For example, the device may be an XR device such as an HMD, a head-up display (HUD) installed in a vehicle, a television, a smartphone, a computer, a wearable device, a home appliance, a digital signage, a vehicle, a robot, etc. In this case, the device may further include at least one of a power supply unit that supplies power and includes a wired / wireless charging circuit, a battery, etc., an input / output unit that obtains control information, data, etc. from the outside and outputs the generated XR object, and a sensor unit that senses status information, environmental information, and user information of the device or the surroundings of the device.

[0097] For example, the device may be a robot that can be classified into industrial, medical, household, military, etc. types depending on the purpose or field of use. In this case, the device may further include at least one of a sensor unit that senses status information, environmental information, and user information of the device or its surroundings, and a driving unit that performs various physical actions, such as moving the robot joints.

[0098] For example, the device may be an AI device such as a TV, a projector, a smartphone, a PC, a laptop, a digital broadcasting terminal, a tablet PC, a wearable device, a set-top box (STB), a radio, a washing machine, a refrigerator, digital signage, a robot, a vehicle, etc. In this case, the device may further include at least one of an input unit that acquires various types of data from the outside, an output unit that generates output related to sight, hearing, or touch, a sensor unit that senses status information, environmental information, and user information of the device or its surroundings, and a training unit that trains a model composed of an artificial neural network using learning data.

[0099] The structure of the wireless device illustrated in FIG. 2 may be understood as a part of a RAN node (e.g., a base station, DU, RU, RRH, etc.). That is, the device illustrated in FIG. 2 may be a RAN node. In this case, the device may further include a wired transceiver for front haul and / or back haul communications. However, if the front haul and / or back haul communications are based on wireless communications, at least one transceiver (206) illustrated in FIG. 2 may be used for front haul and / or back haul communications, and a wired transceiver may not be included.

[0100] FIG. 3 illustrates a method for processing a transmission signal applicable to the present disclosure. For example, the transmission signal may be processed by a signal processing circuit. At this time, the signal processing circuit (300) may include a scrambler (310), a modulator (320), a layer mapper (330), a precoder (340), a resource mapper (350), and a signal generator (360). At this time, as an example, the operations / functions of FIG. 3 may be performed in the processor (202) and / or the transceiver (206) of FIG. 2. Furthermore, as an example, the hardware elements of FIG. 3 may be implemented in the processor (202) and / or the transceiver (206) of FIG. 2. As an example, blocks 310 to 360 may be implemented in the processor (202) of FIG. 2. Additionally, blocks 310 to 350 may be implemented in the processor (202) of FIG. 2, and block 360 may be implemented in the transceiver (206) of FIG. 2, and are not limited to the above-described embodiment.

[0101] The codeword can be converted into a wireless signal through the signal processing circuit (300) of FIG. 3. Here, the codeword is an encoded bit sequence of an information block. The information block may include a transport block (e.g., an UL-SCH transport block, a DL-SCH transport block). The wireless signal may be transmitted through various physical channels (e.g., a PUSCH, a PDSCH). Specifically, the codeword can be converted into a bit sequence scrambled by a scrambler (310). The scramble sequence used for scrambling is generated based on an initialization value, and the initialization value may include ID information of the wireless device, etc. The scrambled bit sequence can be modulated into a modulation symbol sequence by a modulator (320). The modulation scheme may include pi / 2-binary phase shift keying (pi / 2-BPSK), m-phase shift keying (m-PSK), m-quadrature amplitude modulation (m-QAM), etc.

[0102] A complex modulation symbol sequence can be mapped to at least one transmission layer by a layer mapper (330). The modulation symbols of each transmission layer can be mapped to corresponding antenna port(s) by a precoder (340). The output z of the precoder (340) can be obtained by multiplying the output y of the layer mapper (330) by an N×M precoding matrix W. Here, N is the number of antenna ports, and M is the number of transmission layers. Here, the precoder (340) can perform precoding after performing transform precoding (e.g., discrete Fourier transform (DFT) transform) on the complex modulation symbols. Additionally, the precoder (340) can perform precoding without performing transform precoding.

[0103] The resource mapper (350) can map modulation symbols of each antenna port to time-frequency resources. The time-frequency resources can include multiple symbols (e.g., CP-OFDMA symbols, DFT-s-OFDMA symbols) in the time domain and multiple subcarriers in the frequency domain. The signal generator (360) generates a wireless signal from the mapped modulation symbols, and the generated wireless signal can be transmitted to another device through each antenna. To this end, the signal generator (360) can include an inverse fast Fourier transform (IFFT) module, a cyclic prefix (CP) inserter, a digital-to-analog converter (DAC), a frequency uplink converter, and the like.

[0104] The signal processing process for a received signal in a wireless device may be configured in reverse order of the signal processing process (310 to 360) of FIG. 3. For example, a wireless device (e.g., 200 of FIG. 2) may receive a wireless signal from the outside through an antenna port / transceiver. The received wireless signal may be converted into a baseband signal through a signal restorer. For this purpose, the signal restorer may include a frequency downlink converter, an analog-to-digital converter (ADC), a CP remover, and a fast Fourier transform (FFT) module. Thereafter, the baseband signal may be restored to a codeword through a resource demapper process, a postcoding process, a demodulation process, and a descrambling process. The codeword may be restored to the original information block through decoding. Therefore, a signal processing circuit (not shown) for a received signal may include a signal restorer, a resource demapper, a postcoder, a demodulator, a descrambler, and a decoder.

[0105] Figure 4 illustrates a communication procedure between a terminal and a base station applicable to the present disclosure. Figure 4 illustrates operations of a terminal (410) and a base station (420) transmitting and / or receiving data and operations performed prior thereto.

[0106] Referring to FIG. 4, in step 401, the terminal (410) and the base station (420) perform synchronization. For example, the terminal (410) performs an initial cell search operation. Specifically, the terminal (410) can detect at least one synchronization signal transmitted from the base station (420) according to a predefined rule. Here, the synchronization signal can include multiple synchronization signals classified according to structure or purpose (e.g., primary synchronization signal, secondary synchronization signal). Through this, the terminal (410) can check the boundary of the frame, subframe, slot, and / or symbol of the base station (420) and obtain information about the base station (420) (e.g., cell identifier).

[0107] In step 403, the terminal (410) obtains system information transmitted from the base station (420). The system information is information related to the properties, characteristics, and / or capabilities of the base station (420) required to access the base station (420) and use the service, and may be classified according to the content (e.g., whether it is essential for access), transmission structure (e.g., channel used, whether provided on-demand), etc., and may be classified into, for example, a master information block (MIB) and a system information block (SIB). If necessary, the terminal (410) may transmit a signal requesting system information before receiving the system information. However, the request and provision of the system information may be performed after the random access procedure described below.

[0108] In step 405, the terminal (410) and the base station (420) perform a random access procedure. The terminal (410) may transmit and / or receive at least one message (e.g., a random access preamble, a random access response (RAR) message, etc.) for the random access procedure based on information related to the random access channel of the base station (420) obtained through system information (e.g., channel position, channel structure, supported preamble structure, etc.). For example, the terminal (410) may transmit a preamble (e.g., MSG1) through the random access channel, receive an RAR message (e.g., MSG2), transmit a message (e.g., MSG3) including information related to the terminal (410) (e.g., identification information) to the base station (420) using scheduling information included in the RAR message, and receive a message (e.g., MSG4) for contention resolution and / or connection establishment. As another example, MSG1 and MSG3 may be sent and received as one message, or MSG2 and MSG4 may be sent and received as one message.

[0109] In step 407, the terminal (410) and the base station (420) perform signaling of control information. Here, the control information may be defined in various layers, such as a layer that controls a connection (e.g., a radio resource control (RRC) layer), a layer that handles mapping between logical channels and transport channels (e.g., a media access control (MAC) layer), and a layer that handles physical channels (e.g., a physical (PHY) layer). For example, the terminal (410) and the base station (420) may perform at least one of signaling for establishing a connection, signaling for determining settings related to communication, and signaling for indicating allocated resources.

[0110] In step 409, the terminal (410) and the base station (420) transmit and / or receive data. In other words, the terminal (410) and the base station (420) can process, transmit, and / or receive data based on the signaling of the control information. For example, when transmitting data, the terminal (410) or the base station (420) can perform at least one of channel encoding, rate matching, scrambling, constellation mapping, layer mapping, waveform modulation, antenna mapping, and resource mapping on the information bits. Conversely, when receiving data, the terminal (410) or the base station (420) can perform at least one of signal extraction from resources, waveform demodulation for each antenna, signal arrangement considering layer mapping, constellation demapping, descrambling, and channel decoding.

[0111] 6G communication systems and core implementation technologies of 6G systems

[0112] The 5G system defines various operating bands within FR1 (frequency range 1), which covers 410 MHz to 7125 MHz, and FR2 (frequency range 2), which covers 24,250 MHz to 71,000 MHz. Various frequencies are being discussed as operating bands for the subsequent 6G system, and the use of higher frequencies than 5G systems is also being considered for wider bandwidth and higher transmission speeds. One such band is the THz (terahertz) frequency band, which covers approximately 100 GHz to 10 THz. The THz frequency band is a band that has both the transparency of radio waves and the straightness of light waves, and communications using the THz frequency band are expected to play a transitional role from existing radio-centered communications to lightwave-based communications.

[0113] 6G systems utilizing the THz frequency band have the following goals: i) very high data rates per device, ii) a very large number of connected devices, iii) global connectivity, iv) very low latency, v) reduced energy consumption of battery-free IoT devices, vi) ultra-reliable connectivity, and vii) connected intelligence with machine learning capabilities. The vision of the 6G system can be divided into four aspects: “intelligent connectivity,” “deep connectivity,” “holographic connectivity,” and “ubiquitous connectivity,” and the 6G system can be designed to satisfy the requirements as shown in [Table 1] below.

[0114] Per device peak data rate1 TbpsE2E latency1 msMaximum spectral efficiency100 bps / HzMobility supportup to 1000 km / hrSatellite integrationFullyAIFullyAutonomous vehicleFullyXRFullyHaptic CommunicationFully

[0115] At this time, the 6G system may have key factors such as enhanced mobile broadband (eMBB), ultra-reliable low latency communications (URLLC), massive machine type communications (mMTC), AI integrated communication, tactile internet, high throughput, high network capacity, high energy efficiency, low backhaul and access network congestion, and enhanced data security. FIG. 5 illustrates an example of a communication structure that can be provided in a 6G system applicable to the present disclosure. Referring to FIG. 5, the 6G system is expected to have simultaneous wireless communication connectivity that is 50 times higher than that of a 5G wireless communication system. URLLC, a key feature of 5G, is expected to become even more crucial in 6G communications by providing end-to-end latency of less than 1 ms. Furthermore, 6G systems will boast significantly higher volumetric spectral efficiency than the commonly used area spectral efficiency. 6G systems can offer extremely long battery life and advanced battery technologies for energy harvesting, potentially eliminating the need for separate charging for mobile devices in 6G systems.

[0116] As core implementation technologies of the 6G system, technologies such as artificial intelligence (AI), THz (terahertz) communication, optical wireless technology, FSO backhaul network, massive MIMO technology, blockchain, 3D networking, quantum communication, unmanned aerial vehicles, cell-free communication, wireless information and energy transfer (WIET), integration of sensing and communication, integration of access backhaul networks, holographic beamforming, big data analysis, and large intelligent surface (LIS) can be adopted.

[0117] For example, THz communication is a communication that utilizes a spectrum in a frequency band between 0.1 THz and 10 THz with a corresponding wavelength in the range of 0.03 mm to 3 mm as shown in Fig. 6, and can be implemented using circuit elements having a structure as shown in Fig. 7. In addition, optical wireless technology is a technology that generates and modulates THz signals using optical elements, and can be implemented based on devices having structures as shown in Figs. 8, 9, 10, and 11.

[0118] In addition, AI can be implemented based on various models such as neural networks and machine learning (machine models). For example, an AI model of a neural network structure can be based on the structure of a perceptron as shown in Fig. 12. Referring to Fig. 12, an artificial neural network can be composed of multiple perceptrons. According to the structure of the perceptron, when an input vector x={x1, x2, ..., xd} is input, each component is multiplied by a weight {W1, W2, 쪋, Wd}, and all the results are added, and then the activation function σ(·) is applied. A large artificial neural network structure can be formed by extending the simplified perceptron structure illustrated in Fig. 12, and the input vector can be applied to perceptrons of different dimensions. When perceptrons are stacked, a neural network having an input layer, a hidden layer, and an output layer as shown in Fig. 13 can be configured.

[0119] AI technology utilizing the structure of the neural network as described above can be operated based on a functional framework as shown in FIG. 14 below. FIG. 14 illustrates an example of a functional framework for application of AI technology applicable to the present disclosure. First, the data collection block (1410) performs data preparation on input data collected from objects (e.g., UE, RAN node, network node, etc.) to generate training data (1411) and / or inference data (1411) including processed input data. The model training block (1420) performs training on an AI model using the training data (1411) and provides information on the trained model to the model inference block (1430). The model inference block (1430) generates an output (1416) by performing inference and / or prediction using the inference data (1411). Additionally, the model inference block (1430) can provide model performance feedback (1414) to the model training block (1420). Here, the output (1416) refers to the inference output of the AI ​​model generated by the model inference block (1430), and the details of the inference output may vary depending on the use case. The actor block (1440) triggers or performs a specified task / action based on the output (1416). The actor block (1440) can trigger a task / action for another object (e.g., at least one UE, at least one RAN node, at least one network node, etc.) or for itself. Any one of the functions illustrated in FIG. 14 described above may be performed by two or more entities including the RAN, the network node, the network operator's OAM, or the UE in collaboration. This may be referred to as a split AI operation.

[0120] FIG. 15 illustrates an example of a procedure for utilizing an AI model applicable to the present disclosure. FIG. 15 illustrates a case where a model training function (e.g., a function of a model training block (1420)) is included in a network node, and a model inference function (e.g., a function of a model inference block (1430)) is included in a RAN node. Referring to FIG. 15 , in step 1, RAN node 1 and RAN node 2 transmit input data (e.g., training data) for training an AI model to the network node. Here, RAN node 1 and RAN node 2 may also transmit data collected from the UE (e.g., measurements of the UE related to RSRP, RSRQ, SINR of the serving cell and neighboring cells, the UE's position, velocity, etc.) to the network node. In step 2, the network node trains the AI ​​model using the received training data. In step 3, the network node distributes / updates the AI ​​model to RAN node 1 and / or RAN node 2. RAN node 1 and / or RAN node 2 may continue model training based on the received AI model. In this procedure, it is assumed that the AI ​​model is deployed / updated only to RAN node 1. In step 4, RAN node 1 receives input data (e.g., inference data) for AI model inference from the UE and RAN node 2. In step 5, RAN node 1 performs inference based on the AI ​​model using the received inference data to generate output data (e.g., a prediction or a decision). In step 6, if applicable, RAN node 1 may send model performance feedback to network nodes. In step 7, RAN node 1, RAN node 2, and UE (or 'RAN node 1 and UE', or 'RAN node 1 and RAN node 2') perform actions based on the output data. For example, in case of load balancing operation, the UE may move from RAN node 1 to RAN node 2.In step 8, RAN node 1 and RAN node 2 transmit feedback information to the network nodes.

[0121] FIG. 16 illustrates another example of a procedure for utilizing an AI model applicable to the present disclosure. FIG. 16 illustrates a case where a model training function (e.g., a function of a model training block (1420)) and a model inference function (e.g., a function of a model inference block (1430)) are included in a RAN node. Referring to FIG. 16, in step 1, a UE and a RAN node 2 transmit input data (e.g., training data) for training an AI model to a RAN node 1. In step 2, RAN node 1 trains an AI model using the received training data. In step 3, RAN node 1 receives input data (e.g., inference data) for AI model-based inference from the UE and RAN node 2. In step 4, RAN node 1 performs AI model-based inference using the received inference data to generate output data (e.g., a prediction or decision). In step 5, RAN node 1, RAN node 2, and UE (or 'RAN node 1 and UE', or 'RAN node 1 and RAN node 2') perform actions based on the output data. For example, in case of load balancing operation, the UE may move from RAN node 1 to RAN node 2. In step 6, RAN node 2 transmits feedback information to RAN node 1.

[0122] FIG. 17 illustrates another example of a procedure for utilizing an AI model applicable to the present disclosure. FIG. 17 illustrates a case where a model training function (e.g., a function of a model training block (1420)) is included in a RAN node, and a model inference function (e.g., a function of a model inference block (1430)) is included in a UE. Referring to FIG. 17, in step 1, a UE transmits input data (e.g., training data) for training an AI model to a RAN node. Here, the RAN node may collect data from various UEs and / or other RAN nodes. In step 2, the RAN node trains an AI model using the received training data. In step 3, the RAN node distributes / updates the AI ​​model to the UE. The UE may continue model training based on the received AI model. In step 4, input data (e.g., inference data) for inference based on the AI ​​model is received from the UE, the RAN node, and / or other UEs. In step 5, the UE generates output data (e.g., predictions or decisions) by performing AI model-based inference using the received inference data. In step 6, if applicable, the UE may transmit model performance feedback to the RAN node. In step 7, the UE and the RAN node perform actions based on the output data. In step 8, the UE transmits feedback information to the RAN node.

[0123] According to the aforementioned framework and procedures, an AI model can be trained and utilized in a wireless communication system. In the aforementioned framework and procedures, various data, such as input data, training data, and inference data, are introduced. The specific content of the aforementioned data may vary depending on the task for which the AI ​​model is utilized. For example, information used in various embodiments of the present disclosure described below may be included in the aforementioned data.

[0124] Figure 18 illustrates an AI technology-based communication procedure applicable to the present disclosure. The detailed procedures illustrated in Figure 18 can be combined with various embodiments of the present disclosure described below. For example, data generated according to various embodiments of the present disclosure can be used for operations (e.g., configuration, training, inference, and / or data transmission / reception) in at least one of the detailed procedures illustrated in Figure 18. As another example, the results of the inference illustrated in Figure 18 can be used to transmit and / or receive data according to various embodiments of the present disclosure.

[0125] Referring to FIG. 18, in step 1801, at least one of a UE (1810), a RAN node (1820), and a network node (1830) performs an initial access procedure. For example, in this step, at least one of an initial cell search operation, a system information acquisition operation, a random access operation, and a registration operation may be performed. In step 1803, at least one of the UE (1810), the RAN node (1820), and the network node (1830) performs a configuration procedure. Through the configuration procedure, parameters, resources, connections, and / or entities necessary for performing subsequent procedures in layers between the UE (1810) and the RAN node (1820) and / or in at least one layer between the UE (1810) and the network node (1830) may be determined and / or created. In this case, the configuration procedure may be performed based on information, status, and / or characteristics of an AI model used for subsequent training and inference.

[0126] At step 1805, at least one of the UE (1810), the RAN node (1820), and the network node (1830) performs a model training procedure. At least one of the UE (1810), the RAN node (1820), and the network node (1830) may collect training data and perform learning using the training data. For example, the model training procedure may be performed as described with reference to FIG. 15, FIG. 16, or FIG. 17. If an offline-trained model is used, this step may be omitted.

[0127] At step 1807, at least one of the UE (1810), the RAN node (1820), and the network node (1830) performs a task using the trained model. That is, the task may be performed based on the results of inference and / or prediction using the trained model. For example, the task may be a procedure belonging to a communication protocol, and may be a preparatory operation for subsequent data transmission and / or reception, or may be related to data transmission and / or reception, or may be related to data processing (e.g., encoding, decoding, etc.).

[0128] At step 1809, at least one of the UE (1810), the RAN node (1820), and the network node (1830) transmits and / or receives data. At this time, the result of the task performed at step 1807 may be used. In some cases, the task performed at step 1807 may include transmitting and / or receiving data, in which case this step may be omitted as it is part of step 1807.

[0129] Specific embodiments of the present disclosure

[0130] The present disclosure relates to a device and method for detecting semantic errors in a wireless communication system and correcting detected semantic errors. Specifically, the present disclosure relates to a technology for transmitting and receiving semantic representations (SR) and semantic redundancy in a wireless communication system supporting semantic communication, and detecting and correcting semantic errors generated by a channel based on the semantic representation and semantic redundancy. The semantic redundancy may be redundancy data generated for at least one of error checking and error correction for the semantic representation.

[0131] Problems related to communication can be divided into three levels, as illustrated in Figure 19, according to the philosophy of Shannon and Weaver. Figure 19 illustrates a communication model applicable to the present disclosure. The problem at Level A (1910) is a technical problem, which concerns how accurately symbols in communication can be conveyed. The problem at Level B (1920) is a semantic problem, which concerns how accurately the conveyed symbols convey the desired meaning. The problem at Level C (1930) is an effectiveness problem, which concerns how effectively the received meaning influences the operation in the desired manner.

[0132] Shannon's information theory focuses only on technical problems at Level A (1910). However, Weaver argues that Shannon's information theory is general enough to consider problems at Level B (1920) and Level C (1930) by adding semantic transmitters, semantic receivers, and semantic noise to Shannon's communication model.

[0133] Meanwhile, one of the various goals of 6G communication is to enable various new services that interconnect people and machines. Therefore, a semantic communication method that considers the semantic issues of Level B (1920) rather than simply considering the technical issues of Level A (1910) in communication systems needs to be provided. Semantic communication refers to the efficient transmission and reception of semantic information by utilizing common background knowledge between a first device and a second device, respectively corresponding to the source and destination. Referring to the communication model of Figure 11, if the meaning of the intended message sent by the first device, which corresponds to the source, is accurately interpreted by the second device, which corresponds to the destination, then it can be said that correct semantic communication has been performed.

[0134] For semantic communication, a source can generate a semantic representation based on given or collected raw data and transmit the generated semantic representation to a destination. The destination interprets and infers the received semantic representation according to the source's intent. At this time, semantic communication requires an approach not from the perspective of reducing reconstruction errors that occur during the process of restoring the received semantic representation to the original raw data, but from the perspective of ensuring that downstream tasks performed by the destination can operate according to the source's intent using the received semantic representation. Therefore, the destination utilizes its own background knowledge when performing inference operations, and it is desirable that the background knowledge contained in the data transmitted from the source be reflected in the destination's background knowledge to ensure correct interpretation results. In other words, since the destination uses background knowledge to interpret and infer semantic representations, the source and destination must share the same background knowledge.

[0135] For semantic communication, a source can generate a semantic representation containing semantic information about the data it wishes to convey and transmit this representation to a destination. It is crucial for the destination to interpret the semantic information contained in the received semantic representation in accordance with the source's intent. A general framework for this semantic communication can be represented as shown in Figure 20.

[0136] Fig. 20 illustrates an example of a semantic communication framework applicable to the present disclosure. Referring to Fig. 20, a transmitting device (2010) corresponding to a source encodes source data (2001) to be transmitted using a semantic encoder (2013), thereby generating a semantic representation (SR) (2015) containing semantic information of the source data (2011). At this time, the semantic encoder (2013) can generate a semantic representation (2015) for the source data (2011) using background knowledge (2001) shared with the receiving side (Rx). The transmitting device (2010) performs channel encoding on the semantic representation (2015), and then transmits a signal including the channel-encoded semantic representation (2015) to a receiving device (2020) corresponding to a destination through a wireless channel. A receiving device (2020) obtains a semantic representation (2021) by performing channel decoding on a signal received through a wireless channel. The receiving device (2020) restores data corresponding to the semantic representation (2021) using a semantic decoder (2023) and performs a downstream task, which is a task to be ultimately solved, based on the restored data. The semantic decoder (2023) uses background knowledge (2001) shared with the transmitting device (2010) to interpret and reason the semantic representation (2021) in accordance with the intent of the transmitting device (2010).

[0137] In semantic communication, messages received at the destination may contain errors due to channel noise between the source and destination. Errors can occur at either the technical or semantic level. The technical level seeks to preserve the syntactic integrity of the transmitted and received messages. Therefore, errors at the technical level can be defined as syntactic differences between the transmitted and received messages. In contrast, the semantic level does not seek to preserve the syntactic integrity of the transmitted and received messages, but rather seeks to infer similarities between the semantics of the transmitted and received messages. Therefore, errors at the semantic level can be defined as semantic differences between the transmitted and received messages.

[0138] An example of an error is as follows. For example, if a source sends the sentence "Monday ∨ Tuesday ∨ Wednesday ∨ Thursday ∨ Friday" and the destination receives the sentence "￢Saturday ∧ ￢Sunday", the destination can obtain the same semantics based on its background knowledge of the days of the week. In this case, semantic similarity can be understood as being maintained. However, since a syntactic difference has occurred in terms of syntactic preservation, it can be interpreted as a technical error.

[0139] Another example of an error is shown in Figure 21. Figure 21 illustrates an example of a semantic communication error. Figure 21 exemplifies a case where the source is a human and conveys a message through speech. Referring to Figure 21, the source attempts to transmit the message "copy machine" to the destination using sound. However, since the pronunciation of "p" is similar to the pronunciation of "ff," the source may transmit "coffee machine" instead of the intended message. This can be viewed as a semantic error caused by the source performing an incorrect transmission. To prevent this type of semantic error, the source can transmit a different word, "Xerox," which has the same meaning, instead of the word "copy machine." In this case, since the sound is prevented from being mispronounced, the occurrence of a semantic error can be prevented.

[0140] As mentioned above, semantic errors can cause semantic mismatches between the source and destination during semantic communication. To ensure reliable semantic communication, the destination must detect semantic mismatches using semantic error checking, correct semantic errors, or request retransmission of semantic information to the source, thereby obtaining the accurate semantic information intended by the source. Accordingly, deep neural network (DNN)-based semantic encoding / decoding methods for semantic communication and techniques for checking semantic errors have been actively researched recently. Specifically, research is underway to extend classical information theory, such as the definition of semantic entropy and semantic mutual information, to semantic information theory. Additionally, research is being conducted to define loss functions based on information-theoretic metrics such as broad semantic entropy and channel capacity, build DNN-based semantic encoders / decoders based on these, and check for semantic errors. In other words, the main purpose of checking for semantic errors in semantic communication is not to check for syntactic differences like in conventional technical communication, so an error checking technique different from the existing CRC (cyclic redundancy check) technique needs to be provided.

[0141] Figures 22a and 22b illustrate examples of a DNN-based semantic error checking structure. Figures 22a and 22b illustrate a semantic error checking structure that modifies the structure of a conventional CRC technique. Specifically, in the structures of Figures 22a and 22b, the CRC32 encoder (2210) and the CRC decoder (2230) are replaced with a Sim32 encoder (2211) and a Sim32 decoder (2231), and the source and channel encoders (2220) and the source and channel decoders (2240) are replaced with a DNN-based semantic encoder (2221) and a semantic decoder (2241).

[0142] Referring to Figures 22a and 22b, the operating principle of checking semantic errors based on DNN is as follows. First, the source is a message to be transmitted. Input to each of the Sim32 encoder (2211) and the semantic encoder (2221), and a 32-bit header b from each of the Sim32 encoder (2211) and the semantic encoder (2221) sim and obtain the transmission bit b. The source is a 32-bit header b sim and transmits the transmission bit b to the destination. The destination receives a 32-bit header from the received bit stream. , and input some of the received bit streams with the 32-bit header removed to the semantic decoder (2241), and the decoding result . The destination is the decoding result of the semantic decoder (2241). and 32-bit header can be input to the Sim32 decoder (2230), and whether a semantic error occurs can be checked based on the result value of the Sim32 decoder (2231). The Sim32 decoder (2231) can compare the decoding result value with a specified reference value and output a signal indicating whether an error occurs. For example, the Sim32 decoder (2231) If the decoding result value is greater than 0.5, output ACK 1 indicating no semantic error. If the result value is less than or equal to 0.5, ACK 0 can be output, indicating a semantic error.

[0143] For DNN-based semantic error checking, a semantic encoder (2221) and a semantic decoder (2241) may be trained first, and then a Sim32 encoder (2211) and a Sim32 decoder (2231) may be trained. During the training process, and Since is a known value, it is possible to calculate similarity using BERT (bidirectional encoder representation from transformers). BERT is a pre-trained model using over 1 billion words and sentences, and can extract semantic information from sentences. For example, and The similarity value of It can be calculated as follows. If the similarity value is greater than the reference value of 0.98, that is, If the similarity value is less than or equal to the reference value of 0.98, then training of the Sim32 encoder (2211) and Sim32 decoder (2231) can proceed with label 1. If label 0 is used, training of Sim32 encoder (2211) and Sim32 decoder (2231) can proceed. This process is performed when the destination uses DNN to receive the 32-bit header b sim and the message restored through the semantic decoder (2241). It can be interpreted as checking for semantic errors based on semantic similarity with . In this way, the semantic error checking technique differs from the CRC error checking technique that only checks for bit errors in the existing bit sequence. However, the DNN-based semantic error checking technique is only a technique for checking whether semantic errors have occurred in the received data, and does not perform a technique for correcting semantic errors.

[0144] Accordingly, techniques capable of correcting semantic errors are being studied. Fig. 23 illustrates an example of a semantic error correction technique based on a generative adversarial network (GAN). According to the GAN-based semantic error correction technique illustrated in Fig. 23, a correction network (2303) can be trained with a structure similar to a GAN based on an erroneous feature map containing semantic errors. When an erroneous feature map (2301) is input, the correction network (2303) generates a restored feature map (2305) using an error mask (2311). The restored feature map (2305) is a feature map in which errors corresponding to the locations of the error mask (2311) among the errors in the erroneous feature map (2301) are corrected. In this way, the GAN-based error correction technique can correct semantic errors that occur in a feature map. However, GAN-based error correction techniques do not include a verification procedure to verify whether semantic error correction has been performed appropriately. Specifically, GAN-based error correction techniques assume that semantic errors have been perfectly corrected and use the restored feature map data to perform downstream tasks, such as image reconstruction. Therefore, GAN-based error correction techniques face a limitation in that they cannot detect and correct uncorrected semantic errors in the restored feature map.

[0145] Accordingly, the present disclosure proposes a technique for detecting and correcting semantic errors that may occur in a wireless communication channel in semantic communication. According to various embodiments of the present disclosure, when transmitting a semantic expression, a transmitting device transmits semantic redundancy for the semantic expression together. Semantic redundancy is additional data generated for at least one of error checking and error correction for the semantic expression. By transmitting the semantic redundancy, the receiving device can detect and correct semantic errors. That is, the receiving device can detect and / or correct whether a semantic error occurs in the semantic expression based on the semantic redundancy received together with the semantic expression, and perform a downstream task using the corrected semantic expression.

[0146] To improve the reliability of semantic communication, it is essential to detect and correct semantic errors that may occur on the channel. Therefore, the present disclosure proposes a technique for detecting and correcting semantic errors to improve the reliability of semantic communication.

[0147] The embodiments of the present disclosure described below consider a structure in which a semantic encoder and a semantic decoder are configured as DNNs, i.e., a DNN-based semantic communication system, as illustrated in FIG. 20. According to one embodiment of the present disclosure, the semantic encoder and the semantic decoder may be a single integrated encoder and decoder pair trained based on one task or multiple tasks, or may be multiple encoder and decoder pairs. In the case of a semantic encoder and a semantic decoder trained based on a DNN, feature vectors treated as semantic representations have a property that similar data have a small distance due to the Lipschitz continuity or α-Hold continuity assumption. Lipschitz continuity is defined as follows.

[0148] Given two matrix spaces (X,d X ), (Y,d Y ), given function And if there exists a constant K that satisfies the following [Mathematical Formula 1] for all x1,x2∈X, then the function is referred to as the Lipschitz continuous function.

[0149]

[0150] In [Equation 1], is the Lipschitz continuity function, K is the Lipschitz constant, and are different data included in set X, is the distance function for Y, is a distance function with respect to X.

[0151] According to Lipschitz continuity as in [Mathematical Equation 1], different data Similarity to When defined as, semantic representations of data x1 and x2 This SR space small distance d on the top S If you have, The values ​​can also be considered to have small values. That is, semantic expressions with a short distance between them can be considered to have high similarity. The feature spaces of the MNIST (Modified National Institute of Standards and Technology) and MSTAR (Modified National Institute of Standards and Technology) datasets, which are used as datasets for classification tasks, can be visualized as shown in Figures 24 and 25.

[0152] Figure 24 illustrates an example of stochastic neighbor embedding (SNE) for the MNIST dataset, and Figure 25 illustrates an example of SNE for the MSTAR dataset. Figure 24 shows that data of the same type are included in the same cluster. This means that even if an error occurs in a specific data feature, if the erroneous feature is included in the original cluster, a classification task based on the erroneous feature can be performed. Furthermore, Figure 25 shows that when training on a dataset with fewer types, the cluster area of ​​a specific type is more widely distributed. This means that even if a larger error occurs in a specific data feature, if the erroneous feature is included in the original cluster, the correct downstream task can be performed based on the erroneous feature. In other words, the easier the task, the less sensitive it is to errors, and the more difficult the task, the more sensitive it is to errors.

[0153] The present disclosure applies the aforementioned concepts to semantic communication to interpret semantic errors in semantic expressions. Specifically, the higher the difficulty and / or complexity of a downstream task to be performed, the more sensitive the task is to semantic errors. Therefore, the higher the probability of errors occurring in the results of the downstream task, and semantic errors can be interpreted based on this. The concept of semantic error detection and correction according to various embodiments of the present disclosure can be illustrated as shown in FIG. 26.

[0154] Figure 26 illustrates an example of semantic error detection and correction according to one embodiment of the present disclosure. Referring to Figure 26, a semantic expression received by a receiving device If (2603) is out of the cluster area (2611) of the semantic expression s (2601) transmitted from the transmitting device, the receiving device can determine that an error in the semantic expression, i.e., a semantic error, has occurred. If a semantic error has occurred, the receiving device can determine that the semantic expression (2603) Semantic representation included within the cluster area (2611) (2605) can be corrected.

[0155] As described above, if a semantic expression received by a receiving device deviates from the cluster of the original semantic expression, i.e., the semantic expression transmitted by the transmitting device, the receiving device needs to move the semantic expression that deviates from the cluster of the original semantic expression into the corresponding cluster in order to perform a downstream task. To this end, the receiving device according to an embodiment of the present disclosure can detect whether a semantic error has occurred based on the distance between the received semantic expression and the original semantic expression. In this case, whether a semantic error has occurred based on the distance between the semantic expression and the original semantic expression may vary depending on the task to be performed by the receiving device. For example, if the difficulty and / or complexity of the task to be performed by the receiving device is relatively high, a semantic error may be determined to have occurred even if the distance between the received semantic expression and the original semantic expression is relatively short. On the other hand, if the difficulty and / or complexity of the task to be performed by the receiving device is relatively low, it may be determined that no semantic error has occurred even if the distance between the received semantic expression and the original semantic expression is relatively long. In other words, the semantic error addressed in the present disclosure is different from a syntactic error in the transmitted bit data, and means that a spatial distance movement exceeding the allowable range of errors in the semantic expression occurs when performing the task. Here, the spatial distance may be defined based on at least one of various concepts expressing similarity, and may include, for example, at least one of the shortest distance, cosine similarity, and BERT similarity.

[0156] Accordingly, a transmitting device according to an embodiment of the present disclosure can determine additional information about a semantic expression to be transmitted, i.e., the amount of semantic redundancy (e.g., size or length), based on a task to be performed by a receiving device. The transmitting device can generate semantic redundancy corresponding to the determined amount and transmit the semantic expression and semantic redundancy to the receiving device. In addition, the receiving device can determine whether a semantic error has occurred based on the received semantic expression and semantic redundancy, and if a semantic error has occurred, perform a corrective action for the semantic error by correcting the semantic expression.

[0157] FIG. 27 illustrates an example of a structure for transmitting a semantic expression according to one embodiment of the present disclosure.

[0158] Referring to FIG. 27, the transmitting device (2710) includes a semantic encoder (2711) and a semantic redundancy generator (2713), and the receiving device (2720) includes a semantic decoder (2721).

[0159] The transmitter (2710) uses a semantic encoder (2711) to generate a semantic representation (2703) for source data (2701), such as an image or a sentence. Here, the source data is x, and the semantic encoder is Then, the semantic expression s(2703) is can be expressed as. The transmitting device (2710) generates redundancy data (2705) for the semantic expression (2703) using the semantic redundancy generator (2713). Here, the semantic redundancy generator (2713) executes one generation function for semantic redundancy. The generation function executed in the semantic redundancy generator Then, the semantic redundancy r(2705) is can be expressed as

[0160] According to one embodiment, the semantic redundancy generator (2713) comprises a plurality of generating functions for semantic redundancy. may include. j=1, 2, ..., M. Accordingly, the semantic redundancy generator (2713) may select one generation function to be executed from among the plurality of generation functions based on control information received from the receiving device (2720), that is, information about a task to be performed in the receiving device (2720). For example, the semantic redundancy generator (2713) may determine an index j based on a task to be performed in the receiving device (2720), and select a generation function corresponding to the index j from among the plurality of generation functions. The semantic redundancy generator (2713) may generate redundancy data (2705) for the semantic expression (2703) by executing the selected generation function. Here, the redundancy data (2705) includes information for detecting a distance movement in space that exceeds an allowable range of errors in the semantic expression allowed in performing the corresponding task.

[0161] According to one embodiment, the semantic redundancy generator (2713) can generate semantic redundancy having a length corresponding to the length information of the semantic redundancy received from the receiving device (2720). That is, the length of the semantic redundancy generated by the semantic redundancy generator (2713) can be determined based on the length information of the semantic redundancy received from the receiving device (2720). The transmitting device (2710) transmits the semantic expression (2703) generated by the semantic encoder (2711) and the semantic redundancy (2705) generated by the semantic redundancy generator (2713) to the receiving device (2720).

[0162] A semantic representation (2703) and a semantic redundancy (2705) are received by a receiving device (2720) through a channel. A semantic decoder (2721) of the receiving device (2720) obtains a received semantic representation (2707) corresponding to the semantic representation (2703) and a received semantic redundancy (2709) corresponding to the semantic redundancy (2705). When the semantic representation (2703) passes through the channel, distortion due to the channel may occur. Accordingly, the semantic representation (2703) and the received semantic representation (2707) may have a certain distance in space. The semantic decoder (2721) can perform downstream tasks after error detection and / or correction using the received semantic representation (2707) and semantic redundancy (2709).

[0163] In the description with reference to FIG. 27, the semantic redundancy generator (2713) is designed and / or should be designed to enable detection of semantic errors and correction of detected semantic errors in the receiving device (2720). For example, the semantic redundancy generator (2713) and / or a plurality of generating functions included in the semantic redundancy generator (2713) may be configured to generate a semantic expression received by the receiving device (2720). The semantic expression s transmitted should be designed so that it does not converge to a local optima of a value other than the transmitted semantic expression s. In this case, the semantic error correction performance can be guaranteed in the receiving device (2720). According to one embodiment, the generation function used in the semantic redundancy generator (2713) may be a linear function or a quadrature function. This may be a linear function (e.g., for some matrix ) or theories for obtaining global optima by using convexity in quadrature functions have been well studied. According to one embodiment, the generating function used in the semantic redundancy generator (2713) is a differentiable function, and may be defined as a combination of at least one of arithmetic operations and nonlinear operations for elements included in the semantic expression, or may be defined in the form of a neural network. In other words, the generating function may be a function composed of a differentiable DNN structure and a polynomial of degree 3 or higher.

[0164] FIG. 28a illustrates an example of a structure for detecting errors in semantic expressions according to one embodiment of the present disclosure. The receiving device (2720) of FIG. 28a may be the receiving device (2720) of FIG. 27.

[0165] Referring to FIG. 28a, the receiving device (2720) includes a semantic redundancy generator (2811), a semantic error detector (2813), a semantic error corrector (2815), and a semantic decoder (2817).

[0166] The receiving device (2720) can receive data and redundancy data through a channel, and obtain a received semantic representation (2801) and received redundancy data (2803) for the received semantic representation by performing processing such as channel decoding or demodulation. At this time, the received semantic representation (2801) and the received redundancy data (2805) If so, and It could be.

[0167] The receiving device (2720) regenerates redundancy data (2805) for the received semantic representation (2801) using the semantic redundancy generator (2811). The regenerated redundancy data (2805) If so, Here, the semantic redundancy generator (2811) executes one generation function for semantic redundancy. According to one embodiment, the semantic redundancy generator (2811) executes a plurality of generation functions for semantic redundancy. . In this case, j=1, 2, ..., M. Accordingly, the semantic redundancy generator (2811) may select one generation function to be executed from among the plurality of generation functions based on information about a task to be performed in the receiving device (2720). For example, the semantic redundancy generator (2811) may determine an index j based on a task to be performed in the receiving device (2720), and select a generation function corresponding to the determined index j from among the plurality of generation functions. The semantic redundancy generator (2811) may regenerate redundancy data (2805) for the semantic expression (2801), which is the received data, by executing the selected generation function. According to one embodiment, the semantic redundancy regenerated by the semantic redundancy generator (2801) may have a length corresponding to the length information of the semantic redundancy transmitted to the transmitting device (2710). Here, the semantic redundancy generator (2811) and / or the plurality of generation functions included in the semantic redundancy generator (2811) are identical to the generation functions included in the semantic redundancy generator (2713) and / or the semantic redundancy generator (2713) of the transmitting device (2720).

[0168] The receiving device (2720) detects errors in the received semantic expression (2801) using a semantic error detector (2813). To this end, the semantic error detector (2813) first calculates a difference value between redundancy data received through a channel and redundancy data regenerated by a semantic redundancy generator (2801). For example, the semantic error detector (2813) may be configured to detect errors in an error detection function. Using , received redundancy data and regenerated redundancy data The difference value of Calculate . Is and It is a metric that represents the difference or distance, and can be defined as a differentiable metric. For example, is a metric in terms of mean squared error. , a metric from the cosine similarity perspective. , or a metric from the perspective of sentence similarity. can be defined as. According to one embodiment, can be set based on the task to be performed in the receiving device (2720). Here, calculating the difference value between the received redundancy data and the regenerated data is performed by calculating the semantic expression transmitted from the transmitting device (2710) when no semantic error occurs through the channel. (2703) and semantic representation received from the receiving device (2720). (2801) is the same( ) and redundancy data transmitted from the transmitting device (2710). (2705) and redundancy data regenerated at the receiving device. (2803) is the same( ) because it does not cause semantic errors. That is, if no semantic errors occur due to the channel, and If this is satisfied and a semantic error occurs by the channel, and Since this is satisfied, the semantic error detector (2813) calculates the difference between the received redundancy data and the regenerated data. Semantic representation received based on (2801) can be used to determine whether a semantic error has occurred.

[0169] The semantic error detector (2813) measures the difference between the received redundancy data and the regenerated data. Once this is calculated, the calculated difference value and threshold By comparing the threshold values, it is determined whether an error has occurred in the semantic expression. At this time, the threshold value can be determined based on at least one of the types of tasks or semantic error detectors to be performed in the receiving device (2720). For example, a threshold can be set for each task that can be performed on the receiving device (2720). Here, the semantic error detector (2813) compares the difference value and threshold value of the received redundancy data and the regenerated data, and determines whether a semantic error has occurred based on the comparison result. This is because, even if the received redundancy data and the generated redundancy data are not identical, if the difference value is small, the performance of the downstream task to be performed on the receiving device can be guaranteed. Therefore, if the calculated difference value is smaller than the threshold value, the semantic error detector (2813) determines that a semantic error due to the channel, i.e., an error in the received semantic expression, has not occurred. At this time, the semantic error detector (2813) may provide information indicating that a semantic error has not occurred in the received semantic expression or information indicating the performance of semantic decoding to the semantic decoder (2817). On the other hand, if the calculated difference value is greater than or equal to the threshold value, the semantic error detector (2813) determines that the received semantic expression contains a semantic error that makes it impossible to perform a downstream task. At this time, the semantic error detector (2813) may provide information indicating that a semantic error has occurred in the received semantic expression or information indicating that semantic decoding should be suspended to the semantic decoder (2817). In addition, the semantic error detector (2813) provides the calculated difference value to the semantic error corrector (2815).

[0170] The semantic decoder (2817) performs a downstream task using the input semantic expression under the control of the semantic error detector (2813). That is, when the semantic decoder (2817) receives information indicating that no semantic error has occurred from the semantic error detector (2813) or information indicating that semantic decoding should be performed, the semantic decoder (2817) performs a semantic decoding function. Using this, a downstream task is performed based on the input semantic expression. If information indicating that a semantic error has occurred or information indicating that semantic decoding should be suspended is received from the semantic error detector (2813), the semantic decoder (2817) waits without performing the downstream task.

[0171] The semantic error corrector (2817) corrects errors in the received semantic representation. For example, the semantic error corrector (2817) can correct errors in the semantic representation by correcting the received semantic representation based on an optimization technique such as gradient descent. Error correction in the semantic representation is described in detail below with reference to FIG. 28b.

[0172] In the description referring to Fig. 28a, a structure for receiving a semantic expression and detecting errors in the received semantic expression was described. Therefore, below, a structure for correcting errors in a semantic expression will be described with reference to Fig. 28b.

[0173] FIG. 28b illustrates an example of a structure for correcting errors in semantic expressions according to one embodiment of the present disclosure. The receiving device (2720) of FIG. 28b may be the receiving device (2720) of FIG. 27 and / or the receiving device (2720) of FIG. 28a. That is, the components illustrated in FIG. 28b are identical to the components illustrated in FIG. 28a.

[0174] Referring to FIG. 28b, the receiving device (2720) corrects errors in the semantic representation using a semantic error corrector (2815). That is, the semantic error corrector (2815) can correct semantic errors in the semantic representation corresponding to the regenerated redundancy data using a semantic error correction function based on an optimization technique such as gradient descent. For example, the semantic error corrector (2815) may be a semantic redundancy generator. , regenerated redundancy data , received redundancy data , and received or corrected semantic representations Semantic error correction function that uses as input Using , the i-th corrected semantic representation (corrected SR) (2831) can be obtained. The i-th corrected semantic representation (2831) refers to the semantic representation obtained by the ith error correction operation. Specifically, the semantic error corrector (28115) calculates the received redundancy data calculated by the semantic error detector (2813). and regenerated redundancy data The difference value corresponding to the distance between the thresholds The semantic expression can be corrected to be less than the value of .

[0175] When the semantic error corrector uses gradient descent, the gradient for the semantic representation is calculated as shown in [Mathematical Equation 2] below.

[0176]

[0177] In [Equation 2], is the slope for the i-th corrected semantic representation, s is the transmitted data, is a semantic error detection function, is the i-th regenerated redundancy data, is the received redundancy data, is the i-th corrected semantic representation, is a semantic redundancy generation function, is the redundancy data regenerated by the semantic redundancy generation function, i is the number of error corrections, and ∂ is the partial differentiation symbol.

[0178] According to one embodiment of the present disclosure, the semantic error corrector (2815) calculates [Mathematical Formula 2] Based on this, the semantic expression can be corrected as in [Mathematical Formula 3] below.

[0179]

[0180] In [Equation 3], is the i+1th corrected semantic representation, is the i-th corrected semantic representation, is the slope for the i-th corrected semantic representation. Also, is the step-size, is a system parameter to correct. can be a fixed value or a value that changes depending on the number of times error correction is performed. For example, can be a fixed value regardless of the number of error corrections, or a value that decreases as the number of error corrections increases.

[0181] The semantic error corrector (2815) can correct the semantic expression and increase the number of error corrections. For example, the initial value of the number of error corrections i is 0, and it increases by 1 each time the semantic expression is corrected by the error correction operation. ) is performed. Here, increasing the number of error corrections by 1 in the semantic error corrector (2815) is to limit the number of error corrections to the maximum number. That is, the number of error corrections is the maximum number N. iterIf the number of error corrections is less than or equal to the maximum number, the semantic error corrector (2815) may perform error correction operations. On the other hand, if the number of error corrections is greater than or equal to the maximum number, the error correction operation may not be performed and it may be determined that error correction for the semantic expression has failed.

[0182] The semantic error corrector (2815) provides the corrected semantic expression to the semantic redundancy generator (2811). This is to detect semantic errors in the corrected semantic expression and to re-perform the semantic decoding operation or the semantic error correction operation based on the semantic error detection result. That is, the receiving device (2720) uses the semantic redundancy generator (2811) to provide the corrected semantic expression. Redundancy data for Regenerate and regenerate redundancy data using a semantic error detector (2813). and received redundancy data The difference value of calculates the difference value. The receiving device (2720) calculates the difference value and threshold Corrected semantic representation by comparing Determines whether a semantic error has occurred. is the threshold If greater than or equal to, the receiving device (2720) corrects the semantic representation It is determined that the semantic error for is not removed, and the corrected semantic expression is corrected using the semantic error corrector (2815). Error correction actions can be performed. is the threshold If less than, the receiver (2720) corrects the semantic representation It is determined that no semantic error occurred and the corrected semantic expression The final semantic representation with the errors corrected is determined by the receiving device (2720) to be the final semantic representation with the error corrected. is input into the semantic decoder (2817), and the downstream task is performed using the semantic decoder (2817).

[0183] FIG. 29 illustrates an example of a procedure for transmitting a semantic expression according to one embodiment of the present disclosure. FIG. 29 illustrates an operation method of a device transmitting a source (e.g., a transmitting device (2010) and / or a transmitting device (2710)). In the description referring to FIG. 29, the device transmitting the source is referred to as a first device, and the device receiving the source is referred to as a second device. The first device may be a terminal or a base station, and the second device may be a terminal or a base station capable of semantic communication with the first device.

[0184] Referring to FIG. 29, in step S2901, the first device receives control information from the second device. The control information is information for communication with the second device and may include various parameters such as information related to resources, information related to channel structure, information related to transmission method, and information related to communication procedures. According to one embodiment, the control information may include control information for semantic communication. For example, the control information for semantic communication may include at least one of information about a task to be performed by the second device, information about background knowledge, information about a semantic redundancy generator, or information about the length of semantic redundancy. The control information may be received via at least one of a DCI, MAC, or RRC message.

[0185] In step S2903, the first device generates transmission data. At this time, the first device generates the transmission data based on control information. The transmission data includes a semantic representation for semantic communication. Specifically, the first device can generate a semantic representation containing semantic information of source data to be transmitted based on control information for semantic communication received from the second device using a semantic encoder. For example, the semantic encoder can generate a semantic representation for the source data using at least one of information about a task to be performed by the second device or information about background knowledge.

[0186] In step S2905, the first device generates redundancy data. In other words, the first device generates redundancy data for at least one of error checking and error correction of the transmission data. That is, the first device can generate redundancy data for the semantic expression, which is the transmission data, using a semantic redundancy generation function. The semantic redundancy generation function can be determined based on control information received from the second device, i.e., information about the task to be performed by the second device. At this time, the length of the generated semantic redundancy can be determined based on information about the length of the semantic redundancy received from the second device.

[0187] In step S2907, the first device transmits transmission data and redundancy data. In other words, the first device transmits the transmission data and redundancy data to the second device. For example, the first device may transmit a semantic representation and redundancy data for the semantic representation to the second device. Here, the transmission data and redundancy data may be transmitted as a single transport block or a single codeword, or may be transmitted as separate transport blocks or codewords, respectively.

[0188] FIG. 30 illustrates an example of a procedure for receiving a semantic expression according to one embodiment of the present disclosure. FIG. 30 illustrates an operation method of a device receiving a source (e.g., receiving device (2020), receiving device (2720)). In the description referring to FIG. 30, the device receiving the source is referred to as a second device, and the device transmitting the source is referred to as a first device. The second device may be a terminal or a base station, and the first device may be a terminal or a base station capable of semantic communication with the second device.

[0189] Referring to FIG. 30, in step S3001, the second device transmits control information. The control information is information for communication between the first device and the second device, and may include various parameters such as information related to resources, information related to channel structure, information related to transmission methods, and information related to communication procedures. According to one embodiment, the control information may include control information for semantic communication. For example, the control information for semantic communication may include at least one of information about a task to be performed by the second device, information about background knowledge, information about a semantic redundancy generator, or information about the length of semantic redundancy. The information related to semantic communication may be transmitted via at least one of a DCI, MAC, or RRC message.

[0190] In step S3003, the second device receives transmission data and redundancy data. The transmission data may include a semantic representation of the source data. For example, the second device may receive the transmission data and redundancy data via a channel and perform channel decoding to obtain a semantic representation and redundancy data for the semantic representation.

[0191] In step S3005, the second device generates redundancy data based on the received transmission data. That is, the second device can regenerate redundancy data for the semantic expression obtained from the transmission data using a semantic redundancy generation function. The semantic redundancy generation function can be determined based on information about a task to be performed by the second device. That is, the second device can regenerate redundancy data for the received semantic expression using the same semantic redundancy generation function as the semantic redundancy generation function used by the first device. For example, the second device can select one semantic redundancy generation function from among a plurality of semantic redundancy generator functions based on information about the task to be performed by the second device, and regenerate redundancy data for the received semantic expression using the selected semantic redundancy generation function. According to one embodiment, the semantic redundancy regenerated using the selected semantic redundancy generation function may have a length corresponding to the length information of the semantic redundancy transmitted to the first device.

[0192] In step S3007, the second device detects an error in the transmitted data. The second device can determine whether a semantic error has occurred in the semantic representation using the received redundancy data and the regenerated redundancy data. For example, if no semantic error has occurred due to the channel, the semantic representation transmitted from the first device and the semantic representation received from the second device are identical, so the redundancy data transmitted from the first device and the regenerated redundancy data regenerated from the second device must be identical. Accordingly, the second device can determine whether an error has occurred in the semantic representation by checking whether the received redundancy data and the regenerated redundancy data are identical, or by checking whether the difference between the received redundancy data and the regenerated redundancy data is less than a threshold value.

[0193] In the description with reference to Figure 30, if no errors are detected in the transmitted data, the second device can perform a downstream task using the received semantic expression. Conversely, if an error is detected in the transmitted data, the second device may determine that the received semantic expression contains a semantic error that is severe enough to prevent the downstream task from being performed, and may perform an operation to correct the semantic error. The operation to correct the semantic error is described below in Figure 31.

[0194] FIG. 31 illustrates an example of a procedure for detecting and correcting errors in semantic expressions according to one embodiment of the present disclosure. The error detection and correction procedure described with reference to FIG. 31 can be understood as the specific operation of step S3007 of FIG. 30.

[0195] Referring to FIG. 31, in step S3101, the second device calculates a difference value between the received redundancy data and the regenerated redundancy data. For example, the second device may calculate a difference value corresponding to the distance between the redundancy data received from the first device through the channel and the redundancy data regenerated by the second device.

[0196] In step S3103, the second device determines whether the calculated difference value is greater than or equal to a threshold value. That is, the second device determines whether the calculated difference value is greater than or equal to the threshold value by comparing the difference value of the received redundancy data and the regenerated redundancy data with the threshold value. At this time, the threshold value may be determined based on at least one of the types of the task to be performed on the second device or the semantic error detector. For example, if the task to be performed on the second device is the first task, a first threshold value corresponding to the first task may be used, and if the task to be performed on the second device is the second task, a second threshold value corresponding to the second task may be used.

[0197] If the calculated difference value is less than the threshold value, in step S3115, the second device transmits an ACK signal. If the calculated difference value is less than the threshold value, the second device determines that no semantic error has occurred and may transmit an ACK signal to the first device indicating that the semantic expression has been successfully received.

[0198] In step S3117, the second device performs a task based on the semantic representation. The second device can perform a downstream task using the semantic representation corresponding to the regenerated redundancy data. For example, the second device can input the semantic representation corresponding to the regenerated redundancy data into a semantic decoder and use the semantic decoder to perform a downstream task based on the semantic representation.

[0199] If the calculated difference value is greater than or equal to the threshold value, in step S3105, the second device determines whether the number of error corrections is less than the maximum number. If the calculated difference value is greater than or equal to the threshold value, the second device determines that a semantic error has occurred, and can determine whether the number of error corrections is less than the maximum number by comparing the number of error corrections for the received semantic expression with the maximum number. The initial value of the number of error corrections is 0. The maximum number may be determined and / or changed by task information to be performed by the second device, or by the type of semantic error detector.

[0200] If the number of error corrections is less than the maximum number, in step S3107, the second device corrects errors in the semantic representation. The second device can correct semantic errors in the semantic representation by correcting the semantic representation corresponding to the regenerated redundancy data using a semantic error correction function based on an optimization technique. At this time, the semantic error correction function can correct the semantic representation so that the difference value corresponding to the distance between the received redundancy data and the regenerated redundancy data becomes smaller than a threshold value. According to one embodiment, the difference value corresponding to the distance between the received redundancy data and the regenerated redundancy data has been calculated in advance in step S3101, and thus can be used.

[0201] In step S3109, the second device increases the number of error corrections. The second device can increase the number of error corrections for the semantic expression by 1. In step S3111, the second device generates redundancy data for the corrected semantic expression. For example, the second device can regenerate redundancy data for the corrected semantic expression using the selected semantic redundancy generation function. Here, the selected semantic redundancy generation function is the semantic redundancy generation function used to regenerate the regenerated redundancy data described in step S3101. That is, the second device can regenerate redundancy data for the corrected semantic expression using the semantic redundancy generation function selected based on the task information to be performed by the second device, as described in FIG. 31. After generating redundancy data for the corrected semantic expression, the second device can return to step S3101 and re-perform the steps below.

[0202] If the number of error corrections is greater than or equal to the maximum number, in step S3113, the second device transmits a NACK signal to the first device. The second device may determine that error correction is not possible because the number of error corrections for the semantic expression has reached the maximum number, and may transmit a NACK signal to the first device. According to one embodiment, the NACK signal is a signal indicating that the semantic expression was not successfully received, and may request retransmission of the semantic expression.

[0203] Fig. 32 illustrates an example of error correction of a semantic expression according to an embodiment of the present disclosure. Fig. 32 illustrates a case where a first device transmits a semantic expression s (3201), but a second device detects a semantic error due to a channel. (3203) shows the process of detecting and correcting semantic errors when received.

[0204] Referring to Figure 32, the second device is a semantic redundancy generator Received using (3221) Semantic redundancy data corresponding to (3203) (3213) Regenerate and receive semantic redundancy data from the first device (3211) and regenerated semantic redundancy data (3213) distance between calculates the distance. The second device calculates the distance. Based on this, a semantic error is detected and the semantic expression is repeatedly corrected (3223) N times (e.g., 4 times) using a semantic error corrector based on gradient descent. Here, the error correction for the semantic expression is performed by correcting the received semantic redundancy data. and regenerated semantic redundancy data for corrected semantic representations. distance between is the threshold This can be done repeatedly until the received semantic redundancy data becomes smaller. and regenerated semantic redundancy data for corrected semantic representations. distance between If is less than the threshold (3225), the first device receives the semantic representation Treat this transmitted semantic expression s as corrected data with a similar meaning, and the corrected semantic expression (3231) can be used to perform downstream tasks.

[0205] FIG. 33 illustrates an example of a signal exchange for detecting and correcting semantic errors according to one embodiment of the present disclosure. In the description referring to FIG. 33 , the transmitting device (3310) may be a terminal or a base station, and the receiving device (3320) may be a terminal or a base station capable of communicating with the transmitting device (3310). At least some of the operations of FIG. 33 described below may be performed sequentially or in parallel. For example, some of the operations of FIG. 33 may be performed at least temporarily at the same time.

[0206] In the description with reference to FIG. 33, in step S3301, the transmitting device (3310) performs an initialization procedure for semantic communication with the receiving device (3320). The initialization procedure for semantic communication may include an operation of exchanging and / or sharing capability information or control information for semantic communication. For example, the transmitting device (3310) and the receiving device (3320) may exchange and / or share various information, such as resource-related information for semantic communication, channel structure-related information, transmission method-related information, communication procedure-related information, or background knowledge-related information.

[0207] In step S3303, the receiving device (3320) transmits control information including task information to the transmitting device (3310). The task information includes information on a downstream task to be performed by the receiving device (3320). According to one embodiment, the control information may further include at least one of semantic redundancy generator information or length information of semantic redundancy. The semantic redundancy generator information includes information on a semantic redundancy generator to be used by the transmitting device (3310) and the receiving device (3310). It may include an index value j indicating the length of the semantic redundancy. The length information of the semantic redundancy is the length of the redundancy data that must be generated through the semantic redundancy generator in the transmitting device (3310) and the receiving device (3310). At least one of the semantic redundancy generator information or the length information of the semantic redundancy may be determined based on a downstream task to be performed by the receiving device (3320). According to one embodiment, the length information of the semantic redundancy may be determined based on the difficulty or complexity of the downstream task to be performed by the receiving device (3320). For example, the length of the redundancy data The downstream task to be performed by the receiving device (3320) may be determined to have a larger value as the task is more difficult, and the downstream task may be determined to have a smaller value as the task is easier. According to one embodiment, the length of redundancy data can be selected based on channel status information. For example, if the channel status is below a specified level, the length of redundancy data Increase the code rate This is lowered, and if the channel status is above the specified level, the length of the redundancy data By reducing the inefficiency This can be made higher.

[0208] In step S3305, the transmitting device (3310) generates a semantic expression based on the task information. That is, the transmitting device (3310) checks the task information t to be performed in the receiving device (3310) from the control information transmitted from the receiving device (3310), and generates the task information t and the semantic expression generator. Generate a semantic representation s corresponding to source data x using ( ) is done.

[0209] In step S3307, the transmitting device (3310) selects a semantic redundancy generator based on task information. That is, the transmitting device (3310) selects one semantic redundancy generator among a plurality of semantic redundancy generators based on task information to be performed by the receiving device (3310). Select. For example, Semantic Redundancy Generator can be determined by the index j corresponding to the task information. In addition, the inefficiency can be determined by the index j corresponding to the task information.

[0210] In step S3309, the transmitting device (3310) generates semantic redundancy using the selected semantic redundancy generator. That is, the transmitting device (3310) generates semantic redundancy using the selected semantic redundancy generator. Generate redundancy data corresponding to the semantic representation s based on ( ) can be done.

[0211] At step S3311, the transmitting device (3310) provides a semantic representation and semantic redundancy The transmitting device (3310) can channel-encode the semantic representation and semantic redundancy and transmit the channel-encoded semantic representation and semantic redundancy to the receiving device (3320) through the channel. The receiving device (3320) receives the channel-encoded semantic representation and semantic redundancy and performs channel decoding, thereby decoding the received semantic representation. and received semantic redundancy Obtain .

[0212] In step S3313, the receiving device (3320) selects a semantic redundancy generator based on task information. That is, the receiving device (3320) selects one semantic redundancy generator among a plurality of semantic redundancy generators based on task information to be performed. Select. For example, Semantic Redundancy Generator can be determined by the index j corresponding to the task information. In addition, the inefficiency can be determined by the index j corresponding to the task information. Step S3313 can be performed before step S3311.

[0213] At step S3315, the receiving device (3320) receives the received semantic representation Regenerates semantic redundancy for the receiving device (3320). The receiving device (3320) selects a semantic redundancy generator based on task information. Semantic representation received using Semantic redundancy for can be regenerated.

[0214] In step S3317, the receiving device (3320) detects a semantic error. For example, the receiving device (3310) may be a semantic error detector. Semantic redundancy received using and semantic redundancy regenerated in the receiving device (3320). The difference value of and calculate the calculated difference value and threshold are compared. The receiving device (3320) can determine whether semantic occurrence occurs based on the comparison result. The calculated difference value This threshold If the difference is less than the calculated difference value, the receiver (3320) determines that no semantic error has occurred. This threshold If greater than or equal to, the receiving device (3320) determines that a semantic error has occurred.

[0215] If no semantic error occurs, in step S3319, the receiving device (3320) transmits an ACK signal to the transmitting device (3310). The receiving device (3320) transmits a semantic expression Semantic decoder Enter in , and semantic decoder can be used to perform downstream tasks.

[0216] If a semantic error occurs, in step S3321, the receiving device (3320) corrects the semantic expression and checks whether an error occurs in the corrected semantic expression. That is, the receiving device (3320) is a semantic error corrector. Semantic representation received using By performing error correction on the i-th corrected semantic representation can be obtained. For example, the receiving device (3320) can correct the semantic expression as in [Mathematical Expression 2] and [Mathematical Expression 3]. The receiving device (3320) can be a semantic redundancy generator. The i-th corrected semantic representation using Semantic redundancy for Regenerate and regenerate semantic redundancy and received semantic redundancy The difference value of and The receiving device (3320) can check whether an error has occurred in the corrected semantic expression based on the comparison result.

[0217] If no semantic error occurs in the corrected semantic expression as a result of the verification in step S3321, in step S3323, the receiving device (3320) transmits an ACK signal to the transmitting device (3310).

[0218] In step S3325, the receiving device (3320) performs a task based on the corrected semantic representation. That is, the receiving device (3320) performs a task based on the corrected semantic representation. The final semantic representation Set to , and the final semantic representation Semantic decoder Enter in , and semantic decoder can be used to perform downstream tasks.

[0219] If a semantic error occurs as a result of the verification in step S3321, the receiving device (3320) may re-perform a semantic expression correction operation or transmit a NACK signal to the transmitting device (3310) based on the number of error corrections for the semantic expression. For example, if the number of error corrections for the semantic expression is less than the maximum number, the receiving device (3320) determines that additional error correction for the semantic expression is possible and re-performs step S3321. On the other hand, if the number of error corrections is greater than or equal to the maximum number, the receiving device (3320) determines that additional error correction for the semantic expression is impossible and may transmit a NACK signal to the transmitting device (3310) in step S3323.

[0220] In the description with reference to FIG. 33, the transmitting device (3310) and the receiving device (3320) used the same semantic redundancy generator by selecting the semantic redundancy generator (or semantic redundancy generation function) based on task information indicating the downstream task to be performed by the receiving device (3320). However, according to various embodiments, the transmitting device (3310) and the receiving device (3320) can check whether the same semantic redundancy generator has been selected. To this end, the transmitting device (3310) can transmit information about the selected semantic redundancy generator (e.g., index information of the selected semantic redundancy generator) to the receiving device (3320). The receiving device (3320) can compare the information about the selected semantic redundancy generator received from the transmitting device (3310) with the information about the semantic redundancy generator selected based on the task information in the receiving device (3320) to determine whether the transmitting device (3320) and the receiving device (3310) have selected the same semantic redundancy generator. In one embodiment, if the transmitting device (3320) and the receiving device (3310) have not selected the same semantic redundancy generator, the receiving device (3310) can reselect the information about the task in the transmitting device (3320) or the information about the semantic redundancy generator selected in the receiving device (3310). In one embodiment, if the transmitting device (3320) and the receiving device (3310) do not select the same semantic redundancy generator, the receiving device (3310) may select the semantic redundancy generator selected by the transmitting device (3320) based on information about the selected semantic redundancy generator received from the transmitting device (3310).

[0221] As described above, a transmitting device according to one embodiment of the present disclosure transmits a semantic expression and redundancy data for the semantic expression during semantic communication, and a receiving device can detect and correct semantic errors that may occur in a channel based on the received semantic expression and redundancy data.

[0222] Unlike existing semantic error detection techniques, a receiving device according to an embodiment of the present disclosure determines whether a semantic error has occurred using a metric related to the magnitude of the semantic error. That is, while existing semantic error detection techniques can only determine whether a semantic error has occurred, the error detection technique according to an embodiment of the present disclosure can numerically confirm not only whether a semantic error has occurred but also the level of semantic error that has occurred, and has the advantage of enabling additional analysis of the semantic error based on this. In addition, existing semantic error detection techniques can detect semantic errors only by a pre-trained Sim32 encoder and Sim32 decoder using semantic redundancy of a fixed size. On the other hand, the error detection technique according to an embodiment of the present disclosure can adjust the detection intensity of a semantic error by adjusting the size or length of semantic redundancy based on task information and / or channel state information to be performed by the receiving device. That is, the semantic error detection technique proposed in this disclosure enables quantitative analysis of semantic errors occurring in semantic expressions, and can be systematically applied to various communication systems.

[0223] Furthermore, unlike existing semantic error correction techniques, the semantic error correction technique according to the present disclosure can perform additional verification procedures on the results of corrections. Therefore, the semantic error correction technique according to the present disclosure can quantitatively measure the reliability of the ultimately corrected semantic expression.

[0224] Furthermore, existing semantic error correction techniques, such as those illustrated in FIG. 23, have fixed error correction performance by utilizing pre-trained GANs. In contrast, the semantic error correction technique according to the embodiment of the present disclosure can control the code rate by adjusting the size of the transmitted semantic redundancy, thereby enabling semantic error correction with various performances and can be systematically applied to various semantic communication systems.

[0225] Furthermore, the semantic error correction technique of the present disclosure differs significantly from the bit-level error correction employed in existing communication systems. Specifically, existing bit-level error correction techniques fail to correct errors if they are not perfectly corrected, i.e., if the received bit stream is not corrected to be identical to the transmitted bit stream. However, the semantic error correction technique according to the embodiment of the present disclosure corrects the received semantic expression so that it has a high degree of similarity with the transmitted semantic expression, and therefore, even if the received semantic expression is not identical to the transmitted semantic expression, it is not treated as an error correction failure.

[0226] In addition, when a semantic expression is transmitted and received via digital communication, the amount of bit errors due to the channel is not proportional to the size of the error in the semantic expression, so a situation may arise where error correction for the semantic expression cannot be performed using the existing bit-by-bit error correction technique. On the other hand, when using the semantic error correction technique according to the embodiment of the present disclosure, errors in the semantic expression can be corrected, and downstream tasks can be performed well with the corrected semantic expression.

[0227] As described above, the semantic error detection and correction technique according to the embodiment of the present disclosure can detect and correct semantic errors that may occur in semantic expressions transmitted and received between a transmitting device and a receiving device, thereby improving the performance and reliability of downstream tasks performed by the receiving device.

[0228] The proposed methods described above can be implemented independently, but they can also be implemented as a combination (or merge) of some of the proposed methods. Rules can be defined so that the base station notifies the terminal of the applicability of the proposed methods (or information about the rules of the proposed methods) through a predefined signal (e.g., a physical layer signal or a higher layer signal).

[0229] The present disclosure may be embodied in other specific forms without departing from the technical ideas and essential features described herein. Therefore, the above detailed description should not be construed as limiting in all respects but rather as illustrative. The scope of the present disclosure should be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalent scope of the present disclosure are intended to be included within the scope of the present disclosure. Furthermore, claims that are not explicitly cited in the claims may be combined to form an embodiment or incorporated into a new claim through a post-filing amendment.

[0230] Embodiments of the present disclosure can be applied to various wireless access systems. Examples of various wireless access systems include the 3rd Generation Partnership Project (3GPP) or 3GPP2 systems.

[0231] The embodiments of the present disclosure can be applied not only to the various wireless access systems described above, but also to all technical fields that utilize these various wireless access systems. Furthermore, the proposed method can also be applied to mmWave and THz communication systems utilizing ultra-high frequency bands.

[0232] Additionally, embodiments of the present disclosure can be applied to various applications such as autonomous vehicles and drones.

Claims

1. In a method of operating a first device in a wireless communication system, A step of receiving control information for transmission from a second device; A step of generating transmission data based on the above control information; A step of generating redundancy data for at least one of error checking and error correction for the above transmission data; and A step of transmitting the above transmission data and the above redundancy data is included, The above transmission data includes a semantic representation (SR) for semantic communication, A method in which the above redundancy data is generated by a generation function selected from among a plurality of generation functions based on a task to be performed in the second device.

2. In claim 1, A method in which the above redundancy data includes information for detecting a spatial distance movement exceeding the allowable range of SR error in performing a task in the second device.

3. In claim 1, A method in which the above control information includes information related to a task performed in the second device using the SR.

4. In claim 3, A method in which the above redundancy data is generated to have a length determined based on the above task.

5. In claim 1, The above redundancy data is a method of generating based on the above SR.

6. In claim 1, A method further comprising the step of transmitting information related to the selected generating function.

7. In claim 6, A method in which each of the above generating functions is a differentiable function, and includes a function defined by a combination of at least one of arithmetic operations and nonlinear operations on elements included in the SR, or a function defined in the form of a neural network.

8. In a method of operating a second device in a wireless communication system, A step of transmitting control information to a first device; A step of receiving transmission data and redundancy data for the transmission data from the first device; A step of generating redundancy data based on the received transmission data; and A step of checking for an error in the transmitted data based on the difference between the received redundancy data and the generated redundancy data, The above transmission data includes a semantic representation (SR) for semantic communication, A method in which the above-mentioned generated redundancy data is generated by a generation function selected from among a plurality of generation functions based on a task to be performed in the second device.

9. In claim 8, A method wherein the above control information includes information on a task to be performed in the second device.

10. In claim 8, A method further comprising the step of receiving information about a generation function used to generate the received redundancy data from the first device.

11. In claim 8, The step of checking for an error in the transmitted data based on the difference between the received redundancy data and the generated redundancy data is: A method comprising a step of determining whether an error has occurred for the SR based on whether a difference between the received redundancy data and the generated redundancy data is greater than a threshold value.

12. In claim 11, A method wherein the threshold value is determined based on at least one of the types of tasks or semantic error detection functions to be performed in the second device.

13. In claim 11, The difference between the received redundancy data and the generated redundancy data is calculated by a semantic error detection function defined as a differentiable metric, The above semantic error detection function is defined as a mean square error related metric, a cosine similarity related metric, or a sentence similarity related metric.

14. In claim 8, A method further comprising a step of correcting the SR using an optimization technique when an error is found to have occurred in the transmission data as a result of checking the above error.

15. In claim 14, A step of generating redundancy data for the corrected SR based on the corrected SR; and A method further comprising the step of checking for an error in the corrected SR based on a difference between the received redundancy data and the redundancy data generated for the corrected SR.

16. In claim 14, If an error occurs in the transmission data as a result of checking the above error, the method further includes a step of comparing the number of error corrections for the SR with the maximum number of times. If the number of error corrections is less than the maximum number of corrections, the SR is corrected, A method for determining that error correction for the SR is impossible if the number of error corrections is greater than or equal to the maximum number of errors.

17. In claim 16, A method in which the maximum number of times is determined based on at least one of task information to be performed in the second device or a type of semantic error detection function that calculates the difference between the received redundancy data and the generated redundancy data.

18. In a first device in a wireless communication system, Transmitter and receiver; and A processor connected to the above transmitter and receiver is included, The above processor, Receive control information for transmission from a second device, Generate transmission data based on the above control information, Generating redundancy data for at least one of error checking and error correction for the above transmission data, and Control to transmit the above transmission data and the above redundancy data, The above transmission data includes a semantic representation (SR) for semantic communication, The above redundancy data is a first device generated by a generation function selected from among a plurality of generation functions based on a task to be performed in the second device.

19. In a second device in a wireless communication system, Transmitter and receiver; and A processor connected to the above transmitter and receiver is included, The above processor, Transmit control information to the first device, Receive transmission data and redundancy data for the transmission data from the first device, Generate redundancy data based on the received transmission data, Control to check for errors in the transmitted data based on the difference between the received redundancy data and the generated redundancy data, The above transmission data includes a semantic representation (SR) for semantic communication, The above-mentioned generated redundancy data is a second device generated by a generation function selected from among a plurality of generation functions based on a task to be performed in the second device.

20. In communication devices, At least one processor; At least one computer memory connected to said at least one processor and storing instructions that direct operations when executed by said at least one processor, The above actions are, A step of receiving control information for transmission from another communication device; A step of generating transmission data based on the above control information; A step of generating redundancy data for at least one of error checking and error correction for the above transmission data; and A step of transmitting the above transmission data and the above redundancy data is included, The above transmission data includes a semantic representation (SR) for semantic communication, A communication device in which the above redundancy data is generated by a generation function selected from among a plurality of generation functions based on a task to be performed in the other communication device.

21. In a non-transitory computer-readable medium storing at least one instruction, comprising at least one instruction executable by the processor, At least one of the above commands causes the device to: Receive control information for transmission from a second device, Generate transmission data based on the above control information, Generating redundancy data for at least one of error checking and error correction for the above transmission data, and Controls the transmission of the above transmission data and the above redundancy data, The above transmission data includes a semantic representation (SR) for semantic communication, The above redundancy data is a non-transitory computer-readable medium generated by a generation function selected from among a plurality of generation functions based on a task to be performed in the second device.

Images