Semantic communication method, device and system

Abstract

The invention provides a semantic communication method and a related product. The method comprises the following steps: receiving first query information, a first matching score function and a first threshold value broadcasted or multicast by central equipment; according to a first matching score function, determining whether a first matching score between the perception data and the first query information is greater than or equal to a first threshold value; when the first matching score is larger than or equal to a first threshold value, a sensing result is sent to the center equipment, and the sensing result indicates the sensing data. The central equipment broadcasts or multicasts the first query information, the first matching score function and the first threshold value, so that the determination basis of the sensing equipment is the same as that of other sensing equipment, and the response accuracy of the sensing equipment is improved in the horizontal dimension. In addition, sending of the sensing result has a specific condition, and the sensing result is not sent all the time, so that transmission resources are saved. Moreover, the sent sensing result meets the requirement of the first query information, that is, irrelevant information is filtered, and the sent data is required by the central equipment, so that the response accuracy of the sensing equipment is ensured in the vertical dimension.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority to U.S. Provisional Patent Application No. 63 / 509,420, filed June 21, 2023, entitled “Semantic / Task Query Method, Apparatus and System,” the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure generally relates to the field of communication technology, and in particular to a communication method, a communication device, a communication system and related products. Background Technology

[0004] Sensing capabilities will be integrated into the 6th generation (6G) system. A large number of user equipment (UEs) or sensing devices will be densely deployed in cities, factories, farms, and other locations. Besides mobile phones, sensing devices will also become a crucial type of UE or device heralding the arrival of the Internet of Things (IoT) era. Similar to internet search engines, 6G will introduce corresponding IoT search engines in the real physical world. In fact, billions of IoT-based applications, such as driverless cars, automated factories, smart cities, and autonomous farms, will largely rely on efficient, real-time search engines in the physical world.

[0005] In recent years, artificial intelligence (AI) has made breakthroughs in various fields of intelligence and cognition. Some AIs are exploring cutting-edge knowledge in fields such as chemistry, gaming, mathematics, and genetic engineering, while others are providing human-level question-and-answer platforms in the digital world. The area AI has yet to conquer is the real-time physical world. AI in the physical world may be built upon the ubiquitous IoT connectivity brought about by 6G, in which AI technology will permeate all aspects of society and life.

[0006] More challenging than internet search engines are real-world search engines, which must search the physical world in real time over large physical areas and process various types of data and information. Furthermore, green technology, low energy consumption, and low emissions are also key characteristics of 6G. Sensing devices can be battery-powered and / or entirely powered by solar and wind power. In some implementations, sensing devices can be UEs, mobile phones, or handsets, where it is assumed that any two sensing devices are independent of each other; therefore, the wireless system associated with the sensing device can schedule the sensing device individually; the sensing data measured by the sensing device can be the application-level payload of the wireless system and protocols. The above-mentioned sensing device scheduling schemes are inefficient in terms of wireless bandwidth and energy consumption.

[0007] The purpose of this background information is to disclose information that the applicant believes may be relevant to this application. It is not necessarily an admission, nor should any of the foregoing information be construed as constituting prior art in relation to this application. Summary of the Invention

[0008] In a first aspect, this disclosure provides a communication method, the method comprising:

[0009] Receive the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device;

[0010] Based on the first matching score function, determine whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold;

[0011] When the first matching score is greater than or equal to the first threshold, a sensing result is sent to the central device, wherein the sensing result indicates the sensing data.

[0012] The central device broadcasts or multicasts a first query message to wake up the sensing devices and measure the sensing data. Based on the first matching score function broadcast or multicast by the central device, and provided the first matching score is greater than or equal to a first threshold, the sensing devices calculate a first matching score between the sensing data and the first query message. The sensing devices then send the sensing results to the central device. First, the central device broadcasts or multicasts the first query message, the first matching score function, and the first threshold. This ensures that the criteria used to determine the sensing devices are the same as those used by other sensing devices, thereby improving the response accuracy of the sensing devices in the horizontal dimension. Second, the transmission of sensing results is conditional and not continuous, thus saving transmission resources. Third, the transmitted sensing results meet the requirements of the first query message, i.e., irrelevant information is filtered out, and the transmitted data is what the central device needs, thereby ensuring the response accuracy of the sensing devices in the vertical dimension.

[0013] In one possible implementation of the first aspect, the first query information includes query information for a task or query information for a modality.

[0014] A sensing device can serve a single task or a single mode, meaning it is dedicated to a specific task or mode, thus improving its response efficiency.

[0015] In one possible implementation of the first aspect, the method further includes: receiving an indication of a task identifier or a modal identifier from the central device.

[0016] Task identifiers or modal identifiers can be indicated explicitly or implicitly.

[0017] In one possible implementation of the first aspect, the first query information includes multiple query information for multiple tasks and multiple task identifiers respectively associated with the multiple query information.

[0018] Sensing devices can serve multiple tasks simultaneously, thereby improving the resource utilization of sensing devices.

[0019] In one possible implementation of the first aspect, the first query information includes multiple query information for multiple modalities and multiple modality identifiers respectively associated with the multiple query information.

[0020] Sensing devices can serve multiple modalities simultaneously, meaning they can sense and respond to more than one modality (i.e., one type of data), which enriches the types of data that sensing devices respond to.

[0021] In one possible implementation of the first aspect, the first query information includes a first plurality of query information for multiple tasks, a second plurality of query information for multiple modalities, a plurality of task identifiers respectively associated with the first plurality of query information, and a plurality of modal identifiers respectively associated with the second plurality of query information.

[0022] Sensing devices can serve multiple tasks and multiple modalities simultaneously, which not only improves the resource utilization of sensing devices but also enriches the types of data that sensing devices respond to.

[0023] In one possible implementation of the first aspect, receiving the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device includes:

[0024] The central device receives the first query information, the first matching score function, and the first threshold broadcast in the synchronization signal block (SSB) / physical broadcast channel (PBCH) block.

[0025] In one possible implementation of the first aspect, the first query information is carried in a master information block (MIB) message or a system information block (SIB) message, the MIB message or the SIB message including an indication of whether the first query information is presented in the MIB message or the SIB message.

[0026] By indicating whether to present the first query information, signaling overhead can be saved, because if the first query information is not presented, no further processing is required.

[0027] In one possible implementation of the first aspect, the MIB message or the SIB message further includes an indication of the period for indicating the first query information.

[0028] The first query information is valid within the period, but may be invalid outside the period, thus avoiding resource consumption.

[0029] In one possible implementation of the first aspect, receiving the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device includes:

[0030] The central device receives the first query information, the first matching score function, and the first threshold in the form of a multicast message.

[0031] In one possible implementation of the first aspect, the first query information includes at least two query semantics.

[0032] The receipt of the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device includes:

[0033] Receive, via broadcast or multicast from the central device, a first query semantic, a second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, and the format of the first query semantic; receive via broadcast or multicast from the central device, a second query semantic, a third matching score function associated with the second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic; or

[0034] The central device receives, in a multiplexed manner, at least two query semantics, including a first query semantic, a second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, the format of the first query semantic, a second query semantic, a third matching score function associated with the second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic.

[0035] When the first query information includes multiple query semantics, the multiple query semantics and the information related to the multiple query semantics (i.e., the matching score function, threshold, length of the query semantics, and format of the query semantics corresponding to each query semantic) can be received sequentially or simultaneously using a multiplexing method. This provides greater flexibility and can meet different requirements.

[0036] In one possible implementation of the first aspect, the method further includes:

[0037] Acquire the perceived data;

[0038] Convert the perceived data into general perception semantics;

[0039] The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes:

[0040] Based on the second matching score function, determine whether the second matching score between the general-aware semantics and the first query semantics is greater than or equal to the second threshold;

[0041] Based on the third matching score function, determine whether the third matching score between the general perceptual semantics and the second query semantics is greater than or equal to the third threshold.

[0042] When the perceived data is in natural language form and the first query information is in semantic form, the perceived data can be converted into general perceived semantics. For example, general semanticization configuration can be used to generate general perceived semantics, which simplifies the generation of perceived semantics. Then, a comparison is made between the general perceived semantics and each query semantics. That is, both the perceived data and the query information are in a general semantic domain, on which they can be easily compared and fused. Query semantics can retain all the key semantic goals conveyed by the query message, allowing the query semantics to be well converted (de-semantized) back to the query message. Since the semantic form can provide a more accurate true intent, the accuracy of the comparison results is improved.

[0043] In one possible implementation of the first aspect, the method further includes:

[0044] Acquire the perceived data;

[0045] The perceived data is converted into first perceived semantics based on the first semantic configuration, the length of the first query semantics, and the format of the first query semantics;

[0046] The perceived data is converted into second perceived semantics according to the second semantic configuration, the length of the second query semantics, and the format of the second query semantics;

[0047] The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes:

[0048] Based on the second matching score function, determine whether the second matching score between the first perceived semantic and the first query semantic is greater than or equal to the second threshold;

[0049] Based on the third matching score function, determine whether the third matching score between the second perceived semantics and the second query semantics is greater than or equal to the third threshold.

[0050] When the perceived data is in natural language form and the first query information is in semantic form, the perceived data can be converted into first and second perceived semantics respectively. Corresponding semantics are then generated using appropriate semanticization configurations, ensuring the accuracy of the generated perceived semantics. Then, comparisons are made between the first and first perceived semantics, and between the second and second perceived semantics. In other words, both the perceived data and the query information reside in a common semantic domain, within which they can be easily compared and integrated. Query semantics can retain all the key semantic objectives conveyed by the query message, allowing for effective conversion (de-semanticization) back to the query message. Since semantic form can provide a more accurate representation of the true intent, the accuracy of the comparison results is improved.

[0051] In one possible implementation of the first aspect, the method further includes:

[0052] Acquire the perceived data;

[0053] Convert the perceived data into general perception semantics;

[0054] According to the general lexicalization configuration, the general perceptual semantic lexical is converted into a first perceptual lexical, the first query semantic lexical is converted into a first query lexical, the general perceptual semantic lexical is converted into a second perceptual lexical, and the second query semantic lexical is converted into a second query lexical; or according to the first lexicalization configuration, the general perceptual semantic lexical is converted into a first perceptual lexical, the first query semantic lexical is converted into a first query lexical, and according to the second lexicalization configuration, the general perceptual semantic lexical is converted into a second perceptual lexical, and the second query semantic lexical is converted into a second query lexical.

[0055] The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes:

[0056] Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold;

[0057] Based on the third matching score function, determine whether the third matching score between the second perceived word and the second query word is greater than or equal to the third threshold.

[0058] When the perceived data is in natural language form, while the first query information is in semantic form, the perceived data can be processed into first and second perceived tokens using general perceived semantics. Similarly, the first and second query semantics can be processed into first and second query tokens, respectively. Then, a comparison is made between the perceived tokens and their corresponding query tokens. Since tokenization can provide a more accurate representation of the true intent and saves signaling overhead, it improves the accuracy of the comparison results and reduces signaling costs. Furthermore, tokenization can prevent the sensing device from reconstructing the complete query message from the query tokens. Tokenization can also provide a degree of privacy protection for the query message.

[0059] In one possible implementation of the first aspect, the method further includes:

[0060] Acquire the perceived data;

[0061] The perceived data is converted into first perceived semantics based on the first semantic configuration, the length of the first query semantics, and the format of the first query semantics;

[0062] The perceived data is converted into second perceived semantics according to the second semantic configuration, the length of the second query semantics, and the format of the second query semantics;

[0063] According to the general lexicalization configuration, the first perceptual semantic lexical is converted into a first perceptual lexical, the first query semantic lexical is converted into a first query lexical, the second perceptual semantic lexical is converted into a second perceptual lexical, and the second query semantic lexical is converted into a second query lexical; or according to the first lexicalization configuration, the first perceptual semantic lexical is converted into a first perceptual lexical, the first query semantic lexical is converted into a first query lexical, and according to the second lexicalization configuration, the second perceptual semantic lexical is converted into a second perceptual lexical, and the second query semantic lexical is converted into a second query lexical.

[0064] The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes:

[0065] Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold;

[0066] Based on the third matching score function, determine whether the third matching score between the second perceived word and the second query word is greater than or equal to the third threshold.

[0067] When the perceived data is in natural language form, and the first query information is in semantic form, the perceived data can be processed into first perceived tokens and second perceived tokens using first and second perceived semantics, respectively. Similarly, the first and second query semantics can be processed into first query tokens and second query tokens, respectively. Then, a comparison is made between the perceived tokens and their corresponding query tokens. Since tokenization can provide a more accurate representation of the true intent and saves signaling overhead, it improves the accuracy of the comparison results and reduces signaling overhead. Furthermore, tokenization can prevent the sensing device from reconstructing the complete query message from the query tokens. Tokenization can also provide a degree of privacy protection for the query message.

[0068] In one possible implementation of the first aspect, the first query information includes at least two query terms.

[0069] The receipt of the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device includes:

[0070] Receive the first query term from the at least two query terms broadcast or multicast by the central device, the second matching score function associated with the first query term, the second threshold associated with the second matching score function, and the length of the first query term; receive the second query term from the at least two query terms broadcast or multicast by the central device, the third matching score function associated with the second query term, the third threshold associated with the third matching score function, and the length of the second query term; or

[0071] The central device receives, in a multiplexed manner, a first query term from the at least two query terms, a second matching score function associated with the first query term, a second threshold associated with the second matching score function, the length of the first query term, a second query term from the at least two query terms, a third matching score function associated with the second query term, a third threshold associated with the third matching score function, and the length of the second query term.

[0072] When the first query information includes multiple query terms, the multiple query terms and the information related to the multiple query terms (i.e., the matching score function, threshold and length of the query term corresponding to each query semantic) can be received sequentially or simultaneously in a reused manner, which provides greater flexibility and can meet different requirements.

[0073] In one possible implementation of the first aspect, the method further includes:

[0074] Acquire the perceived data;

[0075] Convert the perceived data into general perception semantics;

[0076] According to the general lexicalization configuration, the general perceptual semantic lexical is converted into a first perceptual lexical based on the length of the first query lexical, and the general perceptual semantic lexical is converted into a second perceptual lexical based on the length of the second query lexical; or according to the first lexicalization configuration, the general perceptual semantic lexical is converted into a first perceptual lexical based on the length of the first query lexical, and according to the second lexicalization configuration, the general perceptual semantic lexical is converted into a second perceptual lexical based on the length of the second query lexical.

[0077] The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes:

[0078] Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold;

[0079] Based on the third matching score function, determine whether the third matching score between the second perceived word and the second query word is greater than or equal to the third threshold.

[0080] When the perceived data is in natural language form, and the first query information is in lexical form, the perceived data can be processed into first and second perceived lexical units using general perceptual semantics. Then, a comparison is made between the perceived lexical units and their corresponding query lexical units. Since lexical units can provide a more accurate representation of the true intent and save signaling overhead, the accuracy of the comparison results is improved, further reducing signaling costs. Furthermore, lexicalization can prevent the sensing device from reconstructing the complete query message from the query lexical units. Lexicalization can also provide a degree of privacy protection for the query message.

[0081] In one possible implementation of the first aspect, the method further includes:

[0082] Acquire the perceived data;

[0083] According to the first semantic configuration, the perceived data is converted into the first perceptual semantics;

[0084] According to the second semantic configuration, the perceived data is converted into second perceptual semantics;

[0085] According to the general lexicalization configuration, the first perceptual semantic lexical is converted into a first perceptual lexical based on the length of the first query lexical, and the second perceptual semantic lexical is converted into a second perceptual lexical based on the length of the second query lexical; or according to the first lexicalization configuration, the first perceptual semantic lexical is converted into a first perceptual lexical based on the length of the first query lexical, and according to the second lexicalization configuration, the second perceptual semantic lexical is converted into a second perceptual lexical based on the length of the second query lexical.

[0086] The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes:

[0087] Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold;

[0088] Based on the third matching score function, determine whether the third matching score between the second perceived word and the second query word is greater than or equal to the third threshold.

[0089] When the perceived data is in natural language form, and the first query information is in lexical form, the perceived data can be processed into first-perceived lexical units and second-perceived lexical units respectively through first-perceived semantics and second-perceived semantics. Then, a comparison is made between the perceived lexical units and their corresponding query lexical units. Since lexical units can provide a more accurate representation of the true intent and save signaling overhead, the accuracy of the comparison results is improved, and signaling overhead is reduced. Furthermore, lexicalization can prevent the sensing device from reconstructing the complete query message from the query lexical units. Lexicalization can provide a certain degree of privacy protection for the query message.

[0090] In one possible implementation of the first aspect, the first matching score function, the second matching score function, or the third matching score function includes an inner product or Euclidean distance.

[0091] In one possible implementation of the first aspect, the perception result includes one of the following:

[0092] Raw sensory data;

[0093] Perceptual semantics obtained from raw perceptual data;

[0094] Semi-raw perceptual data and perceptual semantics obtained from raw perceptual data;

[0095] The original perceived data and the first matching score;

[0096] Perceptual semantics obtained from raw perceptual data and the first matching score;

[0097] Semi-raw perceptual data, perceptual semantics obtained from raw perceptual data, and the first matching score.

[0098] The results of perception can take various forms related to the perception data, which provides greater flexibility to meet different requirements.

[0099] In one possible implementation of the first aspect, sending the sensing result to the central device includes:

[0100] Send a first matching perceptual semantic and a fourth matching score related to the first matching perceptual semantic to the central device; send a second matching perceptual semantic and a fifth matching score related to the second matching perceptual semantic to the central device.

[0101] In response to more than one query, a sensing device can send more than one sensing semantic and related matching score to the central device. When multiple sensing devices send such data to the central device, the central device can perform a fusion operation on the data received from multiple sensing devices to improve the accuracy of the fused data.

[0102] In a second aspect, this disclosure provides a communication method, the method comprising:

[0103] Broadcast or multicast the first query information, the first matching score function, and the first threshold to multiple sensing devices;

[0104] A sensing result is received from at least one of the plurality of sensing devices, wherein a first matching score between the sensing data and the first query information is greater than or equal to a first threshold, the first matching score is based on a first matching score function, and the sensing result indicates the sensing data.

[0105] The central device broadcasts or multicasts the first query information, the first matching score function, and the first threshold, thereby improving the transmission efficiency of the central device. When the first matching score between the sensing data from the sensing device and the first query information is greater than or equal to the first threshold, the central device receives the sensing result. The received sensing result meets the requirements of the first query information, i.e., irrelevant information is filtered out, and the received data is what the central device needs, thus ensuring the accuracy of data transmission.

[0106] In one possible implementation of the second aspect, the first query information includes query information for a task or query information for a modality.

[0107] In one possible implementation of the second aspect, the method further includes:

[0108] Send an instruction for the task identifier or the modal identifier.

[0109] In one possible implementation of the second aspect, the first query information includes multiple query information for multiple tasks and multiple task identifiers respectively associated with the multiple query information.

[0110] In one possible implementation of the second aspect, the first query information includes multiple query information for multiple modalities and multiple modality identifiers respectively associated with the multiple query information.

[0111] In one possible implementation of the second aspect, the first query information includes a first plurality of query information for multiple tasks, a second plurality of query information for multiple modalities, a plurality of task identifiers respectively associated with the first plurality of query information, and a plurality of modal identifiers respectively associated with the second plurality of query information.

[0112] In one possible implementation of the second aspect, broadcasting or multicasting the first query information, the first matching score function, and the first threshold to the plurality of sensing devices includes:

[0113] The first query information, the first matching score function, and the first threshold are broadcast to the plurality of sensing devices in the synchronization signal block (SSB) / physical broadcast channel (PBCH) block.

[0114] In one possible implementation of the second aspect, the first query information is carried in a master information block (MIB) message or a system information block (SIB) message, the MIB message or the SIB message including an indication of whether the first query information is presented in the MIB message or the SIB message.

[0115] In one possible implementation of the second aspect, the MIB message or the SIB message further includes an indication of the period for indicating the first query information.

[0116] In one possible implementation of the second aspect, broadcasting or multicasting the first query information, the first matching score function, and the first threshold to the plurality of sensing devices includes:

[0117] The first query information, the first matching score function, and the first threshold are multicast to the multiple sensing devices in the form of a multicast message.

[0118] In one possible implementation of the second aspect, the method further includes:

[0119] Receive the second query information from the generative pre-trained transformer (GPT) device;

[0120] The sensing results are output to the GPT device.

[0121] The central device receives the second query information from the GPT device, provides the first query information, the first matching score function and the first threshold for the sensing device to make a decision, and then receives the sensing result from the sensing device and outputs the sensing result to the GPT device. The central device acts as a bridge between the sensing device and the GPT device, thereby facilitating smooth communication between the sensing device and the GPT device.

[0122] In one possible implementation of the second aspect, receiving the second query information from the GPT device includes:

[0123] Receive at least two query semantics from at least two GPT devices;

[0124] The step of broadcasting or multicasting the first query information, the first matching score function, and the first threshold to the plurality of sensing devices includes:

[0125] Broadcast or multicast to the plurality of sensing devices a first query semantic, a second matching score function related to the first query semantic, a second threshold related to the second matching score function, the length of the first query semantic, and the format of the first query semantic; broadcast or multicast to the plurality of sensing devices a second query semantic, a third matching score function related to the second query semantic, a third threshold related to the third matching score function, the length of the second query semantic, and the format of the second query semantic; or

[0126] The first query semantic, the second matching score function associated with the first query semantic, the second threshold associated with the second matching score function, the length of the first query semantic, the format of the first query semantic, the second query semantic, the third matching score function associated with the second query semantic, the third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic are broadcast or multicast to the plurality of sensing devices in a multiplexed manner.

[0127] When the second query information includes multiple query semantics, the central device can broadcast or multicast multiple query semantics and information related to multiple query semantics (i.e., the matching score function, threshold, length of query semantics and format of query semantics corresponding to each query semantic) in sequence or in a multiplexed manner. This provides greater flexibility and can meet different requirements.

[0128] In one possible implementation of the second aspect, receiving the second query information from the GPT device includes:

[0129] Receive at least two query semantics from at least two GPT devices;

[0130] The method further includes:

[0131] The first query semantic and the second query semantic in the at least two query semantics are lexicalized into first query lexical and second query lexical;

[0132] The step of broadcasting or multicasting the first query information, the first matching score function, and the first threshold to the plurality of sensing devices includes:

[0133] Broadcast or multicast the first query term, the second matching score function associated with the first query term, the second threshold associated with the second matching score function, and the length of the first query term to the plurality of sensing devices; broadcast or multicast the second query term, the third matching score function associated with the second query term, the third threshold associated with the third matching score function, and the length of the second query term to the plurality of sensing devices; or

[0134] The first query term, the second matching score function associated with the first query term, the second threshold associated with the second matching score function, the length of the first query term, the second query term, the third matching score function associated with the second query term, the third threshold associated with the third matching score function, and the length of the second query term are broadcast or multicast to the plurality of sensing devices in a multiplexed manner.

[0135] When the second query information includes multiple query semantics, the central device can convert the query semantic lexicals into corresponding query lexicals, and then broadcast or multicast multiple query lexicals and information related to multiple query lexicals (i.e., the matching score function, threshold, and length of each query lexical) in sequence or in a multiplexed manner. This provides greater flexibility and can meet different requirements.

[0136] In one possible implementation of the second aspect, the step of lexicalizing the first query semantic and the second query semantic in the at least two query semantics into the first query term and the second query term includes:

[0137] According to the first lexicalization configuration, the first query semantic lexicalization is converted into the first query lexical; according to the second lexicalization configuration, the second query semantic lexicalization is converted into the second query lexical; or

[0138] According to the general lexicalization configuration, the first query semantic lexicalization is converted into the first query lexical; according to the general lexicalization configuration, the second query semantic lexicalization is converted into the second query lexical.

[0139] Lexicalization of different query semantics can be achieved through the same lexicalization configuration or different lexicalization configurations, depending on the actual needs.

[0140] In one possible implementation of the second aspect, the first matching score function, the second matching score function, or the third matching score function includes an inner product or Euclidean distance.

[0141] In one possible implementation of the second aspect, the perception result includes one of the following:

[0142] Raw sensory data;

[0143] Perceptual semantics obtained from raw perceptual data;

[0144] Semi-raw perceptual data and perceptual semantics obtained from raw perceptual data;

[0145] The original perceived data and the first matching score;

[0146] Perceptual semantics obtained from raw perceptual data and the first matching score;

[0147] Semi-raw perceptual data, perceptual semantics obtained from raw perceptual data, and the first matching score.

[0148] In one possible implementation of the second aspect, receiving the sensing result from the at least one sensing device includes:

[0149] Receive from at least one of the plurality of sensing devices a plurality of first matching sensing semantics and a plurality of fourth matching scores associated with the plurality of first matching sensing semantics; receive from at least one of the plurality of sensing devices a plurality of second matching sensing semantics and a plurality of fifth matching scores associated with the plurality of second matching sensing semantics;

[0150] The method further includes:

[0151] The perceptual semantics of some or all of the multiple first matches are fused based on the multiple fourth matching scores to obtain the first fused perceptual semantics; the perceptual semantics of some or all of the multiple second matches are fused based on the multiple fifth matching scores to obtain the second fused perceptual semantics.

[0152] If the feedback data from the sensing device includes multiple sensing results from multiple modalities, and a single modality contains multiple sensing semantics, the central device can perform fusion operations on the sensing semantics of the same modality separately. This results in a more comprehensive fused sensing semantic, thereby improving its accuracy. During the fusion operation on the sensing semantics of the same modality, the matching score of each sensing semantic (which indicates the relevance between the sensing semantic and the first query information) is considered. Multiple sensing semantics of the same modality are fused based on their matching scores. For example, sensing semantics with higher matching scores are given greater importance during the fusion process, which can reduce the impact of some less reliable sensing semantics, thereby improving the accuracy of the fused sensing semantic and ensuring its reliability.

[0153] In one possible implementation of the second aspect, receiving the sensing result from the at least one sensing device includes:

[0154] Receive from at least one of the plurality of sensing devices a plurality of sensing semantics, a plurality of fourth matching scores associated with the plurality of sensing semantics, and a plurality of fifth matching scores associated with the plurality of sensing semantics;

[0155] The method further includes:

[0156] The first fused perceptual semantics are obtained by fusing some or all of the multiple fourth matching scores; the second fused perceptual semantics are obtained by fusing some or all of the multiple second perceptual semantics by fusing some or all of the multiple fifth matching scores.

[0157] If the feedback data from the sensing device includes multiple sensing semantics for a single modality but multiple tasks, the central device can perform a fusion operation on the sensing semantics for the same task separately, making the fused sensing semantics more comprehensive and thus improving the accuracy of the fused sensing semantics. During the fusion operation on the sensing semantics for the same task, multiple sensing semantics for the same task are fused according to their matching scores by considering the matching score of each sensing semantic (which can indicate the relevance between the sensing semantic and the first query information). For example, sensing semantics with higher matching scores are given higher importance during the fusion process, thereby reducing the influence of some less reliable sensing semantics, thus improving the accuracy of the fused sensing semantics and ensuring the reliability of the fused sensing semantics.

[0158] In one possible implementation of the second aspect, the method further includes:

[0159] Send the first fused perceptual semantics to the first GPT device among the at least two GPT devices;

[0160] The second fused perceptual semantics is sent to the second GPT device among the at least two GPT devices.

[0161] The first or second fused perceptual semantics can be processed by the first or second GPT device, respectively, to generate the next query based on the fused input.

[0162] In one possible implementation of the second aspect, the method further includes:

[0163] Determine the sixth matching score of the first fused perceptual semantics; determine the seventh matching score of the second fused perceptual semantics.

[0164] The sixth matching score can indicate the correlation between the first fused perceptual semantics and the first query information, and the seventh matching score can indicate the correlation between the second fused perceptual semantics and the first query information. That is, the sixth matching score and the seventh matching score can be used to evaluate the reliability of the corresponding fused perceptual semantics.

[0165] In a third aspect, this disclosure provides a communication apparatus comprising various modules for performing the communication method according to the first aspect or any possible implementation thereof.

[0166] In a fourth aspect, this disclosure provides a communication apparatus comprising various modules for performing the communication method according to the second aspect or any possible implementation thereof.

[0167] In a fifth aspect, this disclosure provides a sensing device including processing circuitry for performing the communication method according to the first aspect or any possible implementation thereof.

[0168] In a sixth aspect, this disclosure provides a central device including processing circuitry for performing the communication method according to the second aspect or any possible implementation thereof.

[0169] In a seventh aspect, this disclosure provides a communication system comprising a sensing device according to the fifth aspect and a central device according to the sixth aspect.

[0170] In an eighth aspect, this disclosure provides a chip including an input / output (I / O) interface and a processor, wherein the processor is configured to invoke and run a computer program stored in a memory, enabling a device equipped with the chip to perform the communication method according to the first or second aspect or any possible implementation thereof.

[0171] In a ninth aspect, this disclosure provides a computer-readable medium storing computer-executable instructions that, when executed by a processor, cause the processor to perform the communication method according to the first or second aspect or any possible implementation thereof.

[0172] In a tenth aspect, this disclosure provides a computer program product including computer-executable instructions that, when executed by a processor, cause the processor to perform the communication method according to the first or second aspect or any possible implementation thereof.

[0173] This disclosure provides a communication method and related products. A central device broadcasts or multicasts a first query information, a first matching score function, and a first threshold to a sensing device. The sensing device determines, based on the first matching score function, whether the first matching score between the sensed data and the first query information is greater than or equal to the first threshold. When the first matching score is greater than or equal to the first threshold, the sensing device sends the sensing result to the central device. Using broadcast or multicast ensures both high transmission efficiency for the central device and that the determining criteria used by the sensing device are the same as those used by other sensing devices, thereby improving the response accuracy of the sensing device in the horizontal dimension. The sensing device has specific conditions for sending the sensing result, and the sensing result is not continuously sent, thus saving transmission resources. Furthermore, the sent sensing result meets the requirements of the first query information, i.e., irrelevant information is filtered out, and the sent data is what the central device needs, thereby ensuring the accuracy of data transmission. Attached Figure Description

[0174] The accompanying drawings are provided to further understand this disclosure and form part of this specification. They are used to explain this disclosure in conjunction with the following specific exemplary embodiments, but should not be construed as limiting this disclosure.

[0175] Figure 1 This is a schematic diagram of one or more examples of a communication system disclosed herein.

[0176] Figure 2 This is another schematic diagram of a communication system, one or more examples of this disclosure.

[0177] Figure 3 This is a schematic diagram of the basic component structure of one or more examples of a communication system disclosed herein.

[0178] Figure 4 A block diagram of a device in a communication system of one or more examples of this disclosure is shown.

[0179] Figure 5 This is a schematic flowchart illustrating one or more examples of communication methods disclosed herein.

[0180] Figure 6 This is a schematic flowchart illustrating another communication method, one or more examples of this disclosure.

[0181] Figure 7 This is a schematic flowchart of yet another communication method of one or more examples of this disclosure.

[0182] Figure 8 This is yet another schematic diagram of a communication system, one or more examples of the present disclosure.

[0183] Figure 9 This is a schematic diagram of the partitioning of sensing devices according to one or more examples of this disclosure.

[0184] Figure 10 This is a schematic diagram illustrating the interaction between devices in one or more example communication systems disclosed herein.

[0185] Figure 11 This is another schematic diagram of the interaction between devices in one or more example communication systems of this disclosure.

[0186] Figure 12 This is a schematic diagram illustrating one or more examples of the present disclosure of generating a series of query messages in a communication system.

[0187] Figure 13 This is a schematic diagram illustrating the interaction between a central device and two sensing devices in one or more example communication systems of this disclosure.

[0188] Figure 14 This is another schematic diagram of the interaction between a central device and two sensing devices in one or more example communication systems of this disclosure.

[0189] Figure 15 This is a schematic diagram of a GPT device in one or more example communication systems of this disclosure generating query semantics.

[0190] Figure 16 This is a schematic diagram illustrating one or more examples of this disclosure of retrieving a query message from query semantics in a communication system.

[0191] Figure 17 This is a schematic diagram illustrating how a GPT device in one or more example communication systems of this disclosure generates query terms.

[0192] Figure 18 This is a schematic diagram of one or more examples of the sensing device responding to query terms from a central device.

[0193] Figure 19 This is a schematic diagram of a scoring operation implemented by a sensing device according to one or more examples of this disclosure.

[0194] Figure 20 This is a schematic diagram of one or more examples of the sensing device responding to query semantics from a central device.

[0195] Figure 21 This is another schematic diagram of a scoring operation implemented by a sensing device according to one or more examples of this disclosure.

[0196] Figure 22 This is another schematic diagram of a sensing device responding to a query semantic from a central device, representing one or more examples of this disclosure.

[0197] Figure 23 This is yet another schematic diagram of a scoring operation implemented by a sensing device according to one or more examples of this disclosure.

[0198] Figure 24 This is a block diagram of one or more examples of a communication device disclosed herein.

[0199] Figure 25 This is a block diagram of another communication device, which is one or more examples of this disclosure. Detailed Implementation

[0200] In the following description, reference is made to the accompanying drawings, which form part of this disclosure, which illustrate by way of description specific aspects of the present disclosure or aspects in which the present disclosure may be used. It should be understood that the present disclosure may be used in other aspects and includes structural or logical variations not depicted in the drawings. Therefore, the following detailed description should not be construed in a limiting sense, and the scope of this disclosure is defined by the appended claims.

[0201] To aid in understanding this disclosure, examples of wireless communication systems and devices are described below.

[0202] Exemplary communication systems and devices

[0203] This disclosure employs at least one UE (i.e., sensing device, also called sensing node) in a wireless system. Figure 1 The interaction and processing between ED (designated as ED), at least one BS (i.e., central device), and at least one GPT device is described as an illustrative example. The exchanged information and protocol flows can also be used between other network nodes described below, such as between ED 110 and TRP 170, between ED 110 and the core network, between ED 110 and ED 110, between TRP 170 and TRP 170, and between TRP 170 and GPT device 180. The UE in the process described in this disclosure can be replaced by the sensing nodes mentioned below. The BS in the process described in this disclosure can be replaced by a sensing coordinator. A sensing coordinator is a node in the network that assists in sensing operations. These nodes can be independent nodes dedicated to sensing operations or other nodes (e.g., TRP 170, ED 110, or core network nodes shown below) that perform sensing operations in parallel with communication transmissions.

[0204] refer to Figure 1 As a non-limiting illustrative example, a simplified schematic diagram of a communication system is provided. Communication system 100 (which may be...) Figure 1The wireless system 100 includes a radio access network 120. The radio access network 120 can be a next-generation (e.g., sixth-generation, 6G or later) radio access network or a traditional (e.g., 5G, 4G, 3G or 2G) radio access network. One or more communication electronic devices (EDs) 110a, 110b, 110c, 110d, 110e, 110f, 110g, 110h, 110i, 110j (generally referred to as 110) can interconnect with each other or connect to one or more network nodes (170a, 170b, generally referred to as 170) within the radio access network 120. The core network 130 can be part of the communication system 100 and can depend on or be independent of the radio access technology used in the communication system 100. Furthermore, the communication system 100 includes a public switched telephone network (PSTN) 140, the Internet 150, and other networks 160.

[0205] Uplink messages / data transmitted between the central device (e.g., network node 170) and the sensing device (e.g., ED 180) can be carried in higher-layer signaling such as RRC signaling or MAC layer signaling. Alternatively, these messages / data can be carried in physical layer signaling (e.g., UCI). Alternatively, these messages / data can be carried in a combination of higher-layer signaling and physical signaling. It should be noted that messages in this disclosure can be replaced by information, which can be carried in a single message or in more than one separate message. Downlink messages / data transmitted between the central device and ED 110 can be carried in higher-layer signaling such as RRC signaling or MAC layer signaling. Alternatively, these messages / data can be carried in physical layer signaling (e.g., DCI). Alternatively, these messages / data can be carried in a combination of higher-layer signaling and physical signaling. It should be noted that messages in this disclosure can be replaced by information, which can be carried in a single message or in more than one separate message.

[0206] Additionally, the communication system 100 includes at least one GPT device 180. The GPT device 180 may be located within one or more network nodes 170. The GPT device 180 may be a standalone device connected to the network 170, such as an ED 110 connected to the network node 170 via a Uu interface. The GPT device 180 may also be a device connected to the network node 170 via the core network 130. When the GPT device 180 is an ED, uplink messages / data transmitted between the central device (e.g., network node 170) and the GPT device 180 may be carried in higher-layer signaling such as RRC signaling or MAC layer signaling. Alternatively, these messages / data may be carried in physical layer signaling (e.g., UCI). Alternatively, these messages / data may be carried in a combination of higher-layer and physical signaling. It should be noted that messages in this disclosure can be replaced by information, which may be carried in a single message or in more than one separate message. Downlink messages / data transmitted between the central device and GPT device 180 can be carried in higher-layer signaling such as RRC signaling or MAC layer signaling. Alternatively, these messages / data can be carried in physical layer signaling (e.g., DCI). Or, these messages / data can be carried in a combination of higher-layer signaling and physical signaling. It should be noted that messages in this disclosure can be replaced by information, which can be carried in a single message or in more than one separate message.

[0207] Figure 2 An exemplary communication system 100 is illustrated. Generally, the communication system 100 enables multiple wireless or wired components to transmit data and other content. The purpose of the communication system 100 may be to provide content such as voice, data, video, signaling, and / or text via broadcast, multicast, and unicast. The communication system 100 can operate by sharing resources (e.g., carrier spectrum bandwidth) among its constituent units. The communication system 100 may include terrestrial communication systems and / or non-terrestrial communication systems. The communication system 100 can provide a wide range of communication services and applications (e.g., earth monitoring, remote sensing, passive sensing and positioning, navigation and tracking, automated delivery and mobility, etc.). The communication system 100 can provide high availability and robustness through the joint operation of terrestrial and non-terrestrial communication systems. For example, integrating a non-terrestrial communication system (or components thereof) into a terrestrial communication system can create a heterogeneous network comprising multiple layers. Compared to traditional communication networks, heterogeneous networks can achieve better overall performance through efficient multi-link joint operation, more flexible function sharing, and faster physical layer link switching between terrestrial and non-terrestrial networks.

[0208] Terrestrial communication systems and non-terrestrial communication systems can be considered subsystems of a communication system. Figure 2In the example shown, communication system 100 includes electronic devices (EDs) 110a, 110b, 110c, and 110d (generally referred to as ED 110), radio access networks (RANs) 120a to 120b, a non-terrestrial communication network 120c, a core network 130, a public switched telephone network (PSTN) 140, the Internet 150, and other networks 160. RANs 120a to 120b include corresponding base stations (BSs) 170a to 170b, which are generally referred to as terrestrial transmit and receive points (T-TRPs) 170a to 170b. The non-terrestrial communication network 120c includes access nodes 172, which are generally referred to as non-terrestrial transmit and receive points (NT-TRPs) 172.

[0209] Alternatively or additionally, any ED 110 can be used to connect, access, or communicate with any T-TRP 170a to 170b, NT-TRP 172, Internet 150, core network 130, PSTN 140, other network 160, or any combination thereof. In some examples, ED 110a can perform uplink and / or downlink transmissions with T-TRP 170a via terrestrial air interface 190a. In some examples, ED 110a, 110b, 110c, and 110d can also communicate directly with each other via one or more sidelink air interfaces 190b. In some examples, ED 110d can perform uplink and / or downlink transmissions with NT-TRP 172 via non-terrestrial air interface 190c.

[0210] Air interfaces 190a and 190b can use similar communication technologies, such as any suitable wireless access technology. For example, communication system 100 can implement one or more channel access methods in air interfaces 190a and 190b, such as code division multiple access (CDMA), space division multiple access (SDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal FDMA (OFDMA), direct Fourier transform spread OFDMA (DFT-OFDMA), or single-carrier FDMA (SC-FDMA). Air interfaces 190a and 190b can utilize other higher-dimensional signal spaces, which may involve combinations of orthogonal and / or non-orthogonal dimensions.

[0211] The non-terrestrial air interface 190c enables communication between the ED 110d and one or more NT-TRP 172s via a wireless link (or simply a link). In some examples, the link is a dedicated connection for unicast transmission, a connection for broadcast transmission, or a connection for multicast transmission between a group of ED 110s and one or more NT-TRP 172s.

[0212] RANs 120a and 120b communicate with core network 130 to provide various services, such as voice, data, and other services, to EDs 110a, 110b, and 110c. RANs 120a and 120b and / or core network 130 may communicate directly or indirectly with one or more other RANs (not shown), which may or may not be directly served by core network 130, and may or may not use the same radio access technology as RANs 120a, RAN 120b, or both. Core network 130 may also serve as a gateway access between (i) RANs 120a and 120b or EDs 110a, 110b, and 110c, or both, and (ii) other networks (e.g., PSTN 140, Internet 150, and other networks 160). Additionally, some or all of EDs 110a, 110b, and 110c may include the ability to communicate with different wireless networks via different radio links using different radio technologies and / or protocols. ED 110a, 110b, and 110c can communicate with a service provider or exchange (not shown) via a wired communication channel and with the Internet 150, rather than wirelessly (or as a supplement to wireless communication). PSTN 140 may include a circuit-switched telephone network for providing plain old telephone service (POTS). The Internet 150 may include computer networks and subnets (corporate intranets) or both, and incorporate protocols such as Internet Protocol (IP), Transmission Control Protocol (TCP), and User Datagram Protocol (UDP). ED 110a, 110b, and 110c may be multimode devices capable of operating under various wireless access technologies and include multiple transceivers required to support these technologies.

[0213] Basic component structure

[0214] Figure 3Another example of the ED 110 and base stations 170a, 170b, and / or 170c is shown. The ED 110 is used to connect people, objects, machines, etc. The ED 110 can be widely used in various scenarios, such as cellular communication, device-to-device (D2D), vehicle-to-everything (V2X), peer-to-peer (P2P), machine-to-machine (M2M), machine-type communications (MTC), Internet of Things (IoT), virtual reality (VR), augmented reality (AR), mixed reality (MR), metaverse, digital twin, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, smart cities, drones, robots, remote sensing, passive sensing, positioning, navigation and tracking, autonomous delivery and mobility, etc.

[0215] Each ED 110 represents any end-user equipment suitable for wireless operation and may include (or be referred to as): user equipment / device (UE), wireless transmit / receive unit (WTRU), mobile station, fixed or mobile subscriber unit, cellular phone, station (STA), machine type communication (MTC) equipment, personal digital assistant (PDA), smartphone, laptop, computer, tablet, wireless sensor, consumer electronics, smartbook, vehicle, automobile, truck, bus, train, or IoT device, wearable device (such as watch, head-mounted device, glasses), industrial equipment, or devices within the aforementioned equipment (e.g., communication module, modem, or chip), etc. Future generations of ED 110 may be referred to using other terms. Each base station 170a and 170b is a T-TRP, referred to below as T-TRP 170. Similarly, Figure 3As shown, NT-TRP is referred to as NT-TRP 172 below. Each ED 110 connected to T-TRP 170 and / or NT-TRP 172 can be dynamically or semi-statically enabled (i.e., established, activated, or enabled), disabled (i.e., released, deactivated, or disabled), and / or configured in response to one or more of connectivity availability and connectivity necessity.

[0216] ED 110 includes a transmitter 201 and a receiver 203 coupled to one or more antennas 204. Only one antenna 204 is shown in the figure. One, some, or all of the antennas 204 may also be panels. The transmitter 201 and receiver 203 may be integrated together, for example, as a transceiver. The transceiver is used to modulate data or other content transmitted by at least one antenna 204 or a network interface controller (NIC). The transceiver is also used to demodulate data or other content received by at least one antenna 204. Each transceiver includes any structure suitable for generating signals for wireless or wired transmission and / or processing signals received wirelessly or wiredly. Each antenna 204 includes any structure suitable for transmitting and / or receiving wireless or wired signals.

[0217] ED 110 includes at least one memory 208. Memory 208 stores instructions and data used, generated, or collected by ED 110. For example, memory 208 may store software instructions or modules executed by one or more processing units (e.g., processor 210) for implementing some or all of the functions and / or embodiments described herein. Each memory 208 includes any suitable volatile and / or non-volatile storage and retrieval device. Any suitable type of memory can be used, such as random access memory (RAM), read-only memory (ROM), hard disk, optical disk, subscriber identity module (SIM) card, memory stick, secure digital (SD) memory card, processor cache, etc.

[0218] ED 110 may also include one or more input / output devices (not shown) or interfaces (e.g., connected to...). Figure 1 The input / output device supports interaction with users or other devices on the network. Each input / output device includes any structure suitable for (e.g., by operation) providing information to or receiving information from a user, such as a speaker, microphone, numeric keypad, keyboard, display, or touchscreen, including network interface communication.

[0219] ED 110 includes processor 210 for performing various operations, including operations related to: preparing to transmit uplink transmissions to NT-TRP 172 and / or T-TRP 170, processing downlink transmissions received from NT-TRP 172 and / or T-TRP 170, and processing lateral link transmissions to and from another ED 110. Processing operations related to preparing to transmit uplink transmissions may include operations such as encoding, modulation, transmit beamforming, and generating symbols for transmission. Processing operations related to processing downlink transmissions may include operations such as receive beamforming, demodulation, and decoding of received symbols. According to an embodiment, receiver 203 may receive downlink transmissions (possibly using receive beamforming), and processor 210 may extract signaling from the downlink transmissions (e.g., by detecting and / or decoding signaling). Examples of signaling may be reference signals transmitted by NT-TRP 172 and / or T-TRP 170. In some embodiments, processor 210 implements transmit beamforming and / or receive beamforming based on beam direction indications received from T-TRP 170, such as beam angle information (BAI). In some embodiments, processor 210 may perform operations related to network access (e.g., initial access) and / or downlink synchronization, such as operations related to detecting synchronization sequences, decoding, and acquiring system information. In some embodiments, processor 210 may perform channel estimation (e.g., using reference signals received from NT-TRP 172 and / or T-TRP 170).

[0220] Although not shown, processor 210 may be part of transmitter 201 and / or receiver 203. Although not shown, memory 208 may be part of processor 210.

[0221] The processing components of processor 210, transmitter 201, and receiver 203 can each be implemented by one or more processors, which may be the same or different, for executing instructions stored in memory (e.g., memory 208). Alternatively, some or all of the processing components of processor 210, transmitter 201, and receiver 203 can be implemented using special-purpose circuits such as a field-programmable gate array (FPGA), graphics processing unit (GPU), central processing unit (CPU), or application-specific integrated circuit (ASIC).

[0222] In some implementations, ED 110 may be a device (also referred to as a component) such as a communication module, modem, chip, or chipset, including at least one processor 210 and an interface or at least one pin. In this scenario, the transmitter 201 and receiver 203 may be replaced by an interface or at least one pin, provided that the interface or at least one pin is used to connect the device (e.g., a chip) and other devices (e.g., a chip, memory, or bus). Therefore, sending information to NT-TRP 172 and / or T-TRP 170 and / or another ED 110 can be referred to as sending information to an interface or at least one pin, or as sending information to NT-TRP 172 and / or T-TRP 170 and / or another ED 110 via an interface or at least one pin, while receiving information from NT-TRP 172 and / or T-TRP 170 and / or another ED 110 can be referred to as receiving information from an interface or at least one pin, or as receiving information from NT-TRP 172 and / or T-TRP 170 and / or another ED 110 via an interface or at least one pin. The information may include control signaling and / or data.

[0223] In some implementations, the T-TRP 170 can be referred to by other names, such as base station, basetransceiver station (BTS), wireless base station, network node, network device, network-side device, transmit / receive node, Node B, evolved Node B (eNodeB or eNB), Home eNodeB, Next Generation Node B (gNB), transmission point (TP), site controller, access point (AP), wireless router, relay station, remote radio head, ground node, ground network device, ground base station, base band unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), positioning node, etc. The T-TRP 170 can be a macro BS, pico BS, relay node, host node, or a combination thereof. T-TRP 170 may refer to the aforementioned device or a component within the aforementioned device (e.g., a communication module, modem, or chip).

[0224] In some embodiments, the various parts of T-TRP 170 may be distributed. For example, some modules in T-TRP 170 may be located remotely from the device housing the antenna 256 of T-TRP 170 and may be coupled to the device housing the antenna 256 via a communication link (not shown) sometimes referred to as a fronthaul (e.g., a common public radio interface (CPRI)). Therefore, in some embodiments, the term T-TRP 170 may also refer to network-side modules that perform processing operations such as location determination, resource allocation (scheduling), message generation, and encoding / decoding of ED 110, and these modules are not necessarily part of the device housing the antenna 256 of T-TRP 170. These modules may also be coupled to other T-TRPs. In some embodiments, T-TRP 170 may actually be multiple T-TRPs that operate together (e.g., by using coordinated multicast) to service ED 110.

[0225] T-TRP 170 includes at least one transmitter 252 and at least one receiver 254 coupled to one or more antennas 256. Only one antenna 256 is shown in the figure. One, some, or all of the antennas 256 may also be panels. The transmitter 252 and receiver 254 may be integrated as a transceiver. T-TRP 170 also includes a processor 260 for performing various operations, including operations related to: preparing to transmit downlink transmissions to ED 110, processing uplink transmissions received from ED 110, preparing to transmit backhaul transmissions to NT-TRP 172, and processing transmissions received from NT-TRP 172 via backhaul. Processing operations related to preparing to transmit downlink or backhaul transmissions may include operations such as encoding, modulation, precoding (e.g., multiple-input multiple-output (MIMO) precoding), transmit beamforming, and generating symbols for transmission. Processing operations related to processing transmissions received in the uplink or backhaul may include operations such as receive beamforming, demodulating received symbols, and decoding received symbols. Processor 260 can also perform operations related to network access (e.g., initial access) and / or downlink synchronization, such as generating the contents of a synchronization signal block (SSB), generating system information, etc. In some embodiments, processor 260 also generates a beam direction indication, such as a BAI, which scheduler 253 can schedule for transmission. Processor 260 performs other network-side processing operations described herein, such as determining the location of ED 110, determining the location for deploying NT-TRP 172, etc. In some embodiments, processor 260 can generate signaling, such as for configuring one or more parameters of ED 110 and / or one or more parameters of NT-TRP 172. Any signaling generated by processor 260 is transmitted by transmitter 252. Note that "signaling" as used herein may also be referred to as control signaling. Dynamic signaling can be transmitted in control channels such as the physical downlink control channel (PDCCH), and static or semi-static higher-layer signaling can be included in packets transmitted in data channels such as the physical downlink shared channel (PDSCH).

[0226] Scheduler 253 may be coupled to processor 260. Scheduler 253 may be included within T-TRP 170 or may operate separately from it. Scheduler 253 may schedule uplink, downlink, and / or backlink transmissions, including issuing scheduling grants and / or configuring unscheduled (“configuration grants”) resources. T-TRP 170 also includes memory 258 for storing information and data. Memory 258 stores instructions and data used, generated, or collected by T-TRP 170. For example, memory 258 may store software instructions or modules for implementing some or all of the functions and / or embodiments described herein, which are executed by processor 260.

[0227] Although not shown, processor 260 may constitute part of transmitter 252 and / or receiver 254. Furthermore, although not shown, processor 260 may implement scheduler 253. Although not shown, memory 258 may constitute part of processor 260.

[0228] The processor 260, the scheduler 253, the processing components of the transmitter 252, and the processing components of the receiver 254 can each be implemented by one or more processors, which may be the same or different, for executing instructions stored in memory (e.g., memory 258). Alternatively, some or all of the processing components of the processor 260, scheduler 253, transmitter 252, and receiver 254 can be implemented using dedicated circuitry such as FPGA, CPU, GPU, or ASIC.

[0229] When T-TRP 170 is a device (also referred to as a component) such as a communication module, modem, chip, or chipset in a device, it includes at least one processor and an interface or at least one pin. In this scenario, transmitter 252 and receiver 254 can be replaced by an interface or at least one pin, requiring that the interface or at least one pin be used to connect the device (e.g., a chip) and other devices (e.g., chips, memory, or buses). Therefore, sending information to NT-TRP 172 and / or T-TRP 170 and / or ED110 can be referred to as sending information to an interface or at least one pin, while receiving information from NT-TRP 172 and / or T-TRP 170 and / or ED 110 can be referred to as receiving information from an interface or at least one pin. The information may include control signaling and / or data.

[0230] Although the NT-TRP 172 is illustrated only as an example of a drone, it can be implemented in any suitable non-terrestrial form, such as an aerial platform, a satellite, an aerial platform as an international mobile telecommunications base station, or an unmanned aerial vehicle, which will be discussed below. Furthermore, in some implementations, the NT-TRP 172 may be referred to by other names, such as a non-terrestrial node, a non-terrestrial network device, or a non-terrestrial base station. The NT-TRP 172 includes a transmitter 272 and a receiver 274 coupled to one or more antennas 280. Only one antenna 280 is shown in the figure. One, some, or all of the antennas may also be panels. The transmitter 272 and receiver 274 may be integrated as a transceiver. The NT-TRP 172 also includes a processor 276 for performing various operations, including operations related to: preparing downlink transmissions to ED 110, processing uplink transmissions received from ED 110, preparing return transmissions to T-TRP 170, and processing transmissions received from T-TRP 170 via return transmissions. Processing operations related to preparing to transmit downlink or backhaul transmissions may include operations such as encoding, modulation, precoding (e.g., MIMO precoding), transmit beamforming, and generating symbols for transmission. Processing operations related to processing transmissions received in the uplink or backhaul may include operations such as receive beamforming, demodulating received symbols, and decoding received symbols. In some embodiments, processor 276 implements transmit beamforming and / or receive beamforming based on beam direction information (e.g., BAI) received from T-TRP 170. In some embodiments, processor 276 may generate signaling, for example, to configure one or more parameters of ED 110. In some embodiments, NT-TRP 172 implements physical layer processing but does not implement higher-layer functions, such as medium access control (MAC) or radio link control (RLC) layer functions. This is just an example; more generally, NT-TRP 172 may implement higher-layer functions in addition to physical layer processing.

[0231] The NT-TRP 172 also includes a memory 278 for storing information and data. Although not shown, a processor 276 may form part of the transmitter 272 and / or the receiver 274. Although not shown, the memory 278 may form part of the processor 276.

[0232] The processing components of processor 276, transmitter 272, and receiver 274 can each be implemented by one or more processors, which may be the same or different, for executing instructions stored in memory (e.g., memory 278). Alternatively, some or all of the processing components of processor 276, transmitter 272, and receiver 274 can be implemented using dedicated circuitry such as a programmable FPGA, GPU, CPU, or ASIC. In some embodiments, NT-TRP 172 may actually be multiple NT-TRPs that operate together (e.g., by using coordinated multicast transmissions) to service ED 110.

[0233] When NT-TRP 172 is a device within a device (e.g., a communication module, modem, chip, or chipset), it includes at least one processor, an interface, or at least one pin. In this scenario, transmitter 272 and receiver 257 can be replaced by an interface or at least one pin, requiring that the interface or at least one pin be used to connect the device (e.g., a chip) and other devices (e.g., a chip, memory, or bus). Therefore, sending information to T-TRP 170 and / or another NT-TRP 172 and / or ED 110 can be referred to as sending information to an interface or at least one pin, while receiving information from T-TRP 170 and / or another NT-TRP 172 and / or ED 110 can be referred to as receiving information from an interface or at least one pin. The information may include control signaling and / or data.

[0234] Note that, as used in this article, "TRP" can refer to either T-TRP or NT-TRP. T-TRP can also be called terrestrial network TRP ("TN TRP"), and NT-TRP can also be called non-terrestrial network TRP ("NTN TRP").

[0235] T-TRP 170, NT-TRP 172 and / or ED 110 may include other components, but these components have been omitted for clarity.

[0236] Any or all of ED 110 and BS 170 can be sensing nodes in system 100. A sensing node is a network entity that senses by sending and receiving sensing signals. Some sensing nodes are communication devices that simultaneously communicate and sense. However, some sensing nodes may not communicate but are dedicated solely to sensing. Sensing agent 174 is an example of a sensing node dedicated solely to sensing. Unlike ED 110 and BS 170, sensing agent 174 does not send or receive communication signals. However, sensing agent 174 can transmit configuration information, sensing information, signaling information, or other information within communication system 100. Sensing agent 174 can communicate with core network 130 to transmit information with the rest of communication system 100. As an example, sensing agent 174 can determine the location of ED 110a and transmit this information to base station 170a via core network 130. Although... Figure 2 Only one sensing agent 174 is shown, but any number of sensing agents can be implemented in the communication system 100. In some embodiments, one or more sensing agents can be implemented at one or more RANs 120.

[0237] Sensing nodes can combine sensing-based technologies with reference signal-based technologies to enhance UE pose determination. This type of sensing node can also be called a sensing management function (SMF). In some networks, the SMF can also be called a location management function (LMF). The SMF can be implemented as a physically independent entity located at core network 130 and connected to multiple BS 170s. In other aspects of this application, the SMF can be implemented as a logical entity co-located within BS 170 through logic executed by processor 260.

[0238] although Figure 3 Not shown, but may include GPT device 180, which has a similar structure to ED 110. For example, GPT device 180 includes at least one processor, transmitter and receiver.

[0239] Basic module structure

[0240] One or more steps of the method provided in this article can be derived from... Figure 4 The corresponding unit or module is executed. Figure 4The diagram illustrates units or modules within a device (e.g., ED 110, T-TRP 170, NT-TRP 172, or GPT device 180). For example, signals may be transmitted by a transmitting unit or transmitting module. Signals may be received by a receiving unit or receiving module. Signals may be processed by a processing unit or processing module. Other steps may be performed by an artificial intelligence (AI) module or a machine learning (ML) module. The corresponding units or modules may be implemented using hardware, one or more components or devices executing software, or a combination thereof. For example, one or more of these units or modules may be a programmable integrated circuit such as an FPGA, GPU, CPU, or ASIC. It should be understood that if these modules are implemented, for example, using software executed by a processor, the processor can retrieve these modules, in whole or in part, as needed, individually or collectively for processing, in one or more instances, and these modules themselves may include instructions for further deployment and instantiation. Reference Figure 3 The sender mentioned could be a specific implementation of the sending module. (See reference.) Figure 3 The receiver mentioned could be a specific implementation of the receiving module. (See reference.) Figure 3 The processor mentioned can be a specific implementation of the processing module.

[0241] Further details regarding ED 110, T-TRP 170, NT-TRP 172, and GPT device 180 are known to those skilled in the art. Therefore, these details are omitted herein.

[0242] Exemplary concepts of some terms

[0243] Message: A payload in a natural language (e.g., English, French, Chinese, etc.).

[0244] Query message: A query statement in natural language.

[0245] Perceptual messages: Descriptions of observed or perceived data in natural language.

[0246] Semantics: Vectors, matrices, and tensors composed of scalars, used to embed messages.

[0247] Query semantics: The semantics of embedded query messages.

[0248] Perceptual semantics: the semantics of embedded perceptual messages.

[0249] Lexicon: A vector of scalars encoded according to semantics.

[0250] Query terms: Terms encoded according to the query semantics.

[0251] Perceptual lexical units: Lexical units encoded based on perceptual semantics.

[0252] GPT device: A device that runs one or more generative AI models to generate one or more query messages based on one or more perception messages.

[0253] Central equipment: As a BS device, it connects multiple terminal devices in DL and UL via wireless access and connects to the core network through the backbone network.

[0254] Sensing device: A device that acts as an end point, connected to one or more BSs, and equipped with sensing gadgets to measure nearby data of interest.

[0255] The foregoing describes possible scenarios or general descriptions of examples of this disclosure, and the motivation and technical concept of this disclosure will be explained below.

[0256] The sixth-generation (6G) system will integrate sensing capabilities. A large number of user equipment (UE) or sensing devices will be densely deployed in cities, factories, farms, and other locations. Besides mobile phones, sensing devices will become a crucial type of UE or device that heralds the arrival of the IoT era.

[0257] Just like internet search engines, 6G will bring corresponding Internet of Things (IoT) search engines to the real physical world. In fact, billions of IoT-based applications, such as self-driving cars, automated factories, smart cities, and autonomous farms, will rely heavily on efficient, real-time search engines in our physical world.

[0258] In recent years, artificial intelligence (AI) has made breakthroughs in various fields of intelligence and cognition. Some AIs are exploring cutting-edge knowledge in fields such as chemistry, gaming, mathematics, and genetic engineering, while others are providing human-level question-and-answer platforms in the digital world. The area AI has yet to conquer is the real-time physical world. AI technology in the physical world may be built upon the ubiquitous IoT connectivity brought about by 6G, in which AI technology will permeate every aspect of our society and life.

[0259] More challenging than internet search engines are real-world search engines, which must search the physical world in real time over vast physical areas and process diverse types of data and information (some novel, others yet to be invented). Furthermore, green technology, low energy consumption, and low emissions are key characteristics of 6G. Sensing devices can be battery-powered and / or entirely powered by solar and wind energy. Requiring all sensing devices on a large scale to simultaneously report what they sense would be prohibitively expensive and impractical. On one hand, frequent sensing and transmission would consume significant amounts of energy from sensing devices, shortening their battery life; on the other hand, such a high-density IoT deployment could clog uplink channels, especially considering that uplink (UL) bandwidth is more expensive than downlink (DL) bandwidth.

[0260] In some implementations, the sensing device can be a UE, a mobile phone, or a handset, where it is assumed that any two sensing devices are independent of each other; therefore, the wireless system associated with the sensing device can schedule the sensing device independently; the sensing data measured by the sensing device can be the application-level payload of the wireless system and protocol.

[0261] The aforementioned sensing device scheduling scheme is inefficient in terms of wireless bandwidth and energy consumption. For example, sensing devices may continuously and blindly send their sensing data to the central device, regardless of whether this sensing data is needed.

[0262] From a higher-level perspective, a better strategy is to only activate sensing devices for measurement and transmission when the sensing data from multiple sensing devices can serve one or more objectives; for example, it would be futile to continuously send irrelevant information to an autonomous vehicle or send all the moving obstacles in its vicinity to an autonomous vehicle parked on the side of the road when the generative pre-trained transformer (GPT) device, such as an autonomous vehicle, might request information about moving obstacles in its vicinity.

[0263] To ensure zero information loss, resources in the wireless system described above may be over-scheduled.

[0264] The basic concept of this disclosure can be summarized as follows. When a query is received (or referred to as first query information, query, query message, etc.), not all sensing devices will respond with the content they have sensed. Only sensing devices whose sensed data is sufficiently relevant to the query information (i.e., the matching score between the sensed data and the query information is greater than or equal to a threshold) will respond and send their sensed data. For example, a central device (or BS) can broadcast a semantic query, and only sensing devices (or UEs) with corresponding results will respond with semantic results, thereby significantly reducing UL transmission overhead. The solution provided in this disclosure can be applied to object detection, sensing tracking, V2X communication, etc.

[0265] The foregoing briefly described the technical concept of this disclosure, and specific examples of this disclosure will then be elaborated in the following description.

[0266] This disclosure provides a communication method, such as Figure 5 As shown, the communication method can be implemented by a sensing device and may include the following steps.

[0267] Step 502: The sensing device receives the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device.

[0268] The first query information is used to retrieve relevant data from the sensing device. Specifically, the first query information can be a question in natural language or machine-readable language; this document does not limit this. For example, the first query information can take the form of a query message, query semantics, query terms, etc. Details regarding the first query information will be described later.

[0269] The first matching score function can be used to calculate a first matching score between the sensing data from the sensing device and the first query information. The sensing device decides whether to send the sensing result based on the comparison between the first matching score and a first threshold. It should be noted that the threshold in this disclosure can be predefined or determined according to actual needs.

[0270] Step 504: The sensing device determines whether the first matching score between the sensing data and the first query information is greater than or equal to the first threshold based on the first matching score function.

[0271] Sensing devices are responsible for measuring and / or collecting local physical world data. These can be sensing UEs, sensing devices, IoT devices, UEs, mobile phones, handsets, or other devices. Sensing devices can be equipped with sensing gadgets or components for measuring local physical world data or information, which can be referred to as sensing data. Furthermore, sensing devices can encode the sensing data and send it to a central device.

[0272] Step 506: When the first matching score is greater than or equal to the first threshold, the sensing device sends the sensing result to the central device, wherein the sensing result indicates the sensing data.

[0273] The central device can be a base station (BS), such as a gNB or eNB, or it can be an access point (AP). The sensing result is related to the sensing data acquired by the sensing device. If the first matching score is greater than or equal to a first threshold, the sensing device will respond with the sensing result. If the first matching score is less than the first threshold, the sensing device will not respond. Details regarding the specific content of the sensing result will be described later.

[0274] It should be noted that in some cases, some sensing devices may proactively send their sensing results without receiving any query information from the central device. Sensing devices that proactively send sensing results may be responding to urgent queries, such as fire alarms or vehicle accidents. In a sense, some query messages have been predefined and configured into the system by default.

[0275] The central device broadcasts or multicasts a first query message to wake up the sensing devices and measure the sensing data. The sensing devices calculate a first matching score between the sensing data and the first query message based on a first matching score function broadcast or multicast by the central device. If the first matching score is greater than or equal to a first threshold, the sensing device sends the sensing result to the central device. First, the central device broadcasts or multicasts the first query message, the first matching score function, and the first threshold, ensuring that the criteria used to determine the sensing devices are the same as those used by other sensing devices, thus improving the response accuracy of the sensing devices in the horizontal dimension. Second, the transmission of the sensing result is conditional and not continuous, thus saving transmission resources. Third, the transmitted sensing result meets the requirements of the first query message, i.e., irrelevant information is filtered out, and the transmitted data is what the central device needs, thus ensuring the response accuracy of the sensing devices in the vertical dimension.

[0276] The first query is goal-oriented or task-oriented, triggering the sensing device to complete one or more tasks, collecting and sending data from one or more modalities. It should be noted that tasks and modalities can be independent or related. For example, the central device (or BS) can broadcast or multicast queries for a single task, a single modality, multiple modalities, or multiple tasks. In this case, the sensing device can respond to a task by feeding back data from one or more modalities; the sensing data for one modality can be the request data corresponding to one or more tasks. Some implementation methods are described below.

[0277] In one possible implementation, the first query information includes query information for a task or query information for a modality. The query information, also referred to as a query, can take the form of a query message, query semantics, or query terms. Specifically, the sensing device can receive instructions from the central device regarding task identifiers or modality identifiers. The instructions for task identifiers and modality identifiers can be one (or more) fields, without limitation here. Task identifiers are used to distinguish specific tasks, and modality identifiers are used to distinguish specific modalities. Queries for a single task or a single modality can be indicated by a single-task / modality message format. For example, a single-task / modality message format could be: for n queries, {q1, q2, ..., q...} n}, where n≥1; each q j From a length of N j vector or N j ×M j Matrix representation. Task / modality identifiers can be explicitly or implicitly indicated. A sensing device can serve a single task or a single modality, meaning it can be dedicated to one task or one modality, thus improving its response efficiency.

[0278] In one possible implementation, the first query information includes multiple query messages for multiple tasks and multiple task identifiers associated with each of the multiple query messages. The queries for multiple tasks can be indicated by a multi-task message format, where each task carries a task identifier. For example, the multi-task message format could be: For m tasks, {{{t1,{q 1,1 , .., q 1,n1}},.., {t m , {q m,1 , .., q 1m,nm}}}, each task t i There are n i There are n queries, m≥1,n i ≥1; t i It is the identifier of the i-th task; each q i,j From a length of N i,j vector or N j ×M j Matrix representation. Sensing devices can serve multiple tasks simultaneously, thereby improving the resource utilization of sensing devices.

[0279] In one possible implementation, the first query information includes multiple query messages for multiple modalities and multiple modality identifiers associated with each of the multiple query messages. The queries for multiple modalities can be indicated by a multimodal message format, where each modality carries a modality identifier. For example, the multimodal message format could be: for m modalities, {{{y1,{q 1,1 , .., q1,n1}},.., {y m , {q m,1 , .., q 1m,nm}}}, each mode y i There are n i There are n queries, m≥1,n i ≥1. y i It is the identifier of the i-th mode; each q i,j From a length of N i,j vector or N j ×M j Matrix representation. Sensing devices can serve multiple modalities simultaneously, meaning they can sense and respond to more than one modality (i.e., one type of data), which enriches the types of data the sensing device responds to.

[0280] In one possible implementation, the first query information includes a first plurality of query information for multiple tasks, a second plurality of query information for multiple modalities, multiple task identifiers associated with the first plurality of query information, and multiple modal identifiers associated with the second plurality of query information. Queries for multiple tasks and multiple modalities can be indicated by a multimodal and multitasking message format, wherein each task has multiple modalities or each modality has multiple tasks. The description of the query information for multiple tasks and multiple modalities is similar to the previous description and will not be repeated here. The sensing device can simultaneously serve multiple tasks and multiple modalities, which not only improves the resource utilization of the sensing device but also enriches the types of data the sensing device responds to.

[0281] The above describes the task- and / or modality-related information that may be included in the first query information. The following describes the transmission method of the first query information.

[0282] The first query information can be sent from the central device to the sensing device via broadcast or multicast. In one example, the sensing device can receive the first query information, the first matching score function, and the first threshold broadcast by the central device in a synchronization signal block (SSB) / physical broadcast channel (PBCH) block. The first query information is carried in a master information block (MIB) message or a system information block (SIB) message, which includes an indication of whether the first query information is presented in the MIB message or SIB message. By indicating whether the first query information is presented, signaling overhead can be saved because no further processing is required if the indication does not show the first query information. In one possible implementation, the MIB message or SIB message also includes an indication of the period of the first query information. The first query information is valid within the period and may be invalid outside the period, thereby avoiding resource consumption. It should be noted that the indication of whether the first query information is presented and the indication of the period can be one or more fields, which are not limited here. For example, a flag can be set to indicate both whether such query messages are presented in SIBx and to indicate the period. In another example, the sensing device can receive first query information, a first matching score function, and a first threshold multicast by the central device in the form of a multicast message.

[0283] As mentioned earlier, the first query information can take the form of a query message, query semantics, or query terms. Besides a single task, the sensing device can also serve two or more tasks, meaning the first query information may include one or more query messages / semantics / terms. Below are some implementations where the first query information includes at least two query semantics or at least two query terms. It should be noted that other implementations are possible in other cases, such as the first query information including at least two query messages, or a single query message / semantics / term; this is not limited to these.

[0284] In one possible implementation, the first query information includes at least two query semantics. The sensing device can receive, from the at least two query semantics broadcast or multicast by the central device, the first query semantic, a second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, and the format of the first query semantic; and receive, from the at least two query semantics broadcast or multicast by the central device, the second query semantic, a third matching score function associated with the second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic. Alternatively, the sensing device can receive, from the at least two query semantics broadcast or multicast by the central device in a multiplexed manner, the first query semantic, the second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, the format of the first query semantic, the second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic. When the first query information includes multiple query semantics, the multiple query semantics and the information related to the multiple query semantics (i.e., the matching score function, threshold, length of the query semantics, and format of the query semantics corresponding to each query semantic) can be received sequentially or simultaneously using a multiplexing method. This provides greater flexibility and can meet different requirements.

[0285] There are multiple ways to determine the matching score in the above scenario. The following implementations are illustrative and not restrictive. In one implementation, the sensing device can acquire sensing data and convert it into general sensing semantics. Then, based on a second matching score function, it determines whether the second matching score between the general sensing semantics and the first query semantics is greater than or equal to a second threshold. Based on a third matching score function, it determines whether the third matching score between the general sensing semantics and the second query semantics is greater than or equal to a third threshold. When the sensing data is in natural language form and the first query information is in semantic form, the sensing data can be converted into general sensing semantics. For example, general semanticization configuration can be used to generate general sensing semantics, which simplifies the generation of sensing semantics. Then, a comparison is made between the general sensing semantics and each query semantic. That is, both the sensing data and the query information are in the general semantic domain, on which the sensing data and query information can be easily compared and fused. Query semantics can retain all the key semantic objectives conveyed by the query message, allowing the query semantics to be well converted (de-semantized) back to the query message. Since the semantic form can provide a more accurate true intent, the accuracy of the comparison results is improved.

[0286] In another implementation, the sensing device can acquire sensing data, convert the sensing data into first sensing semantics according to a first semanticization configuration, the length of a first query semantic, and the format of the first query semantic, and convert the sensing data into second sensing semantics according to a second semanticization configuration, the length of a second query semantic, and the format of a second query semantic. Then, based on a second matching score function, it determines whether the second matching score between the first sensing semantic and the first query semantic is greater than or equal to a second threshold, and based on a third matching score function, it determines whether the third matching score between the second sensing semantic and the second query semantic is greater than or equal to a third threshold. When the sensing data is in natural language form and the first query information is in semantic form, the sensing data can be converted into first and second sensing semantics respectively, and the corresponding sensing semantics can be generated using the corresponding semanticization configuration, thus ensuring the accuracy of the generated sensing semantics. Then, comparisons are made between the first sensing semantic and the first query semantic, as well as between the second sensing semantic and the second query semantic. That is, both the sensing data and the query information are in a common semantic domain, on which the sensing data and query information can be easily compared and fused. The query semantic can retain all the key semantic objectives conveyed by the query message, allowing the query semantic to be well converted (de-semantized) back to the query message. Because semantic form can provide a more accurate representation of the true intent, it improves the accuracy of the comparison results.

[0287] In another implementation, the sensing device can acquire sensing data, convert the sensing data into general sensing semantics according to a general lexicalization configuration, lexicalize the general sensing semantics into first sensing lexicals, lexicalize the first query semantics into first query lexicals, lexicalize the general sensing semantics into second sensing lexicals, and lexicalize the second query semantics into second query lexicals; or, according to the first lexicalization configuration, lexicalize the general sensing semantics into first sensing lexicals, lexicalize the first query semantics into first query lexicals, and according to the second lexicalization configuration, lexicalize the general sensing semantics into second sensing lexicals, and lexicalize the second query semantics into second query lexicals; then, according to a second matching score function, determine whether the second matching score between the first sensing lexical and the first query lexical is greater than or equal to a second threshold, and according to a third matching score function, determine whether the third matching score between the second sensing lexical and the second query lexical is greater than or equal to a third threshold. When the perceived data is in natural language form, while the first query information is in semantic form, the perceived data can be processed into first and second perceived tokens using general perceived semantics. Similarly, the first and second query semantics can be processed into first and second query tokens, respectively. Then, a comparison is made between the perceived tokens and their corresponding query tokens. Since tokenization can provide a more accurate representation of the true intent and saves signaling overhead, it improves the accuracy of the comparison results and reduces signaling costs. Furthermore, tokenization can prevent the sensing device from reconstructing the complete query message from the query tokens. Tokenization can also provide a degree of privacy protection for the query message.

[0288] In another implementation, the sensing device can acquire sensing data, convert the sensing data into first sensing semantics according to a first semanticization configuration, the length of a first query semantic, and the format of the first query semantic, and convert the sensing data into second sensing semantics according to a second semanticization configuration, the length of a second query semantic, and the format of the second query semantic; then, according to a general lexicalization configuration, the first sensing semantic is lexicalized into first sensing lexical, the first query semantic is lexicalized into first query lexical, the second sensing semantic is lexicalized into second sensing lexical, and the second query semantic is lexicalized into second query lexical. Lexical units; or, according to the first lexicalization configuration, the first perceptual semantic lexical unit is converted into a first perceptual lexical unit, and the first query semantic lexical unit is converted into a first query lexical unit; according to the second lexicalization configuration, the second perceptual semantic lexical unit is converted into a second perceptual lexical unit, and the second query semantic lexical unit is converted into a second query lexical unit; then, according to the second matching score function, it is determined whether the second matching score between the first perceptual lexical unit and the first query lexical unit is greater than or equal to the second threshold; according to the third matching score function, it is determined whether the third matching score between the second perceptual lexical unit and the second query lexical unit is greater than or equal to the third threshold. When the perceptual data is in natural language form, and the first query information is in semantic form, the perceptual data can be processed into first perceptual lexical units and second perceptual lexical units respectively through the first perceptual semantics and the second perceptual semantics, and the first query semantics and the second query semantics can be processed into first query lexical units and second query lexical units respectively. Then, a comparison is made between the perceptual lexical units and the corresponding query lexical units. Since the lexical unit form can provide a more accurate true intent and saves signaling overhead, the accuracy of the comparison results is improved, and signaling overhead is saved. Furthermore, lexicalization can be used to prevent sensing devices from reconstructing the complete query message from query terms. Lexicalization can provide a certain degree of privacy protection for query messages.

[0289] In one possible implementation, the first query information includes at least two query terms. The sensing device can receive the first query term, a second matching score function associated with the first query term, a second threshold associated with the second matching score function, and the length of the first query term from at least two query terms broadcast or multicast by the central device; and receive the second query term, a third matching score function associated with the second query term, a third threshold associated with the third matching score function, and the length of the second query term from at least two query terms broadcast or multicast by the central device. Alternatively, the sensing device can receive the first query term, a second matching score function associated with the first query term, a second threshold associated with the second matching score function, the length of the first query term, the second query term, a third matching score function associated with the second query term, a third threshold associated with the third matching score function, and the length of the second query term from at least two query terms broadcast or multicast by the central device in a multiplexed manner. When the first query information includes multiple query terms, the multiple query terms and the information related to the multiple query terms (i.e., the matching score function, threshold and length of the query term corresponding to each query semantic) can be received sequentially or simultaneously in a reused manner, which provides greater flexibility and can meet different requirements.

[0290] Regarding the determination of the matching score in the above situation, multiple methods can be provided. The following implementations are illustrative only and not restrictive. In one implementation, the sensing device can acquire sensing data, convert the sensing data into general sensing semantics, and then, according to the general lexicalization configuration, lexicalize the general sensing semantics into first sensing lexicals based on the length of the first query lexical, and lexicalize the general sensing semantics into second sensing lexicals based on the length of the second query lexical; or, according to the first lexicalization configuration, lexicalize the general sensing semantics into first sensing lexicals based on the length of the first query lexical, and lexicalize the general sensing semantics into second sensing lexicals based on the length of the second query lexical according to the second lexicalization configuration; then, according to the second matching score function, determine whether the second matching score between the first sensing lexical and the first query lexical is greater than or equal to a second threshold, and according to the third matching score function, determine whether the third matching score between the second sensing lexical and the second query lexical is greater than or equal to a third threshold. When the perceived data is in natural language form, and the first query information is in lexical form, the perceived data can be processed into first and second perceived lexical units using general perceptual semantics. Then, a comparison is made between the perceived lexical units and their corresponding query lexical units. Since lexical units can provide a more accurate representation of the true intent and save signaling overhead, the accuracy of the comparison results is improved, further reducing signaling costs. Furthermore, lexicalization can prevent the sensing device from reconstructing the complete query message from the query lexical units. Lexicalization can also provide a degree of privacy protection for the query message.

[0291] In another implementation, the sensing device can acquire sensing data, convert the sensing data into first sensing semantics according to a first semanticization configuration, and convert the sensing data into second sensing semantics according to a second semanticization configuration; then, according to a general lexicalization configuration, the first sensing semantic lexicalization is converted into a first sensing lexical based on the length of a first query lexical, and the second sensing semantic lexicalization is converted into a second sensing lexical based on the length of a second query lexical; or, according to the first lexicalization configuration, the first sensing semantic lexicalization is converted into a first sensing lexical based on the length of a first query lexical, and the second sensing semantic lexicalization is converted into a second sensing lexical based on the length of a second query lexical based on the second lexicalization configuration; next, according to a second matching score function, it is determined whether the second matching score between the first sensing lexical and the first query lexical is greater than or equal to a second threshold, and according to a third matching score function, it is determined whether the third matching score between the second sensing lexical and the second query lexical is greater than or equal to a third threshold. When the perceived data is in natural language form, and the first query information is in lexical form, the perceived data can be processed into first-perceived lexical units and second-perceived lexical units respectively through first-perceived semantics and second-perceived semantics. Then, a comparison is made between the perceived lexical units and their corresponding query lexical units. Since lexical units can provide a more accurate representation of the true intent and save signaling overhead, the accuracy of the comparison results is improved, and signaling overhead is reduced. Furthermore, lexicalization can prevent the sensing device from reconstructing the complete query message from the query lexical units. Lexicalization can provide a certain degree of privacy protection for the query message.

[0292] In the above implementation, the first matching score function, the second matching score function, or the third matching score function can be implemented using inner product or Euclidean distance. Other methods can also be used to implement the scoring function (i.e., the first / second / third matching score function), as long as they can obtain the relevance score between the perceived data and the first query information, or the similarity between the perceived data and the first query information.

[0293] It should be noted that the transformation operation in this disclosure refers to semantic processing, and the transformation operation can be replaced by embedding operation, conversion operation, transformation operation, etc. For example, converting perceptual data into perceptual semantics can be replaced by embedding perceptual data into perceptual semantics, converting perceptual data into perceptual semantics, transforming perceptual data into perceptual semantics, etc. The specific means of transformation operation, embedding operation, conversion operation, and transformation operation are not limited here. For example, the transformation operation can be implemented in existing ways.

[0294] Semanticization of perceived data can be achieved through the same or different semanticization configurations. Similarly, lexicalization of different query semantics or different perceived semantics can be achieved through the same or different lexicalization configurations, depending on the specific requirements. Lexicalization configurations include one of the following: lexicalization model, lexicalization function, projection matrix, graph-based or topology-based pruning algorithms, or compression methods. The availability of multiple lexicalization configurations provides greater flexibility to meet diverse requirements. The above lists some implementation methods for lexicalization; the specific means of implementation are not limited here. For example, lexicalization models can be implemented using existing methods.

[0295] In one possible implementation, the perception result includes one of the following: raw perception data; perception semantics obtained from the raw perception data; semi-raw perception data and perception semantics obtained from the raw perception data; raw perception data and a first matching score; perception semantics obtained from the raw perception data and a first matching score; semi-raw perception data, perception semantics obtained from the raw perception data, and a first matching score. The perception result can take various forms related to the perception data, providing greater flexibility to meet different requirements.

[0296] In one possible implementation, the sensing device can send a first matching sensing semantic and a fourth matching score associated with the first matching sensing semantic to the central device; and a second matching sensing semantic and a fifth matching score associated with the second matching sensing semantic to the central device. As mentioned earlier, a single sensing device can process two or more queries simultaneously; therefore, a responding sensing device can send more than one sensing result to the central device, for example, it can send two sensing semantics to the central device. The sent sensing semantic can be called a matching sensing semantic because the sensing device has determined that the matching score between the sensing data and the corresponding query is greater than or equal to a threshold. For example, the sensing device receives two queries: Q1 and Q2. The sensing device collects and measures its sensing data, and calculates the matching score between the sensing data and Q1, and the matching score between the sensing data and Q2. If the calculated matching score is greater than or equal to the threshold, the sensing device can send Q1 (i.e., the aforementioned first matching sensing semantic), the matching score between the sensing data and Q1 (i.e., the aforementioned fourth matching score), Q2 (i.e., the aforementioned second matching sensing semantic), and the matching score between the sensing data and Q2 (i.e., the aforementioned fifth matching score) to the central device. In response to more than one query, a sensing device can send more than one sensing semantic and related matching score to the central device. When multiple sensing devices send such data to the central device, the central device can perform a fusion operation on the data received from the multiple sensing devices to improve the accuracy of the fused data. A fusion operation will be introduced on the central device side.

[0297] The above text combined Figure 5 The communication method of this disclosure is described from the perspective of the sensing device. The following description will describe the communication method of this disclosure from the perspective of the central device, such as... Figure 6 As shown, the method may include the following steps:

[0298] Step 602: The central device broadcasts or multicasts the first query information, the first matching score function, and the first threshold to multiple sensing devices;

[0299] Step 604: The central device receives a sensing result from at least one of the multiple sensing devices, wherein the first matching score between the sensing data and the first query information is greater than or equal to a first threshold, wherein the first matching score is based on a first matching score function, and the sensing result indicates the sensing data.

[0300] For a description of steps 602 and 606, please refer to the description of steps 502 and 506, which will not be repeated here. The central device broadcasts or multicasts the first query information, the first matching score function, and the first threshold, thereby improving the transmission efficiency of the central device. When the first matching score between the sensing data of the sensing device and the first query information is greater than or equal to the first threshold, the central device receives the sensing result. The received sensing result meets the requirements of the first query information, that is, irrelevant information is filtered out, and the received data is what the central device needs, thereby ensuring the accuracy of data transmission.

[0301] Next, we will briefly describe the task- and / or modality-related information that may be included in the first query information, as well as the transmission method of the first query information. For a detailed description, please refer to the relevant description on the sensing device side, which will not be repeated here.

[0302] In one possible implementation, the first query information includes query information for a task or query information for a modality.

[0303] In one possible implementation, the central device may send an indication of a task identifier or an indication of a modality identifier.

[0304] In one possible implementation, the first query information includes multiple query information for multiple tasks and multiple task identifiers associated with the multiple query information respectively.

[0305] In one possible implementation, the first query information includes multiple query information for multiple modalities and multiple modality identifiers associated with the multiple query information respectively.

[0306] In one possible implementation, the first query information includes a first plurality of query information for multiple tasks, a second plurality of query information for multiple modalities, a plurality of task identifiers associated with the first plurality of query information, and a plurality of modal identifiers associated with the second plurality of query information.

[0307] In one possible implementation, the central device may broadcast a first query information, a first matching score function, and a first threshold to multiple sensing devices in a synchronization signal block / physical broadcast channel block.

[0308] In one possible implementation, the first query information is carried in a master information block (MIB) message or a system information block (SIB) message. The MIB message or SIB message includes an indication indicating whether the first query information is presented in the MIB message or SIB message. By indicating whether the first query information is presented, signaling overhead can be saved because if the indication does not show the first query information, no further processing is required.

[0309] In one possible implementation, the MIB message or SIB message also includes an indication of the period for the first query information. The first query information is valid within the period and may be invalid outside the period, thereby avoiding resource consumption.

[0310] In one possible implementation, the central device can multicast the first query information, the first matching score function, and the first threshold to multiple sensing devices in the form of multicast messages.

[0311] In one possible implementation, such as Figure 7 As shown, the communication method includes:

[0312] Step 702: The central device receives the second query information from the generative pre-trained transformer (GPT) device;

[0313] Step 704: The central device broadcasts or multicasts the first query information, the first matching score function, and the first threshold to multiple sensing devices;

[0314] Step 706: The central device receives a sensing result from at least one of the multiple sensing devices, wherein the first matching score between the sensing data and the first query information is greater than or equal to a first threshold, the first matching score is based on a first matching score function, and the sensing result indicates the sensing data;

[0315] Step 708: The central device outputs the sensing results to the GPT device.

[0316] For a description of steps 704 and 706, please refer to the descriptions of steps 502 and 506, which will not be repeated here. The second query information is used by the central device to generate the first query information. The central device can directly forward the query information received from the GPT device to the sensing device, or it can process the query information received from the GPT device (e.g., semantic processing, lexical processing, etc.) and then send the processed query information to the sensing device. The central device receives the second query information from the GPT device, provides the first query information, the first matching score function, and the first threshold for the sensing device to make a decision, and then receives the sensing result from the sensing device and outputs the sensing result to the GPT device. The central device acts as a bridge between the sensing device and the GPT device, thereby facilitating smooth communication between the sensing device and the GPT device.

[0317] The second query information can take the form of query messages, query semantics, or query terms. Besides a single task, the GPT device can also send two or more tasks to the central device; that is, the second query information may include one or more query messages / semantics / terms. Below are some implementations where the first query information includes at least two query semantics or at least two query terms. It should be noted that other implementations are possible in other cases, such as the first query information including at least two query messages, or a single query message / semantics / term; this is not limited to these.

[0318] In one possible implementation, the central device can receive at least two query semantics from at least two GPT devices; broadcast or multicast a first query semantic, a second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, and the format of the first query semantic to multiple sensing devices; broadcast or multicast a second query semantic, a third matching score function associated with the second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic to multiple sensing devices. Alternatively, the central device can use a multiplexing method to broadcast or multicast the first query semantic, the second matching score function associated with the first query semantic, the second threshold associated with the second matching score function, the length of the first query semantic, the format of the first query semantic, the second query semantic, a third matching score function associated with the second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic to multiple sensing devices. When the second query information includes multiple query semantics, the central device can broadcast or multicast multiple query semantics and information related to multiple query semantics (i.e., the matching score function, threshold, length of query semantics and format of query semantics corresponding to each query semantic) in sequence or in a multiplexed manner. This provides greater flexibility and can meet different requirements.

[0319] In one possible implementation, the central device can receive at least two query semantics from at least two GPT devices, and tokenize the first and second query semantics into first query tokens and second query tokens. Then, the central device can broadcast or multicast the first query token, the second matching score function associated with the first query token, the second threshold associated with the second matching score function, and the length of the first query token to multiple sensing devices; and broadcast or multicast the second query token, the third matching score function associated with the second query token, the third threshold associated with the third matching score function, and the length of the second query token to multiple sensing devices. Alternatively, the central device can use a multiplexing method to broadcast or multicast the first query token, the second matching score function associated with the first query token, the second threshold associated with the second matching score function, the length of the first query token, the second query token, the third matching score function associated with the second query token, the third threshold associated with the third matching score function, and the length of the second query token to multiple sensing devices. When the second query information includes multiple query semantics, the central device can convert the query semantic lexicals into corresponding query lexicals, and then broadcast or multicast multiple query lexicals and information related to multiple query lexicals (i.e., the matching score function, threshold, and length of each query lexical) in sequence or in a multiplexed manner. This provides greater flexibility and can meet different requirements.

[0320] In one possible implementation, the central device can convert the first query semantic token into a first query token based on a first tokenization configuration, and convert the second query semantic token into a second query token based on a second tokenization configuration. Alternatively, the central device can convert the first query semantic token into a first query token based on a general tokenization configuration, and convert the second query semantic token into a second query token based on a general tokenization configuration. Tokenization of different query semantics can be achieved using the same tokenization configuration or different tokenization configurations, depending on the specific requirements. Tokenization configurations include one of the following: tokenization model, tokenization function, projection matrix, graph-based or topology-based pruning algorithm, or compression method. The availability of multiple tokenization configurations provides greater flexibility to meet different requirements. The above lists some implementation methods for tokenization; the specific means of implementation are not limited here. For example, the tokenization model can be implemented using existing methods.

[0321] In one possible implementation, the first matching scoring function, the second matching scoring function, or the third matching scoring function can be implemented using an inner product or Euclidean distance. It should be noted that other methods can also be used to implement the scoring function, as long as they can obtain the relevance score between the perceived data and the first query information, or the similarity between the perceived data and the first query information.

[0322] In one possible implementation, the perception result includes one of the following: raw perception data; perception semantics obtained from the raw perception data; semi-raw perception data and perception semantics obtained from the raw perception data; raw perception data and a first matching score; perception semantics obtained from the raw perception data and a first matching score; semi-raw perception data, perception semantics obtained from the raw perception data, and a first matching score. The perception result can take various forms related to the perception data, providing greater flexibility to meet different requirements.

[0323] In one possible implementation, the central device can receive multiple first-matched perceptual semantics and multiple fourth-matching scores related to the multiple first-matched perceptual semantics from at least one of the multiple sensing devices; it can also receive multiple second-matched perceptual semantics and multiple fifth-matching scores related to the multiple second-matched perceptual semantics from at least one of the multiple sensing devices; then, it fuses some or all of the multiple first-matched perceptual semantics according to the multiple fourth-matching scores to obtain a first-fused perceptual semantics; and it fuses some or all of the multiple second-matched perceptual semantics according to the multiple fifth-matching scores to obtain a second-fused perceptual semantics. If the feedback data from the sensing devices includes multiple sensing results from multiple modalities, and multiple perceptual semantics exist in a single modality, the central device can perform fusion operations on the perceptual semantics of the same modality separately, making the fused perceptual semantics more comprehensive, thereby improving the accuracy of the fused perceptual semantics. For example, the matching score of each perceptual semantic can be regarded as the weight of each perceptual semantic when fusing multiple perceptual semantics. Another example is that the matching score of each perceptual semantic can be regarded as the criterion for selecting the perceptual semantics to be fused. It should be noted that there are other examples of the use of matching scores, which are not limited here. During the fusion operation of perceptual semantics of the same modality, multiple perceptual semantics of the same modality are fused according to their matching scores by considering the matching score of each perceptual semantic (which can indicate the relevance between the perceptual semantic and the first query information). For example, perceptual semantics with higher matching scores are given higher importance in the fusion process, thereby reducing the influence of some less reliable perceptual semantics, thereby improving the accuracy of the fused perceptual semantics and ensuring the reliability of the fused perceptual semantics.

[0324] In one possible implementation, the central device can receive multiple perceptual semantics, multiple fourth matching scores related to the multiple perceptual semantics, and multiple fifth matching scores related to the multiple perceptual semantics from at least one of multiple perceptual devices. Then, it fuses some or all of the multiple perceptual semantics based on the multiple fourth matching scores to obtain a first fused perceptual semantics; and fuses some or all of the multiple second perceptual semantics based on the multiple fifth matching scores to obtain a second fused perceptual semantics. If the feedback data from the perceptual devices includes multiple perceptual semantics for one modality but multiple tasks, the central device can perform fusion operations on the perceptual semantics of the same task separately, making the fused perceptual semantics more comprehensive and thus improving the accuracy of the fused perceptual semantics. During the fusion operation on the perceptual semantics of the same task, multiple perceptual semantics of the same task are fused based on their matching scores by considering the matching score of each perceptual semantic (which can indicate the relevance between the perceptual semantics and the first query information). For example, perceptual semantics with higher matching scores are given higher importance during the fusion process, thereby reducing the influence of some less reliable perceptual semantics, thus improving the accuracy of the fused perceptual semantics and ensuring the reliability of the fused perceptual semantics.

[0325] In one possible implementation, after acquiring the first and second fused perceptual semantics, the central device may send the first fused perceptual semantics to the first GPT device among at least two GPT devices, and send the second fused perceptual semantics to the second GPT device among at least two GPT devices. The first or second fused perceptual semantics may be processed by the first or second GPT device, respectively, to generate the next query based on the fused input. In other words, the GPT devices can generate a series of queries (e.g., query semantics, query terms, or other forms) by interacting with a series of fused perceptual semantics (or fused perceptual messages or other forms) obtained by fusing perceptual data with the central device.

[0326] In one possible implementation, the central device can determine a sixth matching score for the first fused perceptual semantics and a seventh matching score for the second fused perceptual semantics. The sixth matching score can indicate the relevance between the first fused perceptual semantics and the first query information, and the seventh matching score can indicate the relevance between the second fused perceptual semantics and the first query information. That is, the sixth and seventh matching scores can be used to evaluate the reliability of the corresponding fused perceptual semantics.

[0327] To illustrate the communication method of this disclosure more clearly and in detail, the following describes the method in more detail using a communication system including at least one central device, multiple distributed sensing devices, and at least one GPT device as an example, in conjunction with the following exemplary embodiments.

[0328] Example 1

[0329] In this disclosure, the wireless system is also referred to as a communication system or a wireless communication system. In this document, the wireless system includes multiple devices, for example, multiple devices including at least one central device, multiple distributed sensing devices, and at least one GPT device (in...). Figure 8 middle).

[0330] The GPT device is responsible for encoding or decoding query messages and sensed data. Specifically, the GPT device generates a query message for the central device containing one or more natural language targets. The central device semanticizes the query message into a semantic vector (i.e., the aforementioned query semantics), transforms the semantic vector into target semantic units (or vectors) (i.e., the aforementioned query units), and then broadcasts the target semantic units to the sensed device. The sensed device is triggered upon receiving the target semantic units, measures its sensed data, and converts the sensed data into sensed semantic units (i.e., the aforementioned sensed units). The sensed device compares the target semantic units with the sensed semantic units and scores the relevance between them. Only when the relevance score is higher than a threshold will the sensed data be sent in the form of a semantic vector. The central device fuses the sensed data in the form of semantic vectors and outputs the fused sensed data to the GPT device, which then generates the next query message based on the fused input.

[0331] The central device can be a BS, such as a gNB or eNB, or it can be an access point (AP).

[0332] Sensing devices are responsible for measuring and / or collecting local physical world data. These can be sensing UEs, sensing devices, IoT devices, mobile phones, handsets, or other devices. Sensing devices may be equipped with sensing gadgets or components to measure local physical world data in their vicinity as sensing data; the sensing device encodes this sensing data and sends it to a central device.

[0333] The GPT device can generate a series of query messages and receive fused perception messages from the central device. In this disclosure, the GPT device may also be referred to as an AI intelligent agent device, a robotic device, or an intelligent control device.

[0334] In detail, the multiple sensing devices in this paper can be grouped or classified according to the type of sensing data. The first group of sensing devices can measure first-type sensing data (e.g., red, green, blue, RGB images or videos), while the second group of sensing devices can measure second-type sensing data (e.g., RF point clouds or LiDAR point clouds), such as... Figure 9As shown. It should be noted that some sensing devices can be divided into more than one group, that is, some sensing devices can measure more than one type of sensing data.

[0335] The central device actively requests or triggers the sensing device to send its most recent sensing data (in Figure 10 (In the middle). Accordingly, the sensing devices will send their sensing data.

[0336] The central device can send one or more first query messages to one or more sensing devices in one or more DL broadcast, multicast or unicast channels (which can be physical broadcast channels, shared channels or dedicated channels).

[0337] After receiving the first query message, the sensing device decides whether to send its sensing data. Specifically, the sensing device decodes the first query message, measures its data, and decides whether to send its sensing data; this is called responding to the first query message. If the sensing device decides to respond to the first query message, it encodes / encapsulates the sensing data into a payload and then sends the payload to the central device in one or more UL channels (which can be physical UL shared channels or dedicated UL channels).

[0338] After the central device of the wireless system receives all payloads (i.e., the aforementioned sensing results) from the sensing devices responding to the first query message, the central device can fuse all or part of the payloads into a fused payload. Optionally, the central device can input the fused payload into a GPT device, which can process these fused payloads and then generate a second query message.

[0339] The central device can send one or more second query messages to one or more sensing devices in one or more DL broadcast, multicast or unicast channels.

[0340] The GPT device sends query messages to the central device to notify and configure the central device, thereby scheduling when, how, what, and which sensing devices should perform sensing and sending their sensing data to the central device. The GPT device can be implemented / located with the central device to reduce latency, or the GPT device can be implemented in a remote data center (the central device can access the remote data center via the core network), or the GPT device can be located on other connected devices in the same radio system as the central device. Note that in this disclosure, query messages (downlink messages) from the central device to the sensing devices can be carried in higher-layer signaling (e.g., radio resource control (RRC) signaling) or medium access control (MAC) layer signaling. Alternatively, the query message can be carried in physical layer signaling (e.g., downlink control information (DCI)). Alternatively, the query message can be carried in a combination of higher-layer and physical signaling. Other downlink messages / data transmitted from the central device to the sensing devices are similarly carried. Similarly, in this disclosure, uplink messages / data can be carried in higher-layer signaling (e.g., RRC signaling or MAC layer signaling). Alternatively, these messages / data can be carried in physical layer signaling (e.g., uplink control information (UCI)). Alternatively, these messages / data can be carried in a combination of higher-layer signaling and physical signaling. It should be noted that messages in this disclosure can be replaced by information, which can be carried in a single message or in more than one separate message.

[0341] In the preceding text, the GPT device can generate a query message and send it to the central device. In response to the query message from the GPT device, the central device can send a query message to the sensing device. In response to the query message from the central device, the sensing device can collect sensing data and, upon determining that the sensing data matches the query message, send the sensing data to the central device. It should be noted that the query message sent from the GPT device to the central device is a specific form of the aforementioned second query information, while the query message sent from the central device to the sensing device is a specific form of the aforementioned first query information. Furthermore, the first and second query information may include a single query message or more than one query message. Similarly, the fused sensing message is a specific form of the aforementioned fused sensing result, and the fused sensing result may include a single fused sensing message or more than one fused sensing message.

[0342] A wireless system comprising a central device, sensing devices, and GPT devices can form a series of interactions. The GPT devices generate a series of query messages for the sensing devices, the sensing devices collect and return sensing data, and the central device fuses this sensing data and inputs it into the GPT devices, such as... Figure 11 As shown.

[0343] In some cases, sensing devices may proactively send their sensing data without receiving any query messages from the central device. These devices might be responding to urgent queries, such as fire alarms or vehicle accidents. In a sense, some query messages are predefined and configured into the system by default.

[0344] Example 2

[0345] The GPT device in Example 1 can generate a series of query messages based on previously sensed messages, wherein the previously sensed messages are received and / or fused by a central device. The GPT device can infer one or more generative AI models. The one or more generative AI models infer one or more deep neural networks to output one or more query messages. The GPT device generates a series of query messages, referred to as a "thought chain," by interacting with a series of fused sensed messages; these query messages are obtained by the central device fusing the sensed data sent by the sensed device in response. Figure 12 As shown.

[0346] Query messages generated by GPT devices can convey semantic goals, tasks, or purposes. For example, a query message "Locate an approaching pedestrian" explicitly establishes a semantic goal, allowing the sensing device to focus on pedestrians in its vicinity and preventing interference. Because a query message conveys one or more semantic goals, a query message sent by the central device to the sensing devices may trigger a goal-oriented sensing task at each responding sensing device that receives and responds to the query message. Note that a message may convey multiple goals. For example, a message "Find a moving pedestrian wearing a white coat" conveys two semantic goals or tasks: a moving pedestrian and a pedestrian wearing a white coat.

[0347] In one implementation, the central device can broadcast a series of query messages because, in a wireless system with such a high density of sensing devices, individually scheduling each sensing device might be too costly or even impossible. Therefore, once a sensing device receives a query message, it may be woken up, but at this point, it doesn't know whether its sensing data is sufficiently relevant to the target conveyed by the query message. Therefore, the sensing device can activate its sensing tools to perceive the environment near the device as sensing data and compare this data with the query messages. If the sensing device determines that the sensing data is sufficiently relevant to the query message, it encodes the sensing data and sends it to the central device. Figure 13 (Sensing device #1 in the system). Otherwise, the sensor may not respond to the query message at all. Figure 13 (Sensing device #2 in the context). In this sense, the wireless system does not schedule individual sensing devices, but rather schedules common tasks across the entire set of sensing devices.

[0348] The central device can receive multiple sensing data from some or all sensing devices at the end of a predefined response time interval to respond to a query message. The central device can then fuse all the sensing data into a single sensing message and input this message into the GPT device. The GPT device will then generate the next query message based on this sensing message, such as... Figure 14 As shown.

[0349] Since only sensing devices that respond to query messages send sensing data, a significant amount of wireless resources can be saved compared to a one-to-one scheduling algorithm.

[0350] Example 3

[0351] In Example 2, the series of query messages generated by the GPT device and broadcast by the central device are in natural language, i.e., human-readable. The GPT device can use a large-language-model (LLM) to reason about the fused perceptual messages (also in natural language) to generate new query messages. The LLM model can be a "standard" base model like the Transformer model, or a "custom" model built for a smaller vocabulary and specific scenarios. For example, a customized LLM for processing Industry 4.0 or a customized LLM for processing wireless communication signaling and protocols. The GPT device can change, update, reduce, expand, or replace one or more of its LLMs at any time according to its own needs. Note that broadcast, multicast, or unicast is supported.

[0352] In Example 2, the query messages generated by the GPT device use natural language. Due to the randomness of generation, two different query messages may convey very similar semantic goals. For example, "find pedestrians" and "locate walking people" may have the same semantic goal. Therefore, the GPT device can semanticize the query message into query semantics, a process referred to as "embedding," "semanticization," "encoding," "natural language to machine translation," etc. The GPT device can convert the query message into query semantics, which may include vectors, matrices, or tensors composed of scalars. The conversion can be achieved through deep neural networks or other classical functions. Query semantics can preserve all the key semantic goals conveyed by the query message, allowing the query semantics to be well converted (de-semantized) back into the query message. Optionally, the GPT device can send query semantics instead of query messages to a central device, such as... Figure 15 As shown. Figure 16 As shown, query semantics are reversible, meaning that query messages can be recovered from query semantics. Note that if all LLMs output a common natural language (e.g., English), these LLMs are said to be natural language aligned; then, regardless of the LLM used, everyone can smoothly access GPT devices and function well in wireless systems.

[0353] In one implementation, the central device can also tokenize the query semantics into query tokens. Query tokens are fixed-length semantics, but include vectors composed of scalars, making them easier to transmit and compare. The wireless system can pre-specify multiple query token lengths. Therefore, when tokenizing the query semantics, the central device can select an appropriate token length based on the range of query semantic size. Tokenization can be a very strict function to prevent sensing devices from reconstructing the complete query message from the query tokens. Tokenization can provide a degree of privacy protection for the query message. Tokenization can be implemented using deep neural networks or other classic functions, such as... Figure 17 As shown.

[0354] Optionally, the central device receives query semantics from the GPT device, and then converts the query semantics into a fixed-length query term; the central device may broadcast the query term of this length to all sensing devices; the central device may store the query semantics in its own memory or storage to check the feedback sensing data.

[0355] Similar to Example 2, the sensing device can compare its sensing data with the query message; after receiving a query term (with its length or its length indicator), the sensing device is activated to enable its sensing gadgets to measure the physical world environment near the sensing device as sensing data; the sensing device can be equipped with an LLM or more LLMs as a semantic model, and the sensing data is input into the semantic model to output sensing semantics; optionally, the sensing device can select an appropriate sensing semantic length and format; the sensing device can continue to transform the sensing semantic terms into sensing terms of the same length as the query term received by the sensing device; based on the content received by the sensing device, the sensing device compares the query message with the sensing data or scores the relevance between the query message and the sensing data.

[0356] Alternative Solution #1 ( Figure 18 and Figure 19 The sensing device receives query terms and a scoring function; compares query terms with sensing terms and scores the relevance between them; if the relevance score is greater than or equal to a predefined threshold, the sensing device determines that the sensing data is sufficiently relevant to the query message from the central device.

[0357] Alternative Solution #2 Figure 20 and Figure 21 The sensing device receives the query semantics and the scoring function; if the query semantics and the sensing semantics are similar in size and format, the two semantics are compared and the relevance between them is scored; if the relevance score is greater than or equal to a predefined threshold, the sensing device determines that the sensing data is sufficiently relevant to the query message from the central device.

[0358] Alternative solution #3 Figure 22 and Figure 23 The sensing device receives the query semantics and scoring function; first, it converts the query semantics into query terms through a local lexicalization model; it compares the query terms with the sensing terms and scores the relevance between the query terms and the sensing terms; if the relevance score is greater than or equal to a predefined threshold, the sensing device determines that the sensing data is sufficiently relevant to the query message from the central device.

[0359] Therefore, the central device can send information about the scoring function and one or more thresholds related to the query semantics (or message or term) to the sensing device. In one implementation, the scoring function can be a "standard" function, such as a simple inner product, so that only the thresholds related to the query term need to be sent to the sensing device along with the query term.

[0360] Example 4

[0361] To achieve low latency, i.e., fast physical world search, the transmission of query terms (or semantics) in Example 3 can be integrated into the physical layer of the wireless communication system. For example, query terms (or semantics) can be carried in physical layer signaling.

[0362] As described in Example 3, the query term can be a fixed-length vector composed of scalars, while the length of the perceptual term generated by the perceptual device can be the same as that of the query term; the query term can be encoded in the following possible ways.

[0363] Alternative #1: Query terms can be encoded into a sequence of predefined length; in wireless systems, a sequence of predefined length can be applied (transmitted) in various forms, such as (pre-, post-, or in-) preambles, masks (acting on preambles, pilots, spreading codes), interleaved sequences, or other forms.

[0364] Alternative solution #2: Query terms can be encoded into special payloads using a fixed MCS scheme; for example, the length of a term can be a power of 2, so punctured polar codes can be used directly to encode the query terms.

[0365] Alternative #3: If the inner product is used as the scoring function, long query terms can be divided into several sub-blocks for incremental transmission; if the inner product of the first sub-block already shows strong relevance, the transmission of subsequent sub-blocks can be omitted.

[0366] Next, exemplary embodiments of products related to the communication method will be described.

[0367] Figure 24 A block diagram of the communication device 2400 is shown. (e.g.) Figure 24 As shown, the device 2400 includes:

[0368] The first receiving module 2402 is used to receive the first query information, the first matching score function and the first threshold broadcast or multicast by the central device;

[0369] The determining module 2404 is used to determine whether the first matching score between the perceived data and the first query information is greater than or equal to a first threshold based on the first matching score function.

[0370] The sending module 2406 is used to send the sensing result to the central device when the first matching score is greater than or equal to the first threshold, wherein the sensing result indicates the sensing data.

[0371] In one possible implementation, the first query information includes query information for a task or query information for a modality.

[0372] In one possible implementation, the method further includes: a second receiving module for receiving an indication of a task identifier or a modal identifier from a central device.

[0373] In one possible implementation, the first query information includes multiple query information for multiple tasks and multiple task identifiers associated with the multiple query information respectively.

[0374] In one possible implementation, the first query information includes multiple query information for multiple modalities and multiple modality identifiers associated with the multiple query information respectively.

[0375] In one possible implementation, the first query information includes a first plurality of query information for multiple tasks, a second plurality of query information for multiple modalities, a plurality of task identifiers associated with the first plurality of query information, and a plurality of modal identifiers associated with the second plurality of query information.

[0376] In one possible implementation, the first receiving module is specifically used to: receive the first query information, the first matching score function, and the first threshold broadcast by the central device in the synchronization signal block / physical broadcast channel block.

[0377] In one possible implementation, the first query information is carried in a master information block (MIB) message or a system information block (SIB) message, and the MIB message or SIB message includes an indication of whether the first query information is presented in the MIB message or SIB message.

[0378] In one possible implementation, the MIB message or SIB message may also include an indication of the period for the first query information.

[0379] In one possible implementation, the first receiving module is specifically used to: receive the first query information, the first matching score function, and the first threshold multicast by the central device in the form of multicast messages.

[0380] In one possible implementation, the first query information includes at least two query semantics.

[0381] The first receiving module is specifically used for:

[0382] The system receives at least two query semantics broadcast or multicast from a central device, including a first query semantic, a second matching score function related to the first query semantic, a second threshold related to the second matching score function, the length of the first query semantic, and the format of the first query semantic; it also receives at least two query semantics broadcast or multicast from a central device, including a second query semantic, a third matching score function related to the second query semantic, a third threshold related to the third matching score function, the length of the second query semantic, and the format of the second query semantic; or

[0383] The receiving center device broadcasts or multicasts at least two query semantics in a multiplexed manner: a first query semantic, a second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, the format of the first query semantic, a second query semantic in at least two query semantics, a third threshold associated with a third matching score function, the length of the second query semantic, and the format of the second query semantic.

[0384] In one possible implementation, the device further includes:

[0385] The acquisition module is used to acquire sensor data;

[0386] The conversion module is used to convert perceptual data into general perceptual semantics;

[0387] The module is specifically used for:

[0388] Based on the second matching score function, determine whether the second matching score between the general perceptual semantics and the first query semantics is greater than or equal to the second threshold;

[0389] Based on the third matching score function, determine whether the third matching score between the general perceptual semantics and the second query semantics is greater than or equal to the third threshold.

[0390] In one possible implementation, the device further includes:

[0391] The acquisition module is used to acquire sensor data;

[0392] The first conversion module is used to convert the perceived data into the first perceived semantics according to the first semantic configuration, the length of the first query semantics, and the format of the first query semantics.

[0393] The second conversion module is used to convert the perceived data into second perceived semantics according to the second semantic configuration, the length of the second query semantics, and the format of the second query semantics.

[0394] The module is specifically used for:

[0395] Based on the second matching score function, determine whether the second matching score between the first perceptual semantics and the first query semantics is greater than or equal to the second threshold;

[0396] Based on the third matching score function, determine whether the third matching score between the second perceptual semantics and the second query semantics is greater than or equal to the third threshold.

[0397] In one possible implementation, the method further includes:

[0398] The acquisition module is used to acquire sensor data;

[0399] The conversion module is used to convert perceptual data into general perceptual semantics;

[0400] The lexicalization module is used to convert general-aware semantic lexical units into first-aware lexical units, first-query semantic lexical units into first-query lexical units, general-aware semantic lexical units into second-aware lexical units, and second-query semantic lexical units into second-query lexical units, according to the general lexicalization configuration; or to convert general-aware semantic lexical units into first-aware lexical units and first-query semantic lexical units into first-query lexical units, according to the first lexicalization configuration, and to convert general-aware semantic lexical units into second-aware lexical units and second-query semantic lexical units into second-query lexical units, according to the second lexicalization configuration.

[0401] The module is specifically used for:

[0402] Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold.

[0403] Based on the third matching score function, determine whether the third matching score between the second perceptual term and the second query term is greater than or equal to the third threshold.

[0404] In one possible implementation, the device further includes:

[0405] The acquisition module is used to acquire sensor data;

[0406] The first conversion module is used to convert the perceived data into the first perceived semantics according to the first semantic configuration, the length of the first query semantics, and the format of the first query semantics.

[0407] The second conversion module is used to convert the perceived data into second perceived semantics according to the second semantic configuration, the length of the second query semantics, and the format of the second query semantics.

[0408] The lexicalization module is used to convert first-perception semantic lexical units into first-perception lexical units, first-query semantic lexical units into first-query lexical units, second-perception semantic lexical units into second-perception lexical units, and second-query semantic lexical units into second-query lexical units, according to a general lexicalization configuration; or to convert first-perception semantic lexical units into first-perception lexical units and first-query semantic lexical units into first-query lexical units, according to a first-perception lexicalization configuration, and to convert second-perception semantic lexical units into second-perception lexical units and second-query semantic lexical units into second-query lexical units, according to a second-perception lexicalization configuration.

[0409] The module is specifically used for:

[0410] Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold.

[0411] Based on the third matching score function, determine whether the third matching score between the second perceptual term and the second query term is greater than or equal to the third threshold.

[0412] In one possible implementation, the first query information includes at least two query terms.

[0413] The first receiving module is specifically used for:

[0414] The receiving center device broadcasts or multicasts at least two query terms, including a first query term, a second matching score function associated with the first query term, a second threshold associated with the second matching score function, and the length of the first query term; and receives, from the center device broadcasts or multicasts at least two query terms, a second query term, a third matching score function associated with the second query term, a third threshold associated with the third matching score function, and the length of the second query term; or

[0415] The receiving center equipment broadcasts or multicasts at least two query terms, including a first query term, a second matching score function associated with the first query term, a second threshold associated with the second matching score function, the length of the first query term, a second query term among at least two query terms, a third matching score function associated with the second query term, a third threshold associated with the third matching score function, and the length of the second query term.

[0416] In one possible implementation, the device further includes:

[0417] The acquisition module is used to acquire sensor data;

[0418] The conversion module is used to convert perceptual data into general perceptual semantics;

[0419] The lexicalization module is used to convert general-aware semantic lexical units into first-aware lexical units based on the length of the first query lexical unit, and to convert general-aware semantic lexical units into second-aware lexical units based on the length of the second query lexical unit, according to the general lexicalization configuration; or to convert general-aware semantic lexical units into first-aware lexical units based on the length of the first query lexical unit, according to the first lexicalization configuration, and to convert general-aware semantic lexical units into second-aware lexical units based on the length of the second query lexical unit, according to the second lexicalization configuration.

[0420] The module is specifically used for:

[0421] Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold.

[0422] Based on the third matching score function, determine whether the third matching score between the second perceptual term and the second query term is greater than or equal to the third threshold.

[0423] In one possible implementation, the device further includes:

[0424] The acquisition module is used to acquire sensor data;

[0425] The first conversion module is used to convert the perceived data into the first perceived semantics according to the first semantic configuration;

[0426] The second conversion module is used to convert the perceptual data into second perceptual semantics according to the second semantic configuration;

[0427] The lexicalization module is used to convert a first perceptual semantic lexical into a first perceptual lexical based on the length of a first query lexical, and a second perceptual semantic lexical into a second perceptual lexical based on the length of a second query lexical, according to a general lexicalization configuration; or to convert a first perceptual semantic lexical into a first perceptual lexical based on the length of a first query lexical, according to a first lexicalization configuration, and a second perceptual semantic lexical into a second perceptual lexical based on the length of a second query lexical, according to a second lexicalization configuration.

[0428] The module is specifically used for:

[0429] Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold.

[0430] Based on the third matching score function, determine whether the third matching score between the second perceptual term and the second query term is greater than or equal to the third threshold.

[0431] In one possible implementation, the first matching score function, the second matching score function, or the third matching score function includes an inner product or Euclidean distance.

[0432] In one possible implementation, the perception result includes one of the following: raw perception data; perception semantics obtained from the raw perception data; semi-raw perception data and perception semantics obtained from the raw perception data; raw perception data and a first matching score; perception semantics obtained from the raw perception data and a first matching score; semi-raw perception data, perception semantics obtained from the raw perception data, and a first matching score.

[0433] In one possible implementation, the sending module is specifically configured to: send a first matching perceptual semantic and a fourth matching score related to the first matching perceptual semantic to the central device; and send a second matching perceptual semantic and a fifth matching score related to the second matching perceptual semantic to the central device.

[0434] The communication device can be applied to the sensing device described in the above method examples, or it can be a sensing device described in the above method examples. Those skilled in the art should understand that the relevant descriptions of the modules in this disclosure examples can be understood by referring to the relevant descriptions of the communication methods in the examples of this disclosure.

[0435] like Figure 25 As shown, this disclosure provides a communication device 2500, comprising:

[0436] The first sending module 2502 is used to broadcast or multicast the first query information, the first matching score function and the first threshold to multiple sensing devices;

[0437] The first receiving module 2504 is configured to receive a sensing result from at least one of a plurality of sensing devices, wherein a first matching score between the sensing data and the first query information is greater than or equal to a first threshold, the first matching score is based on a first matching score function, and the sensing result indicates the sensing data.

[0438] In one possible implementation, the first query information includes query information for a task or query information for a modality.

[0439] In one possible implementation, the apparatus further includes a second sending module for sending an indication of a task identifier or an indication of a modality identifier.

[0440] In one possible implementation, the first query information includes multiple query information for multiple tasks and multiple task identifiers associated with the multiple query information respectively.

[0441] In one possible implementation, the first query information includes multiple query information for multiple modalities and multiple modality identifiers associated with the multiple query information respectively.

[0442] In one possible implementation, the first query information includes a first plurality of query information for multiple tasks, a second plurality of query information for multiple modalities, a plurality of task identifiers associated with the first plurality of query information, and a plurality of modal identifiers associated with the second plurality of query information.

[0443] In one possible implementation, the first transmitting module is specifically used to: broadcast a first query information, a first matching score function, and a first threshold to multiple sensing devices in a synchronization signal block / physical broadcast channel block.

[0444] In one possible implementation, the first query information is carried in a master information block (MIB) message or a system information block (SIB) message, and the MIB message or SIB message includes an indication of whether the first query information is presented in the MIB message or SIB message.

[0445] In one possible implementation, the MIB message or SIB message may also include an indication of the period for the first query information.

[0446] In one possible implementation, the first sending module is specifically used to: multicast the first query information, the first matching score function, and the first threshold to multiple sensing devices in the form of multicast messages.

[0447] In one possible implementation, the device further includes a second receiving module and an output module, wherein the second receiving module is used to receive second query information from a generative pre-trained transformer (GPT) device; and the output module is used to output the perception result to the GPT device.

[0448] In one possible implementation, the second receiving module is specifically used to: receive at least two query semantics from at least two GPT devices;

[0449] The first sending module is specifically used for:

[0450] Broadcast or multicast to multiple sensing devices a first query semantic, a second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, and the format of the first query semantic; broadcast or multicast to multiple sensing devices a second query semantic, a third matching score function associated with the second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic; or

[0451] Broadcast or multicast at least two query semantics to multiple sensing devices in a multiplexed manner: a first query semantic, a second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, the format of the first query semantic, a second query semantic in at least two query semantics, a third matching score function associated with the second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic.

[0452] In one possible implementation, the second receiving module is specifically used to: receive at least two query semantics from at least two GPT devices;

[0453] The device further includes a lexicalization module, used to lexicalize the first query semantic and the second query semantic in at least two query semantics into a first query lexical and a second query lexical;

[0454] The first sending module is specifically used for:

[0455] Broadcast or multicast a first query term, a second matching score function associated with the first query term, a second threshold associated with the second matching score function, and the length of the first query term to multiple sensing devices; broadcast or multicast a second query term, a third matching score function associated with the second query term, a third threshold associated with the third matching score function, and the length of the second query term to multiple sensing devices; or

[0456] The first query term, the second matching score function associated with the first query term, the second threshold associated with the second matching score function, the length of the first query term, the second query term, the third matching score function associated with the second query term, the third threshold associated with the third matching score function, and the length of the second query term are broadcast or multicast to multiple sensing devices in a multiplexed manner.

[0457] In one possible implementation, the lexicalization module is specifically used for:

[0458] Based on the first lexicalization configuration, the first query semantic lexicalization is converted into the first query lexicalization; based on the second lexicalization configuration, the second query semantic lexicalization is converted into the second query lexicalization; or

[0459] Based on the general lexicalization configuration, the first query semantic lexicalization is converted into the first query lexicalization; based on the general lexicalization configuration, the second query semantic lexicalization is converted into the second query lexicalization.

[0460] In one possible implementation, the first matching score function, the second matching score function, or the third matching score function includes an inner product or Euclidean distance.

[0461] In one possible implementation, the perception result includes one of the following: raw perception data; perception semantics obtained from the raw perception data; semi-raw perception data and perception semantics obtained from the raw perception data; raw perception data and a first matching score; perception semantics obtained from the raw perception data and a first matching score; semi-raw perception data, perception semantics obtained from the raw perception data, and a first matching score.

[0462] In one possible implementation, the first receiving module is specifically configured to: receive from at least one of the multiple sensing devices a plurality of first-matching sensing semantics and a plurality of fourth-matching scores associated with the plurality of first-matching sensing semantics; and receive from at least one of the multiple sensing devices a plurality of second-matching sensing semantics and a plurality of fifth-matching scores associated with the plurality of second-matching sensing semantics.

[0463] The device further includes an acquisition module, configured to acquire a first fused perceptual semantics by fusing some or all of the perceptual semantics of the first matches based on multiple fourth matching scores; and to acquire a second fused perceptual semantics by fusing some or all of the perceptual semantics of the second matches based on multiple fifth matching scores.

[0464] In one possible implementation, the first receiving module is specifically configured to: receive multiple perceptual semantics, multiple fourth matching scores associated with the multiple perceptual semantics, and multiple fifth matching scores associated with the multiple perceptual semantics from at least one of the multiple perceptual devices;

[0465] The device further includes an acquisition module, configured to fuse some or all of the multiple perceptual semantics based on multiple fourth matching scores to acquire a first fused perceptual semantic; and to fuse some or all of the multiple second perceptual semantics based on multiple fifth matching scores to acquire a second fused perceptual semantic.

[0466] In one possible implementation, the apparatus further includes a third transmitting module, configured to transmit a first fused perception semantic to a first GPT device among at least two GPT devices, and to transmit a second fused perception semantic to a second GPT device among at least two GPT devices.

[0467] In one possible implementation, the apparatus further includes: a determining module, configured to determine a sixth matching score of the first fused perceptual semantics; and to determine a seventh matching score of the second fused perceptual semantics.

[0468] The communication device can be applied to the central device described in the above method examples, or it can be the central device described in the above method examples. Those skilled in the art should understand that the descriptions of the modules in this disclosure examples can be understood by referring to the descriptions of the communication methods in the examples of this disclosure.

[0469] This disclosure provides a sensing device including processing circuitry for performing any of the communication methods described above. It should be understood that the sensing device is capable of performing the steps described in the method examples above, which will not be elaborated further here.

[0470] This disclosure provides a central device including processing circuitry for performing any of the communication methods described above. It should be understood that the central device is capable of performing the steps shown in the method examples, which will not be elaborated further here.

[0471] This disclosure provides a communication system comprising a central device and a sensing device. The sensing device is used to perform steps in any communication method performed by the sensing device, and the central device is used to perform steps in any communication method performed by the central device.

[0472] This disclosure provides a communication system comprising a sensing device and at least one of a central device and a GPT device. The sensing device is used to perform steps in any communication method performed by the sensing device, and the central device / GPT device is used to perform steps in any communication method performed by the central device.

[0473] This disclosure provides a chip including an input / output (I / O) interface and a processor, wherein the processor is used to call and execute computer execution instructions stored in memory to enable a device equipped with the chip to perform any of the above-described communication methods.

[0474] This disclosure provides a computer-readable medium that stores computer-executable instructions that, when executed by a processor, cause the processor to perform any of the above-described communication methods.

[0475] This disclosure provides a computer program product including computer-executable instructions that, when executed by a processor, cause the processor to perform any of the above-described communication methods.

[0476] This disclosure provides a computer program that includes computer-executable instructions that, when executed by a processor, cause the processor to perform any of the above-described communication methods.

[0477] Some aspects of this disclosure relate to a semantic-based communication scheme for managing and scheduling a large number of sensing devices, which may be of different types. The query semantics are target-oriented; only sensing devices whose sensing data is sufficiently relevant to the semantic message will respond and transmit their sensing data, which is also preferably in semantic form.

[0478] Some aspects of this disclosure relate to a collective semantic lexical-based scheduling scheme for a large number of sensing devices, rather than one-to-one individual scheduling.

[0479] Some aspects of this disclosure relate to a scheme for using a large-language-model (LLM) to bring query and sense data into a common semantic domain, on which such query and sense data can be easily compared and fused.

[0480] The above-described aspects of this disclosure may have at least one of the following benefits:

[0481] Scheduling can be task-oriented or goal-oriented; only sensing devices that contribute to the scheduled task or goal will respond and transmit their sensing data.

[0482] Privacy is protected: task, target, or query and perception data are well protected; no raw data is transmitted over the wireless transmission, or only a very small amount of raw data or messages are transmitted.

[0483] Forward compatibility: The semantic-based perception system in this disclosure is forward compatible, that is, it can support any new perception mechanism.

[0484] In some aspects of this disclosure, a computer program comprising instructions is provided. When executed by a processor, the instructions cause the processor to implement the methods of this disclosure.

[0485] In some aspects of this disclosure, a non-transitory computer-readable medium is provided that stores instructions which, when executed by a processor, cause the processor to implement the methods of this disclosure.

[0486] In some aspects of this disclosure, an apparatus / chipset system is provided, including means for implementing the sensing device implementation method of this disclosure.

[0487] In some aspects of this disclosure, an apparatus / chipset system is provided, including means for implementing a central device implementation method of this disclosure.

[0488] In some aspects of this disclosure, an apparatus / chipset system is provided, including means for implementing the GPT device implementation method of this disclosure.

[0489] In some aspects of this disclosure, a system is provided that includes at least two of the means in the sensing device of this disclosure, the means in the central device of this disclosure, and the means in the GPT device of this disclosure.

[0490] In some aspects of this disclosure, an apparatus / chipset system is provided, including at least one processor that executes instructions stored in a computer-readable medium to implement the sensing device implementation method of this disclosure.

[0491] In some aspects of this disclosure, an apparatus / chipset system is provided, including at least one processor that executes instructions stored in a computer-readable medium to implement the methods implemented by the central device of this disclosure.

[0492] In some aspects of this disclosure, an apparatus / chipset system is provided, including at least one processor that executes instructions stored in a computer-readable medium to implement the GPT device implementation method of this disclosure.

[0493] Please note that different embodiments can be implemented individually or in combination. Although combinations of features are shown in the illustrated embodiments, not all features need to be combined to achieve the benefits of the various examples of this disclosure. In other words, the system or method designed in this disclosure does not necessarily include all features shown in any of the figures, nor does it necessarily include all parts schematically shown in the figures. Furthermore, selected features of one exemplary embodiment may be combined with selected features of other exemplary embodiments.

[0494] Although this disclosure has been described with reference to illustrative embodiments, it is not intended to be interpreted in a limiting sense. Various modifications and combinations of the illustrative embodiments, as well as other examples of this disclosure, will be apparent to those skilled in the art upon reference to this specification. Therefore, the appended claims are intended to cover any such modifications or examples.

[0495] Although this disclosure describes methods and processes in a specific order, one or more steps in the methods and processes may be omitted or changed as needed. One or more steps may be performed in an order other than that described, as required.

[0496] Note that, as used herein, the expression "at least one of A or B" is interchangeable with the expression "A and / or B". This expression refers to a list in which you can choose either A or B, or A and B. Similarly, as used herein, "at least one of A, B, or C" is interchangeable with "A and / or B and / or C" or "A, B, and / or C". This expression refers to a list in which you can choose: A or B or C, or A and B, or A and C, or B and C, or all of A, B, and C. The same principle applies to longer lists with the same format.

[0497] Although this disclosure has been described at least in part as a method, those skilled in the art will understand that this disclosure also relates to various components for performing at least some aspects and features of the described methods, whether by hardware components, software, or any combination of both. Therefore, the technical solutions of this disclosure can be embodied in the form of a software product. Suitable software products may be stored in pre-recorded storage devices or other similar non-volatile or non-transitory computer-readable media, including, for example, DVDs, CD-ROMs, USB flash drives, removable hard drives, or other storage media. The software product includes instructions tangibly stored thereon that enable a processing device (e.g., a personal computer, server, or network device) to perform the method examples disclosed herein. Machine-executable instructions may be in the form of sequences of code, configuration information, or other data, which, when executed, cause a machine (e.g., a processor or other processing device) to perform the steps of the methods executable in this disclosure.

[0498] This disclosure may be embodied in other specific forms without departing from the subject matter of the claims. The examples described are to be regarded in all respects as illustrative rather than restrictive. Features selected from one or more of the foregoing examples may be combined to create alternative examples not explicitly described, and features suitable for such combinations are to be understood within the scope of this disclosure.

[0499] All values ​​and sub-ranges within the scope of the disclosure are also disclosed. Furthermore, although the systems, devices, and processes disclosed and illustrated herein may include a specific number of elements / components, these systems, devices, and components may be modified to include more or fewer such elements / components. For example, although any element / component disclosed may be referred to as singular, the examples disclosed herein may be modified to include multiple such elements / components. The subject matter described herein is intended to cover and include all suitable technical changes.

[0500] Although examples have been described above with reference to the accompanying drawings, those skilled in the art will understand that variations and modifications may be made without departing from the scope defined by the appended claims.

Claims

1. A communication method, characterized in that, include: Receive the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device; Based on the first matching score function, determine whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold; When the first matching score is greater than or equal to the first threshold, a sensing result is sent to the central device, wherein the sensing result indicates the sensing data.

2. The method according to claim 1, characterized in that, The first query information includes query information for a task or query information for a modality.

3. The method according to claim 2, characterized in that, The method further includes: Receive instructions from the central device regarding the task identifier or the modal identifier.

4. The method according to claim 1, characterized in that, The first query information includes multiple query information for multiple tasks and multiple task identifiers associated with the multiple query information respectively.

5. The method according to claim 1, characterized in that, The first query information includes multiple query information for multiple modalities and multiple modal identifiers associated with the multiple query information respectively.

6. The method according to claim 1, characterized in that, The first query information includes a first plurality of query information for multiple tasks, a second plurality of query information for multiple modalities, a plurality of task identifiers associated with the first plurality of query information, and a plurality of modal identifiers associated with the second plurality of query information.

7. The method according to any one of claims 1 to 6, characterized in that, The receipt of the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device includes: The central device receives the first query information, the first matching score function, and the first threshold broadcast in the Synchronization Signal Block (SSB) / Physical Broadcast Channel (PBCH) block.

8. The method according to claim 7, characterized in that, The first query information is carried in a main information block (MIB) message or a system information block (SIB) message, and the MIB message or the SIB message includes an indication for whether the first query information is presented in the MIB message or the SIB message.

9. The method according to claim 8, characterized in that, The MIB message or the SIB message also includes an indication of the period for the first query information.

10. The method according to any one of claims 1 to 6, characterized in that, The receipt of the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device includes: The central device receives the first query information, the first matching score function, and the first threshold in the form of a multicast message.

11. The method according to any one of claims 1 to 10, characterized in that, The first query information includes at least two query semantics. The receipt of the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device includes: Receive, via broadcast or multicast from the central device, a first query semantic, a second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, and the format of the first query semantic; receive via broadcast or multicast from the central device, a second query semantic, a third matching score function associated with the second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic; or The central device receives, in a multiplexed manner, a first query semantic, a second matching score function associated with the first query semantic, a second threshold associated with the second matching score function, the length of the first query semantic, the format of the first query semantic, a second query semantic, a third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic from at least two query semantics.

12. The method according to claim 11, characterized in that, The method further includes: Acquire the perceived data; Convert the perceived data into general perception semantics; The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes: Based on the second matching score function, determine whether the second matching score between the general-aware semantics and the first query semantics is greater than or equal to the second threshold; Based on the third matching score function, determine whether the third matching score between the general perceptual semantics and the second query semantics is greater than or equal to the third threshold.

13. The method according to claim 11, characterized in that, The method further includes: Acquire the perceived data; The perceived data is converted into first perceived semantics based on the first semantic configuration, the length of the first query semantics, and the format of the first query semantics; The perceived data is converted into second perceived semantics according to the second semantic configuration, the length of the second query semantics, and the format of the second query semantics; The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes: Based on the second matching score function, determine whether the second matching score between the first perceived semantic and the first query semantic is greater than or equal to the second threshold; Based on the third matching score function, determine whether the third matching score between the second perceived semantics and the second query semantics is greater than or equal to the third threshold.

14. The method according to claim 11, characterized in that, The method further includes: Acquire the perceived data; Convert the perceived data into general perception semantics; According to the general lexicalization configuration, the general perceptual semantic lexical is converted into a first perceptual lexical, the first query semantic lexical is converted into a first query lexical, the general perceptual semantic lexical is converted into a second perceptual lexical, and the second query semantic lexical is converted into a second query lexical; or according to the first lexicalization configuration, the general perceptual semantic lexical is converted into a first perceptual lexical, the first query semantic lexical is converted into a first query lexical, and according to the second lexicalization configuration, the general perceptual semantic lexical is converted into a second perceptual lexical, and the second query semantic lexical is converted into a second query lexical. The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes: Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold; Based on the third matching score function, determine whether the third matching score between the second perceived word and the second query word is greater than or equal to the third threshold.

15. The method according to claim 11, characterized in that, The method further includes: Acquire the perceived data; The perceived data is converted into first perceived semantics based on the first semantic configuration, the length of the first query semantics, and the format of the first query semantics; The perceived data is converted into second perceived semantics according to the second semantic configuration, the length of the second query semantics, and the format of the second query semantics; According to the general lexicalization configuration, the first perceptual semantic lexical is converted into a first perceptual lexical, the first query semantic lexical is converted into a first query lexical, the second perceptual semantic lexical is converted into a second perceptual lexical, and the second query semantic lexical is converted into a second query lexical; or according to the first lexicalization configuration, the first perceptual semantic lexical is converted into a first perceptual lexical, the first query semantic lexical is converted into a first query lexical, and according to the second lexicalization configuration, the second perceptual semantic lexical is converted into a second perceptual lexical, and the second query semantic lexical is converted into a second query lexical. The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes: Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold; Based on the third matching score function, determine whether the third matching score between the second perceived word and the second query word is greater than or equal to the third threshold.

16. The method according to any one of claims 1 to 10, characterized in that, The first query information includes at least two query terms. The receipt of the first query information, the first matching score function, and the first threshold broadcast or multicast by the central device includes: Receive the first query term from the at least two query terms broadcast or multicast by the central device, the second matching score function associated with the first query term, the second threshold associated with the second matching score function, and the length of the first query term; receive the second query term from the at least two query terms broadcast or multicast by the central device, the third matching score function associated with the second query term, the third threshold associated with the third matching score function, and the length of the second query term; or The central device receives, in a multiplexed manner, a first query term from the at least two query terms, a second matching score function associated with the first query term, a second threshold associated with the second matching score function, the length of the first query term, a second query term from the at least two query terms, a third matching score function associated with the second query term, a third threshold associated with the third matching score function, and the length of the second query term.

17. The method according to claim 16, characterized in that, The method further includes: Acquire the perceived data; Convert the perceived data into general perception semantics; According to the general lexicalization configuration, the general perceptual semantic lexical is converted into a first perceptual lexical based on the length of the first query lexical, and the general perceptual semantic lexical is converted into a second perceptual lexical based on the length of the second query lexical; or according to the first lexicalization configuration, the general perceptual semantic lexical is converted into a first perceptual lexical based on the length of the first query lexical, and according to the second lexicalization configuration, the general perceptual semantic lexical is converted into a second perceptual lexical based on the length of the second query lexical. The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes: Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold; Based on the third matching score function, determine whether the third matching score between the second perceived word and the second query word is greater than or equal to the third threshold.

18. The method according to claim 16, characterized in that, The method further includes: Acquire the perceived data; According to the first semantic configuration, the perceived data is converted into the first perceptual semantics; According to the second semantic configuration, the perceived data is converted into second perceptual semantics; According to the general lexicalization configuration, the first perceptual semantic lexical is converted into a first perceptual lexical based on the length of the first query lexical, and the second perceptual semantic lexical is converted into a second perceptual lexical based on the length of the second query lexical; or according to the first lexicalization configuration, the first perceptual semantic lexical is converted into a first perceptual lexical based on the length of the first query lexical, and according to the second lexicalization configuration, the second perceptual semantic lexical is converted into a second perceptual lexical based on the length of the second query lexical. The step of determining whether the first matching score between the perceived data and the first query information is greater than or equal to the first threshold based on the first matching score function includes: Based on the second matching score function, determine whether the second matching score between the first perceived word and the first query word is greater than or equal to the second threshold; Based on the third matching score function, determine whether the third matching score between the second perceived word and the second query word is greater than or equal to the third threshold.

19. The method according to any one of claims 11 to 18, characterized in that, The first matching score function, the second matching score function, or the third matching score function includes an inner product or Euclidean distance.

20. The method according to any one of claims 1 to 19, characterized in that, The perception result includes one of the following: Raw sensory data; Perceptual semantics obtained from raw perceptual data; Semi-raw perceptual data and perceptual semantics obtained from raw perceptual data; The original perceived data and the first matching score; Perceptual semantics obtained from raw perceptual data and the first matching score; Semi-raw perceptual data, perceptual semantics obtained from raw perceptual data, and the first matching score.

21. The method according to any one of claims 11 to 20, characterized in that, Sending the sensing result to the central device includes: Send a first matching perceptual semantic and a fourth matching score related to the first matching perceptual semantic to the central device; send a second matching perceptual semantic and a fifth matching score related to the second matching perceptual semantic to the central device.

22. A communication method, characterized in that, include: Broadcast or multicast the first query information, the first matching score function, and the first threshold to multiple sensing devices; A sensing result is received from at least one of the plurality of sensing devices, wherein a first matching score between the sensing data and the first query information is greater than or equal to a first threshold, the first matching score is based on a first matching score function, and the sensing result indicates the sensing data.

23. The method according to claim 22, characterized in that, The first query information includes query information for a task or query information for a modality.

24. The method according to claim 23, characterized in that, The method further includes: Send an instruction for the task identifier or the modal identifier.

25. The method according to claim 22, characterized in that, The first query information includes multiple query information for multiple tasks and multiple task identifiers associated with the multiple query information respectively.

26. The method according to claim 22, characterized in that, The first query information includes multiple query information for multiple modalities and multiple modal identifiers associated with the multiple query information respectively.

27. The method according to claim 22, characterized in that, The first query information includes a first plurality of query information for multiple tasks, a second plurality of query information for multiple modalities, a plurality of task identifiers associated with the first plurality of query information, and a plurality of modal identifiers associated with the second plurality of query information.

28. The method according to any one of claims 22 to 27, characterized in that, The step of broadcasting or multicasting the first query information, the first matching score function, and the first threshold to the plurality of sensing devices includes: The first query information, the first matching score function, and the first threshold are broadcast to the plurality of sensing devices in the Synchronization Signal Block (SSB) / Physical Broadcast Channel (PBCH) block.

29. The method according to claim 28, characterized in that, The first query information is carried in a main information block (MIB) message or a system information block (SIB) message, and the MIB message or the SIB message includes an indication for whether the first query information is presented in the MIB message or the SIB message.

30. The method according to claim 29, characterized in that, The MIB message or the SIB message also includes an indication of the period for the first query information.

31. The method according to any one of claims 22 to 27, characterized in that, The step of broadcasting or multicasting the first query information, the first matching score function, and the first threshold to the plurality of sensing devices includes: The first query information, the first matching score function, and the first threshold are multicast to the multiple sensing devices in the form of a multicast message.

32. The method according to any one of claims 22 to 31, characterized in that, The method further includes: Receive the second query information from the generative pre-trained converter model GPT device; The sensing results are output to the GPT device.

33. The method according to claim 32, characterized in that, Receiving the second query information from the GPT device includes: Receive at least two query semantics from at least two GPT devices; The step of broadcasting or multicasting the first query information, the first matching score function, and the first threshold to the plurality of sensing devices includes: Broadcast or multicast to the plurality of sensing devices a first query semantic, a second matching score function related to the first query semantic, a second threshold related to the second matching score function, the length of the first query semantic, and the format of the first query semantic; broadcast or multicast to the plurality of sensing devices a second query semantic, a third matching score function related to the second query semantic, a third threshold related to the third matching score function, the length of the second query semantic, and the format of the second query semantic; or The first query semantic, the second matching score function associated with the first query semantic, the second threshold associated with the second matching score function, the length of the first query semantic, the format of the first query semantic, the second query semantic, the third matching score function associated with the second query semantic, the third threshold associated with the third matching score function, the length of the second query semantic, and the format of the second query semantic are broadcast or multicast to the plurality of sensing devices in a multiplexed manner.

34. The method according to claim 32, characterized in that, Receiving the second query information from the GPT device includes: Receive at least two query semantics from at least two GPT devices; The method further includes: The first query semantic and the second query semantic in the at least two query semantics are lexicalized into first query lexical and second query lexical; The step of broadcasting or multicasting the first query information, the first matching score function, and the first threshold to the plurality of sensing devices includes: Broadcast or multicast the first query term, the second matching score function associated with the first query term, the second threshold associated with the second matching score function, and the length of the first query term to the plurality of sensing devices; broadcast or multicast the second query term, the third matching score function associated with the second query term, the third threshold associated with the third matching score function, and the length of the second query term to the plurality of sensing devices; or The first query term, the second matching score function associated with the first query term, the second threshold associated with the second matching score function, the length of the first query term, the second query term, the third matching score function associated with the second query term, the third threshold associated with the third matching score function, and the length of the second query term are broadcast or multicast to the plurality of sensing devices in a multiplexed manner.

35. The method according to claim 34, characterized in that, The step of transforming the first query semantic and the second query semantic in the at least two query semantics into the first query term and the second query term includes: According to the first lexicalization configuration, the first query semantic lexicalization is converted into the first query lexical; according to the second lexicalization configuration, the second query semantic lexicalization is converted into the second query lexical; or According to the general lexicalization configuration, the first query semantic lexicalization is converted into the first query lexical; according to the general lexicalization configuration, the second query semantic lexicalization is converted into the second query lexical.

36. The method according to any one of claims 33 to 35, characterized in that, The first matching score function, the second matching score function, or the third matching score function includes an inner product or Euclidean distance.

37. The method according to any one of claims 22 to 36, characterized in that, The perception result includes one of the following: Raw sensory data; Perceptual semantics obtained from raw perceptual data; Semi-raw perceptual data and perceptual semantics obtained from raw perceptual data; The original perceived data and the first matching score; Perceptual semantics obtained from raw perceptual data and the first matching score; Semi-raw perceptual data, perceptual semantics obtained from raw perceptual data, and the first matching score.

38. The method according to any one of claims 22 to 37, characterized in that, Receiving the sensing result from the at least one sensing device includes: Receive from at least one of the plurality of sensing devices a plurality of first matching sensing semantics and a plurality of fourth matching scores associated with the plurality of first matching sensing semantics; receive from at least one of the plurality of sensing devices a plurality of second matching sensing semantics and a plurality of fifth matching scores associated with the plurality of second matching sensing semantics; The method further includes: The perceptual semantics of some or all of the multiple first matches are fused based on the multiple fourth matching scores to obtain the first fused perceptual semantics; the perceptual semantics of some or all of the multiple second matches are fused based on the multiple fifth matching scores to obtain the second fused perceptual semantics.

39. The method according to any one of claims 22 to 37, characterized in that, Receiving the sensing result from the at least one sensing device includes: Receive from at least one of the plurality of sensing devices a plurality of sensing semantics, a plurality of fourth matching scores associated with the plurality of sensing semantics, and a plurality of fifth matching scores associated with the plurality of sensing semantics; The method further includes: The first fused perceptual semantics are obtained by fusing some or all of the multiple fourth matching scores; the second fused perceptual semantics are obtained by fusing some or all of the multiple second perceptual semantics by fusing some or all of the multiple fifth matching scores.

40. The method according to claim 38 or 39, characterized in that, Also includes: Send the first fused perceptual semantics to the first GPT device among the at least two GPT devices; The second fused perceptual semantics is sent to the second GPT device among the at least two GPT devices.

41. The method according to any one of claims 38 to 40, characterized in that, Also includes: Determine the sixth matching score of the first fused perceptual semantics; Determine the seventh matching score of the second fused perceptual semantics.

42. A communication device, characterized in that, It includes modules for performing the method according to any one of claims 1 to 21, or modules for performing the method according to any one of claims 22 to 41.

43. An electronic device, characterized in that, It includes processing circuitry for performing the method according to any one of claims 1 to 21, or processing circuitry for performing the method according to any one of claims 22 to 41.

44. A chip, characterized in that, It includes an input / output I / O interface and a processor, wherein the processor is configured to invoke and run a computer program stored in a memory, enabling a device on which the chip is mounted to perform the method according to any one of claims 1 to 21, or to perform the method according to any one of claims 22 to 41.

45. A sensing device, characterized in that, include: One or more processors; A non-transitory computer-readable storage medium coupled to the one or more processors and storing a program for execution by the processors, wherein, when executed by the processors, the program configures the sensing device to perform the method according to any one of claims 1 to 21.

46. ​​A central device, characterized in that, include: One or more processors; A non-transitory computer-readable storage medium is coupled to the processor and stores a program for execution by the processor, wherein, when executed by the processor, the program configures the central device to perform the method according to any one of claims 22 to 41.

47. A communication system, characterized in that, include: The sensing device according to claim 45 and the central device according to claim 46.

48. A non-transitory computer-readable medium, characterized in that, The device carries program code that, when executed by a computer device, causes the computer device to perform the method according to any one of claims 1 to 21 or the method according to any one of claims 22 to 41.

49. A computer program product, characterized in that, Includes program code, which, when executed on a computer or processor, is used to perform the method according to any one of claims 1 to 21 or the method according to any one of claims 22 to 41.

Images