Semantic communication method and related device

Abstract

The invention discloses a semantic communication method and a related device, and the method comprises the steps: determining a first set, the first set comprises at least one triple corresponding to to-be-transmitted data, and each triple comprises a first head entity, a first tail entity, and a first relation between the first head entity and the first tail entity; a third set is determined according to the probability corresponding to the second relation in the second set, the second set comprises at least one tetrad, each tetrad comprises a second head entity, a second tail entity, the second relation existing between the second head entity and the second tail entity and the probability corresponding to the second relation, and the third set comprises at least one two-tuple; the two-tuple comprises a first head entity and a first tail entity; and sending the third set to the receiving end, the third set being used by the receiving end to determine the first set. Therefore, the data transmission quantity is reduced, and the communication time delay is reduced.

Description

A semantic communication method and related device Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a semantic communication method and related devices. Background Art

[0002] Compared to traditional communication technologies, semantic communication introduces semantic encoding when transmitting the source's semantic information. Semantic communication captures the semantic features of the sender's data based on the task to be performed by the receiver. Only the important semantic features are transmitted, significantly reducing the resources required for communication. The receiver's task can be data reconstruction or more intelligent tasks such as image classification and language translation.

[0003] A knowledge graph (KG), as a structured form of knowledge, represents a collection of semantic information. A knowledge graph consists of entities and relations. Entities can be real-world objects or abstract concepts, while relations represent the specific relationships between entities. Knowledge in a knowledge graph can be represented by triples (head entity, relation, tail entity), such as (radar, used for target detection) and (federated learning, subordinate to, distributed learning). Unstructured text data can be used to construct a structured knowledge graph by extracting entity and relation information.

[0004] Currently, when transmitting data, the sender can represent the data based on a knowledge graph and send the knowledge graph to the receiver. After receiving the knowledge graph, the receiver can restore the original data or perform subsequent intelligent tasks. In the above solution, all information in the knowledge graph is generally sent to the receiver, resulting in a large amount of data transmission.

[0005] Summary of the Invention

[0006] The present application provides a semantic communication method and related devices, which can reduce the amount of transmitted data.

[0007] In a first aspect, the present application provides a semantic communication method that can be applied in the field of communication technology, including:

[0008] First, a first set is determined, wherein the first set includes at least one triple corresponding to the data to be transmitted, and each triple includes a first head entity, a first tail entity, and a first relationship between the first head entity and the first tail entity; a third set is determined based on the probability corresponding to the second relationship in the second set, wherein the second set includes at least one quadruple, each quadruple includes a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and a probability corresponding to the second relationship, and the third set includes at least one binary, and the binary includes the first head entity and the first tail entity; finally, the third set is sent to the receiving end, wherein the third set is used by the receiving end to determine the first set.

[0009] In this application, compared to directly sending the first set corresponding to the data to be transmitted, the first relations that can be omitted in the first set are determined based on the probabilities corresponding to the second relations, and a third set with the omitted first relations is sent, thereby reducing the amount of data transmitted and lowering communication latency. In addition, the same second set is shared with the receiving end, so that the receiving end can recover the omitted first relations in the third set based on the second set, thereby obtaining a complete first set.

[0010] A possible implementation also includes:

[0011] A second set is received.

[0012] In this implementation, by receiving the second set, part of the information is omitted based on the same background knowledge as the receiving end.

[0013] A possible implementation also includes:

[0014] The second set is sent to the receiving end.

[0015] In this implementation, the second set is sent to the receiving end so that the receiving end can restore the omitted information based on the background knowledge shared with the receiving end.

[0016] A possible implementation also includes:

[0017] Determining first information, where the first information indicates a maximum number of omitted information in the first set;

[0018] Send the first information to a receiving end.

[0019] A possible implementation also includes:

[0020] First information is received from a receiving end, where the first information indicates a maximum number of omitted information in a first set.

[0021] In one possible implementation, determining the third set according to the probability corresponding to the second relationship in the second set includes:

[0022] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0023] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0024] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0025] In this implementation, compared with the first set, the third set includes triplets with the first relation omitted and triplets with the first relation not omitted, thereby further reducing the amount of data transmission.

[0026] In one possible implementation, determining the third set according to the probability corresponding to the second relationship in the second set includes:

[0027] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0028] For each triple, if a quadruple matching the triple is selected from the second set, then when the probability corresponding to the second relationship in the quadruple satisfies the first condition, the first relationship in the triple is omitted to obtain a triple with the first relationship omitted;

[0029] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0030] In this implementation, quads matching the first set are selected from the second set. When the probability corresponding to the second relationship in the quadruple satisfies the first condition, the first relationship in the triple is omitted to obtain a third set that omits some of the first relationships, thereby reducing the amount of data transmitted. Furthermore, selecting the third set based on the second set allows the receiving end to restore the omitted first relationships based on the shared second set, thereby completing the first set.

[0031] In one possible implementation, determining the third set according to the probability corresponding to the second relationship in the second set includes:

[0032] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0033] For each triple, if the first head entity in the triple is the same as the second head entity in the quadruple; and the first tail entity in the triple is the same as the second tail entity in the quadruple; and the first relationship in the triple is the same as the second relationship in the quadruple, then when the probability corresponding to the second relationship in the quadruple meets the first condition, the first relationship in the triple is omitted to obtain the triple with the first relationship omitted;

[0034] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0035] In this implementation method, only when the first head entity, first relationship, and first tail entity in the triplet all match the second head entity, first relationship, and second tail entity in the quadruple, can the first relationship in the triplet be omitted based on the probability corresponding to the second relationship in the quadruple, and only the first head entity and the first tail entity are transmitted to reduce the amount of data transmission.

[0036] In one possible implementation, determining the third set according to the probability corresponding to the second relationship in the second set includes:

[0037] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0038] For each triple, if the first head entity in the triple is the same as the second head entity in the quadruple; and the first tail entity in the triple is the same as the second tail entity in the quadruple; and the first relationship in the triple is the same as the second relationship in the quadruple, then when the probability corresponding to the second relationship in the quadruple is the maximum among all probabilities, the first relationship in the triple is omitted, and the triple with the first relationship omitted is obtained, and all probabilities are the probabilities corresponding to all second relationships between the second head entity and the second tail entity in the quadruple;

[0039] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0040] In this implementation method, the first relationship in the corresponding triplet can only be omitted when the probability corresponding to the second relationship in the quadruple is the maximum value among all probabilities, so that the receiving end can restore the omitted first relationship in the triplet according to the same rule, which helps to reduce the amount of data transmission and assists in data recovery.

[0041] In one possible implementation, determining the third set according to the probability corresponding to the second relationship in the second set includes:

[0042] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0043] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0044] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0045] For the plurality of triples in the fifth set, the triples with the first relation omitted and the triples with the first relation not omitted are screened out from the plurality of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set;

[0046] The third set is obtained according to the updated fourth set and fifth set.

[0047] In this implementation method, after obtaining the triples with the first relationship omitted and the triples with the first relationship not omitted, based on all the triples with the first relationship omitted in the first set as conditions, the triples with the first relationship omitted and the triples with the first relationship not omitted can be further filtered out from the fifth set to filter out more triplets with the first relationship omitted, thereby further reducing the amount of data transmission.

[0048] In one possible implementation, determining the third set according to the probability corresponding to the second relationship in the second set includes:

[0049] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0050] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0051] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0052] For the triples in the fifth set, for each triple in the triples, when the triple matches any quadruple in the second set, calculate the conditional probability of the quadruple under the condition that any n triples in the fourth set exist, where n is a positive integer and its initial value is 1;

[0053] When the conditional probability satisfies the second condition, the first relation in the triple is omitted to obtain the updated fourth set and fifth set;

[0054] When the third condition is satisfied, n is incremented by 1, and the process returns to the step of calculating, for each triple in the plurality of triples, a conditional probability of the quadruple under the condition that any n triples in the fourth set exist when the triple matches any quadruple in the second set;

[0055] The third set is obtained according to the updated fourth set and fifth set.

[0056] In this implementation, after the set omitting part of the first relation is screened out, the triples omitting the first relation are used as conditions for the second round of conditional probability search to further reduce the amount of data transmission.

[0057] In one possible implementation, the second condition includes:

[0058] The conditional probability corresponding to the quadruple is the maximum value among all probabilities, and all probabilities are the conditional probabilities corresponding to all second relations between the second head entity and the second tail entity in the quadruple.

[0059] In this implementation, the conditional probability corresponding to the first relationship is the maximum probability among all conditional probabilities as an omission condition, so that the receiving end can recover the first relationship with the highest probability from the same second set according to this condition, thereby facilitating the recovery of the first set.

[0060] In one possible implementation, the third condition includes at least one of the following:

[0061] n is less than the first value;

[0062] The number of triplets in the fourth set is less than the second value.

[0063] In one possible implementation, determining the third set according to the probability corresponding to the second relationship in the second set includes:

[0064] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0065] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0066] When the number of all omitted first relations in the first set is less than a second value, the execution continues, where the second value is the maximum number of omitted information in the first set, and the second value is obtained based on device capabilities, time thresholds, or channel conditions;

[0067] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0068] For the plurality of triples in the fifth set, the triples with the first relation omitted and the triples with the first relation not omitted are screened out from the plurality of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set;

[0069] The third set is obtained according to the updated fourth set and fifth set.

[0070] In this implementation, the number of omitted first relationships in the first set can be obtained based on device capabilities, time thresholds or channel conditions, so that the number of omitted first relationships in the first set can be limited according to actual needs to meet different transmission requirements.

[0071] A possible implementation also includes:

[0072] Second information is determined, where the second information is used to indicate that the third set is sent using transmit power.

[0073] A possible implementation also includes:

[0074] Second information is received from the receiving end, where the second information indicates that the third set is sent using the transmission power.

[0075] In a second aspect, the present application provides a semantic communication method, comprising:

[0076] Determine a first set, wherein the first set includes at least one triple, each triple includes a first head entity, a first tail entity, and a first relationship between the first head entity and the first tail entity;

[0077] Determining the omitted information in the first set based on the probability corresponding to the second relationship in the second set to obtain a third set, wherein the second set includes at least one quadruple, each quadruple includes the second head entity, the second tail entity, the second relationship between the second head entity and the second tail entity, and the probability corresponding to the second relationship;

[0078] The third set is sent to the receiving end, where the third set is used by the receiving end to determine the first set.

[0079] A possible implementation also includes:

[0080] A second set is received.

[0081] A possible implementation also includes:

[0082] The second set is sent to the receiving end.

[0083] A possible implementation also includes:

[0084] Determining first information, where the first information indicates a maximum number of omitted information in the first set;

[0085] Send the first information to a receiving end.

[0086] A possible implementation also includes:

[0087] First information is received from a receiving end, where the first information indicates a maximum number of omitted information in a first set.

[0088] In one possible implementation, determining the third set according to the probability corresponding to the second relationship in the second set includes:

[0089] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0090] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0091] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0092] In one possible implementation, all triples whose first relation is not omitted are obtained based on the first set and all triples whose first relation is omitted in the first set, including:

[0093] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0094] For each triple, if a quadruple matching the triple is selected from the second set, then when the probability corresponding to the second relationship in the quadruple satisfies the first condition, the first relationship in the triple is omitted to obtain a triple with the first relationship omitted;

[0095] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0096] In one possible implementation, the condition for selecting a quadruple matching the triple from the second set is:

[0097] The first head entity in the triple is the same as the second head entity in the quad; and,

[0098] The first last entity in the triple is the same as the second last entity in the quad; and

[0099] The first relation in the triple is the same as the second relation in the quad.

[0100] In one possible implementation, the first condition includes:

[0101] The probability corresponding to the second relationship in the quadruple is the maximum value among all probabilities, and all probabilities are the probabilities corresponding to all second relationships between the second head entity and the second tail entity in the quadruple.

[0102] In one possible implementation, a third set is obtained based on all triples whose first relation is omitted and all triples whose first relation is not omitted, including:

[0103] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0104] For the plurality of triples in the fifth set, the triples with the first relation omitted and the triples with the first relation not omitted are screened out from the plurality of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set;

[0105] The third set is obtained according to the updated fourth set and fifth set.

[0106] In one possible implementation, for the plurality of triples in the fifth set, triples with the first relation omitted and triples with the first relation not omitted are filtered out from the plurality of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set, including:

[0107] For the triples in the fifth set, for each triple in the triples, when the triple matches any quadruple in the second set, calculate the conditional probability of the quadruple under the condition that any n triples in the fourth set exist, where n is a positive integer and its initial value is 1;

[0108] When the conditional probability satisfies the second condition, the first relation in the triple is omitted to obtain the updated fourth set and fifth set;

[0109] When the third condition is met, let n be increased by 1, and return to the step of calculating the conditional probability corresponding to the existence of any n triplets in the fourth set for each triple in the several triplets when the triplet matches any quadruple in the second set.

[0110] In one possible implementation, the second condition includes:

[0111] The conditional probability corresponding to the quadruple is the maximum value among all probabilities, and all probabilities are the conditional probabilities corresponding to all second relations between the second head entity and the second tail entity in the quadruple.

[0112] In one possible implementation, the third condition includes at least one of the following:

[0113] n is less than the first value;

[0114] The number of triplets in the fourth set is less than the second value.

[0115] Before getting the fourth and fifth sets, also include:

[0116] Obtain a second value, where the second value is the maximum value of the number of omitted information in the first set;

[0117] When the number of all omitted first relations in the first set is less than the second value, execution continues.

[0118] In a third aspect, the present application provides a semantic communication method, comprising:

[0119] receiving a third set from a sending end, the third set including at least one two-tuple, the two-tuple including a first header entity and a first tail entity;

[0120] The first set is determined based on the probability corresponding to the second relationship in the second set, the second set includes at least one quadruple, each quadruple includes the second head entity, the second tail entity, the second relationship between the second head entity and the second tail entity, and the probability corresponding to the second relationship, the first set includes at least one triple, each triple includes the first head entity, the first tail entity, the first relationship between the first head entity and the first tail entity.

[0121] In the present application, after receiving the third set, the first relationship omitted in the third set can be restored based on the second set shared with the transmitting end to obtain the complete first set.

[0122] A possible implementation also includes:

[0123] The second set is received.

[0124] A possible implementation also includes:

[0125] The second set is sent to the sending end.

[0126] A possible implementation also includes:

[0127] Determining first information, where the first information indicates a maximum number of omitted information in the first set;

[0128] Send the first message.

[0129] A possible implementation also includes:

[0130] First information is received from a transmitting end, where the first information indicates a maximum number of omitted information in a first set.

[0131] In one possible implementation, determining the first set according to the probability corresponding to the second relationship in the second set includes:

[0132] For at least one bigram in the third set, obtaining, according to the probability of the second relation, a first relation in which each of the at least one bigram is omitted, to obtain at least one triple;

[0133] The first set is obtained according to all triples and at least one triple in the third set.

[0134] In a possible implementation, obtaining the first relation in which at least one binary tuple is omitted according to the probability of the second relation to obtain at least one triple includes:

[0135] For each binary group in at least one binary group, a quadruple matching the binary group is selected from the second set. When the probability corresponding to the second relationship in the quadruple group satisfies the first condition, the second relationship of the quadruple group is added to the binary group to obtain a triple.

[0136] In one possible implementation, the conditions for selecting a quadruple matching the binary from the second set include:

[0137] The first head entity in the two-tuple is the same as the second head entity in the four-tuple, and the first tail entity in the two-tuple is the same as the second tail entity in the four-tuple.

[0138] In one possible implementation, the first condition includes:

[0139] The probability corresponding to the second relationship in the quadruple is the maximum value among all probabilities, and all probabilities are the probabilities corresponding to all second relationships between the second head entity and the second tail entity in the quadruple.

[0140] In one possible implementation, obtaining the first set according to all triples and at least one triple in the third set includes:

[0141] The conditional probability of any unrecovered dyad under the condition that there is a dyad that has restored the triple in the third set is calculated;

[0142] For any unrecovered binary in the third set, when the conditional probability of the unrecovered binary is the maximum value among all conditional probabilities, obtain the first relation corresponding to the unrecovered binary to obtain the recovered triple, where all conditional probabilities are the conditional probabilities corresponding to all unrecovered binary in the third set under the same conditions;

[0143] A first set is obtained according to the restored triples, all triples in the third set, and at least one triple.

[0144] In a fourth aspect, the present application provides a communication device, characterized by comprising:

[0145] a processing module, configured to obtain a first set, the first set including at least one triple corresponding to the data to be transmitted, each triple including a first head entity, a first tail entity, and a first relationship between the first head entity and the first tail entity;

[0146] The processing module is further configured to determine a third set based on a probability corresponding to the second relationship in the second set, the second set including at least one quadruple, each quadruple including a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and a probability corresponding to the second relationship, the third set including at least one bigram, the bigram including the first head entity and the first tail entity;

[0147] The transceiver module is used to send the third set to the receiving end, and the third set is used by the receiving end to determine the first set.

[0148] In one possible implementation, the processing module is further configured to:

[0149] A second set is received.

[0150] In one possible implementation, the transceiver module is further configured to:

[0151] The second set is sent to the receiving end.

[0152] In a possible implementation, the processing module is further configured to determine first information, where the first information indicates a maximum number of omitted information in the first set;

[0153] The transceiver module is further configured to send the first information to a receiving end.

[0154] In one possible implementation, the transceiver module is further configured to:

[0155] First information is received from a receiving end, where the first information indicates a maximum number of omitted information in a first set.

[0156] In one possible implementation, the processing module is further configured to:

[0157] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0158] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0159] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0160] In one possible implementation, the processing module is further configured to:

[0161] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0162] For each triple, if a quadruple matching the triple is selected from the second set, then when the probability corresponding to the second relationship in the quadruple satisfies the first condition, the first relationship in the triple is omitted to obtain a triple with the first relationship omitted;

[0163] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0164] In one possible implementation, the processing module is further configured to:

[0165] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0166] For each triple, if the first head entity in the triple is the same as the second head entity in the quadruple; and the first tail entity in the triple is the same as the second tail entity in the quadruple; and the first relationship in the triple is the same as the second relationship in the quadruple, then when the probability corresponding to the second relationship in the quadruple meets the first condition, the first relationship in the triple is omitted to obtain the triple with the first relationship omitted;

[0167] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0168] In one possible implementation, the processing module is further configured to:

[0169] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0170] For each triple, if the first head entity in the triple is the same as the second head entity in the quadruple; and the first tail entity in the triple is the same as the second tail entity in the quadruple; and the first relationship in the triple is the same as the second relationship in the quadruple, then when the probability corresponding to the second relationship in the quadruple is the maximum among all probabilities, the first relationship in the triple is omitted, and the triple with the first relationship omitted is obtained, and all probabilities are the probabilities corresponding to all second relationships between the second head entity and the second tail entity in the quadruple;

[0171] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0172] In one possible implementation, the processing module is further configured to:

[0173] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0174] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0175] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0176] For the plurality of triples in the fifth set, the triples with the first relation omitted and the triples with the first relation not omitted are screened out from the plurality of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set;

[0177] The third set is obtained according to the updated fourth set and fifth set.

[0178] In one possible implementation, the processing module is further configured to:

[0179] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0180] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0181] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0182] For the triples in the fifth set, for each triple in the triples, when the triple matches any quadruple in the second set, calculate the conditional probability of the quadruple under the condition that any n triples in the fourth set exist, where n is a positive integer and its initial value is 1;

[0183] When the conditional probability satisfies the second condition, the first relation in the triple is omitted to obtain the updated fourth set and fifth set;

[0184] When the third condition is satisfied, n is incremented by 1, and the process returns to the step of calculating, for each triple in the plurality of triples, a conditional probability of the quadruple under the condition that any n triples in the fourth set exist when the triple matches any quadruple in the second set;

[0185] The third set is obtained according to the updated fourth set and fifth set.

[0186] In one possible implementation, the second condition includes:

[0187] The conditional probability corresponding to the quadruple is the maximum value among all probabilities, and all probabilities are the conditional probabilities corresponding to all second relations between the second head entity and the second tail entity in the quadruple.

[0188] In this implementation, the conditional probability corresponding to the first relationship is the maximum probability among all conditional probabilities as an omission condition, so that the receiving end can recover the first relationship with the highest probability from the same second set according to this condition, thereby facilitating data recovery.

[0189] In one possible implementation, the third condition includes at least one of the following:

[0190] n is less than the first value;

[0191] The number of triplets in the fourth set is less than the second value.

[0192] In one possible implementation, the processing module is further configured to:

[0193] Determining the third set according to the probability corresponding to the second relationship in the second set includes:

[0194] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0195] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0196] When the number of all omitted first relations in the first set is less than a second value, the execution continues, where the second value is the maximum number of omitted information in the first set, and the second value is obtained based on device capabilities, time thresholds, or channel conditions;

[0197] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0198] For the plurality of triples in the fifth set, the triples with the first relation omitted and the triples with the first relation not omitted are screened out from the plurality of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set;

[0199] The third set is obtained according to the updated fourth set and fifth set.

[0200] In one possible implementation, the processing module is further configured to:

[0201] Second information is determined, where the second information is used to indicate that the third set is sent using transmit power.

[0202] In one possible implementation, the transceiver module is further configured to:

[0203] Second information is received from the receiving end, where the second information indicates that the third set is sent using the transmission power.

[0204] In a fifth aspect, the present application provides a communication device, including:

[0205] a transceiver module, configured to receive a third set from a transmitting end, where the third set includes at least one two-tuple, where the two-tuple includes a first header entity and a first tail entity;

[0206] A processing module is used to determine a first set based on the probability corresponding to the second relationship in the second set, where the second set includes at least one quadruple, each quadruple includes a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and the probability corresponding to the second relationship, and the first set includes at least one triple, each triple includes a first head entity, a first tail entity, and the first relationship between the first head entity and the first tail entity.

[0207] In one possible implementation, the processing module is further configured to:

[0208] A second set is received.

[0209] In one possible implementation, the transceiver module is further configured to:

[0210] The second set is sent to the sending end.

[0211] In a possible implementation, the processing module is further configured to determine first information, where the first information indicates a maximum number of omitted information in the first set;

[0212] The transceiver module is further configured to send the first information to the receiving end.

[0213] In one possible implementation, the transceiver module is further configured to:

[0214] First information is received from a receiving end, where the first information indicates a maximum number of omitted information in a first set.

[0215] In one possible implementation, determining the first set according to the probability corresponding to the second relationship includes:

[0216] For at least one bigram in the third set, obtaining, according to the probability of the second relation, a first relation in which each of the at least one bigram is omitted, to obtain at least one triple;

[0217] The first set is obtained according to all triples and at least one triple in the third set.

[0218] In one possible implementation, the processing module is further configured to:

[0219] For each binary group in at least one binary group, if a quadruple matching the binary group is selected from the second set, when the probability corresponding to the second relationship in the quadruple group satisfies the first condition, the second relationship of the quadruple group is added to the binary group to obtain a triple.

[0220] In one possible implementation, the conditions for selecting a quadruple matching the binary from the second set include:

[0221] The first head entity in the two-tuple is the same as the second head entity in the four-tuple, and the first tail entity in the two-tuple is the same as the second tail entity in the four-tuple.

[0222] In one possible implementation, the first condition includes:

[0223] The probability corresponding to the second relationship in the quadruple is the maximum value among all probabilities, and all probabilities are the probabilities corresponding to all second relationships between the second head entity and the second tail entity in the quadruple.

[0224] In the sixth aspect, the present application provides a communication device, characterized in that it includes a processor and an interface circuit, the interface circuit is used to receive signals from other communication devices and transmit them to the processor or send signals from the processor to other communication devices, and the processor is used to implement the methods of the first to third aspects above through logic circuits or execution code instructions.

[0225] In a seventh aspect, the present application provides a communication system comprising at least one communication device, which executes the methods of the first to third aspects above.

[0226] In an eighth aspect, the present application provides a computer program product, which, when executed on a computer, enables the computer to execute the methods of the first to third aspects above.

[0227] In a ninth aspect, the present application provides a chip system comprising a processor for implementing the methods of aspects 1 to 3 above. In one possible design, the chip system further comprises a memory for storing program instructions and / or data. The chip system may be composed of a chip alone or may include a chip and other discrete components.

[0228] In a tenth aspect, the present application provides a computer-readable storage medium, in which a computer program or instruction is stored. When the computer program or instruction is executed by a communication device, the methods of the first to third aspects mentioned above are implemented.

[0229] In an eleventh aspect, the present application provides a communication device comprising a processor, the processor being configured to execute computer programs or instructions in a memory to implement the methods of the first to third aspects above.

[0230] The solutions of the second to eleventh aspects mentioned above are used to implement or cooperate with the method of the first aspect mentioned above, and therefore can achieve the same or corresponding beneficial effects as the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0231] FIG1a is a schematic diagram of the architecture of a communication system used in an embodiment of the present application;

[0232] FIG1b is another schematic diagram of the architecture of a communication system used in an embodiment of the present application;

[0233] Figure 1c is a schematic diagram of an architecture of semantic communication technology;

[0234] Figure 1d is a schematic diagram of a knowledge graph;

[0235] Figure 1e is a schematic diagram of an architecture for semantic communication based on knowledge graph;

[0236] FIG2 is a flow chart of a semantic communication method provided in an embodiment of the present application;

[0237] FIG3 is another flow chart of the semantic communication method provided in an embodiment of the present application;

[0238] FIG4 is a schematic diagram of a probability map provided in an embodiment of the present application;

[0239] FIG5 is a schematic diagram of omitting the first relationship provided in an embodiment of the present application;

[0240] FIG6 is a schematic diagram of a flow chart of semantic information processing performed by a sending end according to an embodiment of the present application;

[0241] FIG7 is a schematic structural diagram of a communication device provided in an embodiment of the present application;

[0242] FIG8 is another schematic structural diagram of a communication device provided in an embodiment of the present application;

[0243] FIG9 is another schematic structural diagram of a communication device provided in an embodiment of the present application;

[0244] FIG10 is another schematic structural diagram of a communication device provided in an embodiment of the present application;

[0245] FIG11 is another structural diagram of a communication device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0246] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application. It is known to those skilled in the art that with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0247] The terms "first," "second," and the like in the specification and claims of this application and in the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0248] The term "and / or" as used in this application can be used to describe an association relationship between related objects, indicating that three possible relationships exist. For example, A and / or B can represent the following three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this application generally indicates that the related objects are in an "or" relationship.

[0249] It should also be noted that, in some alternative implementations, the functions / acts noted may occur out of the order of the drawings. For example, two figures shown in succession may in fact occur substantially concurrently or may sometimes be executed in the reverse order, depending on the functionality / acts involved.

[0250] In the embodiments of the present application, unless otherwise specified, the meaning of "at least one" refers to one or more, and the meaning of "a plurality of" refers to two or more. It is understood that in the present application, "when...", "if" and "if" all mean that the device will make corresponding processing under certain objective circumstances, and do not limit the time, nor do they require that there must be a judgment action when the device is implemented, nor do they mean that there are other limitations. In addition, the special word "exemplary" means "used as an example, embodiment or illustrative". Any embodiment described as "exemplary" is not necessarily interpreted as being superior to or better than other embodiments.

[0251] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0252] First, in order to better understand the solution of the embodiment of the present application, the application scenario of the embodiment of the present application is described below.

[0253] Please refer to Figure 1a, which is a schematic diagram of the architecture of a communication system used in an embodiment of the present application. As shown in Figure 1a, communication system 1000 includes a radio access network 100 and a core network 200. Optionally, communication system 1000 may also include the Internet 300. The radio access network 100 may include at least one network device (such as 110a and 110b in Figure 1a) and at least one terminal device (such as 120a-120j in Figure 1a). The terminal device is wirelessly connected to the network device, and the network device is wirelessly or wiredly connected to the core network. The core network device and the network device may be independent and distinct physical devices, or the functions of the core network device and the logical functions of the network device may be integrated into the same physical device, or a single physical device may integrate some of the functions of the core network device and some of the functions of the network device. Terminal devices and network devices may be interconnected via wired or wireless connections. Figure 1a is merely a schematic diagram. The communication system may also include other network devices, such as wireless relay devices and wireless backhaul devices, which are not shown in Figure 1a.

[0254] A network device is an access device that allows a terminal device to wirelessly access a communication system. A network device can be a base station, an evolved NodeB (eNodeB), a transmission reception point (TRP), a next-generation NodeB (gNB) in a fifth-generation (5G) mobile communication system, a next-generation base station in a sixth-generation (6G) mobile communication system, a base station in a future mobile communication system, or an access node in a Wi-Fi system. It can also be a module or unit that performs some of the functions of a base station, such as a centralized unit (CU) or a distributed unit (DU). The CU here completes the functions of the radio resource control protocol and the packet data convergence protocol (PDCP) of the base station, and can also complete the function of the service data adaptation protocol (SDAP); the DU completes the functions of the radio link control layer and the medium access control (MAC) layer of the base station, and can also complete the functions of part of the physical layer or all of the physical layer. For the specific description of the above-mentioned various protocol layers, please refer to the relevant technical specifications of the 3rd Generation Partnership Project (3GPP). The network device can be a macro base station (such as 110a in Figure 1a), a micro base station or an indoor station (such as 110b in Figure 1a), or a relay node or a donor node, etc. The embodiments of the present application do not limit the specific technology and specific device form adopted by the network device.

[0255] A terminal device is a device with wireless transceiver capabilities that can send signals to or receive signals from network devices. A terminal device may also be referred to as user equipment (UE), a mobile station, or a mobile terminal device. Terminal devices can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), the Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, and smart cities. A terminal device may be a mobile phone, a tablet computer (Pad), a computer with wireless transceiver capabilities, a wearable device, a vehicle, a vehicle-mounted mobile device, an airplane, a ship, a robot, a robotic arm, a personal digital assistant, a wearable device, a virtual reality (VR) device, an augmented reality (AR) device, or various IoT devices. The embodiments of this application do not limit the specific technology and device form factors used by the terminal device.

[0256] Network devices and terminal devices can be fixed or mobile. They can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; they can also be deployed on aircraft, balloons, and artificial satellites. The embodiments of this application do not limit the application scenarios of network devices and terminal devices.

[0257] The roles of network devices and terminal devices can be relative. For example, the helicopter or drone 120i in Figure 1a can be configured as a mobile network device. To terminal devices 120j accessing the wireless access network 100 via 120i, terminal device 120i is a network device. However, to network device 110a, 120i is a terminal device, meaning that communication between 110a and 120i occurs via a wireless air interface protocol. Of course, communication between 110a and 120i can also occur via an interface protocol between network devices. In this case, 120i is also a network device relative to 110a. Therefore, both network devices and terminal devices can be collectively referred to as communication devices. 110a and 110b in Figure 1a can be referred to as communication devices with network device functionality, while 120a-120j in Figure 1a can be referred to as communication devices with terminal device functionality.

[0258] Network devices and terminal devices, network devices and network devices, and terminal devices and terminal devices can communicate through authorized spectrum, unauthorized spectrum, or both; can communicate through spectrum below 6 gigahertz (GHz), spectrum above 6 GHz, or spectrum below 6 GHz and spectrum above 6 GHz simultaneously. The embodiments of the present application do not limit the spectrum resources used for wireless communications.

[0259] In the embodiments of the present application, the functions of the network device may also be performed by a module (such as a chip) in the network device, or by a control subsystem that includes the network device functions. The control subsystem that includes the network device functions here may be a control center in the above-mentioned application scenarios such as smart grid, industrial control, smart transportation, and smart city. The functions of the terminal device may also be performed by a module (such as a chip or a modem) in the terminal device, or by a device that includes the terminal device functions.

[0260] The semantic communication method provided in the embodiments of the present application can be applicable to downlink (DL) data transmission, uplink (UL) data transmission, and device-to-device (D2D) data transmission. For downlink data transmission, the sending device is a network device, and the corresponding receiving device is a terminal device. For uplink data transmission, the sending device is a terminal device, and the corresponding receiving device is a network device. For D2D data transmission, the sending device is a terminal device, and the corresponding receiving device is also a terminal device. The embodiments of the present application do not limit the devices corresponding to the sending end and the receiving end, that is, the communication between the sending end and the receiving end can be between network devices and network devices, between terminal devices and terminal devices, or between network devices and terminal devices.

[0261] The semantic communication method provided in the embodiment of the present application can be applicable to a communication system that adopts hybrid automatic repeat request (HARQ) technology, can be applicable to a frequency division duplex (FDD) system, and can also be applicable to a time division duplex (TDD) system. The semantic communication method provided in the embodiment of the present application described in the embodiment of the present application can be applicable to any communication system that adopts, for example, low density parity check (LDPC) code or polar code as a data channel coding method, and can also be applicable to 5G communication systems and other wireless communication systems. The embodiment of the present application does not limit this.

[0262] Network devices and terminal devices, as well as terminal devices and terminal devices, can communicate through licensed spectrum, or through unlicensed spectrum, or through both licensed spectrum and unlicensed spectrum. Access network devices and terminal devices, as well as terminal devices and terminal devices, can communicate through spectrum below 6G, or through spectrum above 6G, or through spectrum below 6G and spectrum above 6G at the same time. The embodiments of the present application do not limit the spectrum resources used between network devices and terminal devices.

[0263] In addition, in the communication system shown in Figure 1a, the communication between each network device and each terminal device can also be represented in another form. Please refer to Figure 1b, which is another architectural diagram of the communication system used in an embodiment of the present application. Terminal device 10 includes a processor 101, a memory 102, and a transceiver 103. Transceiver 103 includes a transmitter 1031, a receiver 1032, and an antenna 1033. Network device 20 includes a processor 201, a memory 202, and a transceiver 203. Transceiver 203 includes a transmitter 2031, a receiver 2032, and an antenna 2033. Receiver 1032 can be configured to receive transmission control information via antenna 1033, and transmitter 1031 can be configured to send transmission feedback information to network device 20 via antenna 1033. Transmitter 2031 can be configured to send transmission control information to terminal device 10 via antenna 2033, and receiver 2032 can be configured to receive transmission feedback information sent by terminal device 10 via antenna 2033.

[0264] It should be understood that the communication scenarios shown in Figures 1a and 1b are merely examples, and this application does not impose any specific restrictions on the number of terminal devices and network devices in the communication system. The semantic communication method provided in the embodiments of this application can be applied to the structures described above, but is not limited to this.

[0265] In the communication scenarios shown in Figures 1a and 1b above, the main application scenario of the embodiments of the present application is the transmission of data between the sender and the receiver based on semantic communication technology. Among them, semantic communication is used to transmit semantic information. For example, please refer to Figure 1c, which is an architectural diagram of semantic communication technology. As shown in Figure 1c, a key difference between semantic communication and traditional communication technology is that it introduces semantic coding, which captures the semantic features of the sender's data based on the tasks to be performed by the receiver. Only important semantic features will be transmitted, which greatly reduces the required communication resources. The sender semantically encodes the data and sends it to the receiver, and the receiver uses semantic decoding to extract the data. In addition, semantic communication requires the sender and the receiver to have shared background knowledge and a common understanding of the extracted semantic information.

[0266] It should be understood that transmission in this application may refer to sending or receiving.

[0267] As a structured form of knowledge, the knowledge graph (KG) is an important technical means for representing semantic information. A knowledge graph consists of entities and relations. Entities can be real-world objects or abstract concepts, while relations represent specific relationships between entities. A specific example is shown in Figure 1d, which is a schematic diagram of a knowledge graph. Knowledge in a knowledge graph can be represented by triples (head entity, relation, tail entity), such as (Hangzhou, scenic spot, West Lake) and (Zhejiang, provincial capital, Hangzhou). Unstructured text data can be used to construct a structured knowledge graph by extracting entity and relation information.

[0268] For example, in current semantic communication technology, see Figure 1e, which illustrates an architecture for semantic communication based on a knowledge graph. In Figure 1e, when transmitting data, the semantic information corresponding to the data can be modeled using a knowledge graph. The sender can represent the data based on the knowledge graph and send the knowledge graph to the receiver. After receiving the knowledge graph, the receiver can restore the original data or perform subsequent intelligent tasks.

[0269] However, in the above scheme, all the information in the knowledge graph is generally sent to the receiving end, resulting in a large amount of data transmitted. Specifically, corresponding to different semantic communication scenarios, the relationships between different entities in the triples have different frequencies or certainties. If all the extracted semantic information (i.e., the knowledge graph) is transmitted directly, there will be a certain redundancy, resulting in some semantic information in certain specific scenarios being transmitted unnecessarily, resulting in a large amount of data transmitted. Among them, the higher the frequency or the greater the probability of the relationship between entities, the greater the certainty.

[0270] In order to solve the above problems, an embodiment of the present application provides a semantic communication method and related devices, wherein the semantic communication method is applied to a network device (110a and 110b) or a terminal device (120a-120j) as shown in Figure 1a, or it can be a chip, chip system, processor, etc. that supports the above network device or terminal device to implement the above method.

[0271] Specifically, please refer to Figure 2, which is a flow chart of a semantic communication method provided in an embodiment of the present application. The method is mainly applied to network devices or terminal devices. The method specifically includes: S1. Determining a first set, the first set includes at least one triple corresponding to the data to be transmitted, each triple including a first head entity, a first tail entity, and a first relationship between the first head entity and the first tail entity. S2. Obtaining a second set, the second set includes at least one quadruple, each quadruple including a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and a probability corresponding to the second relationship. Optionally, the second set can be obtained from the receiving end or other node, or the second set can be determined based on sample data and sent to the sending end. S3. Determining a third set based on the probability corresponding to the second relationship, the third set includes at least one binary tuple, the binary tuple including the first head entity and the first tail entity. S4. Sending the third set to the receiving end, the third set is used by the receiving end to determine the first set. Optionally, the third set is used by the receiving end to determine the first set based on the second set.

[0272] In this application, compared to directly sending the first set corresponding to the data to be transmitted, the first relations that can be omitted from the first set are determined based on the probabilities corresponding to the second relations, and a third set with the omitted first relations is sent, thereby reducing the amount of data transmitted and lowering communication latency. In addition, the same second set is shared with the receiving end, so that the receiving end can recover the omitted first relations in the third set based on the second set, thereby obtaining a complete first set.

[0273] For ease of understanding, the semantic communication method provided by the embodiment of the present application is specifically described below in conjunction with the accompanying drawings and application scenarios. Those skilled in the art will appreciate that with the development of technology and the emergence of new scenarios, the technical solutions provided by the embodiment of the present application are also applicable to similar technical problems.

[0274] Please refer to Figure 3, which is another flow chart of the semantic communication method provided in an embodiment of the present application. The data processing method can be executed by a network device or terminal device as shown in Figure 1a. For ease of explanation, the sending end and the receiving end are used as the execution entities for explanation, wherein the sending end can be a network device or a terminal device, and the receiving end can also be a network device or a terminal device, which is not limited here. As shown in Figure 3, the following steps 301 to 308 are included.

[0275] Step 301: The sending end obtains data to be transmitted.

[0276] In this embodiment, the data type to be transmitted can be various types of data such as text, image, etc. The sending end can use different types of data for transmission based on specific application scenarios and requirements, which is not limited here.

[0277] Step 302: The sending end determines a first set.

[0278] In this embodiment, the sending end can determine the first set from the data to be transmitted. The first set can be understood as a constructed knowledge graph, or a set of triples. The semantic information of the knowledge graph is expressed by triples, and the form of triples is generally: (first head entity, first relationship, first tail entity), such as (vehicle, left neighbor, pedestrian). The sending end can extract several such triples from source data such as text or images, and these several triples can be represented as a knowledge graph.

[0279] Step 303: The transmitting end obtains a second set, where the second set includes probabilities corresponding to the second relationship.

[0280] In this embodiment, the second set can be understood as a set of quads, also known as a shared probability map, shared between the transmitter and receiver. A quad is a triple that adds the probability of the relationship between the head entity and the tail entity. The quad corresponding to the second set is generally in the form of: (second head entity, second relationship, probability of the second relationship, second tail entity).

[0281] The second set may be constructed locally by the sending end, or may be received from an external source. Optionally, the sending end receives the second set from the receiving end or another node.

[0282] In an optional implementation, the second set can be constructed from sample data.

[0283] In this implementation, when there are many texts with similar topics, the head entity and tail entity of some of their corresponding triples are the same, but the specific relationships may be different. A specific example is shown in Figure 4, which is a schematic diagram of a probability graph provided by an embodiment of the present application. For example, in text data 1, there is a triple (pedestrian, on the left, vehicle), and in text data 2, there is a triple (pedestrian, on the right, vehicle). It is advisable to merge the above two triples and express them as {pedestrian, [(on the left, 0.4), (on the right, 0.6)], vehicle}, where the number next to the relationship represents the frequency of occurrence of the relationship. When there are many text data with similar topics, the same method can be used to merge their corresponding knowledge graphs into a knowledge graph with a probability map. At this time, the edges in the knowledge graph are not uniquely determined relationship edges, but probabilistic edges composed of the probabilities of occurrence of different relationships.

[0284] Specifically, before formal communication begins, a probability map can be constructed based on a large number of samples. Once constructed, this probability map is shared between the sender and receiver as common background knowledge. This process only needs to be performed once, and after completion, formal communication and data transmission can begin.

[0285] Optionally, the following example illustrates a method for constructing a probability map.

[0286] The sample data is represented as:

[0287] Where N represents the total number of sample data. Represents the sample data set. The knowledge graph extracted from the sample data n is expressed as:

[0288] in, Represents the knowledge graph G n The mth triple in , where M represents the total number of triples. It can be written in the following form:

[0289] in, It is a triple The head entity, yes relationship, yes The tail entity.

[0290] The shared probability graph is written as: GS = {δ1,δ2,…,δ s ,…,δ S}

[0291] Among them, δ sis the quadruple that increases the probability of relationship, S is the total number of quadruple, δ s Specifically:

[0292] Among them, h s is the quadruple δ s The head entity, t s is δ s The tail entity, is the relationship between the head and tail entities, I is the total number of relationships between the head and tail entities, It is a triple The set of labels for the samples that appear.

[0293] The sending end can directly construct the second set based on the sample data and send the second set to the receiving end. In other words, the sending end is responsible for constructing the second set. After constructing the second set using the above method, the sending end sends the second set to the receiving end so that both ends share the same probability map.

[0294] Alternatively, the sending end may receive the second set from the receiving end. That is, the receiving end is responsible for constructing the second set. After constructing the second set using the above method, the receiving end sends the second set to the sending end so that both ends share the same probability map.

[0295] There is no requirement for the order in which step 302 and step 303 are executed.

[0296] Step 304: The transmitting end determines a third set according to the probability corresponding to the second relationship in the second set.

[0297] In this embodiment, after obtaining the first relationship and the second relationship, the sending end uses the second set as shared knowledge to compress the semantic information. Specifically, the sending end may determine the third set based on the probability corresponding to the second relationship in the second set. In an optional implementation, the sending end determines the third set based on the probability corresponding to the second relationship in the second set, including:

[0298] For at least one triple, the sending end filters out all triples in which the first relation is omitted from the at least one triple according to the probability corresponding to the second relation;

[0299] The sending end obtains all triples whose first relation is not omitted according to the first set and all triples whose first relation is omitted in the first set;

[0300] The sending end obtains a third set according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0301] In this implementation, after the sending end obtains the second set, the omitted triples can be filtered out from the first set according to the probability corresponding to the second relationship in the second set, and the remaining non-omitted triples and the omitted triples form a third set.

[0302] For example, before transmitting data, semantic information is first extracted from the data, that is, its corresponding knowledge graph, denoted as G, which can be written as: G = {ε1,ε2,…,ε j ,…,ε J}

[0303] Among them, ε j Represents the jth triple in the knowledge graph G, and J represents the total number of triples. j It can be specifically written in the following form: j =(h j ,r j ,t j )

[0304] For details, please refer to Figure 5, which is a schematic diagram of omitting the first relationship provided by an embodiment of the present application. As shown in Figure 5, the first set includes the following triples: (tree, outside, sidewalk), (tree, front, vehicle), (pedestrian, right, vehicle), (pedestrian, in hand, hat). The sending end omits the first relationship of some triples in the knowledge graph G, for example, the outside and the right, by comparing the first set (knowledge graph G) with the second set (shared probability graph GS), and obtains a third set including binary and triple groups. The third set can be in the form of: (tree, sidewalk), (tree, front, vehicle), (pedestrian, vehicle), (pedestrian, in hand, hat).

[0305] Among them, the probability of the second relationship can be obtained by Calculated. Represents the second relationship The set of labels of existing or known samples. For triples with the same second head entity and second tail entity, there may be multiple relationships. The sender counts the set of sample labels of the second relationship between the same second head entity and the second tail entity, and The probability or frequency of the second relationship can be obtained by dividing the number of existing or known samples by the total number of samples of the second relationship.

[0306] In an optional implementation, for each triple, if the sending end filters out a quadruple that matches the triple from the second set, then when the probability corresponding to the second relationship of the sending end in the quadruple meets the first condition, the first relationship in the triple is omitted to obtain a triple with the first relationship omitted.

[0307] In this implementation, for each triple in the first set, if a matching quadruple is selected from the second set, the sender further determines whether the probability corresponding to the second relationship in the quadruple satisfies the first condition. If the condition is satisfied, the sender may omit the first relationship in the triple. Optionally, the conditions for the sender to select a quadruple matching the triple from the second set include:

[0308] The first head entity in the triple is the same as the second head entity in the quad; and,

[0309] The first last entity in the triple is the same as the second last entity in the quad; and

[0310] The first relation in the triple is the same as the second relation in the quad.

[0311] Optionally, the first condition includes:

[0312] The probability corresponding to the second relationship in the quadruple is the maximum value among all probabilities, and all probabilities are the probabilities corresponding to all second relationships between the second head entity and the second tail entity in the quadruple.

[0313] Specifically, the sending end omits the maximum value in the probability of the second relationship so that when the receiving end only receives a binary group, that is, only receives the first head entity and the first tail entity, and does not receive the first relationship, it defaults to generating a triplet with the second relationship with the highest probability corresponding to the second head entity and the second tail entity that are the same as the first head entity and the first tail entity in the quadruple.

[0314] It is understandable that the reason why the sending end omits the maximum value of the probability of the second relationship instead of the minimum value is that there may be no relationship between some entities in the second set. At this time, the probability of the second relationship is zero. If the minimum value is used as the first condition, the unique second relationship cannot be screened out, and the receiving end can also restore the omitted first relationship based on this condition.

[0315] For example, in the first round, for any ε j , if there is δ in GS s , so that h j =h s And t j =t s , then search for δ s Is there a relationship in the relationship set Make If it exists, continue to judge its corresponding relationship probability Is it a relation probability set P s The largest one, P s It can be written as:

[0316] if is the relation probability set P s The largest of the three, then omit the triple ε during transmission j =(h j ,r j ,t j ) in r j , only the head entity and the tail entity are transmitted to reduce the amount of transmitted data. If there is no δ in GS s , so that h j =h s And t j =t s , or although there is δ in GS s , so that h j =h s And t j =t s , but δ s There is no relationship in the relationship set Make Then this triplet must be transmitted in its entirety.

[0317] After completing the first round of relation screening, the sender can use all triples with the first relation omitted as conditions to further screen triples with the first relation omitted from all triples with the first relation not omitted, and continue the second round of operations:

[0318] Optionally, the sending end obtains a fourth set and a fifth set, where the fourth set includes all triples in the first set with the first relation omitted, and the fifth set includes triples in the first set except the fourth set;

[0319] For the plurality of triples in the fifth set, the sender filters out the triples with the first relation omitted and the triples with the first relation not omitted from the plurality of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set;

[0320] The sending end obtains the third set according to the updated fourth set and fifth set.

[0321] Specifically, after the sending end filters out some triplets that omit the first relationship from the first set according to the first condition, it can use the two tuples that omit the first relationship that were screened out in the first round as the fourth set, and use the fourth set as a condition to further filter out triplets that can omit the first relationship from the fifth set to further reduce the amount of data transmission.

[0322] Optionally, if the sending end does not filter out a quadruple matching the triplet from the second set, the sending end sends the first set to the receiving end.

[0323] Exemplarily, after obtaining the above-mentioned set of four-tuples, the sending end may calculate the multi-dimensional conditional probability distribution in the presence of multiple information, using the following method.

[0324] In the absence of any known information, the triple The probability of existence or knowledge can be written as:

[0325] in, is a collection The total number of sample labels in .

[0326] If the triple is known exists or is known, then the triple The conditional probability of existence or being known can be written as:

[0327] in, is a collection With collection The intersection of is a collection To the collection The union of .

[0328] By analogy, it is not difficult to get the triples when it is known that N triples exist or are known. Multidimensional conditional probability exists or is known:

[0329] in represents the set of conditions under which N triples exist. Represents all eligible The intersection of the sample label sets.

[0330] Optionally, for each triple in the plurality of triples, when the triple matches any quadruple in the second set, the sender calculates a conditional probability corresponding to the quadruple under the condition that any n triplets in the fourth set exist, where n is a positive integer and an initial value of n is 1;

[0331] When the conditional probability satisfies the second condition, the sending end omits the first relation in the triplet to obtain the updated fourth set and fifth set;

[0332] When the third condition is met, the sending end adds 1 to n and returns to the step of calculating the conditional probability corresponding to the existence of any n triplets in the fourth set for each triple in the plurality of triplets when the triplet matches any quadruple in the second set.

[0333] Specifically, since the semantic relations of some triples have been omitted in the first round, these triples can be used as conditions for the search of conditional probabilities in the second round. The second round of comparison is similar to the first round. The set of triples that are not omitted in the first round and do not necessarily need to be fully transmitted is written as: G 1 ={ε1,ε2,…,ε a ,…,ε A}

[0334] Write the set of triples that are omitted in the first round as: 1 ={ε1,ε2,…,ε b ,…,ε B}

[0335] For G 1 Any ε in a , respectively in O 1 Any triple ε in b Calculate the conditional probability under known or existing conditions. Assume ε a Corresponding to the shared probability graph Then its conditional probability can be written as Determine whether it is in the following conditional probability set:

[0336] If it is the largest, its semantic relationship will be omitted during transmission. If it is not the largest, change the condition to O 1 Repeat the above operation until O 1 The triples in are used as conditions. This operation is recorded as the first round in the second round of comparison. After the above operation is completed, the number of triples that have not been omitted will become smaller, and the triples that have not been omitted will form a new G 1 , update the newly omitted triples to O 1 , continue to perform the above operation until no new omissible triples are added.

[0337] It's understandable that as the rounds progress, the number of triplets used as conditions increases accordingly. For example, in the third round, two triplets are used as conditions for calculating the conditional probability, in the fourth round, three triplets are used as conditions for calculating the conditional probability, and so on. The comparison process is essentially the same as in the previous rounds, so I won't elaborate on it here.

[0338] It's worth noting that deeper rounds of comparison require more computing resources and don't necessarily yield greater benefits. Therefore, the number of comparison rounds required can be determined as needed during communication. The termination conditions for the aforementioned loop rounds can be set based on actual needs and are not limited here. In one optional implementation, the sender can set a maximum value for the number of omitted information in the first set and terminate the loop when the number of binary pairs in the second set is greater than or equal to the maximum value for the number of omitted information in the first set.

[0339] In this implementation, the maximum value of the number of omitted information in the first set is referred to as the first value, and the first information includes information indicating the first value. The first information can be determined by the transmitting end and notified to the receiving end, or determined by the receiving end and notified to the transmitting end. If the transmitting end makes the decision, or uses the performance of the transmitting end as a reference object, the transmitting end can obtain the channel conditions and the device capabilities and time thresholds of itself or the receiving end, calculate the first value E, and inform the receiving end of E. If the receiving end makes the decision, or uses the performance of the receiving end as a reference object, the receiving end can calculate the first value E based on the channel conditions and the device capabilities and time thresholds of itself or the transmitting end, and then inform the transmitting end.

[0340] Optionally, the first value can be calculated based on device capabilities, time thresholds, or channel conditions.

[0341] The first value can be calculated by the transmitting end based on its own device capabilities, time thresholds, or channel conditions, or it can be calculated by the receiving end and then transmitted to the transmitting end. In actual applications, whether the transmitting end or the receiving end performs the calculation depends on actual needs. If the decision or performance of the transmitting end is the primary consideration, the first value can be directly calculated by the transmitting end; otherwise, the first value can be calculated by the receiving end.

[0342] By performing the above calculation process, the sender omits some information from the transmitted triplet, thereby reducing communication latency. However, this calculation process also increases transmission time and energy. To achieve better results and reduce the total system energy consumption during communication, it is necessary to jointly schedule communication and computing resources and allocate them appropriately. The following describes the specific process of this allocation.

[0343] The total delay of a transmission process generally consists of two parts: communication delay t1 and computation delay t2. Assuming that the delay of transmitting a piece of data information is limited to T, the total process must meet the delay constraint t1+t2≤T.

[0344] The communication delay t1 is calculated as follows. Assume that during the communication process, the channel path loss between the transmitter and the receiver is h and the noise power is σ 2 , the transmission bandwidth is B, and the base station transmission power is p. Then the channel capacity between the base station and the user can be written as:

[0345] Assuming that the first entity, relation, and last entity in a triple are all represented by the same number of bits R, the total number of bits of the data semantic information D can be written as: size(D) = R(3M-E)

[0346] Where M is the total number of triplets in D, and E is the number of triplets with omitted relations. Therefore, the communication delay can be written as:

[0347] The calculation method of the time delay t2 is as follows.

[0348] It is understandable that the computational delay t2 primarily arises from the process of comparing the first set with the second set. For ease of explanation, this embodiment only illustrates the first and second rounds of comparison. Because it is impossible to pre-determine the computational resources required to omit E relationships for a specific data set, this embodiment utilizes statistical methods to determine the ratio of the number of triples that can be omitted per comparison round to the total number of triples remaining. This is used to calculate the computational resources required to statistically omit E relationships.

[0349] The ratio of the number of triplets that can be omitted in each round of comparison to the total number of triplets before this round of comparison is written as:

[0350] Among them, q1 is the ratio of the number of triples that can be omitted in the first round of comparison to the total number of triples, q2 is the ratio of the number of triples that can be omitted in the first round of comparison in the second round to the total number of triples before this round of comparison, and q n It is the ratio of the number of triples that can be omitted in the (n-1)th round of comparison in the second round to the total number of triples before this round of comparison, and N is the total number of statistical rounds.

[0351] The number of semantic relations that can be omitted in each round of comparison can be written as a recursive formula:

[0352] Where M is the total number of triplets, E N is the number of triples in which relations can be omitted in the Nth round of comparison. Based on the above formula, the number of calculations required to omit E relations in M ​​triples can be written as:

[0353] It can be observed that c(E) is an increasing piecewise function, each of which is a linear function of E.

[0354] Therefore, the calculation delay can be written as:

[0355] Where τ1 is a constant coefficient and f is the frequency of the central processing unit (CPU). The value of τ1 is related to the time it takes to perform a comparison.

[0356] Furthermore, the energy consumed by communication can be written as: e1=t1·p

[0357] Where t1 is the communication time and p is the base station transmission power.

[0358] The energy consumed can be calculated as: e2=τ1τ2c(E)f 2

[0359] Where τ1 and τ2 are constant coefficients, and f is the CPU frequency.

[0360] At this point, the following joint optimization problem is constructed:

[0361] e1+e2,

[0362] st 0≤p≤p max , (2)

[0363] Among them, constraint (1) limits the total communication and computation delay, and constraint (2) limits the base station transmission power p to be non-negative and not more than p max ,Constraint (3) limits the total number of triples E with omitted semantic relations to a natural number, and must not exceed the total number of triples M. The goal of this optimization problem is to minimize the total energy consumption of the system.

[0364] The parameters to be optimized in this optimization problem are the base station transmit power p and the total number of triplets with omitted semantic relationships, E. Increasing the base station transmit power p reduces communication latency, but not necessarily reduces communication energy consumption. Increasing the total number of triplets with omitted semantic relationships, E, reduces communication latency but increases computational latency, and reduces communication energy consumption but increases computational energy. Therefore, the values ​​of base station transmit power p and the total number of triplets with omitted semantic relationships, E, must be carefully selected to minimize total system energy consumption while satisfying all constraints.

[0365] Assuming that the total number of triplets with omitted semantic relations E is fixed, we analyze the impact of the base station transmission power p on the overall situation. Simplifying constraint (1) and separating the base station transmission power p, we obtain:

[0366] Observe the right side of the formula, it is easy to know that When, that is:

[0367] When , the right side of the formula:

[0368] when:

[0369] When , we can combine constraints (1) and (2) to obtain:

[0370] Expand the objective function of the optimization problem and get:

[0371] When E is fixed, the objective function becomes a univariate function with respect to p. At this time, the objective function is denoted as f(p), and its monotonicity on the domain is analyzed.

[0372] When E is fixed, the second term of f(p) τ1τ2c(E)f 2 is a constant, so we only need to analyze the monotonicity of its first term. Multiply the first term of f(p) by the constant Denoted as g(p), we get:

[0373] It is easy to see that the monotonicity of g(p) is the same as that of f(p). Let it be x, and write the new function h(x) as:

[0374] It is easy to see that the monotonicity of the function f(p) on p>0 is the same as the monotonicity of the function h(x) on x>0.

[0375] When E is fixed, the objective function is:

[0376] It increases monotonically when p>0. When the total number of triples with omitted semantic relations E is fixed, the smaller the base station transmission power p, the better. When the constraints are met, the base station transmission power that minimizes the total system energy consumption output by the objective function is:

[0377] If constraint (1) cannot be satisfied, it means that the value of E is too large, and the calculation delay alone exceeds the limit of the total delay. In this case, this E is not desirable. Moreover, because c(E) is a monotonically increasing function, all values ​​larger than this E are also not desirable. Only a smaller value of E can be selected.

[0378] If the delay constraint (1) is satisfied but the power constraint (2) is not satisfied, it means that the maximum transmit power of the base station becomes a limitation. In this case, poptimiz ed = p max .

[0379] Therefore, for each fixed value E, we can determine the value of base station transmit power p that minimizes the total system energy consumption. Because the total number of triplets E that omit semantic relations is a natural number and has an upper bound, we can design an algorithm based on the traversal of E. The specific algorithm flow is shown below.

[0380] (1) Parameter initialization: p optimized For p max , E optimized is 0 and the minimum value of the objective function obj min is infinite or a sufficiently large number;

[0381] (2) Let E traverse from 0 to M:

[0382] (2.1)Judgment Is it satisfied? If not, skip the remaining steps and select the next E.

[0383] (2.2) If satisfied, then judge Whether it meets:

[0384] (2.2.1) If it is satisfied, then when the total number of triplets with omitted semantic relations E is fixed, the smaller the base station transmission power p is, the better.

[0385] Correspondingly, the base station transmission power that minimizes the total system energy consumption output by the objective function is:

[0386] (2.2.2) If not satisfied, let p be p max .

[0387] calculate

[0388] (2.3) Determine whether obj is less than the minimum value obj of the previous round of objective function min If it is less than, then let the minimum value of the objective function obj min Update to the calculated obj, p optimized is p, E optimized For E.

[0389] The time threshold mentioned in this embodiment can be understood as T, the channel condition refers to the channel information transmitted by the transmitter and the receiver, and the device capability refers to the computing and transmission capability of the transmitter or the receiver itself. The computing capability of the device can be understood as

[0390] Optionally, the transmitting end may determine second information, where the second information is used to instruct the transmitting end to use the transmission power to send the third set.

[0391] Optionally, the sending end may also receive second information from the receiving end.

[0392] The second information may include transmit power or an identifier or index information indicating transmit power. If the transmitting end makes the decision, or uses the performance of the transmitting end as a reference object, the transmitting end may obtain the channel conditions and the device capabilities and time thresholds of itself or the receiving end, and calculate the first value E and transmit power p. Wherein, E needs to be informed to the receiving end. If the receiving end makes the decision, or uses the performance of the receiving end as a reference object, the receiving end may calculate the first value E and transmit power p based on the channel conditions and the device capabilities and time thresholds of itself or the transmitting end, and then send or inform the transmitting end.

[0393] It can be understood that, in order to optimize the problem of minimizing the total energy consumption of the communication system, this embodiment jointly optimizes the resource allocation of communication and computing, effectively reducing the total energy overhead of the system under the condition that the total communication and computing delay limit remains unchanged.

[0394] Step 305: The sending end sends the third set to the receiving end.

[0395] In this embodiment, after the transmitting end determines the third set that omits the first relation, it sends the third set to the receiving end. Because the third set includes two-tuples that omit some of the first relations, the amount of data transmission is further reduced compared to directly sending a set containing all triples of the data to be transmitted.

[0396] For example, please refer to Figure 6, which is a flow chart of semantic information processing performed by the sending end provided in an embodiment of the present application. As shown in Figure 6, taking text data as an example, the sending end extracts semantic information from the received text data to obtain a knowledge graph. Based on the shared probability graph, a probability graph search is performed from the knowledge graph, that is, matching quadruplets are found, and the first relationship corresponding to the maximum value of the probability corresponding to the second relationship is omitted based on the condition, and finally a compressed knowledge graph is output.

[0397] Step 306: The receiving end receives the third set from the sending end.

[0398] Step 307: The receiving end obtains a second set, where the second set includes probabilities corresponding to the second relationship.

[0399] In this embodiment, the second set is a four-tuple shared by the receiving end and the sending end, and its main form is: (second head entity, second relation, probability corresponding to the second relation, second tail entity). The second set can be constructed by the receiving end or the sending end, and accordingly, the following two implementation methods will be generated.

[0400] In one implementation, the receiving end receives the second set sent from the sending end.

[0401] In another implementation, the receiving end constructs a second set and sends the second set to the sending end.

[0402] Optionally, the second set can be constructed based on sample data.

[0403] Before executing step 308, in an optional implementation, the receiving end may also obtain first information, wherein the first information includes device capabilities, time thresholds, or channel conditions; then, the receiving end obtains a first value based on the first information, and the first value includes the maximum value of the number of omitted information in the first set; finally, the receiving end sends the first value to the sending end.

[0404] In this implementation, the first information mainly includes the transmit power and the first value. The receiving end can obtain the maximum number of omitted information based on the first value.

[0405] Optionally, the first information obtained by the receiving end may be received from the sending end, or may be obtained by calculation by the receiving end according to device capabilities, time thresholds, or channel conditions.

[0406] Step 308: The receiving end determines a first set according to the probability corresponding to the second relationship.

[0407] In this embodiment, after receiving the third set, the receiving end can recover the omitted first relationship from the third set based on the probability corresponding to the second relationship in the second set in the same way as the transmitting end to recover the first set.

[0408] In one possible implementation, the receiving end determines the first set according to the probability corresponding to the second relationship, including:

[0409] For at least one binary group in the third set, the receiving end obtains, based on the probability of the second relationship, the first relationship in which each of the at least one binary group is omitted, to obtain at least one triple;

[0410] The receiving end obtains the first set according to all triples and at least one triple in the third set.

[0411] In a possible implementation, the receiving end obtains the first relation in which each of the at least one two-tuple is omitted according to the probability of the second relation, to obtain at least one triple, including:

[0412] For each binary group in at least one binary group, the receiving end filters out a quadruple that matches the binary group from the second set. When the probability corresponding to the second relationship in the quadruple meets the first condition, the receiving end adds the second relationship of the quadruple to the binary group to obtain a triple.

[0413] In one possible implementation, the conditions for the receiving end to filter out the quadruple matching the binary from the second set include:

[0414] The first head entity in the two-tuple is the same as the second head entity in the four-tuple, and the first tail entity in the two-tuple is the same as the second tail entity in the four-tuple.

[0415] In one possible implementation, the first condition includes:

[0416] The probability corresponding to the second relationship in the quadruple is the maximum value among all probabilities, and all probabilities are the probabilities corresponding to all second relationships between the second head entity and the second tail entity in the quadruple.

[0417] After the restored triples are obtained based on the above method, a second round of conditional probability search can be performed with the binary tuples in the third set that have restored the triples as conditions. Similar to the operation at the sending end, the conditional probability of any unrestored binary tuple under the condition that the binary tuple that has restored the triples exists is calculated with any binary tuple that has restored the triples as conditions. Then, when the conditional probability of the first unrestored binary tuple is the maximum value of all the conditional probabilities corresponding to all the unrestored binary tuples under the same conditions, the first relationship in the first unrestored binary tuple is restored to obtain the restored triples. After traversing all the unrestored binary tuples in the third set to obtain all the restored triplets that meet the above requirements, the first set is obtained based on all the restored binary tuples, at least one triple restored in the first round, and the triples that originally existed in the third set.

[0418] For specific operations, please refer to the specific content of calculating the conditional probability in the second round at the sending end, which will not be repeated here.

[0419] Optionally, the sending end and the receiving end may respectively omit and restore the first relationship according to the identifier.

[0420] The sending end can identify the triplets that can omit the first relation in each round of screening, that is, the binary tuples in the third set. After receiving the third set, the receiving end can restore the binary tuples obtained in each round according to the identification. For example, the receiving end first determines the binary tuples that the sending end omitted in the first round according to the identification, and then restores these binary tuples into triplets according to the first condition. Subsequently, the receiving end uses the binary tuples in the third set that have restored the triples as conditions to calculate the conditional probability of the binary tuples that the sending end omitted in the second round under the condition that the binary tuples that have restored the triples exist, so as to restore the binary tuples that omitted the first relation in the second round into triplets.

[0421] It's understandable that as the rounds progress, the number of bigrams used as conditions increases accordingly. For example, in the third round, two bigrams will be used as conditions for calculating the conditional probability, in the fourth round, three bigrams will be used as conditions for calculating the conditional probability, and so on. The comparison process is essentially the same as in the previous rounds, so I won't elaborate on it here.

[0422] It is worth noting that by sharing the same second set (probability map) with the sending end, the receiving end can restore the compressed semantic information based on the same method after the sending end compresses the semantic information, and obtain the complete first set, which not only reduces the amount of data transmission but also ensures the complete transmission of data.

[0423] Based on the embodiments corresponding to Figures 1a to 6, in order to better implement the above solutions of the embodiments of the present application, the following also provides related devices for implementing the above solutions. Please refer to Figure 7, which is a structural diagram of a communication device provided in the embodiments of the present application.

[0424] The communication device 7000 may be one of the terminal devices 120a-120j shown in FIG1a, or a component of a terminal device (e.g., a processor, chip, or chip system), or the network device 110a or 110b shown in FIG1a, or a component of a network device (e.g., a processor, chip, or chip system). The communication device 7000 may be used to execute the process executed by the transmitting end in the embodiments shown in FIG3 to FIG6. The communication device 7000 includes:

[0425] Processing module 7001 is configured to obtain a first set, where the first set includes at least one triple corresponding to data to be transmitted, each triple including a first head entity, a first tail entity, and a first relationship between the first head entity and the first tail entity;

[0426] Processing module 7001 is further configured to determine a third set based on a probability corresponding to a second relationship in the second set, where the second set includes at least one quadruple, each quadruple including a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and a probability corresponding to the second relationship; and the third set includes at least one bigram, where the bigram includes the first head entity and the first tail entity.

[0427] The transceiver module 7002 is configured to send a third set to a receiving end, where the third set is used by the receiving end to determine the first set.

[0428] In a possible implementation, the processing module 7001 is further configured to:

[0429] A second set is received from a receiving end.

[0430] In one possible implementation, the transceiver module 7002 is further configured to:

[0431] The second set is sent to the receiving end.

[0432] In a possible implementation, the processing module 7001 is further configured to determine first information, where the first information indicates a maximum number of omitted information in the first set;

[0433] The transceiver module 7002 is further configured to send the first information to a receiving end.

[0434] In one possible implementation, the transceiver module 7002 is further configured to:

[0435] First information is received from a receiving end, where the first information indicates a maximum number of omitted information in a first set.

[0436] In a possible implementation, the processing module 7001 is further configured to:

[0437] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0438] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0439] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0440] In a possible implementation, the processing module 7001 is further configured to:

[0441] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0442] For each triple, if a quadruple matching the triple is selected from the second set, then when the probability corresponding to the second relationship in the quadruple satisfies the first condition, the first relationship in the triple is omitted to obtain a triple with the first relationship omitted;

[0443] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0444] In a possible implementation, the processing module 7001 is further configured to:

[0445] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0446] For each triple, if the first head entity in the triple is the same as the second head entity in the quadruple; and the first tail entity in the triple is the same as the second tail entity in the quadruple; and the first relationship in the triple is the same as the second relationship in the quadruple, then when the probability corresponding to the second relationship in the quadruple meets the first condition, the first relationship in the triple is omitted to obtain the triple with the first relationship omitted;

[0447] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0448] In a possible implementation, the processing module 7001 is further configured to:

[0449] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0450] For each triple, if the first head entity in the triple is the same as the second head entity in the quadruple; and the first tail entity in the triple is the same as the second tail entity in the quadruple; and the first relationship in the triple is the same as the second relationship in the quadruple, then when the probability corresponding to the second relationship in the quadruple is the maximum among all probabilities, the first relationship in the triple is omitted, and the triple with the first relationship omitted is obtained, and all probabilities are the probabilities corresponding to all second relationships between the second head entity and the second tail entity in the quadruple;

[0451] A third set is obtained according to all triples whose first relation is omitted and all triples whose first relation is not omitted.

[0452] In a possible implementation, the processing module 7001 is further configured to:

[0453] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0454] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0455] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0456] For the plurality of triples in the fifth set, the triples with the first relation omitted and the triples with the first relation not omitted are screened out from the plurality of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set;

[0457] The third set is obtained according to the updated fourth set and fifth set.

[0458] In a possible implementation, the processing module 7001 is further configured to:

[0459] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0460] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0461] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0462] For the triples in the fifth set, for each triple in the triples, when the triple matches any quadruple in the second set, calculate the conditional probability of the quadruple under the condition that any n triples in the fourth set exist, where n is a positive integer and its initial value is 1;

[0463] When the conditional probability satisfies the second condition, the first relation in the triple is omitted to obtain the updated fourth set and fifth set;

[0464] When the third condition is satisfied, n is incremented by 1, and the process returns to the step of calculating, for each triple in the plurality of triples, a conditional probability of the quadruple under the condition that any n triples in the fourth set exist when the triple matches any quadruple in the second set;

[0465] The third set is obtained according to the updated fourth set and fifth set.

[0466] In one possible implementation, the second condition includes:

[0467] The conditional probability corresponding to the quadruple is the maximum value among all probabilities, and all probabilities are the conditional probabilities corresponding to all second relations between the second head entity and the second tail entity in the quadruple.

[0468] In this implementation, the conditional probability corresponding to the first relationship is the maximum probability among all conditional probabilities as an omission condition, so that the receiving end can recover the first relationship with the highest probability from the same second set according to this condition, thereby facilitating data recovery.

[0469] In one possible implementation, the third condition includes at least one of the following:

[0470] n is less than the first value;

[0471] The number of triplets in the fourth set is less than the second value.

[0472] In a possible implementation, the processing module 7001 is further configured to:

[0473] For at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation;

[0474] According to the first set and all triples in the first set whose first relation is omitted, all triples whose first relation is not omitted are obtained;

[0475] When the number of all omitted first relations in the first set is less than a second value, the execution continues, where the second value is the maximum number of omitted information in the first set, and the second value is obtained based on device capabilities, time thresholds, or channel conditions;

[0476] Obtain a fourth set and a fifth set, the fourth set including all triples in the first set with the first relation omitted, and the fifth set including the triples in the first set except the fourth set;

[0477] For the plurality of triples in the fifth set, the triples with the first relation omitted and the triples with the first relation not omitted are screened out from the plurality of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set;

[0478] The third set is obtained according to the updated fourth set and fifth set.

[0479] In a possible implementation, the processing module 7001 is further configured to:

[0480] Second information is determined, where the second information is used to indicate that the third set is sent using transmit power.

[0481] In one possible implementation, the transceiver module 7002 is further configured to:

[0482] Second information is received from the receiving end, where the second information indicates that the third set is sent using the transmission power.

[0483] In this embodiment, the operations performed by each unit in the communication device 7000 are similar to those described in the method embodiments shown in Figures 3 to 6 above, and can be used to implement the functions of the sending end in the above method embodiments, and can also achieve the beneficial effects of the above method embodiments, which will not be repeated here.

[0484] Please refer to FIG8 , which is another structural diagram of a communication device provided in an embodiment of the present application.

[0485] The communication device 8000 may be one of the terminal devices 120a-120j shown in FIG1a, or a component of a terminal device (e.g., a processor, chip, or chip system), or the network device 110a or 110b shown in FIG1a, or a component of a network device (e.g., a processor, chip, or chip system). The communication device 8000 may be used to execute the process executed by the receiving end in the embodiments shown in FIG3 to FIG6. The communication device 8000 includes:

[0486] The transceiver module 8001 is configured to receive a third set from a transmitting end, where the third set includes at least one two-tuple, and the two-tuple includes a first header entity and a first tail entity;

[0487] Processing module 8002 is used to determine the first set based on the probability corresponding to the second relationship in the second set, where the second set includes at least one quadruple, each quadruple includes a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and the probability corresponding to the second relationship.

[0488] In a possible implementation, the processing module 8002 is further configured to:

[0489] A second set is received from the sending end.

[0490] In one possible implementation, the transceiver module 8001 is further configured to:

[0491] The second set is sent to the sending end.

[0492] In a possible implementation, the processing module 8002 is further configured to determine first information, where the first information indicates a maximum number of omitted information in the first set;

[0493] The transceiver module 8001 is further configured to send the first information to the receiving end.

[0494] In one possible implementation, the transceiver module 8001 is further configured to:

[0495] First information is received from a receiving end, where the first information indicates a maximum number of omitted information in a first set.

[0496] In a possible implementation, the processing module 8002 is further configured to:

[0497] Acquiring first information, where the first information includes device capabilities, time thresholds, or channel conditions;

[0498] Obtaining a first value according to the first information, where the first value includes a maximum value of the number of omitted information in the first set;

[0499] Send the first value.

[0500] In a possible implementation, the processing module 8002 is further configured to:

[0501] First information is received from a sending end.

[0502] In one possible implementation, determining the first set according to the probability corresponding to the second relationship includes:

[0503] For at least one bigram in the third set, obtaining, according to the probability of the second relation, a first relation in which each of the at least one bigram is omitted, to obtain at least one triple;

[0504] The first set is obtained according to all triples and at least one triple in the third set.

[0505] In a possible implementation, the processing module 8002 is further configured to:

[0506] For each binary group in at least one binary group, if a quadruple matching the binary group is selected from the second set, when the probability corresponding to the second relationship in the quadruple group satisfies the first condition, the second relationship of the quadruple group is added to the binary group to obtain a triple.

[0507] In one possible implementation, the conditions for selecting a quadruple matching the binary from the second set include:

[0508] The first head entity in the two-tuple is the same as the second head entity in the four-tuple, and the first tail entity in the two-tuple is the same as the second tail entity in the four-tuple.

[0509] In one possible implementation, the first condition includes:

[0510] The probability corresponding to the second relationship in the quadruple is the maximum value among all probabilities, and all probabilities are the probabilities corresponding to all second relationships between the second head entity and the second tail entity in the quadruple.

[0511] In this embodiment, the operations performed by each unit in the communication device 8000 are similar to those described in the method embodiments shown in Figures 3 to 6 above, and can be used to implement the functions of the computing device in the above method embodiments, and can also achieve the beneficial effects of the above method embodiments, which will not be repeated here.

[0512] Please refer to Figure 9, which is a schematic diagram of another structure of a communication device provided in an embodiment of the present application. Communication device 9000 can be a network device, a terminal device, or a chip, chip system, processor, etc. that supports the terminal device or network device to implement the above-mentioned method. This device can be used to implement the method described in Figures 3 to 6 of the embodiments of the present application. For details, please refer to the description of the above-mentioned method embodiment.

[0513] The communication device 9000 includes one or more processors 901 and an interface circuit 902. The processor 901 and the interface circuit 902 are coupled to each other. It is understood that the interface circuit 902 can be a transceiver or an input / output interface.

[0514] Optionally, the communication device 9000 may include one or more memories 903 for storing instructions executed by the processor 901, or storing input data required by the processor 901 to execute instructions, or storing data generated after the processor 901 executes instructions. The memory 903 may be provided inside the communication device 9000 or outside the communication device 9000.

[0515] In another optional design, processor 901 may include a transceiver unit for implementing receiving and transmitting functions. For example, the transceiver unit may be a transceiver circuit, an interface, or an interface circuit. The transceiver circuit, interface, or interface circuit for implementing the receiving and transmitting functions may be separate or integrated. The transceiver circuit, interface, or interface circuit may be used for reading and writing code / data, or the transceiver circuit, interface, or interface circuit may be used for transmitting or delivering signals.

[0516] In another possible design, the communication device 9000 may include a circuit that can implement the functions of sending, receiving, or communicating in the aforementioned method embodiments.

[0517] When the communication device 9000 is a chip used in a terminal device, the terminal device chip implements the functions of the terminal device in the above method embodiments. The terminal device chip receives information from other modules in the terminal device (such as a radio frequency module or antenna), and the information is sent by the network device to the terminal device; or the terminal device chip sends information to other modules in the terminal device (such as a radio frequency module or antenna), and the information is sent by the terminal device to the network device.

[0518] When the above-mentioned communication device 9000 is a module applied to a network device, the network device module of the network device implements the functions of the network device in the above-mentioned method embodiment. The network device module receives information from other modules in the network device (such as a radio frequency module or an antenna), and the information is sent by the terminal device to the network device; or the network device module sends information to other modules in the network device (such as a radio frequency module or an antenna), and the information is sent by the network device to the terminal device. The network device module here can be a baseband chip of the network device, or it can be a DU or other module. The DU here can be a DU under the open radio access network (O-RAN) architecture.

[0519] It is understood that the processor 901 in the embodiments of the present application may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0520] The method steps in the embodiments of the present application can be implemented in hardware or in software instructions that can be executed by a processor. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disks, mobile hard disks, compact disc read-only memory (CD-ROM) or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. The storage medium can also be an integral part of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a network device or a terminal device. The processor and the storage medium can also be present in a network device or a terminal device as discrete components.

[0521] The device described in the above embodiments may be a network device or a terminal device, but the scope of the device described in this application is not limited thereto, and the structure of the device may not be limited to FIG9 . The device may be an independent device or may be part of a larger device. For example, the device may be:

[0522] (1) An independent integrated circuit (IC), chip, or chip system or subsystem;

[0523] (2) having a set of one or more ICs, optionally including a storage component for storing data and / or instructions;

[0524] (3) ASIC, such as modem (MSM);

[0525] (4) Modules that can be embedded in other devices;

[0526] (5) Receivers, terminal devices, intelligent terminal devices, cellular phones, wireless devices, handheld devices, mobile units, vehicle-mounted devices, network devices, cloud devices, artificial intelligence devices, machine devices, home devices, medical devices, industrial equipment, etc.;

[0527] (6)Others, etc.

[0528] Please refer to Figure 10, which is a schematic diagram of another structure of a communication device provided in an embodiment of the present application. Communication device 1000 can be a network device, a terminal device, or a chip, chip system, processor, etc. that supports a terminal device or network device to implement the above-mentioned method. This device can be used to implement the method described in Figures 3 to 6 of the embodiments of the present application. For details, please refer to the description of the above-mentioned method embodiment.

[0529] As shown in Figure 10, it can be understood that the communication device 1000 includes necessary forms of means such as modules, units, elements, circuits, or interfaces, which are appropriately configured together to implement the present solution. The communication device 1000 can be a radio access network (RAN) node, terminal device, core network device or other network device in Figure 1a, or a component (such as a chip) in these devices, used to implement the method described in the following method embodiment. The communication device 1000 includes one or more processors 1010. The processor 1010 can be a general-purpose processor or a dedicated processor. For example, it can be a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control the communication device (such as a RAN node, terminal device, or chip, etc.), execute software programs, and process data of software programs.

[0530] Optionally, in one design, the processor 1010 may include a program 1030 (sometimes also referred to as code or instructions), which may be executed on the processor 1010 to cause the communication device 1000 to perform the methods described in the following embodiments. In another possible design, the communication device 1000 includes a circuit (not shown in FIG10 ) configured to implement the functions described in FIG3 to FIG6 in the embodiments.

[0531] Optionally, the communication device 1000 may include one or more memories 1020 on which a program 1040 (sometimes also referred to as code or instructions) is stored. The program 1040 can be run on the processor 1010, so that the communication device 1000 performs the method described in the method embodiments described in Figures 3 to 6.

[0532] Optionally, the processor 1010 and / or the memory 1020 may include an artificial intelligence (AI) module 1070, 1080, which is used to implement AI-related functions. The AI ​​module can be implemented through software, hardware, or a combination of software and hardware. For example, the AI ​​module may include a radio access network intelligent controller (RIC) module. For example, the AI ​​module may be a near real-time RIC or a non-real-time RIC.

[0533] Optionally, data may be stored in the processor 1010 and / or the memory 1020. The processor and memory may be provided separately or integrated together.

[0534] Optionally, the communication device 1000 may further include a transceiver 1050 and / or an antenna 1060. The processor 1010 may also be sometimes referred to as a processing unit, and controls the communication device (e.g., a RAN node or a terminal device). The transceiver 1050 may also be sometimes referred to as a transceiver unit, a transceiver, a transceiver circuit, or a transceiver, and is configured to implement the transceiver function of the communication device through the antenna 1060.

[0535] When the communication device in the embodiment of the present application is a network device, the communication device can be as shown in Figure 11, which is another structural schematic diagram of the communication device provided in the embodiment of the present application. The communication device may include one or more radio frequency units, such as a remote radio unit (RRU) 1110 and one or more baseband units (BBU) (also known as digital units, DU) 1120. The RRU 1110 can be called a transceiver module, which may include a transmitting module and a receiving module, or the transceiver module can be a module that can realize the sending and receiving functions. The transceiver module can correspond to the transceiver module 7002 in Figure 7, that is, the transceiver module can perform the actions performed by the transceiver module 7002. Optionally, the transceiver module can also be called a transceiver, a transceiver circuit, or a transceiver, etc., and may include at least one antenna 1111 and a radio frequency unit 1112. The RRU 1110 part is mainly used for transmitting and receiving radio frequency signals and converting radio frequency signals into baseband signals. The BBU 1120 is mainly used for baseband processing, base station control, etc. The RRU 1110 and the BBU 1120 may be physically arranged together or physically separated, that is, a distributed base station.

[0536] The BBU 1120 is the control center of the base station, also known as a processing module, and is primarily responsible for performing baseband processing functions such as channel coding, multiplexing, modulation, and spread spectrum. Furthermore, the processing module can control the actions performed by the processing module 7002 in FIG. 7 . For example, the BBU (processing module) can be used to control the base station to execute the network device operation procedures described in the aforementioned method embodiments.

[0537] In one example, the BBU 1120 can be composed of one or more single boards, and multiple single boards can jointly support a wireless access network with a single access standard (such as a long term evolution (LTE) network), or can separately support wireless access networks with different access standards (such as an LTE network, a 5G network, or other networks). The BBU 1120 also includes a memory 1121 and a processor 1122. The memory 1121 is used to store necessary instructions and data. The processor 1122 is used to control the base station to perform necessary actions, such as controlling the base station to execute the operation process of the network device in the above method embodiment. The memory 1121 and the processor 1122 can serve one or more single boards. That is, a memory and a processor can be set separately on each single board. Alternatively, multiple single boards can share the same memory and processor. In addition, necessary circuits can also be set on each single board.

[0538] An embodiment of the present application further provides a computer-readable medium on which a computer program is stored. When the computer program is executed by a computer, the functions of any of the above method embodiments are realized.

[0539] The embodiments of the present application also provide a computer program product, which implements the functions of any of the above method embodiments when executed by a computer.

[0540] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions according to the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated therein. Available media may be magnetic media (eg, floppy disks, hard disks, tapes), optical media (eg, high-density digital video discs (DVDs)), or semiconductor media (eg, solid state disks (SSDs)).

[0541] It is understood that the systems, devices, and methods described in this application may also be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not performed. In addition, the coupling or direct coupling or communication connection shown or discussed may be through some interface, indirect coupling or communication connection of devices or units, and may be electrical, mechanical, or other forms.

[0542] The above is only a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the scope of protection of the present application. The various numerical numbers involved in the embodiments of the present application are only for the convenience of description and are not used to limit the scope of the embodiments of the present application. The size of the sequence number of each process does not mean the order of execution. The order of execution of each process should be determined by its function and internal logic.

Claims

1. A semantic communication method, characterized in that: include: Determine a first set, the first set including at least one triple, each triple including a first head entity, a first tail entity, and a first relationship between the first head entity and the first tail entity; Determine a third set according to the probability corresponding to the second relationship in the second set, the second set includes at least one quadruple, each quadruple includes a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and the probability corresponding to the second relationship, the third set includes at least one bigram, the bigram includes the first head entity and the first tail entity; The third set is sent to a receiving end, where the third set is used by the receiving end to determine the first set.

2. The method according to claim 1, characterized in that Also includes: The second set is received.

3. The method according to claim 1, characterized in that Also includes: The second set is sent to the receiving end.

4. The method according to any one of claims 1 to 3, characterized in that Also includes: Determining first information, where the first information indicates a maximum value of the number of omitted information in the first set; Sending the first information to the receiving end.

5. The method according to any one of claims 1 to 3, characterized in that Also includes: First information is received from the receiving end, where the first information indicates a maximum value of the number of omitted information in the first set.

6. The method according to any one of claims 1 to 5, characterized in that Determining the third set according to the probability corresponding to the second relationship in the second set includes: For the at least one triple, all triples in which the first relation is omitted are filtered out from the at least one triple according to the probability corresponding to the second relation; Obtain all triples in which the first relation is not omitted according to the first set and all triples in which the first relation is omitted in the first set; The third set is obtained according to all triples in which the first relation is omitted and all triples in which the first relation is not omitted.

7. The method according to claim 6, characterized in that According to all triples in which the first relation is omitted and all triples in which the first relation is not omitted, the third set is obtained, including: Acquire a fourth set and a fifth set, wherein the fourth set includes all triples in the first set in which the first relation is omitted, and the fifth set includes triples in the first set except the fourth set; For a number of triples in the fifth set, filtering out triples in which the first relation is omitted and triples in which the first relation is not omitted from the number of triples based on the fourth set as a condition, so as to update the fourth set and the fifth set; The third set is obtained according to the updated fourth set and the fifth set.

8. The method according to any one of claims 1 to 7, characterized in that Also includes: Determine second information, where the second information is used to indicate the use of transmit power to send the third set.

9. The method according to any one of claims 1 to 7, characterized in that Also includes: Second information is received from a receiving end, where the second information indicates that the third set is sent with a transmit power.

10. A semantic communication method, characterized in that: include: receiving a third set from a transmitting end, the third set including at least one two-tuple, the two-tuple including the first header entity and the first tail entity; The first set is determined according to the probability corresponding to the second relationship in the second set, the second set includes at least one quadruple, each quadruple includes a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and the probability corresponding to the second relationship, the first set includes at least one triple, each triple includes a first head entity, a first tail entity, and a first relationship between the first head entity and the first tail entity.

11. The method according to claim 10, characterized in that Also includes: The second set is received.

12. The method according to claim 10, characterized in that Also includes: The second set is sent to a sending end.

13. The method according to any one of claims 10 to 12, characterized in that Also includes: Determining first information, where the first information indicates a maximum value of the number of omitted information in the first set; The first information is sent.

14. The method according to any one of claims 10 to 12, characterized in that Also includes: The first information is received from a transmitting end, where the first information indicates a maximum value of the number of omitted information in the first set.

15. The method according to any one of claims 10 to 14, characterized in that Determining the first set according to the probability corresponding to the second relationship in the second set includes: For at least one bigram in the third set, obtaining first relations in which each of the at least one bigram is omitted according to the probability of the second relation, to obtain at least one triple; The first set is obtained according to all triples in the third set and the at least one triple.

16. A communication device, characterized in that: include: a processing module, configured to determine a first set, wherein the first set includes at least one triple corresponding to the data to be transmitted, each triple including a first header entity, a first tail entity, and a first relationship between the first header entity and the first tail entity; The processing module is further configured to determine a third set according to the probability corresponding to the second relationship in the second set, the second set comprising at least one quadruple, each quadruple comprising a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and the probability corresponding to the second relationship, the third set comprising at least one bigram, the bigram comprising the first head entity and the first tail entity; The transceiver module is used to send the third set to the receiving end, and the third set is used by the receiving end to determine the first set.

17. A communication device, characterized in that: include: A transceiver module, configured to receive a third set from a transmitting end, wherein the third set includes at least one two-tuple, and the two-tuple includes the first head entity and the first tail entity; A processing module, used to determine a first set according to the probability corresponding to the second relationship in the second set, the second set includes at least one quadruple, each quadruple includes a second head entity, a second tail entity, a second relationship between the second head entity and the second tail entity, and the probability corresponding to the second relationship, the first set includes at least one triple, each triple includes a first head entity, a first tail entity, and a first relationship between the first head entity and the first tail entity.

18. A communication device, characterized in that: The invention comprises a processor and an interface circuit, wherein the interface circuit is used to receive signals from other communication devices and transmit them to the processor or send signals from the processor to other communication devices, and the processor is used to implement the method as claimed in any one of claims 1 to 9 or the method as claimed in any one of claims 10 to 15 through a logic circuit or by executing code instructions.

19. A communication system, comprising at least one communication device, wherein the communication device executes the method according to any one of claims 1 to 9, or executes the method according to any one of claims 10 to 15.

20. A computer-readable storage medium, characterized in that: The storage medium stores a computer program or an instruction. When the computer program or the instruction is executed by the communication device, the method according to any one of claims 1 to 9 or the method according to any one of claims 10 to 15 is implemented.

21. A communication device, characterized in that: The device comprises a processor, wherein the processor is used to execute a computer program or instruction in a memory to implement the method according to any one of claims 1 to 9, or to implement the method according to any one of claims 10 to 15.

Images