Structured information source semantic transmission method and system for power wireless communication and storage medium

Abstract

The invention discloses a structured information source semantic transmission method for power wireless communication. The method comprises the following steps: acquiring parameter information of a target power wireless communication network; acquiring data information and channel information of a structured information source in a target power wireless communication network in real time during communication; training a semantic coding and decoding network according to an information bottleneck theory; carrying out semantic coding by adopting the obtained semantic coding network and carrying out adaptive semantic feature transmission; and after a receiving end receives the data information, performing corresponding decoding by adopting the obtained semantic decoding network to obtain the data information of the structured information source so as to complete structured information source semantic transmission for power wireless communication. The invention also discloses a system for realizing the structured information source semantic transmission method for power wireless communication, and a storage medium comprising the structured information source semantic transmission method for power wireless communication. According to the method, the structured information source semantic transmission of the target power wireless communication is realized, the reliability is higher, the accuracy is better, and the overall efficiency is higher.

Description

Technical Field

[0001] The present invention belongs to the technical field of power communication, and in particular relates to a structured source semantic transmission method, system and storage medium for power wireless communication. Background Art

[0002] With the development of economy and technology and the improvement of people's living standards, electricity has become an indispensable secondary energy source in people's production and life, bringing endless convenience to people's production and life. Therefore, ensuring a stable and reliable supply of electricity has become one of the most important tasks of the power system.

[0003] With the rapid development of new power systems, more and more intelligent terminal devices are being connected to power wireless communication networks. The amount of structured signal transmission in power wireless communication is also increasing rapidly. Therefore, ensuring the reliability and accuracy of data transmission in power wireless communication networks is particularly important.

[0004] Currently, traditional power wireless communication primarily focuses on syntax, typically achieving reliable and accurate data transmission through improved data encoding and decoding technologies. However, with the continuous advancement of encoding and decoding technologies and the massive influx of intelligent terminal devices in power systems, traditional power wireless communication solutions have become overly redundant, causing the communication system to approach the Shannon limit. This makes it impossible to guarantee the stable and reliable operation of power wireless communication networks, and the reliability and accuracy of data transmission processes are gradually deteriorating. Summary of the Invention

[0005] One of the objectives of the present invention is to provide a structured source semantic transmission method for power wireless communication with high reliability, good accuracy and high efficiency.

[0006] A second object of the present invention is to provide a system for implementing the structured source semantic transmission method for power wireless communication.

[0007] A third object of the present invention is to provide a storage medium having a computer program stored thereon; when the computer program is executed by a processor, the structured source semantic transmission method for power wireless communication is implemented.

[0008] The structured source semantic transmission method for power wireless communication provided by the present invention comprises the following steps:

[0009] S1 obtains parameter information of the target power wireless communication network;

[0010] S2. During communication, real-time acquisition of data information and channel information of the structured source in the target power wireless communication network;

[0011] S3. Based on the mutual information between the semantic features and the data information obtained in step S2, according to the information bottleneck theory, the semantic encoding and decoding network is trained;

[0012] S4. Based on the real-time signal-to-interference-and-noise ratio of the target power wireless communication network, the semantic coding network obtained in step S3 is used for semantic coding and adaptive semantic feature transmission;

[0013] S5. After receiving the data information, the receiving end uses the semantic decoding network obtained in step S3 to perform corresponding decoding to obtain the data information of the structured source, so as to complete the semantic transmission of the structured source for power wireless communication.

[0014] During the communication described in step S2, data information and channel information of the structured signal source in the target power wireless communication network are acquired in real time, specifically including the following steps:

[0015] Assume that the total number of data X of the structured information source of communication node i is N, and the data X of the structured information source obtained is in is the Nth set of power business data X in the region, For collection The nth original data in for The mth original data in for The kth original data in ;

[0016] The signal-to-interference-and-noise ratio (SINR) of the channel of communication node i at the current moment is calculated using the following formula: i :

[0017]

[0018] Where S is the power of the useful signal, I is the power of the channel interference signal, and NN is the noise power.

[0019] The mutual information between the semantic features and the data information obtained in step S2 described in step S3 is used to train the semantic encoding and decoding network according to the information bottleneck theory, which specifically includes the following steps:

[0020] Select a semantic encoding and decoding network (such as the JSCC network);

[0021] Set the target output semantic information to in for The Lth semantic information data in , j = 1, 2, ..., N; the semantic features to be extracted are in T li The cth semantic feature in l = 1, 2, ..., C, where C is the feature dimension;

[0022] The following formula is used as the loss function during training:

[0023]

[0024] In the formula is the loss function value; I(T;X) is the mutual information between the semantic feature T and the structured source X, and p(t,x) is the joint probability distribution of T and X; p(t) is the marginal probability distribution of T, p(x) is the marginal probability distribution of X, t is the semantic feature data, x is the structured source data; I(T;Y) is the mutual information between the semantic feature T and the target output semantic information Y, and p(y) is the marginal probability distribution of Y, y is the target output semantic information data; β is the first parameter set to balance compression and semantic information retention; λ is the weight of the entropy regularization term; H(T) is the regularization term, and P(t) is the marginal probability distribution of T;

[0025] During training, the SGD optimizer is used to optimize the model.

[0026] The step S4, based on the real-time signal-to-interference-and-noise ratio of the target power wireless communication network, uses the semantic coding network obtained in step S3 to perform semantic coding and adaptive semantic feature transmission, specifically includes the following steps:

[0027] Using the semantic encoding network obtained in step S3 to perform semantic encoding;

[0028] Based on the real-time signal-to-interference-and-noise ratio of the target power wireless communication network, the feature dimension C is dynamically adjusted using the following rules:

[0029] Set the low SINR threshold low SINR low =α(dB), high signal-to-interference-and-noise ratio threshold SINR high SINR high =β(dB), semantic feature dimension C under low signal-to-interference-noise ratio low and semantic feature dimension C under high signal-to-interference-noise ratio high ;

[0030] If SINR i <SINR low , then set C = C low ;

[0031] If SINR i >SINR high , then set C = Chigh ;

[0032] If SINR low ≤SINR i ≤SINR high , then set

[0033]

[0034] After the receiving end receives the data information in step S5, the semantic decoding network obtained in step S3 is used to perform corresponding decoding to obtain the data information of the structured source, which specifically includes the following steps:

[0035] After the receiving end receives the data information, the corresponding semantic decoding network performs corresponding decoding according to the size of the feature dimension C to obtain the data information of the structured source.

[0036] The present invention also provides a system for implementing the structured source semantic transmission method for electric power wireless communication, comprising a parameter acquisition module, a data acquisition module, a network training module, a feature transmission module and a semantic transmission module; the parameter acquisition module, the data acquisition module, the network training module, the feature transmission module and the semantic transmission module are connected in series in sequence; the parameter acquisition module is used to obtain parameter information of the target electric power wireless communication network and upload the data information to the data acquisition module; the data acquisition module is used to obtain data information and channel information of the structured source in the target electric power wireless communication network in real time during communication based on the received data information, and upload the data information to the network training module; the network training module is used to obtain the data information and channel information of the structured source in the target electric power wireless communication network in real time based on the received data information The semantic encoding and decoding network is trained based on the mutual information between the semantic features and the acquired data information according to the information bottleneck theory, and the data information is uploaded to the feature transmission module; the feature transmission module is used to perform semantic encoding based on the received data information and the real-time signal-to-interference-noise ratio of the target power wireless communication network, and perform adaptive semantic feature transmission, and upload the data information to the semantic transmission module; the semantic transmission module is used to perform corresponding decoding based on the received data information and the data information received by the receiving end using the obtained semantic decoding network to obtain the data information of the structured source, so as to complete the semantic transmission of the structured source for power wireless communication.

[0037] The present invention also provides a storage medium on which a computer program is stored; when the computer program is executed by a processor, the structured source semantic transmission method for power wireless communication is implemented.

[0038] The structured source semantic transmission method, system and storage medium for power wireless communication provided by the present invention not only realize the structured source semantic transmission of the target power wireless communication by acquiring the parameter information and data information of the target power wireless communication network, as well as training the semantic encoding and decoding network and adaptive semantic feature transmission, but also have higher reliability, better accuracy and higher overall efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of the process flow of the present invention.

[0040] Figure 2 Schematic diagram of the functional modules of the system of the present invention. DETAILED DESCRIPTION

[0041] like Figure 1 The method flow diagram of the present invention is shown as follows: The structured source semantic transmission method for power wireless communication disclosed in the present invention includes the following steps:

[0042] S1 obtains parameter information of the target power wireless communication network;

[0043] S2. During communication, real-time acquisition of data information and channel information of the structured signal source in the target power wireless communication network; specifically comprising the following steps:

[0044] Assume that the total number of data X of the structured information source of communication node i is N, and the data X of the structured information source obtained is in is the Nth set of power business data X in the region, For collection The nth original data in for The mth original data in for The kth original data in ;

[0045] The signal-to-interference-and-noise ratio (SINR) of the channel of communication node i at the current moment is calculated using the following formula: i :

[0046]

[0047] Where S is the power of the useful signal, I is the power of the channel interference signal, and NN is the noise power.

[0048] The mutual information between the semantic features and the data information obtained in step S2 described in step S3 is used to train the semantic encoding and decoding network according to the information bottleneck theory, which specifically includes the following steps:

[0049] Select a semantic encoding and decoding network (such as the JSCC network);

[0050] Set the target output semantic information to in for The Lth semantic information data in , j = 1, 2, ..., N; the semantic features to be extracted are in T l i The cth semantic feature in l = 1, 2, ..., C, where C is the feature dimension;

[0051] The following formula is used as the loss function during training:

[0052]

[0053] In the formula is the loss function value; I(T;X) is the mutual information between the semantic feature T and the structured source X, and p(t,x) is the joint probability distribution of T and X; p(t) is the marginal probability distribution of T, p(x) is the marginal probability distribution of X, t is the semantic feature data, x is the structured source data; I(T;Y) is the mutual information between the semantic feature T and the target output semantic information Y, and p(y) is the marginal probability distribution of Y, y is the target output semantic information data; β is the first parameter set to balance compression and semantic information retention; λ is the weight of the entropy regularization term; H(T) is the regularization term, and P(t) is the marginal probability distribution of T;

[0054] During training, the SGD optimizer is used to optimize the model.

[0055] The step S4, based on the real-time signal-to-interference-and-noise ratio of the target power wireless communication network, uses the semantic coding network obtained in step S3 to perform semantic coding and adaptive semantic feature transmission, specifically includes the following steps:

[0056] Using the semantic encoding network obtained in step S3 to perform semantic encoding;

[0057] Based on the real-time signal-to-interference-and-noise ratio of the target power wireless communication network, the feature dimension C is dynamically adjusted using the following rules:

[0058] Set the low SINR threshold low SINR low =α(dB), high signal-to-interference-and-noise ratio threshold SINR high SINR high =β(dB), semantic feature dimension C under low signal-to-interference-noise ratio lowand semantic feature dimension C under high signal-to-interference-noise ratio high ;

[0059] If SINR i <SINR low , then set C = C low ;

[0060] If SINR i >SINR high , then set C = C high ;

[0061] If SINR low ≤SINR i ≤SINR high , then set

[0062]

[0063] After the receiving end receives the data information in step S5, the semantic decoding network obtained in step S3 is used to perform corresponding decoding to obtain the data information of the structured source, which specifically includes the following steps:

[0064] After the receiving end receives the data information, the corresponding semantic decoding network performs corresponding decoding according to the size of the feature dimension C to obtain the data information of the structured source.

[0065] The method of the present invention takes the transmission of semantic information as the core, obtains high-frequency collected data and real-time channel information from the electric power structured information source, trains the network to extract semantic features by minimizing mutual information loss based on the information bottleneck theory, and transmits adaptive semantic features based on the real-time signal-to-interference-noise ratio. The receiving end recovers the high-frequency collected data from the electric power structured information source and only transmits semantic information that is valuable to the receiver, effectively solving the redundant transmission of similar content and reducing the data transmission volume to ensure that the structured information source of the electric power high-frequency collected data can be transmitted safely and stably.

[0066] like Figure 2The figure shows a schematic diagram of the functional modules of the system of the present invention: the system disclosed in the present invention for realizing the structured source semantic transmission method for electric power wireless communication comprises a parameter acquisition module, a data acquisition module, a network training module, a feature transmission module and a semantic transmission module; the parameter acquisition module, the data acquisition module, the network training module, the feature transmission module and the semantic transmission module are connected in series in sequence; the parameter acquisition module is used to obtain the parameter information of the target electric power wireless communication network and upload the data information to the data acquisition module; the data acquisition module is used to obtain the data information and channel information of the structured source in the target electric power wireless communication network in real time during communication based on the received data information, and upload the data information to the network training module; the network training module The module is used to train the semantic encoding and decoding network according to the received data information, based on the mutual information between the semantic features and the acquired data information, according to the information bottleneck theory, and upload the data information to the feature transmission module; the feature transmission module is used to perform semantic encoding based on the received data information, based on the real-time signal-to-interference-noise ratio of the target power wireless communication network, using the obtained semantic encoding network, and perform adaptive semantic feature transmission, and upload the data information to the semantic transmission module; the semantic transmission module is used to perform corresponding decoding based on the received data information, after the receiving end receives the data information, using the obtained semantic decoding network to obtain the data information of the structured source, so as to complete the structured source semantic transmission for power wireless communication.

[0067] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiment of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal translation scripting language JavaScript, etc.

[0068] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0069] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0070] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0071] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0072] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A structured source semantic transmission method for power wireless communication, comprising the following steps: S1 obtains parameter information of the target power wireless communication network; S2. During communication, real-time acquisition of data information and channel information of the structured source in the target power wireless communication network; S3. Based on the mutual information between the semantic features and the data information obtained in step S2, according to the information bottleneck theory, the semantic encoding and decoding network is trained; S4. Based on the real-time signal-to-interference-and-noise ratio of the target power wireless communication network, the semantic coding network obtained in step S3 is used for semantic coding and adaptive semantic feature transmission; S5. After receiving the data information, the receiving end uses the semantic decoding network obtained in step S3 to perform corresponding decoding to obtain the data information of the structured source, so as to complete the semantic transmission of the structured source for power wireless communication.

2. The structured source semantic transmission method for power wireless communication according to claim 1 is characterized in that During the communication described in step S2, data information and channel information of the structured signal source in the target power wireless communication network are acquired in real time, specifically including the following steps: Assume that the total number of data X of the structured information source of communication node i is N, and the data X of the structured information source obtained is in is the Nth set of power business data X in the region, For collection The nth original data in for The mth original data in for The kth original data in ; The signal-to-interference-and-noise ratio (SINR) of the channel of communication node i at the current moment is calculated using the following formula: i : Where S is the power of the useful signal, I is the power of the channel interference signal, and NN is the noise power.

3. The structured source semantic transmission method for power wireless communication according to claim 2 is characterized in that The mutual information between the semantic features and the data information obtained in step S2 described in step S3 is used to train the semantic encoding and decoding network according to the information bottleneck theory, which specifically includes the following steps: Select the semantic encoding and decoding network; Set the target output semantic information to in for The Lth semantic information data in , j = 1, 2, ..., N; the semantic features to be extracted are in for The cth semantic feature in l = 1, 2, ..., C, where C is the feature dimension; The following formula is used as the loss function during training: In the formula is the loss function value; I(T;X) is the mutual information between the semantic feature T and the structured source X, and p(t,x) is the joint probability distribution of T and X; p(t) is the marginal probability distribution of T, p(x) is the marginal probability distribution of X, t is the semantic feature data, x is the structured source data; I(T;Y) is the mutual information between the semantic feature T and the target output semantic information Y, and p(y) is the marginal probability distribution of Y, y is the target output semantic information data; β is the first parameter set to balance compression and semantic information retention; λ is the weight of the entropy regularization term; H(T) is the regularization term, and P(t) is the marginal probability distribution of T; During training, the SGD optimizer is used to optimize the model.

4. The structured source semantic transmission method for power wireless communication according to claim 3 is characterized in that The step S4, based on the real-time signal-to-interference-and-noise ratio of the target power wireless communication network, uses the semantic coding network obtained in step S3 to perform semantic coding and adaptive semantic feature transmission, specifically includes the following steps: Using the semantic encoding network obtained in step S3 to perform semantic encoding; Based on the real-time signal-to-interference-and-noise ratio of the target power wireless communication network, the feature dimension C is dynamically adjusted using the following rules: Set the low SINR threshold low SINR low =α(dB), high signal-to-interference-and-noise ratio threshold SINR high SINR high =β(dB), semantic feature dimension C under low signal-to-interference-noise ratio low and semantic feature dimension C under high signal-to-interference-noise ratio high ; If SINR i <SINR low , then set C = C low ; If SINR i >SINR high , then set C = C high ; If SINR low ≤SINR i ≤SINR high , then set 5. The structured source semantic transmission method for power wireless communication according to claim 4 is characterized in that After the receiving end receives the data information in step S5, the semantic decoding network obtained in step S3 is used to perform corresponding decoding to obtain the data information of the structured source, which specifically includes the following steps: After the receiving end receives the data information, the corresponding semantic decoding network performs corresponding decoding according to the size of the feature dimension C to obtain the data information of the structured source.

6. A system for implementing the structured source semantic transmission method for power wireless communication according to any one of claims 1 to 5, characterized in that It includes a parameter acquisition module, a data acquisition module, a network training module, a feature transmission module and a semantic transmission module; the parameter acquisition module, the data acquisition module, the network training module, the feature transmission module and the semantic transmission module are connected in series in sequence; the parameter acquisition module is used to obtain parameter information of the target power wireless communication network and upload the data information to the data acquisition module; the data acquisition module is used to obtain data information and channel information of the structured signal source in the target power wireless communication network in real time during communication based on the received data information, and upload the data information to the network training module; The network training module is used to train the semantic encoding and decoding network based on the received data information, the mutual information between the semantic features and the acquired data information, and the information bottleneck theory, and upload the data information to the feature transmission module; The feature transmission module is used to perform semantic encoding based on the received data information and the real-time signal-to-interference-noise ratio of the target power wireless communication network, and to perform adaptive semantic feature transmission, and upload the data information to the semantic transmission module; The semantic transmission module is used to perform corresponding decoding using the obtained semantic decoding network according to the received data information. After the receiving end receives the data information, it obtains the data information of the structured source to complete the semantic transmission of the structured source for power wireless communication.

7. A storage medium having a computer program stored thereon; when the computer program is executed by a processor, the structured source semantic transmission method for power wireless communication according to any one of claims 1 to 5 is implemented.

Images