Model semantic orthogonality-based multiple access method and related device

Abstract

The invention relates to the technical field of communication, in particular to a multiple access method based on model semantic orthogonality and a related device, which are applied to a scene in which a sending end transmits a first quantity of service information to a receiving end, and comprises the following steps: the sending end converts the first quantity of first service information into a first quantity of semantic codes; the sending end quantizes the semantic code to obtain an orthogonal code, and the orthogonal code is used as a to-be-transmitted signal; the sending end carries out superposition transmission on the to-be-transmitted signal at an air interface; and the receiving end decodes the received signal to obtain second service information corresponding to the first number of first service information.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a multiple access method based on model semantic orthogonality and a related device. Background Art

[0002] With the development of information technology and the maturity of Internet of Things technology, the number of communication users and smart devices has experienced explosive growth. In the future intelligent network of everything, network nodes tend to be intelligent. The extreme intelligence of network nodes leads to the rapid expansion of information space and even dimensional disaster, which increases the difficulty of representing information carrying space, making it difficult to match traditional network service capabilities with high-dimensional information space, and the amount of data transmitted in communication is too large. The information business service system cannot continue to meet people's needs for complex, diverse and intelligent information transmission.

[0003] In order to transmit more information, the source end generally compresses the business information to be transmitted in the prior art, for example, a high compression rate image encoding format is used when transmitting images. At present, various types of business information (such as text, images, voice, video, etc.) have mature compression technologies. The above technology belongs to the source end encoding technology.

[0004] In addition, source coding and channel coding can be considered jointly to improve the overall performance of the communication system. This technology can improve the communication quality by optimizing the coding scheme without adding additional communication bandwidth. Specifically, the optimization is achieved by utilizing the statistical characteristics of the source and channel. It compresses the source signal appropriately to adapt to the capabilities and characteristics of the channel while maintaining the quality of the signal. The above technology belongs to the source channel joint coding technology.

[0005] In order to access more users on the channel, multiple access technology needs to be used on the channel. From the first generation (1G) to the fifth generation (5G) of mobile communication technology, multiple access schemes have always been the representative technology of each generation of mobile communication and the key technology to distinguish the capacity of wireless systems. Time division multiple access (TDMA), frequency division multiple access (FDMA) and code division multiple access (CDMA) in the 1G to 3G era avoid interference between users by allocating communication physical resources, and realize flexible allocation in the time domain, frequency domain and code domain. In addition, multiple-input multiple-output technology creates spatial physical resources and produces space division multiple access (SDMA) technology. In the 4G era, orthogonal frequency division multiple access achieves more efficient utilization of frequency division multiple access by allocating a subset of carriers to each user. In the 5G era, in order to further adjust the spectrum efficiency, different powers are allocated to each user by allocating power domain resources (NOMA). The transmitter superimposes and transmits them, and the receiver uses interference elimination technology to separate the signals of each user.

[0006] The semantic communication system is considered to be a potential next-generation communication paradigm. It differs from the classical communication system in that it considers both the semantic information representation of the source and the influence of the channel on the semantics. The multiple access technology in the semantic communication paradigm can further improve the system capacity based on the source-channel joint coding technology and has higher information transmission efficiency. Summary of the invention

[0007] In view of the above problems existing in the prior art, a multiple access method based on model semantic orthogonality and a related device are now provided.

[0008] The present invention includes a multiple access method based on model semantic orthogonality, which is applied to a scenario in which a transmitting end transmits a first amount of first service information to a receiving end, the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices, including:

[0009] The sending end converts the first quantity of the first service information into a first quantity of semantic codes;

[0010] The transmitting end transmits the signal to be transmitted by superimposing it on the air interface.

[0011] Optionally, the transmitting end converts the first quantity of the first service information into a first quantity of semantic codes, including:

[0012] The sending end converts the first service information into a semantic code based on a preset semantic coder; different semantic coders are orthogonal to each other.

[0013] Optionally, the transmitting end quantizes the semantic code to obtain an orthogonal code, including:

[0014] The transmitting end quantizes the semantic code based on a preset quantizer.

[0015] Optionally, the transmitting end performs superposition transmission of the signal to be transmitted on an air interface, including:

[0016] Each sending device corresponding to each service information has its own transmission resources;

[0017] During superposition transmission, the signal to be transmitted is weighted with the corresponding transmission resource and then superimposed and transmitted.

[0018] Optionally, the receiving end decodes the received signal to obtain second service information corresponding to the first quantity of first service information, including:

[0019] The receiving end decodes the received signal based on a preset semantic decoder;

[0020] The semantic decoder corresponds to the semantic encoder of the sending end.

[0021] Optionally, the sending end generates the semantic encoder locally, or the sending end obtains the semantic encoder from other devices.

[0022] Optionally, the receiving end generates the semantic decoder locally, or the receiving end obtains the semantic decoder from other devices.

[0023] Optionally, the semantic encoder or the semantic decoder is divided into several model components for storage or transmission according to a segmentation rule; the segmentation rule includes at least one of the following: horizontal segmentation by layer; vertical segmentation across layers; random segmentation; segmentation based on data volume; segmentation based on function; segmentation based on usage.

[0024] Optionally, the business information includes at least one of the following: text, audio, image, video, point cloud, one-dimensional waveform data, radar data.

[0025] Optionally, the semantic coding adopts a joint source-channel coding technology.

[0026] The present invention also includes a multiple access method based on model semantic orthogonality, which is applied to a transmitting end, wherein the transmitting end transmits a first quantity of first service information to a receiving end, wherein the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices, and is characterized in that it includes:

[0027] The sending end converts the first quantity of the first service information into a first quantity of semantic codes;

[0028] The transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted;

[0029] The transmitting end transmits the signal to be transmitted in a superimposed manner on the air interface; the signal to be transmitted is transmitted in a superimposed manner to the receiving end, i.e., the signal received by the receiving end; the received signal is decoded at the receiving end to obtain second business information corresponding to the first business information of the first quantity.

[0030] The present invention also includes a multiple access method based on model semantic orthogonality, which is applied to a receiving end, wherein the receiving end receives a first quantity of first service information transmitted from a sending end, wherein the sending end includes one or more sending devices, and the receiving end includes one or more receiving devices, and is characterized in that it includes:

[0031] The receiving end receives a signal; the signal is a signal that the transmitting end transmits by superimposing the signal to be transmitted on the air interface; the signal to be transmitted is an orthogonal code obtained by quantizing the semantic code at the transmitting end; the semantic code is a code obtained by converting the first quantity of the first service information at the transmitting end;

[0032] The receiving end decodes the received signal to obtain second service information corresponding to the first quantity of first service information.

[0033] The present invention also includes a multiple access device based on model semantic orthogonality, which is applied to a scenario where a transmitting end transmits a first quantity of first service information to a receiving end, the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices, and is characterized in that it includes:

[0034] An encoding module, wherein the transmitting end converts a first amount of the first service information into a first amount of semantic codes;

[0035] A quantization module, wherein the transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted;

[0036] A transmission module, wherein the transmitting end performs superposition transmission of the signal to be transmitted on an air interface;

[0037] The decoding module is used for the receiving end to decode the received signal to obtain the second service information corresponding to the first service information of the first quantity.

[0038] Optionally, the device includes modules corresponding to the subdivided steps of the corresponding method described above.

[0039] The present invention also includes a multiple access device based on model semantic orthogonality, applied to a receiving end, the receiving end receiving a first quantity of first service information transmitted from a sending end, the sending end including one or more sending devices, the receiving end including one or more receiving devices, characterized in that:

[0040] The third transmission module, the receiving end receives a signal; the signal is the signal that the transmitting end transmits the signal to be transmitted by superimposing it on the air interface; the signal to be transmitted is the orthogonal code obtained by quantizing the semantic code by the transmitting end; the semantic code is the code obtained by the transmitting end converting the first quantity of the first service information;

[0041] The third decoding module is used for the receiving end to decode the received signal to obtain the second service information corresponding to the first quantity of the first service information.

[0042] Optionally, the system includes modules corresponding to the subdivided steps of the corresponding method described above.

[0043] The present invention also includes an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the communication bus is used to complete mutual communication between the processor, the communication interface and the memory;

[0044] The memory is used to store computer programs;

[0045] The processor is used to execute the computer program stored in the memory to implement the multiple access method based on model semantic orthogonality as described in any one of the above technical solutions.

[0046] The present invention also includes a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the multiple access method based on model semantic orthogonality described in any one of the above technical solutions is implemented.

[0047] The present invention also includes a computer program product. When the computer program product is executed on a computer, the computer is enabled to execute the multiple access method based on model semantic orthogonality described in any one of the above technical solutions.

[0048] The multiple access method and related device based on model semantic orthogonality provided by the present invention, combined with the advantages of joint source and channel coding (JSCC), proposes an orthogonal model multiple access (O-MDMA) technology that can be applied to any semantic model. According to the derivation of capacity area, combined with the traditional multiple access technology, the capacity domain is doubled, which can significantly reduce the data transmission volume in communication services and greatly improve the information transmission efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The embodiments of the present invention will be described more fully with reference to the attached drawings, which are provided for illustration and description only and are not intended to limit the scope of the present invention.

[0050] Figure 1 A schematic diagram of a multiple access method provided by an embodiment of the present invention;

[0051] Figure 2 A schematic diagram of an application scenario provided by an embodiment of the present invention;

[0052] Figure 3 A flowchart of a multiple access method provided by an embodiment of the present invention;

[0053] Figure 4 A schematic diagram of a multiple access method provided by an embodiment of the present invention;

[0054] Figure 5A schematic diagram of a multiple access method provided by an embodiment of the present invention;

[0055] Figure 6 A schematic diagram of a multiple access device provided by an embodiment of the present invention;

[0056] Figure 7 A schematic diagram of a multiple access device provided by an embodiment of the present invention;

[0057] Figure 8 A schematic diagram of a multiple access device provided by an embodiment of the present invention;

[0058] Fig. 9 It is a schematic diagram of an electronic device used to implement the method of the embodiment of the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0061] Transmitter and receiver: There is a channel connection between the transmitter and the receiver, the transmitter (transmitter) has an encoder, and the receiver (receiver) has a decoder. The present invention is based on semantic coding and semantic decoding, so the transmitter of the present invention has a semantic encoder; the receiver of the present invention has a semantic decoder. The transmitter and receiver of the present invention work at the semantic layer, and the bottom layer is still the Shannon physical layer. The transmitter described in the present invention sends multiple business information, and the transmitter does not specifically refer to one transmitting device, but can be one or more transmitting devices. The receiving end described in the present invention receives multiple business information, and the receiving end is not one receiving device, but can be multiple receiving devices. There are multiple communication connections between the transmitter and the receiver, and multiple business information is sent and received in parallel. The transmitting device and the receiving device can be physical devices or virtual devices.

[0062] Business information: information involved in information business services. Depending on the type of information, it can be subdivided into text business information, image business information, voice business information, video business information, point cloud business information, etc. It can also be subdivided into structured information and unstructured information. It can also be all possible types of business information transmitted by communication.

[0063] Time-frequency resources: When data is transmitted between the transmitter and receiver in a multiple access manner, the multiple devices participating in the transmission are divided into their own transmission intervals according to time division, frequency division, etc. When a device is assigned to a specific time domain interval and / or frequency domain interval, the time domain interval and / or frequency domain interval becomes the time-frequency resource available to the device. This means that the device performs multiple access transmission on the time-frequency resource.

[0064] Multiple access technology: multiple access (MA). When multiple users are connected to a common transmission medium to communicate with each other, it is necessary to distinguish users in a certain domain. Common domains of distinction include time domain, which distinguishes users by time; frequency domain, which distinguishes users by frequency; and space domain, which distinguishes users by space.

[0065] Orthogonal multiple access: convert multiple business information to be transmitted into multiple orthogonal codes, so as to achieve the ability of multiple users to communicate simultaneously. For example, code division multiple access (CDMA), orthogonal frequency division multiple access (OFDM) technology, etc. The present invention is a model semantic orthogonal multiple access.

[0066] Non-orthogonal multiple access: Convert multiple business information to be transmitted into multiple non-orthogonal codes to achieve the ability of multiple users to communicate simultaneously. For example, non-orthogonal multiple access (NOMA) technology.

[0067] Joint Source Channel Coding: Joint Channel Source Coding (JSCC) is an important technology in semantic communication. The classic JSCC strategy is based on the statistical probability of the source and ignores its semantic features. The present invention is a JSCC based on semantic features.

[0068] Artificial intelligence model: includes an artificial intelligence semantic encoder and an artificial intelligence semantic decoder. The artificial intelligence semantic encoder is used to encode the service information of the future source into semantic information. The semantic information is transmitted on the channel. The artificial intelligence semantic decoder is used to decode the semantic information transmitted on the channel into service information. According to the different semantic types, it can be divided into text artificial intelligence model, audio artificial intelligence model, image artificial intelligence model, video artificial intelligence model, point cloud artificial intelligence model, one-dimensional waveform model, radar data model, etc. The category of artificial intelligence model is related to the category of semantic coding and the category of service information, and can be a model of all possible types of communication transmission.

[0069] Semantic coding: semantic information contained in business information. According to different semantic types, it can be subdivided into text semantic coding, image semantic coding, audio semantic coding, video semantic coding, point cloud semantic coding, etc. The possible categories of semantic coding are related to the possible categories of business information, and can be all possible types of semantic coding for communication transmission. The generation of semantic coding is also related to the semantic encoder, which can adopt all possible models of artificial intelligence, deep learning, pattern recognition and other disciplines.

[0070] Model Division Multiple Access: The semantics extracted by the model is used to distinguish the business information of multiple users. This distinction method can be called model division multiple access, or MDMA for short.

[0071] Orthogonal Mode Division Multiple Access: Orthogonal mode division multiple access technology, referred to as Orthogonal Model Division Multiple Access (O-MDMA).

[0072] Orthogonal semantic coding: Multiple semantic codings that are orthogonal to each other, that is, multiple semantic codings do not interfere with each other in the semantic space. From the perspective of communication transmission, after the orthogonal semantic codings are superimposed and transmitted to the receiving end, multiple semantic codings can still be distinguished by decoding in the receiving section.

[0073] Orthogonal semantic model: multiple semantic models that are orthogonal to each other, that is, multiple semantic encodings obtained after multiple semantic models process the same source information do not interfere with each other in the semantic domain (are orthogonal to each other).

[0074] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0075] Embodiment 1

[0076] The embodiment of the present invention provides a multiple access method based on model semantic orthogonality, which uses the model to provide information business services. The method steps are as shown in the attached figure. Figure 1 As shown, including:

[0077] Step S101, the sending end converts a first number of the first service information into a first number of semantic codes;

[0078] Step S102, the transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted;

[0079] Step S103, the transmitting end performs superposition transmission of the signal to be transmitted on the air interface;

[0080] Step S104: the receiving end decodes the received signal to obtain second service information corresponding to the first quantity of first service information.

[0081] The general application scenario of this embodiment is a transmission link scenario between a transmitting end and a receiving end. In the scenario where the transmitting end transmits a first amount of first service information to the receiving end, the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices. Without loss of generality, the present invention does not limit the number of transmitting ends and the number of receiving ends.

[0082] One implementation manner is that a sending device sends the first service information to a receiving device.

[0083] Another implementation manner is that a sending device sends the first service information to multiple receiving devices.

[0084] Another implementation manner is that multiple sending devices send the first service information to one receiving device.

[0085] Another implementation manner is that multiple sending devices send the first service information to multiple receiving devices.

[0086] Attached Figure 2 In the exemplary scenario shown, there is a channel 204 between the transmitter and the receiver. Assume that there are m users at the transmitter, and the first service information that needs to be transmitted through the channel 204 can be represented as x1, x2, ... x m .

[0087] (i) The sending end converts each first service information into a semantic code based on the semantic encoder 201.

[0088] Assume that the semantic encoders 201 used by the sender are f e1 ,f e2 ,…f em . The semantic encoder 201 comes from a set of n semantic models, that is, m semantic encoders are selected from the set of n semantic models for m business information of m users. These semantic encoders may be the same or different. For example, the three business information of three users are image 1, image 2, and text 1, and the semantic encoders corresponding to the three business information may be image encoder 1, image encoder 1, and text encoder 1. That is, image encoder 1 is selected for both image 1 and image 2.

[0089] The semantic models in this set are orthogonal to each other. Assume that there are m semantic encoders f e1 ,f e2 ,…f em The model parameters in are denoted as θ1, θ2, ... θ m, these model parameters satisfy the following constraints:

[0090]

[0091] i and j are the ordinal numbers of the semantic encoders in the semantic model set. That is, the intersection of the two model parameters is empty, which is used to indicate that the semantic codes generated by the i-th model and the j-th model are orthogonal to each other.

[0092] m semantic encoders f e1 ,f e2 ,...f em The m business information χ1, χ2, ... χ n Convert to semantic code y1,y2,…y n .

[0093] Optionally, the sending end converts the first quantity of the first business information into a first quantity of semantic codes, including: the sending end uses a pre-configured semantic encoder to convert the business information to be transmitted into semantic codes; the sending end generates the semantic encoder locally, or the sending end obtains the semantic encoder from other devices.

[0094] Optionally, the pre-configured semantic encoder can be classified according to the service information type (ie, the encoded semantic type). Taking the service information type as image as an example, the semantic encoder is an image semantic encoder, and the corresponding semantic decoder is also an image semantic decoder.

[0095] Optionally, the semantic encoder is configured in pairs with a semantic decoder, and the semantic decoder can decode the semantic encoding generated according to the original business information according to the semantic encoding to obtain the restored business information. The network devices at both ends (sender and receiver) of the channel transmitting a business information should be configured with semantic encoders and semantic decoders with corresponding functions (i.e., encoding and decoding are a reversible process).

[0096] Specifically, when the business information is an image, the semantic encoder in this embodiment can adopt an image semantic encoder based on a convolutional neural network algorithm; the semantic decoder can adopt an image semantic decoder based on a generative adversarial network technology or other deep learning network technology. It should be noted that the second business information obtained after the semantic code is restored by the semantic decoder is highly similar to the business information of the sender. For example, the business information is a cat image, and the semantic information (semantic code) is obtained by semantically segmenting the cat image using the semantic encoder; after a series of processing and transmission steps, the receiving end uses the semantic decoder to reconstruct the semantic information to obtain the cat image (second business information). The cat image restored by the semantic decoder will have some differences in image quality compared to the original cat image, but the two are consistent in content, and the visual experience for the user is almost the same.

[0097] Optionally, the semantic encoder is pre-configured in the transmitting end; the semantic decoder corresponding to the semantic encoder is pre-configured in the receiving end. For network devices that often transmit common types of business information (such as text, images, audio, video, etc.), it is necessary to pre-configure semantic encoders and semantic decoders of these categories.

[0098] Optionally, the semantic encoder or the semantic decoder is divided into several model components for storage or transmission according to a segmentation rule; the segmentation rule includes at least one of the following: horizontal segmentation by layer; vertical segmentation across layers; random segmentation; segmentation based on data volume; segmentation based on function; segmentation based on usage.

[0099] Optionally, the image semantic encoder encodes the input image service data into a semantic code. The length of the semantic code can be determined according to the range of the transmitted image service data (which can be determined from the richness of the image semantics and the scale of the training data set). If the image transmitted on the channel contains less semantic content, the length of the semantic code of the image can be relatively small; if the image transmitted on the channel involves more categories and more semantic content, the length of the semantic code of the image can be relatively large.

[0100] For example, a general image semantic encoder is trained based on the OpenImage dataset to encode a 512*512*3 (RGB three image channels) color input image, and a 4096-bit semantic code can be obtained. After the general image semantic encoder is transferred to the cats and dogs dataset (Dogs VS Cats), a cats and dogs image semantic encoder is obtained. The cats and dogs image semantic encoder only needs to recognize two categories, cats and dogs, so its semantic code length is relatively small, for example, a 1024-bit semantic code is obtained.

[0101] The semantic encoder and the semantic decoder are trained based on a domain-wide dataset, so they are compatible with all business information in the same domain. Taking the OpenImage dataset as an example, its data labels are fully covered and the data volume is huge, which is enough to cover the encoding requirements of all business information in the image field.

[0102] Similarly, for different types of business information, including but not limited to images, voice, text, video, point cloud, one-dimensional waveform data, and radar data, each type of semantic encoder and semantic decoder can be trained separately, so that the model deployed on the communication equipment can normally process all business information that needs to be transmitted in the communication.

[0103] (ii) To obtain orthogonal semantic codes, the quantizer 202 quantizes the semantic codes to obtain a first number of orthogonal codes

[0104] That is, m semantic codes y1, y2, ...y n Quantize to obtain m orthogonal codes s1, s2, ...s corresponding to m business information n .

[0105] The effect of quantizer 202 is to discretize the input data and map the continuous numerical value to the discrete coding, thereby obtaining the orthogonal semantic coding. In addition, the quantization step can also reduce the complexity of the model, improve the generalization ability of the model, improve the computational efficiency of the model, and make the model process large-scale data faster. The quantization method that quantizer 202 can choose includes equal width quantization, arithmetic difference quantization, vector quantization, etc.

[0106] Orthogonal Coding i and j It can be expressed by the following formula.

[0107] <f d (s i ,θ i ),f d (s j ,θ j )>=0

[0108] Among them, f d represents the semantic decoder, s i and j It is obtained by encoding by the i-th and j-th semantic encoders.<a,b> =0 means that in the semantic space, semantic coding a will not be affected by semantic coding b, that is, the two semantic codings are orthogonal.

[0109] (iii) At the transmitting end, the transmitting device corresponding to each service information has its own transmission resources (eg, time domain, frequency domain, spatial domain, power, etc.).

[0110] Optionally, the transmitting end performs superposition transmission of the signal to be transmitted on an air interface, including:

[0111] Each sending device corresponding to each service information has its own transmission resources;

[0112] During superposition transmission, the signal to be transmitted is weighted with the corresponding transmission resource and then superimposed and transmitted.

[0113] When superimposed transmission is used, the transmission resources corresponding to the m service information can be expressed as p1, p2, … p m When m signals to be transmitted are superimposed and transmitted, the superimposed and transmitted signal t can be expressed as the weighted sum of the m user signals:

[0114] t=p1s1+p2s2+p3s3+…+p m s m

[0115] The signal t to be transmitted after superposition coding is transmitted from the transmitter 203 to the receiver 205 at the receiving end through the channel 204.

[0116] (iv) The semantic decoder 206 decodes the received signal to restore the signal of each user, that is, obtains the second service information corresponding to the first service information.

[0117] Optionally, the pre-configured semantic decoder can be classified according to the service information type (ie, the encoded semantic type). Taking the service information type as image as an example, the semantic encoder is an image semantic encoder, and the corresponding semantic decoder is also an image semantic decoder.

[0118] Optionally, the semantic encoder is configured in pairs with a semantic decoder, and the semantic decoder can decode the semantic encoding generated according to the original business information according to the semantic encoding to obtain the restored business information. The network devices at both ends (sender and receiver) of the channel transmitting a business information should be configured with semantic encoders and semantic decoders with corresponding functions (i.e., encoding and decoding are a reversible process).

[0119] The semantic decoder can restore the signal processed by the quantizer and the semantic encoder at the source end to the source information.

[0120] Specifically, taking the business information as an image as an example, the semantic decoder in this embodiment can adopt an image semantic decoder based on a convolutional neural network algorithm; the semantic decoder can also adopt an image semantic decoder based on generative adversarial network technology or other deep learning network technology.

[0121] Optionally, the semantic decoders are pre-configured in the receiving end. For network devices that often transmit common types of service information (such as text, image, audio, video, etc.), it is necessary to pre-configure semantic encoders and semantic decoders of these types.

[0122] Optionally, the semantic encoder or the semantic decoder is divided into several model components for storage or transmission according to a segmentation rule; the segmentation rule includes at least one of the following: horizontal segmentation by layer; vertical segmentation across layers; random segmentation; segmentation based on data volume; segmentation based on function; segmentation based on usage.

[0123] Similarly, for different types of business information, including but not limited to images, voice, text, video, point cloud, one-dimensional waveform data, and radar data, each type of semantic encoder and semantic decoder can be trained separately, so that the model deployed on the communication equipment can normally process all business information that needs to be transmitted in the communication.

[0124] The semantic decoder 206 has the same function as the semantic encoder 201 at the transmitting end. For example, for service information x, the semantic encoder 201 encodes x into semantic code y, and the semantic decoder 206 decodes the semantic code y to restore the original service information x.

[0125] (V) Model semantic orthogonality has doubled the transmission efficiency compared to traditional multiple access technology. Figure 3 As shown in the figure, users 1-4 encode semantic signals using four semantic models respectively, and perform signal superposition in the model semantic space and then in the physical domain to achieve multi-user multiple access. By fully utilizing the advantages of JSCC, the bandwidth is saved by half compared to traditional multiple access technology.

[0126] Therefore, the capacity regions and sum-rate capacity of the uplink are given below for O-MDMA, respectively.

[0127]

[0128]

[0129] Where g1, g2, B, N0 are the channel gain, bandwidth and noise power spectral density of each user respectively.

[0130] Similarly, the capacity regions and sum-rate capacity of the uplink are,

[0131]

[0132] The content of this embodiment can be freely combined with the content of other embodiments to form an implementation scheme.

[0133] Embodiment 2

[0134] The present invention also includes a multiple access method based on model semantic orthogonality, which is applied to a transmitting end, wherein the transmitting end transmits a first quantity of first service information to a receiving end, wherein the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices, and the method steps are as shown in the attached figure. Figure 4 As shown, including:

[0135] Step S201, the sending end converts a first number of the first service information into a first number of semantic codes;

[0136] Step S202, the transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted;

[0137] In step S203, the transmitting end transmits the signal to be transmitted in a superimposed manner on the air interface; the signal to be transmitted is transmitted in a superimposed manner to the receiving end, i.e., the signal received by the receiving end; the received signal is decoded at the receiving end to obtain second service information corresponding to the first service information of the first quantity.

[0138] The content of this embodiment can be freely combined with the content of other embodiments to form an implementation scheme.

[0139] Embodiment 3

[0140] The present invention also includes a multiple access method based on model semantic orthogonality, which is applied to a receiving end, wherein the receiving end receives a first quantity of first service information transmitted from a sending end, wherein the sending end includes one or more sending devices, and the receiving end includes one or more receiving devices. The method steps are as shown in the attached Figure 5 As shown, including:

[0141] Step S301, the receiving end receives a signal; the signal is a signal that the transmitting end transmits by superimposing the signal to be transmitted on the air interface; the signal to be transmitted is an orthogonal code obtained by quantizing the semantic code by the transmitting end; the semantic code is a code obtained by the transmitting end converting the first quantity of the first service information;

[0142] Step S302: the receiving end decodes the received signal to obtain second service information corresponding to the first quantity of first service information.

[0143] The content of this embodiment can be freely combined with the content of other embodiments to form an implementation scheme.

[0144] Embodiment 4

[0145] The embodiment of the present invention provides a multiple access device based on model semantic orthogonality, which is applied to a scenario where a transmitting end transmits a first quantity of first service information to a receiving end, wherein the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices. The method steps are as shown in the attached figure. Figure 6 As shown, including:

[0146] Coding module 101, the sending end converts the first quantity of the first service information into a first quantity of semantic codes;

[0147] Quantization module 102, the transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted;

[0148] Transmission module 103, the transmitting end performs superposition transmission of the signal to be transmitted on the air interface;

[0149] The decoding module 104, the receiving end decodes the received signal to obtain the second service information corresponding to the first quantity of the first service information.

[0150] The content of this embodiment can be freely combined with the content of other embodiments to form an implementation scheme.

[0151] Embodiment 5

[0152] The present invention also includes a multiple access device based on model semantic orthogonality, which is applied to a transmitting end, wherein the transmitting end transmits a first quantity of first service information to a receiving end, wherein the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices, and the method steps are as shown in the attached Figure 7 As shown, including:

[0153] A second encoding module 201, the sending end converts a first number of the first service information into a first number of semantic codes;

[0154] A second quantization module 202, wherein the transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted;

[0155] The second transmission module 203, the transmitting end superimposes and transmits the signal to be transmitted on the air interface; the signal to be transmitted is superimposed and transmitted to the receiving end, that is, the signal received by the receiving end; the received signal is decoded at the receiving end to obtain the second business information corresponding to the first quantity of the first business information.

[0156] The content of this embodiment can be freely combined with the content of other embodiments to form an implementation scheme.

[0157] Embodiment 6

[0158] The present invention also includes a multiple access device based on model semantic orthogonality, which is applied to a receiving end, wherein the receiving end receives a first quantity of first service information transmitted from a sending end, wherein the sending end includes one or more sending devices, and the receiving end includes one or more receiving devices, wherein the method steps are as shown in the attached Figure 8 As shown, including:

[0159] The third transmission module 301, the receiving end receives a signal; the signal is the signal that the transmitting end transmits the signal to be transmitted by superimposing it on the air interface; the signal to be transmitted is the orthogonal code obtained by quantizing the semantic code by the transmitting end; the semantic code is the code obtained by the transmitting end converting the first quantity of the first service information;

[0160] The third decoding module 302, the receiving end decodes the received signal to obtain the second service information corresponding to the first quantity of the first service information.

[0161] The content of this embodiment can be freely combined with the content of other embodiments to form an implementation scheme.

[0162] Embodiment 7

[0163] According to an embodiment of the present invention, the present invention also provides an electronic device, a readable storage medium and a computer program product.

[0164] The present invention also includes an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the communication bus is used to complete mutual communication between the processor, the communication interface and the memory;

[0165] The memory is used to store computer programs;

[0166] The processor is used to execute the computer program stored in the memory to implement the multiple access method based on model semantic orthogonality as described in any one of the above technical solutions.

[0167] The present invention also includes a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the multiple access method based on model semantic orthogonality described in any one of the above technical solutions is implemented.

[0168] The present invention also includes a computer program product. When the computer program product is executed on a computer, the computer is enabled to execute the multiple access method based on model semantic orthogonality described in any one of the above technical solutions.

[0169] Fig. 9A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0170] The device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 803 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0171] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0172] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as the test method. For example, in some embodiments, the test method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the test method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the test method in any other appropriate manner (e.g., by means of firmware).

[0173] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0174] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0175] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0176] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0177] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0178] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0179] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0180] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A multiple access method based on model semantic orthogonality, applied to a scenario where a transmitting end transmits a first amount of first service information to a receiving end, the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices, characterized in that: include: The sending end converts the first quantity of the first service information into a first quantity of semantic codes; The transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted; The transmitting end performs superposition transmission of the signal to be transmitted on the air interface; The receiving end decodes the received signal to obtain second service information corresponding to the first quantity of first service information.

2. The method according to claim 1, characterized in that The transmitting end converts the first quantity of the first service information into a first quantity of semantic codes, including: The sending end converts the first service information into a semantic code based on a preset semantic coder; different semantic coders are orthogonal to each other.

3. The method according to claim 1, characterized in that: The transmitting end quantizes the semantic code to obtain an orthogonal code, including: The transmitting end quantizes the semantic code based on a preset quantizer.

4. The method according to claim 1, characterized in that The transmitting end performs superposition transmission of the signal to be transmitted on the air interface, including: Each sending device corresponding to each service information has its own transmission resources; During superposition transmission, the signal to be transmitted is weighted with the corresponding transmission resource and then superimposed and transmitted.

5. The method according to claim 2, characterized in that: The receiving end decodes the received signal to obtain second service information corresponding to the first service information of the first quantity, including: The receiving end decodes the received signal based on a preset semantic decoder; The semantic decoder corresponds to the semantic encoder of the sending end.

6. The method according to claim 3, characterized in that include: The sending end generates the semantic encoder locally, or the sending end obtains the semantic encoder from other devices; The receiving end generates the semantic decoder locally, or the receiving end obtains the semantic decoder from other devices.

7. A multiple access method based on model semantic orthogonality, applied to a transmitting end, wherein the transmitting end transmits a first quantity of first service information to a receiving end, wherein the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices, characterized in that: include: The sending end converts the first quantity of the first service information into a first quantity of semantic codes; The transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted; The transmitting end performs superposition transmission of the signal to be transmitted on the air interface; The signal to be transmitted is superimposed and transmitted to the receiving end, that is, the signal received by the receiving end; the received signal is decoded at the receiving end to obtain second service information corresponding to the first service information of the first quantity.

8. A multiple access method based on model semantic orthogonality, applied to a receiving end, the receiving end receiving a first quantity of first service information transmitted from a sending end, the sending end including one or more sending devices, the receiving end including one or more receiving devices, characterized in that: include: The receiving end decodes the received signal to obtain second business information corresponding to the first quantity of first business information; the signal is the signal that the transmitting end transmits by superimposing the signal to be transmitted on the air interface; the signal to be transmitted is the orthogonal code obtained by quantizing the semantic code at the transmitting end; the semantic code is the code obtained by converting the first quantity of the first business information at the transmitting end.

9. A multiple access device based on model semantic orthogonality, applied to a scenario where a transmitting end transmits a first quantity of first service information to a receiving end, the transmitting end includes one or more transmitting devices, and the receiving end includes one or more receiving devices, characterized in that: include: An encoding module, wherein the transmitting end converts a first amount of the first service information into a first amount of semantic codes; A quantization module, wherein the transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted; A transmission module, wherein the transmitting end performs superposition transmission of the signal to be transmitted on an air interface; The decoding module is used for the receiving end to decode the received signal to obtain the second service information corresponding to the first service information of the first quantity.

10. A multiple access device based on model semantic orthogonality, applied to a transmitting end, the transmitting end transmits a first quantity of first service information to a receiving end, the transmitting end includes one or more transmitting devices, the receiving end includes one or more receiving devices, characterized in that: A second encoding module, the sending end converting the first quantity of the first service information into a first quantity of semantic codes; A second quantization module, wherein the transmitting end quantizes the semantic code to obtain an orthogonal code as a signal to be transmitted; A second transmission module, wherein the transmitting end performs superposition transmission of the signal to be transmitted on the air interface; The signal to be transmitted is superimposed and transmitted to the receiving end, that is, the signal received by the receiving end; the received signal is decoded at the receiving end to obtain second service information corresponding to the first service information of the first quantity.

11. A multiple access device based on model semantic orthogonality, applied to a receiving end, the receiving end receiving a first quantity of first service information transmitted from a sending end, the sending end including one or more sending devices, the receiving end including one or more receiving devices, characterized in that: The third transmission module, the receiving end receives a signal; the signal is the signal that the transmitting end transmits the signal to be transmitted by superimposing it on the air interface; the signal to be transmitted is the orthogonal code obtained by quantizing the semantic code by the transmitting end; the semantic code is the code obtained by the transmitting end converting the first quantity of the first service information; The third decoding module is used for the receiving end to decode the received signal to obtain the second service information corresponding to the first quantity of the first service information.

12. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the communication bus is used to complete the mutual communication between the processor, the communication interface and the memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory to implement the multiple access method based on model semantic orthogonality as described in any one of claims 1-8.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the multiple access method based on model semantic orthogonality according to any one of claims 1 to 8.

14. A computer program product, characterized in that When the computer program product is executed on a computer, the computer is enabled to execute the multiple access method based on model semantic orthogonality according to any one of claims 1 to 8.

Images