WO2022174642A1 - 基于空间划分的数据处理方法和通信装置 - Google Patents

基于空间划分的数据处理方法和通信装置 Download PDF

Info

Publication number
WO2022174642A1
WO2022174642A1 PCT/CN2021/131856 CN2021131856W WO2022174642A1 WO 2022174642 A1 WO2022174642 A1 WO 2022174642A1 CN 2021131856 W CN2021131856 W CN 2021131856W WO 2022174642 A1 WO2022174642 A1 WO 2022174642A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
coverage area
coverage areas
information
signal
Prior art date
Application number
PCT/CN2021/131856
Other languages
English (en)
French (fr)
Inventor
钟财军
高佳宝
于天航
张公正
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP21926356.3A priority Critical patent/EP4266603A4/en
Publication of WO2022174642A1 publication Critical patent/WO2022174642A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/391—Modelling the propagation channel
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/20—Monitoring; Testing of receivers
    • H04B17/24—Monitoring; Testing of receivers with feedback of measurements to the transmitter
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/20—Monitoring; Testing of receivers
    • H04B17/27—Monitoring; Testing of receivers for locating or positioning the transmitter
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/391—Modelling the propagation channel
    • H04B17/3913—Predictive models, e.g. based on neural network models
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/024—Channel estimation channel estimation algorithms
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/024—Channel estimation channel estimation algorithms
    • H04L25/0254—Channel estimation channel estimation algorithms using neural network algorithms
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • the present application relates to the field of communication, and more particularly, to a data processing method and communication device based on space division.
  • large-scale antennas have many advantages such as high spectral efficiency, high energy efficiency, and high reliability.
  • large-scale antenna systems usually adopt a hybrid analog-digital architecture, that is, the number of radio frequency links at the base station is less than the number of antennas.
  • the received signal at the antenna is processed in the analog domain composed of phase shifters and then processed in the digital domain.
  • the application of large-scale antennas brings great challenges to signal processing.
  • the present application provides a data processing method and communication device based on space division, which can improve the reliability of communication.
  • a communication method is provided, and the method can be executed by a first device or a module (eg, a chip) configured in (or used for) the first device.
  • a module eg, a chip
  • the method includes: a first device obtains a first signal from a second device; the first device infers the first signal to obtain first channel information according to a first model, and the second device is in multiple coverages of the first device The first coverage area in the area, and the first model is a channel information inference model corresponding to the first coverage area.
  • the coverage of the signal receiving device is divided into multiple coverage areas, and the signal receiving device uses the artificial intelligence inference model corresponding to the coverage area according to the coverage area where the signal transmitting device is located to infer the received signal to obtain channel information.
  • the accuracy of the received signal can be improved. Thereby improving the reliability of communication.
  • the first channel information is angle domain channel information.
  • the channel data processing performance can be improved and the estimation overhead can be reduced by utilizing the sparse structure of the angle domain channel.
  • inferring the first signal according to the first model to obtain the first channel information includes: inferring and obtaining measurement information according to the first model; according to the measurement information Processing the first signal to obtain a second signal; according to the first model, inferring the second signal to obtain the first channel information.
  • the first device can infer the measurement information used to measure the signal through the first model, and further infer the channel information experienced by the signal according to the first model. It can improve the channel estimation performance and reduce the estimation overhead.
  • the first model is an autoencoder model
  • the first model includes an encoder part and a decoder part
  • inference is obtained to measure
  • the information includes: the first device deduces the measurement information according to the encoder part; and, according to the first model, deduces the measurement information, including: the first device deduces the second signal according to the decoder part Obtain the first channel information.
  • the method further includes: the first device acquires a training signal set corresponding to the first coverage area, the first training set includes multiple training signals, and the first device is based on The training signal set is used to train the training model to obtain the first model.
  • the first device trains the channel information models of different coverage areas based on the training data in the channel environment, which increases the matching degree between the model and the coverage area and improves the accuracy of channel estimation by the model.
  • the training model includes an encoder part and a decoder part
  • the training of the training model according to the training signal set includes: using the first training set
  • the angle domain channel information sample is input into the encoder part to obtain the target encoded signal
  • the target encoded signal is input into the decoder part to obtain the target decoded signal
  • the training is obtained
  • the parameter information of the model; according to the parameter information, the parameters of the encoder part and the parameters of the decoder part are adjusted.
  • one of the multiple coverage areas belongs to a sector covered by a central angle centered on the first device, and the first coverage area belongs to the first device.
  • the first coverage area belongs to the first device.
  • any two of the multiple coverage areas belong to different sectors, or the distance between any two of the multiple coverage areas and the first device is different.
  • the adjacent areas are divided into the same coverage area.
  • Using the corresponding channel information inference model according to different coverage areas can improve the accuracy of channel information acquisition, so as to restore the received signal more accurately.
  • the method further includes: acquiring, by the first device, location information from the second device; the first device determining, according to the location information, that the second device is in the first coverage area.
  • the first device may determine that the second device is in the first coverage area according to the location information sent by the second device. Therefore, the acquired signal of the second device is processed by using the first model corresponding to the first coverage area. Improve the accuracy of channel estimation, thereby improving the reliability of communication.
  • the method further includes: the first device sending first indication information, where the first indication information is used to indicate the multiple coverage areas, the multiple coverage areas Corresponding to multiple channel information inference models; the first device acquires second indication information, where the second indication information is used to indicate that the second device is in the first coverage area or to indicate the first model.
  • the first device can notify the second device of multiple coverage areas through the first indication information, and the second device determines the first coverage area to which it belongs and notifies the first device through the second indication information, so that the first device can The acquired signal of the second device is processed by using the first model corresponding to the first coverage area.
  • the first indication information is broadcast information.
  • the first device is a network device or a terminal device.
  • a communication method is provided, and the method can be executed by a second device or a module (eg, a chip) configured in (or used in) the second device.
  • a module eg, a chip
  • the method includes: the second device acquires first indication information, the first indication information is used to indicate multiple coverage areas, the multiple coverage areas correspond to multiple channel information inference models; the second device sends the second indication information , the second indication information is used to indicate that the second device is in the first coverage area or to indicate the first model, the first coverage area is one of the multiple coverage areas, and the first model is the first coverage area corresponding model.
  • one of the multiple coverage areas belongs to a sector covered by a central angle centered on the first device, and the first coverage area belongs to the first
  • two of the multiple coverage areas belong to different sectors, or two coverage areas of the multiple coverage areas have different distances from the first device.
  • the first indication information includes one or more of the following: reference location information of each coverage area in the multiple coverage areas; Identification information of each coverage area or identification information of each model in the plurality of channel information inference models.
  • the second indication information includes identification information of the first coverage area or identification information of the first model.
  • a communication apparatus is provided, and the apparatus is a first device or a module (eg, a chip) configured in (or used for) the first device.
  • a module eg, a chip
  • the communication device includes: a transceiver unit for acquiring a first signal from a second device;
  • a processing unit configured to infer a first signal to obtain first channel information according to a first model
  • the second device is located in a first coverage area among multiple coverage areas of the first device
  • the first model is the first coverage
  • the channel information inference model corresponding to the region.
  • the first channel information is angle domain channel information.
  • the processing unit is specifically configured to: obtain measurement information by reasoning according to the first model; process the first signal according to the measurement information to obtain the second signal; According to the first model, the second signal is inferred to obtain the first channel information.
  • the first model is an auto-encoder model
  • the first model includes an encoder part and a decoder part
  • the processing unit is specifically configured to:
  • the measurement information is obtained by inference; and, according to the decoder part, the first channel information is obtained by inferring the second signal.
  • the transceiver unit before the transceiver unit acquires the first signal from the second device, the transceiver unit is further configured to acquire the first training set of the first coverage area,
  • the first training set includes a plurality of training signals;
  • the processing unit is further configured to train a training model based on the first training set to obtain the first model.
  • the training model includes an encoder part and a decoder part, and the processing unit is specifically used for:
  • one of the multiple coverage areas belongs to a sector covered by a central angle centered on the first device, and the first coverage area belongs to the first device.
  • the first coverage area belongs to the first device.
  • any two of the multiple coverage areas belong to different sectors, or the distance between any two of the multiple coverage areas and the first device is different.
  • the transceiver unit is further configured to acquire location information from the second device; the processing unit is further configured to determine that the second device is in the first device according to the location information a coverage area.
  • a communication apparatus is provided, and the apparatus is a second device or a module (eg, a chip) configured in (or used for) the second device.
  • the communication device includes: a transceiver unit for acquiring first indication information, where the first indication information is used to indicate multiple coverage areas, the multiple coverage areas correspond to multiple channel information inference models; a processing unit for The first indication information determines that the second device is in a first coverage area of the multiple coverage areas; the transceiver unit is further configured to send second indication information, where the second indication information is used to indicate that the second device is in the The first coverage area or indicates a first model, where the first model is a model corresponding to the first coverage area.
  • one of the multiple coverage areas belongs to a sector covered by a central angle centered on the first device, and the first coverage area belongs to the first device.
  • the first coverage area belongs to the first device.
  • two of the multiple coverage areas belong to different sectors, or the distance between the two coverage areas of the multiple coverage areas and the first device is different.
  • the first indication information includes one or more of the following:
  • the second indication information includes identification information of the first coverage area or identification information of the first model.
  • a communication apparatus including a processor.
  • the processor may implement the first aspect and the method in any possible implementation manner of the first aspect.
  • the communication device further includes a memory, and the processor is coupled to the memory and can be configured to execute instructions in the memory, so as to implement the first aspect and the method in any possible implementation manner of the first aspect.
  • the communication device further includes a communication interface, and the processor is coupled to the communication interface.
  • the communication interface may be a transceiver, a pin, a circuit, a bus, a module, or other types of communication interfaces, which are not limited.
  • the communication apparatus is a first device.
  • the communication interface may be a transceiver, or an input/output interface.
  • the communication device is a chip configured in the first device.
  • the communication interface may be an input/output interface
  • the processor may be a logic circuit.
  • the input/output interface is used for inputting the first signal from the second device; the logic circuit is used for inferring the first signal according to the first model to obtain the first channel information, the second device is in a plurality of the first device The first coverage area in the coverage area, where the first model is a channel information inference model corresponding to the first coverage area.
  • the transceiver may be a transceiver circuit.
  • the input/output interface may be an input/output circuit.
  • a communication apparatus including a processor.
  • the processor may implement the method in the second aspect and any possible implementation manner of the second aspect.
  • the communication device further includes a memory, and the processor is coupled to the memory and can be configured to execute instructions in the memory, so as to implement the second aspect and the method in any possible implementation manner of the second aspect.
  • the communication device further includes a communication interface, and the processor is coupled to the communication interface.
  • the communication apparatus is a second device.
  • the communication interface may be a transceiver, or an input/output interface.
  • the communication device is a chip configured in the second device.
  • the communication interface may be an input/output interface
  • the processor may be a logic circuit.
  • the input/output interface is used for inputting first indication information, and the first indication information is used to indicate a plurality of coverage areas, and the plurality of coverage areas correspond to a plurality of channel information inference models; the logic circuit is used for according to the first indication information.
  • An indication information determines that the second device is in a first coverage area of the plurality of coverage areas; the input/output interface is further configured to output second indication information, the second indication information is used to indicate that the second device is in the first coverage area A coverage area or an indication of a first model, the first model being a model corresponding to the first coverage area.
  • the transceiver may be a transceiver circuit.
  • the input/output interface may be an input/output circuit.
  • a processor including: an input circuit, an output circuit, and a processing circuit.
  • the processing circuit is configured to receive a signal through the input circuit and transmit a signal through the output circuit, so that the processor performs the method of the first aspect or the second aspect and any possible implementation of the first aspect or the second aspect .
  • the above-mentioned processor may be one or more chips
  • the input circuit may be input pins
  • the output circuit may be output pins
  • the processing circuit may be transistors, gate circuits, flip-flops and various logic circuits, etc. .
  • the input signal received by the input circuit may be received and input by, for example, but not limited to, a receiver
  • the signal output by the output circuit may be, for example, but not limited to, output to and transmitted by a transmitter
  • the circuit can be the same circuit that acts as an input circuit and an output circuit at different times.
  • the embodiments of the present application do not limit the specific implementation manners of the processor and various circuits.
  • a computer program product comprising: a computer program (also referred to as code, or instructions), which, when the computer program is executed, causes the computer to execute the above-mentioned first aspect or the second aspect and the method in any possible implementation manner of the first aspect or the second aspect.
  • a computer program also referred to as code, or instructions
  • a computer-readable storage medium stores a computer program (also referred to as code, or instruction), when it is run on a computer, causing the computer to execute the above-mentioned first aspect or A method of the second aspect and any possible implementation of the first aspect or the second aspect.
  • a computer program also referred to as code, or instruction
  • a communication system including the aforementioned at least one first device and at least one second device.
  • FIG. 1 is a schematic architecture of a communication system applicable to an embodiment of the present application
  • FIG. 2 is a schematic diagram of a hybrid analog-digital large-scale antenna system architecture provided by the present application.
  • FIG. 3 is a schematic flowchart of a communication method provided by an embodiment of the present application.
  • FIG. 4 is a schematic diagram of a self-encoding model training process provided by an embodiment of the present application.
  • FIG. 5 is a schematic diagram of multiple coverage areas of a first device provided by an embodiment of the present application.
  • FIG. 6 is another schematic diagram of multiple coverage areas of a first device provided by an embodiment of the present application.
  • FIG. 7 is another schematic diagram of multiple coverage areas of a first device provided by an embodiment of the present application.
  • FIG. 8 is another schematic flowchart of the communication method provided by the embodiment of the present application.
  • FIG. 9 is a schematic block diagram of an example of a communication device of the present application.
  • FIG. 10 is a schematic configuration diagram of an example of a communication device of the present application.
  • LTE long term evolution
  • FDD frequency division duplex
  • TDD time division duplex
  • 5th generation, 5G fifth generation
  • future communication system such as sixth generation (6th generation, 6G) communication system
  • 6G sixth generation
  • NR new radio
  • FIG. 1 is a schematic diagram of a communication system 100 suitable for an embodiment of the present application.
  • the communication system applicable to this embodiment of the application may include at least one network device, such as the network device 110 in the communication system 100 shown in FIG. 1 .
  • the wireless communication system may further include at least one terminal device, such as the terminal device 120 in the communication system 100 shown in FIG. 1 .
  • Communication between the network device 110 and the terminal device 120 may be through a wireless link.
  • the network device 110 can process the signal from the terminal device 120 through the communication method provided in this application.
  • the terminal device 120 can also process the signal from the network device 110 through the communication method provided in this application. This application does not limit this.
  • the technical solutions provided in the embodiments of the present application can be applied to various communication scenarios, for example, can be applied to one or more of the following communication scenarios: enhanced mobile broadband (eMBB) communication, ultra-reliable and low-latency communication (ultra reliable low latency communication, URLLC), machine type communication (MTC), massive machine type communication (mMTC), device-to-device (D2D) communication, vehicle Outreach (vehicle to everything, V2X) communication, vehicle to vehicle (V2V) communication, vehicle to network (V2N), vehicle to infrastructure (V2I), vehicle to pedestrian ( vehicle to pedestrian, V2P), Internet of things (Internet of things, IoT), drone communication and satellite communication, etc.
  • the mMTC may include one or more of the following communications: communications in industrial wireless sensor networks (IWSN), communications in video surveillance (video surveillance) scenarios, and wearable device communications Wait.
  • IWSN industrial wireless sensor networks
  • video surveillance video surveillance
  • wearable device communications Wait wearable device communications Wait.
  • Communication between communication devices may include: communication between a network device and a terminal device, communication between a network device and a network device, and/or communication between a terminal device and a terminal device.
  • the term “communication” may also be described as "transmission”, “information transmission”, or “signal transmission”, or the like. Transmission can include sending and/or receiving.
  • the technical solution of the embodiments of the present application is described by taking the communication between the network device and the terminal device as an example, and those skilled in the art can also use the technical solution for communication between other scheduling entities and subordinate entities, such as between a macro base station and a micro base station.
  • the scheduling entity may allocate radio resources, such as air interface resources, to the subordinate entities.
  • Air interface resources include one or more of the following resources: time domain resources, frequency domain resources, code resources and space resources.
  • the communication between the network device and the terminal device includes: the network device sends a downlink signal to the terminal device, and/or the terminal device sends an uplink signal to the network device.
  • the signal can also be replaced with information or data, etc.
  • the terminal device involved in the embodiments of the present application may also be referred to as a terminal.
  • the terminal may be a device with wireless transceiving function. Terminals can be deployed on land, including indoors, outdoors, handheld, and/or vehicle; can also be deployed on water (such as ships, etc.); and can also be deployed in the air (such as aircraft, balloons, and satellites, etc.).
  • the terminal equipment may be user equipment (user equipment, UE). UEs include handheld devices, in-vehicle devices, wearable devices, or computing devices with wireless communication capabilities. Exemplarily, the UE may be a mobile phone, a tablet computer, or a computer with a wireless transceiver function.
  • the terminal device may also be a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in unmanned driving, a wireless terminal in telemedicine, intelligent A wireless terminal in a power grid, a wireless terminal in a smart city, and/or a wireless terminal in a smart home, and so on.
  • VR virtual reality
  • AR augmented reality
  • a wireless terminal in a power grid a wireless terminal in a smart city
  • a wireless terminal in a smart home and so on.
  • the network device involved in the embodiments of the present application includes a base station (base station, BS), which may be a device deployed in a wireless access network and capable of wirelessly communicating with a terminal device.
  • the base station may have various forms, such as macro base station, micro base station, relay station or access point.
  • the base station involved in the embodiments of the present application may be a base station in a 5G system, a base station in an LTE system, or a base station in other systems, which is not limited.
  • the base station in the 5G system can also be called a transmission reception point (TRP) or a next generation Node B (generation Node B, gNB or gNodeB).
  • TRP transmission reception point
  • gNB next generation Node B
  • the base station may be an integrated base station, or may be a base station separated into multiple network elements, which is not limited.
  • the base station is a base station in which a centralized unit (centralized unit, CU) and a distributed unit (distributed unit, DU) are separated, that is, the base station includes a CU and a DU.
  • the apparatus for implementing the function of the terminal device may be a terminal device; it may also be an apparatus capable of supporting the terminal device to implement the function, such as a chip system.
  • the device can be installed in the terminal equipment or used in combination with the terminal equipment.
  • the chip system may be composed of chips, or may include chips and other discrete devices.
  • the apparatus for implementing the function of the network device may be a network device; it may also be an apparatus capable of supporting the network device to implement the function, such as a chip system.
  • the apparatus can be installed in network equipment or used in combination with network equipment.
  • “/” may indicate that the objects associated before and after are an “or” relationship, for example, A/B may indicate A or B; “and/or” may be used to describe that there are three types of associated objects A relationship, for example, A and/or B, can mean that A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural.
  • words such as “first” and “second” may be used to distinguish technical features with the same or similar functions. The words “first”, “second” and the like do not limit the quantity and execution order, and the words “first”, “second” and the like do not limit the difference.
  • words such as “exemplary” or “for example” are used to represent examples, illustrations or illustrations, and any embodiment or design solution described as “exemplary” or “for example” should not be construed are preferred or advantageous over other embodiments or designs.
  • the use of words such as “exemplary” or “such as” is intended to present the relevant concepts in a specific manner to facilitate understanding.
  • At least one (species) may also be described as one (species) or multiple (species), and the multiple (species) may be two (species), three (species), four (species) ) or more (species), which is not limited in this application.
  • an autoencoder is a neural network that can be applied end-to-end (eg, network devices and end devices). It can jointly optimize the transceiver ends to improve the overall performance.
  • the network device and the terminal device can jointly train the neural networks on both sides to obtain a better algorithm. Therefore, how to make artificial intelligence operate effectively in mobile communication networks is a problem worth studying.
  • AI Artificial intelligence
  • Neural network As an important branch of artificial intelligence, it is a network structure that imitates the behavioral characteristics of animal neural networks for information processing.
  • the structure of a neural network consists of a large number of nodes (or neurons) connected to each other.
  • the neural network is based on a specific operation model, and achieves the purpose of processing information by learning and training the input information.
  • a neural network consists of input layer, hidden layer and output layer.
  • the input layer is responsible for receiving input signals
  • the output layer is responsible for outputting the calculation results of the neural network
  • the hidden layer is responsible for complex functions such as feature expression.
  • the function of the hidden layer is characterized by the weight matrix and the corresponding activation function.
  • a deep neural network is generally a multi-layer structure. Increasing the depth and width of a neural network can improve its expressiveness and provide more powerful information extraction and abstract modeling capabilities for complex systems.
  • the depth of a neural network can be expressed as the number of layers of the neural network. For one of the layers, the width of the neural network can be expressed as the number of neurons included in the layer.
  • DNNs can be constructed in various ways, including, but not limited to, recurrent neural networks (RNNs), convolutional neural networks (CNNs), and fully connected neural networks.
  • RNNs recurrent neural networks
  • CNNs convolutional neural networks
  • fully connected neural networks including, but not limited to, recurrent neural networks (RNNs), convolutional neural networks (CNNs), and fully connected neural networks.
  • Training refers to the process of processing a model (or training a model). During this process, the model learns to perform a specific task by optimizing parameters in the model, such as weights.
  • the embodiments of the present application are applicable to, but not limited to, one or more of the following training methods: supervised learning, unsupervised learning, reinforcement learning, transfer learning, and the like.
  • Supervised learning is trained using a set of training samples that have been correctly labeled. Among them, having been correctly labeled means that each sample has an expected output value.
  • unsupervised learning refers to a method that automatically classifies or groups incoming data without given pre-labeled training samples.
  • Inference refers to performing data processing using a trained model (the trained model may be referred to as an inference model).
  • the actual data is input into the inference model for processing, and the corresponding inference result is obtained.
  • Inference may also be referred to as prediction or decision, and inference results may also be referred to as prediction results, decision results, and the like.
  • FIG. 2 is a schematic diagram of the hybrid analog-digital large-scale antenna system architecture provided by the present application.
  • the antenna system architecture shown in Figure 2 is configured in a network device, the front end of the antenna system is configured with a uniform linear array (ULA) of N antennas, and the number R of back-end RF links is much smaller than the antennas The number is N, and the network device simultaneously serves K terminal devices through the antenna system.
  • the network device performs uplink channel estimation from the terminal device to the network device, the network device processes the signal received by the antenna array in the analog domain and the digital domain, and the processed received signal Y is:
  • W BB is the combination matrix of the digital domain, and the dimension is R ⁇ R.
  • W RF is an analog domain combination matrix with dimension R ⁇ N.
  • N is additive white Gaussian noise.
  • p k is the pilot sequence sent by the kth terminal device, the dimension is 1 ⁇ L p , and L p is the length of the pilot sequence.
  • h k is the uplink channel from the kth terminal device to the network device, and the dimension is N ⁇ 1.
  • a channel is formed by the superposition of multiple paths:
  • N p is the number of paths
  • ⁇ i is the complex gain of the ith path
  • ⁇ i is the arrival angle of the ith path reaching the network device.
  • the channel information in the angular domain has a structure. Sparse properties. Specifically, the number of non-zero elements of the channel information in the angle domain is much smaller than the number of antennas, and is concentrated in a small range. Using this structural sparse feature, the channel data processing performance can be improved. For example, it can help improve channel estimation performance and reduce estimation overhead.
  • the angle domain channel information can be obtained by transforming the original channel through discrete Fourier transform (DFT):
  • the received signal corresponding to the normalized transmitted signal of the kth terminal device can be extracted as:
  • the channel estimation problem of a large-scale antenna system is to design an appropriate measurement matrix ⁇ and a channel estimation algorithm according to the received signal y k to estimate the channel information in the angle domain.
  • the error between it and the real angle domain channel information x k is as small as possible.
  • This application proposes that by using the sparse characteristics of channel information in the angle domain, the coverage of the signal receiving device is divided into multiple coverage areas, and the signal receiving device adopts the corresponding artificial intelligence inference model according to the coverage area where the signal transmitting device is located. , inferring the received signal to obtain the channel information. The accuracy of the received signal can be improved. Thereby improving the reliability of communication.
  • FIG. 3 is a schematic flowchart of a communication method 300 provided by an embodiment of the present application.
  • the communication method shown in FIG. 3 is performed by the first device or an apparatus configured with the first device, and the following description will be given by taking the communication method performed by the first device as an example.
  • the first device is a signal receiving device, which may be a network device or a terminal device, which is not limited in this application.
  • the first device acquires the first signal from the second device.
  • the second device sends a reference signal (the reference signal may also be referred to as a pilot signal) to the first device, and the first device can receive the reference signal after channel transmission through the array antenna.
  • the reference signal is the first signal.
  • the present application is not limited to this.
  • the first device obtains first channel information by inferring the first signal according to the first model.
  • the coverage of the first device is divided into multiple coverage areas, and each coverage area in the multiple coverage areas corresponds to a channel information inference model.
  • the second device is in a first coverage area among the multiple coverage areas of the first device.
  • the first model is a channel information inference model corresponding to the first coverage area. Since environmental data at similar locations are usually correlated, using corresponding channel information inference models according to different coverage areas can improve the accuracy of channel information acquisition, thereby more accurately restoring received signals.
  • the multiple coverage areas may be multiple coverage areas within the same cell, or may be coverage areas within multiple cells, which are not limited in this application.
  • the first channel information is angle domain channel information.
  • the first device obtains the first channel information by inferring the first signal according to the first model.
  • the first device may obtain the measurement information by inference according to the first model, and then process the first signal according to the measurement information. Get the second signal. After obtaining the second signal, the first device infers the second signal to obtain the first channel information according to the first model.
  • the first model may be a neural network model.
  • the first model is an auto-encoder model in a neural network model
  • the auto-encoder model includes an encoder part and a decoder part.
  • the first device can obtain the measurement through reasoning of the encoder part of the first model matrix ⁇ (that is, an example of measurement information), and through this measurement matrix ⁇ , the digital combination matrix W BB is obtained according to the following formula,
  • the W BB is the digital domain combination matrix corresponding to the signal processing the first coverage area.
  • the pseudo-inverse matrix representing W RF B may be, for example, a Moore-Penrose pseudo-inverse matrix.
  • the present application is not limited to this. Since the analog domain combining matrix W RF is determined by the phase of the phase shifters, and the phase shifters that receive signals from each communication device are shared. Therefore, when the first device adopts an omnidirectional antenna, the W RF can be an arbitrary row full-rank matrix. Alternatively, when the first device is not an omnidirectional antenna, the W RF may be an analog domain combining matrix corresponding to the trained analog domain beam.
  • the first device receives the first signal, processes the first signal through the phase shifter, and uses the W BB obtained by measuring the matrix ⁇ to process the second signal to obtain the second signal.
  • the first device then infers the second signal according to the decoder part of the first model to obtain estimated channel information in the angle domain
  • the accuracy of estimating the angle channel can be improved. Decoding the data signal according to the angle domain channel information with higher accuracy can reduce the probability of decoding failure and improve the reliability of communication.
  • the channel information inference models corresponding to the multiple coverage areas of the first device may be preconfigured in the first device.
  • the first device uses the training signal to train the training model to obtain a trained training model, and the trained training model is the above-mentioned reasoning model.
  • the first device may collect a first training signal set for training, the first training signal set includes a plurality of training signals, each training signal includes data and a label, and the data may also be referred to as a sample , the data can be the angle domain channel information with noise, and the label is the angle domain channel information, which is used to calculate the accuracy of the training model output.
  • the angle domain channel information may be calculated based on an actual received signal of the first device, where the actual received signal is a received signal sent by a communication device in the first coverage area, or may be based on a channel model of the first coverage area Signals synthesized from computer simulations.
  • the present application is not limited to this.
  • FIG. 4 is a schematic diagram of a training process of an auto-encoding model provided by an embodiment of the present application.
  • the self-encoder model shown in Figure 4 includes an encoder part and a decoder part, wherein the encoder part may include convolutional layer 1 and convolutional layer 2, and the decoder part includes convolutional layer 3, convolutional layer 4. Fully connected layer 1 and fully connected layer 2.
  • the input dimensions of the convolutional layer 1 and the convolutional layer 2 of the encoder part are both 2N ⁇ 1
  • the dimension of the convolution kernel (size) is N ⁇ 1
  • the convolution strides (strides) is N, where N is number of antennas.
  • the convolution layer 1 includes a total of R convolution kernels (for example, the convolution kernel can be implemented by a filter), and the parameters of each convolution kernel may be different, where R is the number of radio frequency channels.
  • the output dimension of each convolution kernel of convolution layer 1 is 2, and the first dimension realizes the real part and the input vector of the rth row of the measurement matrix.
  • the real part is multiplied, and the second dimension realizes the multiplication of the real part of the rth row of the measurement matrix and the imaginary part of the input vector;
  • the output dimension of each convolution kernel of convolution layer 2 is 2, and the first dimension realizes the measurement matrix.
  • the imaginary part of the rth row is multiplied by the real part of the input vector, and the second dimension realizes that the imaginary part of the rth row of the measurement matrix is multiplied by the imaginary part of the input vector and the result of the multiplication takes a negative value, where r is less than or equal to R.
  • the result of the first dimension output by convolution layer 1 and the result of the second dimension output by convolution layer 2 are added to obtain the real part of the measured data y; the result of the second dimension output by convolution layer 1 is the same as the result of convolution
  • the results of the first dimension output by layer 2 are added to obtain the imaginary part of the measured data y.
  • the measured training signal y is input to the decoder part.
  • the input of the convolutional layer 3 in the decoder part is 2 sets of vectors with dimension R ⁇ 1, which are equivalent to 2 channels, and each channel consists of 32 dimensions.
  • the 3 ⁇ 1 convolution kernel is calculated, and the calculation results of the 2 channels are added as the output.
  • the input of convolutional layer 4 is 32 sets of vectors with dimension R ⁇ 1, equivalent to 32 channels, each channel is calculated by 32 convolution kernels with dimension 3 ⁇ 1, and the calculation results of 32 channels are added as output , the output dimension is 32 sets of vectors with dimension R ⁇ 1, the output is reconstructed into a one-dimensional vector and sent to the fully connected layer, and the fully connected layer 2 outputs the angle domain channel estimated from the measured data y with the estimated angle domain channel information the mean square error (MSE) with the label x,
  • MSE mean square error
  • the first device adjusts the weights of the convolution layer and the fully connected layer according to the mean square error calculation result.
  • the mean square error MSE with x is used as the loss function, and the stochastic gradient descent algorithm is used to train the autoencoder model until the loss function converges, and the trained model is obtained as the angle domain channel inference model corresponding to the first coverage area (that is, the first model). ).
  • the present application is not limited to this.
  • the first device trains the channel information models of different coverage areas based on the training data in the channel environment, which increases the matching degree between the model and the coverage area and improves the accuracy of channel estimation by the model.
  • the division of the multiple coverage areas of the first device may include, but is not limited to, the following manners.
  • one of the multiple coverage areas is a sector covered by a central angle with the first device as the center.
  • FIG. 5 is a schematic diagram of multiple coverage areas of the first device.
  • Each coverage area corresponds to a central angle range, and the central angle may also be referred to as a horizontal azimuth angle.
  • Table 1 is an example of the range of horizontal azimuth angles of multiple coverage areas. As shown in Table 1, the horizontal azimuth range included in the coverage area 1 is 0 ⁇ 1 , and the horizontal azimuth angle range included in the coverage area 2 is ⁇ 1 ⁇ 2 and so on. However, the present application is not limited to this.
  • the central angles of any two coverage areas in the multiple coverage areas may be equal in size. For example, if the central angle of each coverage area is ⁇ , the coverage area of the first device may include: coverage area.
  • ⁇ may be 5°, 10°, etc., but the present application is not limited thereto.
  • the sizes of the central angles of the two coverage areas in the multiple coverage areas may be unequal.
  • the first device may divide the range of the central angle with greater channel correlation into the same coverage area according to the channel characteristics. Therefore, the multiple coverage areas The size of the coverage center angle can be different for the two coverage areas in .
  • one coverage area in the multiple coverage areas belongs to a sector covered by a central angle with the first device as the center, wherein any two coverage areas belong to different sectors, or any two coverage areas are different from all the coverage areas.
  • the distances between the first devices are different.
  • FIG. 6 is another schematic diagram of multiple coverage areas of a first device.
  • any two coverage areas belong to different sectors, or the distances between any two coverage areas and the first device are different.
  • the coverage area 1 and the coverage area 2 belong to the same sector, but the distance between the coverage area 1 and the first device is different from the distance between the coverage area 2 and the first device.
  • the distance from the coverage area 2 to the first device is the same as the distance from the coverage area N to the first device, but the coverage area 2 and the coverage area N belong to different sectors.
  • Table 2 is an example of the range of the horizontal azimuth angle ⁇ of the coverage area and the range of the distance d.
  • the coverage area 1 includes a distance to the first device in a range of 0 ⁇ d ⁇ d 1 and a horizontal azimuth in a range of 0 ⁇ 1 .
  • the coverage area 2 includes distances to the first device in the range d 1 ⁇ d ⁇ d 2 and horizontal azimuth angles in the range 0 ⁇ 1 .
  • the coverage area 1 includes the distance to the first device in the range of 0 ⁇ d ⁇ d 1 and the horizontal azimuth in the range of ⁇ 1 ⁇ 2 and so on.
  • Table 2 is only an example, and the coverage area may also be determined by using the distance range table shown in Table 3 and the horizontal azimuth angle range table shown in Table 4. This application does not limit this.
  • one coverage area in the multiple coverage areas is an area determined by a horizontal azimuth angle and a vertical azimuth angle with the first device as the center.
  • FIG. 7 is a schematic diagram of a coverage area.
  • the coverage area shown in Figure 7 is defined by the horizontal azimuth ⁇ and the vertical azimuth a defined area.
  • Horizontal azimuth of coverage area The range of , may be as shown in Table 4, and the range of the vertical azimuth may be as shown in Table 5.
  • the range of the horizontal azimuth angle ⁇ of the coverage area 3 can be determined from Table 4 as ⁇ 1 ⁇ 2
  • the vertical azimuth angle of the coverage area 3 can be determined from Table 5
  • the range is
  • the vertical azimuth angle of the coverage area the vertical azimuth angle range of the first device can be divided into a range of 0 to 180° of the vertical azimuth angle of multiple coverage areas.
  • the angular range of 0 to 180° delimits the vertical azimuth range of the multiple coverage areas.
  • the vertical azimuth angle range of 0 to 360° may be divided into vertical azimuth angle ranges of multiple coverage areas.
  • the vertical azimuth angle is about 90°, it may be the coverage area for communication with equipment such as aircraft and satellites.
  • the present application is not limited to this.
  • the vertical azimuth angles may be uniformly divided into multiple coverage areas, or the vertical azimuth angles of multiple coverage areas may be divided according to different channel characteristics, which are not limited in this application.
  • the horizontal azimuth angle of the coverage area is similar to that in the first mode, and is not repeated here for brevity.
  • the following introduces an optional implementation manner in which the first device determines that the second device is in the first coverage area.
  • FIG. 8 is another schematic flowchart of the communication method provided by the present application.
  • the first device sends first indication information to the second device, where the first indication information is used to indicate multiple coverage areas of the first device, where the multiple coverage areas correspond to multiple channel information inference models.
  • the second device obtains the first indication information from the first device.
  • the first indication information may be broadcast information. That is, the first indication information can be received by a plurality of devices that communicate with the first device.
  • the first indication information includes positioning information of a boundary of each of the multiple coverage areas.
  • the positioning information may be global positioning system (global positioning system, GPS) information, and the multiple coverage areas of the first device may be as shown in FIG. 5 .
  • the first indication information includes an identifier (ID) of each coverage area and GPS information of two boundaries of the coverage area corresponding to the ID.
  • the first indication information includes ⁇ GPS0, GPS1, ID1, GPS2, GPS3, ID2 ⁇ , etc., wherein the first two GPS information of the ID are the GPS information of the two boundaries of the coverage area identified by the ID, and the ID1 identified
  • the GPS information of the two boundaries of the coverage area 1 are GPS0 and GPS1, respectively.
  • the second device can determine the horizontal azimuth of the boundary of the first coverage area according to GPS0 and GPS1, respectively. Then the first coverage area The range is between the horizontal azimuth determined by GPS0 and GPS1.
  • the second device may determine each coverage area based on the first indication information in this manner. Thus, the coverage area where the second device is located, that is, the first coverage area is determined.
  • the first indication information further includes identification information of each coverage area and/or identification information of a model corresponding to each coverage area.
  • the second device After the second device determines the first coverage area in which it is located, it may indicate the identification information of the first coverage area or the identification information of the model corresponding to the first coverage area through the second indication information in S820.
  • the second device sends second indication information to the first device, where the second indication information is used to indicate that the second device is in the first coverage area or to indicate the first model.
  • the first device receives the second indication information from the second device.
  • the second device may notify the first device of identification information of the first coverage area through the second indication information in S820.
  • the first device is caused to determine, according to the identification information of the first coverage area, the signal of the second device acquired by processing the first model corresponding to the first coverage area.
  • the second device determines the identification information of the model corresponding to the coverage area in S810, then the second device can notify the first device of the identification information of the first model through the second indication information.
  • the first device may determine, according to the identifier of the first model in the second indication information, to process the acquired signal of the second device by using the first model.
  • the first device may acquire location information from the second device, and determine that the second device is in the first coverage area according to the location information.
  • the performance of the method provided by the present application and the traditional channel estimation method is compared through experiments.
  • SNR signal noise ratio
  • the experimental result is the normalized mean square error (NMSE) of the channel estimation.
  • NMSE normalized mean square error
  • the received signal is measured (or called as processed) according to the measurement matrix ⁇ A and ⁇ B , respectively, and compared through sparse bayesian learning (SBL), orthogonal matching pursuit (OMP) , Variational Bayesian inference (variational bayesian inference, VBI) algorithm NMSE results of channel estimation for the measured received signal.
  • SBL sparse bayesian learning
  • OMP orthogonal matching pursuit
  • VBI Variational Bayesian inference
  • VBI Variational bayesian inference
  • the channel estimation NMSE results based on ⁇ B measurement are all smaller than the channel based on ⁇ A measurement.
  • Estimated NMSE It can be seen that the channel estimation performance when the measurement matrix ⁇ B obtained by the first model of the present application is applied is better than the channel estimation performance when the Zadoff-Chu measurement matrix is applied.
  • the NMSE results of the channel information obtained by the first model inference when using the same measurement matrix are significantly better than the channel estimation performance of the traditional channel estimation algorithms SBL, OMP, and VBI.
  • the method provided by the present application has a remarkable performance in channel estimation. outperforms traditional algorithms.
  • the present application proposes that by using the sparse characteristics of channel information in the angle domain, the coverage of the signal receiving device is divided into multiple coverage areas, and the signal receiving device adopts the artificial intelligence inference model corresponding to the coverage area according to the coverage area where the signal transmitting device is located. , inferring the received signal to obtain the channel information. The accuracy of the received signal can be improved. Thereby improving the reliability of communication.
  • each network element may include a hardware structure and/or a software module, and implement the above functions in the form of a hardware structure, a software module, or a hardware structure plus a software module. Whether one of the above functions is performed in the form of a hardware structure, a software module, or a hardware structure plus a software module depends on the specific application and design constraints of the technical solution.
  • FIG. 9 is a schematic block diagram of a communication apparatus provided by an embodiment of the present application.
  • the communication apparatus 900 may include a processing unit 910 and a transceiver unit 920 .
  • the communication apparatus 900 may correspond to the first device in the above method embodiments, or a chip configured (or used for) in the first device, or any other device capable of implementing the first device.
  • the first device may be a network device or a terminal device.
  • the communication apparatus 900 may correspond to the first device in the methods 300 and 800 in the embodiments of the present application, and the communication apparatus 900 may include a device for executing the first device in the methods 300 and 800 in FIG. 3 and FIG. 8 . method unit.
  • each unit in the communication device 900 and the other operations and/or functions mentioned above are to implement the corresponding processes of the methods 300 and 800 in FIG. 3 and FIG. 8 , respectively.
  • the transceiver unit 920 is used to acquire the first signal from the second device; the processing unit 910 is used to obtain the first signal according to the first model , infer the first signal to obtain the first channel information.
  • the processing unit 910 is specifically configured to infer measurement information according to the first model; process the first signal according to the measurement information to obtain the second signal; and infer according to the first model The second signal obtains the first channel information.
  • the processing unit 910 is specifically configured to obtain the measurement information by inference according to the encoder part, and obtain the first channel information by inferring the second signal according to the decoder part.
  • the transceiver unit 920 before the transceiver unit 920 acquires the first signal from the second device, the transceiver unit 920 is further configured to acquire a first training set of the first coverage area, the first training set Including a plurality of training signals; the processing unit 910 is further configured to train a training model based on the first training set to obtain the first model.
  • the training model includes an encoder part and a decoder part
  • the processing unit 910 is specifically configured to input the angle domain channel information samples in the first training set into the encoder part to obtain a target encoded signal , and input the target encoded signal into the decoder part to obtain the target decoded signal.
  • the processing unit 910 further obtains parameter information of the training model based on the angle domain channel information label and the target decoded signal, and adjusts the parameters of the encoder part and the decoder part according to the parameter information.
  • the transceiver unit 920 is further configured to acquire location information from the second device; the processing unit 910 is further configured to determine that the second device is in the first coverage area according to the location information.
  • the transceiver unit 920 in the communication apparatus 900 may be an input/output interface or circuit of the chip, and the The processing unit 910 may be a logic circuit in a chip.
  • the communication apparatus 900 may correspond to the second device in the above method embodiments, for example, or a chip configured (or used in) the second device, or other devices capable of implementing the first 2. Apparatus, module, circuit or unit, etc. of the method performed by the apparatus.
  • the second device is a device that communicates with the first device through a wireless link, and optionally, the second device may be a network device or a terminal device.
  • the communication apparatus 900 may correspond to the second device in the methods 300 and 800 according to the embodiments of the present application, and the communication apparatus 900 may include the second device for executing the methods 300 and 800 in FIG. 3 and FIG. 8 .
  • each unit in the communication device 900 and the other operations and/or functions mentioned above are to implement the corresponding processes of the methods 300 and 800 in FIG. 3 and FIG. 8 , respectively.
  • the transceiver unit 920 is used to obtain the first indication information; the processing unit 910 is used to determine according to the first indication information The first coverage area, the transceiver unit 920, is further configured to send second indication information.
  • the transceiver unit 920 in the communication apparatus 900 may be an input/output interface or circuit of the chip, and the The processing unit 910 may be a logic circuit in a chip.
  • the communication device 900 may further include a storage unit 930, the storage unit 930 may be used to store instructions or data, and the processing unit 910 may execute the instructions or data stored in the storage unit, so as to enable the communication device to implement corresponding operations .
  • the transceiver unit 920 in the communication apparatus 900 may be implemented through a communication interface (such as a transceiver or an input/output interface), for example, may correspond to the transceiver 1010 in the communication device 1000 shown in FIG. 10 .
  • the processing unit 910 in the communication apparatus 900 may be implemented by at least one processor, for example, may correspond to the processor 1020 in the communication device 1000 shown in FIG. 10 .
  • the processing unit 910 in the communication device 900 may also be implemented by at least one logic circuit.
  • the storage unit 930 in the communication apparatus 900 may correspond to the memory in the communication apparatus 1000 shown in FIG. 10 .
  • FIG. 10 is a schematic structural diagram of a terminal device 1000 provided by an embodiment of the present application.
  • the communication device 1000 may correspond to the first device in the above method embodiment.
  • the first device 1000 includes a processor 1020 and a transceiver 1010 .
  • the first device 1000 further includes a memory.
  • the processor 1020, the transceiver 1010 and the memory can communicate with each other through an internal connection path to transmit control and/or data signals.
  • the memory is used for storing a computer program, and the processor 1020 is used for executing the computer program in the memory to control the transceiver 1010 to send and receive signals.
  • the communication device 1000 shown in FIG. 10 can implement the process involving the first device in the method embodiments shown in FIG. 3 and FIG. 8 .
  • the operations and/or functions of each module in the first device 1000 are respectively to implement the corresponding processes in the foregoing method embodiments.
  • the communication device 1000 may correspond to the second device in the above method embodiment.
  • the second device 1000 includes a processor 1020 and a transceiver 1010 .
  • the second device 1000 further includes a memory.
  • the processor 1020, the transceiver 1010 and the memory can communicate with each other through an internal connection path to transmit control and/or data signals.
  • the memory is used for storing a computer program, and the processor 1020 is used for executing the computer program in the memory to control the transceiver 1010 to send and receive signals.
  • the communication device 1000 shown in FIG. 10 can implement the process involving the second device in the method embodiments shown in FIG. 3 and FIG. 8 .
  • the operations and/or functions of each module in the second device 1000 are respectively to implement the corresponding processes in the foregoing method embodiments.
  • the first device may correspond to a terminal device in the system as shown in FIG. 1
  • the second device may correspond to a network device in the system as shown in FIG. 1
  • the first device may correspond to a network device in the system as shown in FIG. 1
  • the second device may correspond to a terminal device in the system as shown in FIG. 1 .
  • the above-mentioned processor 1020 and the memory can be combined into a processing device, and the processor 1020 is configured to execute the program codes stored in the memory to realize the above-mentioned functions.
  • the memory may also be integrated in the processor 1020, or be independent of the processor 1020.
  • the processor 1020 may correspond to the processing unit in FIG. 9 .
  • the transceiver 1010 described above may correspond to the transceiver unit in FIG. 9 .
  • the transceiver 1010 may include a receiver (or receiver, receiving circuit) and a transmitter (or transmitter, transmitting circuit). Among them, the receiver is used for receiving signals, and the transmitter is used for transmitting signals.
  • the communication device 1000 shown in FIG. 10 can implement the processes involving the terminal device in the method embodiments shown in FIG. 3 and FIG. 8 .
  • the operations and/or functions of each module in the terminal device 1000 are respectively to implement the corresponding processes in the foregoing method embodiments.
  • An embodiment of the present application further provides a processing apparatus, including a processor and a (communication) interface; the processor is configured to execute the method in any of the above method embodiments.
  • the above-mentioned processing device may be one or more chips.
  • the processing device may be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on chip (SoC), or a It is a central processing unit (CPU), a network processor (NP), a digital signal processing circuit (DSP), or a microcontroller (microcontroller unit). , MCU), it can also be a programmable logic device (PLD) or other integrated chips.
  • FPGA field programmable gate array
  • ASIC application specific integrated circuit
  • SoC system on chip
  • MCU microcontroller unit
  • MCU programmable logic device
  • PLD programmable logic device
  • the present application further provides a computer program product, the computer program product includes: computer program code, when the computer program code is executed by one or more processors, the computer program code including the processor
  • the apparatus executes the methods in the embodiments shown in FIG. 3 and FIG. 8 .
  • the technical solutions provided in the embodiments of the present application may be implemented in whole or in part by software, hardware, firmware, or any combination thereof.
  • software When implemented in software, it can be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, a network device, a terminal device, a core network device, a machine learning device, or other programmable devices.
  • the computer instructions may be stored in or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server or data center Transmission to another website site, computer, server, or data center by wire (eg, coaxial cable, optical fiber, digital subscriber line, DSL) or wireless (eg, infrared, wireless, microwave, etc.).
  • the computer-readable storage medium can be any available media that can be accessed by a computer, or a data storage device such as a server, data center, etc. that includes one or more available media integrated.
  • the usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, digital video discs (DVDs)), or semiconductor media, and the like.
  • the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores program codes, and when the program codes are executed by one or more processors, the processing includes the processing
  • the device of the controller executes the method in the embodiment shown in FIG. 3 and FIG. 8 .
  • the present application further provides a system, which includes the aforementioned one or more first devices. Also the system may further comprise one or more of the aforementioned second devices.
  • the first device may be a network device or a terminal device
  • the second device may be a device that communicates with the first device through a wireless link.
  • the disclosed system, apparatus and method may be implemented in other manners.
  • the apparatus embodiments described above are only illustrative.
  • the division of the units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented.
  • the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Power Engineering (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请提供了一种基于空间划分的数据处理方法和通信装置。该方法包括：第一设备获取来自第二设备的第一信号；该第一设备根据第一模型，推理该第一信号得到第一信道信息，该第二设备处于该第一设备的多个覆盖区域中的第一覆盖区域，该第一模型为该第一覆盖区域对应的信道信息推理模型。以期提高通信的可靠性。

Description

基于空间划分的数据处理方法和通信装置
本申请要求于2021年02月22日提交中国专利局、申请号为202110197721.6、申请名称为“基于空间划分的数据处理方法和通信装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信领域，并且更具体地，涉及一种基于空间划分的数据处理方法和通信装置。
背景技术
作为第五代移动通信核心技术之一，大规模天线具有高频谱效率、高能量效率、高可靠性等诸多优点。为了节省硬件与能耗开销，大规模天线系统在实践中，通常采用混合模拟数字架构，即基站端配备的射频链路数小于天线数。天线处的接收信号经由移相器组成的模拟域处理后再经过数字域处理。大规模天线的应用对信号处理带来了很大的挑战。
发明内容
本申请提供了一种基于空间划分的数据处理方法和通信装置，能够提高通信的可靠性。
第一方面，提供了一种通信方法，该方法可以由第一设备或配置于(或用于)第一设备的模块(如芯片)执行，以下以该方法由网络设备执行为例进行说明。
该方法包括：第一设备获取来自第二设备的第一信号；该第一设备根据第一模型，推理该第一信号得到第一信道信息，该第二设备处于该第一设备的多个覆盖区域中的第一覆盖区域，该第一模型为该第一覆盖区域对应的信道信息推理模型。
根据上述方案，信号接收设备的覆盖范围划分为多个覆盖区域，信号接收设备根据信号发送设备所处的覆盖区域，采用该覆盖区域相对应的人工智能推理模型，推理接收信号得到信道信息。能够提高接收信号的准确度。从而提高通信的可靠性。
结合第一方面，在第一方面的某些实现方式中，该第一信道信息为角度域信道信息。
根据上述方案，利用角度域信道的结构稀疏特性，可以提升信道数据处理性能、降低估计开销。
结合第一方面，在第一方面的某些实现方式中，该根据第一模型，推理该第一信号得到第一信道信息，包括：根据该第一模型，推理得到测量信息；根据该测量信息处理该第一信号，得到第二信号；根据该第一模型，推理该第二信号得到该第一信道信息。
根据上述方案，第一设备可以通过第一模型可以推理用于测量信号的测量信息，进而根据该第一模型还可以推理得到信号经历的信道信息。能够提升信道估计性能、降低估计开销。
结合第一方面，在第一方面的某些实现方式中，该第一模型为自编码器模型，该第一模型包括编码器部分和译码器部分，该根据该第一模型，推理得到测量信息，包括：该第一设备根据该编码器部分，推理得到测量信息；以及，根据该第一模型，推理得到测量信息，包括：该第一设备根据该译码器部分，推理该第二信号得到该第一信道信息。
结合第一方面，在第一方面的某些实现方式中，该方法还包括：第一设备获取第一覆盖区域对应的训练信号集合，该第一训练集中包括多个训练信号，第一设备基于训练信号集合，对训练模型进行训练，得到第一模型。
根据上述方案，第一设备基于信道环境中的训练数据对不同覆盖区域的信道信息模型进行训练，增加了模型与覆盖区域的匹配度，提高了模型进行信道估计的准确度。
结合第一方面，在第一方面的某些实现方式中，该训练模型包括编码器部分和译码器部分，该根据训练信号集合，对训练模型进行训练，包括：将该第一训练集中的角度域信道信息样本输入该编码器部分，得到目标编码信号；将该目标编码信号输入该译码器部分，得到目标译码信号；根据该角度域信道信息标签和该目标译码信号，得到训练模型的参数信息；根据该参数信息，调节该编码器部分的参数和该译码器部分的参数。结合第一方面，在第一方面的某些实现方式中，该多个覆盖区域中的一个覆盖区域属于以该第一设备为圆心的一个圆心角覆盖的扇区，该第一覆盖区域属于第一圆心角覆盖的扇区，该多个覆盖区域中的任意两个覆盖区域所属的扇区不同，或该多个覆盖区域中的任意两个覆盖区域与该第一设备的间距不同。
根据上述方案，由于相近位置环境数据通常具有相关性，将相近区域划分为同一覆盖区域。根据不同覆盖区域采用相应的信道信息推理模型能够提高信道信息获取的准确度，从而较准确地还原接收信号。
结合第一方面，在第一方面的某些实现方式中，该方法还包括：该第一设备获取来自该第二设备的位置信息；该第一设备根据该位置信息确定该第二设备处于该第一覆盖区域。
根据上述方案，第一设备可以根据第二设备发送的位置信息确定第二设备处于第一覆盖区域。从而采用第一覆盖区域相对应的第一模型处理获取的第二设备的信号。提高信道估计的准确性，从而提高通信的可靠性。
结合第一方面，在第一方面的某些实现方式中，该方法还包括：该第一设备发送第一指示信息，该第一指示信息用于指示该多个覆盖区域，该多个覆盖区域与多个信道信息推理模型对应；该第一设备获取第二指示信息，该第二指示信息用于指示该第二设备处于该第一覆盖区域或指示该第一模型。
根据上述方案，第一设备可以通过第一指示信息通知第二设备多个覆盖区域，由第二设备确定其所属的第一覆盖区域并通过第二指示信息通知第一设备，以便第一设备可以采用第一覆盖区域相对应的第一模型处理获取的第二设备的信号。提高信道估计的准确性，从而提高通信的可靠性。
结合第一方面，在第一方面的某些实现方式中，该第一指示信息为广播信息。
结合第一方面，在第一方面的某些实现方式中，第一设备为网络设备或终端设备。
第二方面，提供了一种通信方法，该方法可以由第二设备或配置于(或用于)第二设备的模块(如芯片)执行，以下以该方法由终端设备执行为例进行说明。
该方法包括:第二设备获取第一指示信息，该第一指示信息用于指示多个覆盖区域，该多个覆盖区域与多个信道信息推理模型相对应；该第二设备发送第二指示信息，该第二指示信息用于指示该第二设备处于该第一覆盖区域或指示第一模型，该第一覆盖区域为该多个覆盖区域中的一个，该第一模型为该第一覆盖区域对应的模型。
结合第二方面，在第二方面的某些实现方式中，该多个覆盖区域中的一个覆盖区域属于以第一设备为圆心的一个圆心角覆盖的扇区，该第一覆盖区域属于第一圆心角覆盖的扇 区，该多个覆盖区域中的两个覆盖区域所属的扇区不同，或该多个覆盖区域中的两个覆盖区域与该第一设备的间距不同。
结合第二方面，在第二方面的某些实现方式中，该第一指示信息包括以下一项或多项：该多个覆盖区域中每个覆盖区域的参考位置信息、该多个覆盖区域中每个覆盖区域的标识信息或该多个信道信息推理模型中每个模型的标识信息。
结合第二方面，在第二方面的某些实现方式中，该第二指示信息中包括该第一覆盖区域的标识信息或该第一模型的标识信息。
第三方面，提供了一种通信装置，该装置是第一设备或配置于(或用于)第一设备的模块(如芯片)。
该通信装置包括：收发单元，用于获取来自第二设备的第一信号；
处理单元，用于根据第一模型，推理第一信号得到第一信道信息，该第二设备处于该第一设备的多个覆盖区域中的第一覆盖区域，该第一模型为该第一覆盖区域对应的信道信息推理模型。
结合第三方面，在第三方面的某些实现方式中，该第一信道信息为角度域信道信息。
结合第三方面，在第三方面的某些实现方式中，该处理单元具体用于：根据该第一模型，推理得到测量信息；根据该测量信息处理该第一信号，得到该第二信号；根据该第一模型，推理该第二信号得到该第一信道信息。
结合第三方面，在第三方面的某些实现方式中，该第一模型为自编码器模型，该第一模型包括编码器部分和译码器部分，该处理单元具体用于，
根据该编码器部分，推理得到测量信息；以及，根据该译码器部分，推理该第二信号得到该第一信道信息。
结合第三方面，在第三方面的某些实现方式中，在该收发单元获取来自该第二设备的该第一信号之前，该收发单元还用于获取第一覆盖区域的第一训练集，该第一训练集中包括多个训练信号；该处理单元还用于基于该第一训练集对训练模型进行训练，得到该第一模型。
结合第三方面，在第三方面的某些实现方式中，该训练模型包括编码器部分和译码器部分，该处理单元具体用于：
将该第一训练集中的角度域信道信息样本输入该编码器部分，得到目标编码信号；将该目标编码信号输入该译码器部分，得到目标译码信号；根据该角度域信道信息标签和该目标译码信号，得到训练模型的参数信息；以及，根据该参数信息，调节该编码器部分的参数和该译码器部分的参数。
结合第三方面，在第三方面的某些实现方式中，该多个覆盖区域中的一个覆盖区域属于以该第一设备为圆心的一个圆心角覆盖的扇区，该第一覆盖区域属于第一圆心角覆盖的扇区，该多个覆盖区域中的任意两个覆盖区域所属的扇区不同，或该多个覆盖区域中的任意两个覆盖区域与该第一设备的间距不同。
结合第三方面，在第三方面的某些实现方式中，该收发单元还用于获取来自该第二设备的位置信息；该处理单元还用于根据该位置信息确定该第二设备处于该第一覆盖区域。
第四方面，提供了一种通信装置，该装置是第二设备或配置于(或用于)第二设备的模块(如芯片)。
该通信装置包括：收发单元，用于获取第一指示信息，该第一指示信息用于指示多个 覆盖区域，该多个覆盖区域与多个信道信息推理模型相对应；处理单元，用于根据该第一指示信息确定该第二设备处于该多个覆盖区域中的第一覆盖区域；该收发单元，还用于发送第二指示信息，该第二指示信息用于指示该第二设备处于该第一覆盖区域或指示第一模型，该第一模型为该第一覆盖区域对应的模型。
结合第四方面，在第四方面的某些实现方式中，该多个覆盖区域中的一个覆盖区域属于以该第一设备为圆心的一个圆心角覆盖的扇区，该第一覆盖区域属于第一圆心角覆盖的扇区，该多个覆盖区域中的两个覆盖区域所属的扇区不同，或该多个覆盖区域中的两个覆盖区域与该第一设备的间距不同。
结合第四方面，在第四方面的某些实现方式中，该第一指示信息包括以下一项或多项：
该多个覆盖区域中每个覆盖区域的参考位置信息、该多个覆盖区域中每个覆盖区域的标识信息或该多个信道信息推理模型中每个模型的标识信息。
结合第四方面，在第四方面的某些实现方式中，该第二指示信息中包括该第一覆盖区域的标识信息或该第一模型的标识信息。
第五方面，提供了一种通信装置，包括处理器。该处理器可以实现上述第一方面以及第一方面中任一种可能实现方式中的方法。可选地，该通信装置还包括存储器，该处理器与该存储器耦合，可用于执行存储器中的指令，以实现上述第一方面以及第一方面中任一种可能实现方式中的方法。可选地，该通信装置还包括通信接口，处理器与通信接口耦合。本申请实施例中，通信接口可以是收发器、管脚、电路、总线、模块或其它类型的通信接口，不予限制。
在一种实现方式中，该通信装置为第一设备。当该通信装置为第一设备时，该通信接口可以是收发器，或，输入/输出接口。
在另一种实现方式中，该通信装置为配置于第一设备中的芯片。当该通信装置为配置于第一设备中的芯片时，该通信接口可以是输入/输出接口，该处理器可以是逻辑电路。
该输入/输出接口用于输入来自第二设备的第一信号；该逻辑电路用于根据第一模型，推理该第一信号得到第一信道信息，该第二设备处于该第一设备的多个覆盖区域中的第一覆盖区域，该第一模型为所述第一覆盖区域对应的信道信息推理模型。
可选地，该收发器可以为收发电路。可选地，该输入/输出接口可以为输入/输出电路。
第六方面，提供了一种通信装置，包括处理器。该处理器可以实现上述第二方面以及第二方面中任一种可能实现方式中的方法。可选地，该通信装置还包括存储器，该处理器与该存储器耦合，可用于执行存储器中的指令，以实现上述第二方面以及第二方面中任一种可能实现方式中的方法。可选地，该通信装置还包括通信接口，处理器与通信接口耦合。
在一种实现方式中，该通信装置为第二设备。当该通信装置为第二设备时，该通信接口可以是收发器，或，输入/输出接口。
在另一种实现方式中，该通信装置为配置于第二设备中的芯片。当该通信装置为配置于第二设备中的芯片时，该通信接口可以是输入/输出接口，该处理器可以是逻辑电路。
该输入/输出接口用于输入第一指示信息，该第一指示信息用于指示多个覆盖区域，该多个覆盖区域与多个信道信息推理模型相对应；该逻辑电路，用于根据该第一指示信息确定该第二设备处于该多个覆盖区域中的第一覆盖区域；该输入/输出接口还用于输出第二指示信息，该第二指示信息用于指示该第二设备处于该第一覆盖区域或指示第一模型，该第一模型为该第一覆盖区域对应的模型。可选地，该收发器可以为收发电路。可选地，该输 入/输出接口可以为输入/输出电路。
第七方面，提供了一种处理器，包括：输入电路、输出电路和处理电路。该处理电路用于通过该输入电路接收信号，并通过该输出电路发射信号，使得该处理器执行第一方面或第二方面以及第一方面或第二方面中任一种可能实现方式中的方法。
在具体实现过程中，上述处理器可以为一个或多个芯片，输入电路可以为输入管脚，输出电路可以为输出管脚，处理电路可以为晶体管、门电路、触发器和各种逻辑电路等。输入电路所接收的输入的信号可以是由例如但不限于接收器接收并输入的，输出电路所输出的信号可以是例如但不限于输出给发射器并由发射器发射的，且输入电路和输出电路可以是同一电路，该电路在不同的时刻分别用作输入电路和输出电路。本申请实施例对处理器及各种电路的具体实现方式不做限定。
第八方面，提供了一种计算机程序产品，该计算机程序产品包括：计算机程序(也可以称为代码，或指令)，当该计算机程序被运行时，使得计算机执行上述第一方面或第二方面以及第一方面或第二方面中任一种可能实现方式中的方法。
第九方面，提供了一种计算机可读存储介质，该计算机可读存储介质存储有计算机程序(也可以称为代码，或指令)当其在计算机上运行时，使得计算机执行上述第一方面或第二方面以及第一方面或第二方面中任一种可能实现方式中的方法。
第十方面，提供了一种通信系统，包括前述的至少一个第一设备和至少一个第二设备。
上述第二方面至第十方面中任一方面及其任一方面中任意一种可能的实现可以达到的技术效果，请参照上述第一方面及其第一方面中相应实现可以带来的技术效果描述，这里不再重复赘述。
附图说明
图1是适用于本申请实施例的通信系统的一个示意性架构；
图2是本申请提供的混合模拟数字的大规模天线系统架构的一个示意图；
图3是本申请实施例提供的通信方法的一个示意性流程图；
图4是本申请实施例提供的自编码模型训练过程的一个示意图；
图5是本申请实施例提供的第一设备的多个覆盖范围的一个示意图；
图6是本申请实施例提供的第一设备的多个覆盖范围的另一个示意图；
图7是本申请实施例提供的第一设备的多个覆盖范围的另一个示意图；
图8是本申请实施例提供的通信方法的另一个示意性流程图；
图9是本申请的通信装置的一例的示意性框图；
图10是本申请的通信设备的一例的示意性结构图。
具体实施方式
下面将结合附图，对本申请中的技术方案进行描述。
本申请实施例的技术方案可以应用于各种通信系统，例如：长期演进(long term evolution，LTE)系统、LTE频分双工(frequency division duplex，FDD)系统、LTE时分双工(time division duplex，TDD)、第五代(5th generation，5G)通信系统、未来的通信系统(如第六代(6th generation，6G)通信系统)、或者多种通信系统融合的系统等，本申请实施例不做限定。其中，5G还可以称为新无线(new radio,NR)。
图1是适用于本申请实施例的通信系统100的一示意图。
本适用于申请实施例的通信系统可以包括至少一个网络设备，如图1所示的通信系统100中的网络设备110。该无线通信系统还可以包括至少一个终端设备，如图1所示的通信系统100中的终端设备120。网络设备110和终端设备120之间可以通过无线链路通信。其中，网络设备110可以通过本申请提供的通信方法处理来自终端设备120的信号。以及，终端设备120也可以通过本申请提供的通信方法处理来自网络设备110的信号。本申请对此不做限定。
本申请实施例提供的技术方案可以应用于各种通信场景，例如可以应用于以下通信场景中的一种或多种：增强移动带宽(enhanced mobile broadband，eMBB)通信、超高可靠低时延通信(ultra reliable low latency communication，URLLC)、机器类型通信(machine type communication，MTC)、大规模机器类型通信(massive machine type communication，mMTC)、设备到设备(device-to-device，D2D)通信、车辆外联(vehicle to everything，V2X)通信、车辆到车辆(vehicle to vehicle，V2V)通信、车到互联网(vehicle to network，V2N)、车到基础设施(vehicle to infrastructure，V2I)、车到行人(vehicle to pedestrian，V2P)、物联网(internet of things，IoT)、无人机通信和卫星通信等。可选地，mMTC可以包括以下通信中的一种或多种：工业无线传感器网络(industrial wireless sens or network，IWSN)的通信、视频监控(video surveillance)场景中的通信、和可穿戴设备的通信等。
本申请实施例提供的技术方案可以应用于通信设备间的通信。通信设备间的通信可以包括：网络设备和终端设备间的通信、网络设备和网络设备间的通信、和/或终端设备和终端设备间的通信。在本申请实施例中，术语“通信”还可以描述为“传输”、“信息传输”、或“信号传输”等。传输可以包括发送和/或接收。以网络设备和终端设备间的通信为例描述本申请实施例的技术方案，本领域技术人员也可以将该技术方案用于进行其它调度实体和从属实体间的通信，例如宏基站和微基站之间的通信，例如第一终端设备和第二终端设备间的通信。其中，调度实体可以为从属实体分配无线资源，例如空口资源。空口资源包括以下资源中的一种或多种：时域资源、频域资源、码资源和空间资源。
在本申请实施例中，网络设备和终端设备间的通信包括：网络设备向终端设备发送下行信号，和/或终端设备向网络设备发送上行信号。其中，信号还可以被替换为信息或数据等。
本申请实施例涉及到的终端设备还可以称为终端。终端可以是一种具有无线收发功能的设备。终端可以被部署在陆地上，包括室内、室外、手持、和/或车载；也可以被部署在水面上(如轮船等)；还可以被部署在空中(例如飞机、气球和卫星上等)。终端设备可以是用户设备(user equipment，UE)。UE包括具有无线通信功能的手持式设备、车载设备、可穿戴设备或计算设备。示例性地，UE可以是手机(mobile phone)、平板电脑或带无线收发功能的电脑。终端设备还可以是虚拟现实(virtual reality，VR)终端设备、增强现实(augmented reality，AR)终端设备、工业控制中的无线终端、无人驾驶中的无线终端、远程医疗中的无线终端、智能电网中的无线终端、智慧城市(smart city)中的无线终端、和/或智慧家庭(smart home)中的无线终端等等。
本申请实施例涉及到的网络设备包括基站(base station，BS)，可以是一种部署在无线接入网中能够和终端设备进行无线通信的设备。基站可能有多种形式，比如宏基站、微基站、中继站或接入点等。本申请实施例涉及到的基站可以是5G系统中的基站、LTE系 统中的基站或其它系统中的基站，不做限制。其中，5G系统中的基站还可以称为发送接收点(transmission reception point，TRP)或下一代节点B(generation Node B，gNB或gNodeB)。其中，基站可以是一体化的基站，也可以是分离成多个网元的基站，不予限制。例如，基站是集中式单元(centralized unit，CU)和分布式单元(distributed unit，DU)分离的基站，即基站包括CU和DU。
本申请实施例中，用于实现终端设备的功能的装置可以是终端设备；也可以是能够支持终端设备实现该功能的装置，例如芯片系统。该装置可以被安装在终端设备中或者和终端设备匹配使用。本申请实施例中，芯片系统可以由芯片构成，也可以包括芯片和其他分立器件。本申请实施例提供的技术方案中，以用于实现终端设备的功能的装置是终端设备为例，描述本申请实施例提供的技术方案。
本申请实施例中，用于实现网络设备的功能的装置可以是网络设备；也可以是能够支持网络设备实现该功能的装置，例如芯片系统。该装置可以被安装在网络设备中或者和网络设备匹配使用。在本申请实施例提供的技术方案中，以用于实现网络设备的功能的装置是网络设备为例，描述本申请实施例提供的技术方案。
在本申请实施例中，“/”可以表示前后关联的对象是一种“或”的关系，例如，A/B可以表示A或B；“和/或”可以用于描述关联对象存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况，其中A，B可以是单数或者复数。为了便于描述本申请实施例的技术方案，在本申请实施例中，可以采用“第一”、“第二”等字样对功能相同或相似的技术特征进行区分。该“第一”、“第二”等字样并不对数量和执行次序进行限定，并且“第一”、“第二”等字样也并不限定一定不同。在本申请实施例中，“示例性的”或者“例如”等词用于表示例子、例证或说明，被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念，便于理解。
在本申请实施例中，至少一个(种)还可以描述为一个(种)或多个(种)，多个(种)可以是两个(种)、三个(种)、四个(种)或者更多个(种)，本申请不做限制。
为了满足对移动通信系统的高性能要求，可以考虑将人工智能(artificial intelligence，AI)应用于移动通信网络以提升移动通信网络的性能。例如，自编码器(autoencoder)是一种可以应用到端到端(例如，网络设备和终端设备)的神经网络。它可以对收发两端进行联合优化，以提高整体性能。例如网络设备和终端设备可以联合训练两侧的神经网络以获取较优的算法。因此，如何使人工智能在移动通信网络中有效运行起来是值得研究的问题。
为了更好地理解本申请实施例，下面对本文中涉及到的术语做简单说明。
1、人工智能(artificial intelligence，AI)
人工智能AI是让机器具有学习能力，能够积累经验，从而能够解决人类通过经验可以解决的诸如自然语言理解、图像识别和/或下棋等问题。
2、神经网络(neural network，NN)：作为人工智能的重要分支，是一种模仿动物神经网络行为特征进行信息处理的网络结构。神经网络的结构由大量的节点(或称神经元)相互联接构成。神经网络基于特定运算模型，通过对输入信息进行学习和训练达到处理信息的目的。一个神经网络包括输入层、隐藏层及输出层。输入层负责接收输入信号，输出 层负责输出神经网络的计算结果，隐藏层负责特征表达等复杂的功能。隐藏层的功能由权重矩阵和对应的激活函数来表征。
深度神经网络(deep neural network，DNN)一般为多层结构。增加神经网络的深度和宽度，可以提高它的表达能力，为复杂系统提供更强大的信息提取和抽象建模能力。神经网络的深度可以表示为神经网络的层数。对于其中一层，神经网络的宽度可以表示为该层包括的神经元的个数。
DNN可以有多种构建方式，例如包括但不限于，递归神经网络(recurrent neural network，RNN)、卷积神经网络(convolutional neural network，CNN)以及全连接神经网络等。
3、训练(training)或学习
训练是指对模型(或称为训练模型)的处理过程。在该处理过程中通过优化该模型中的参数，如加权值，使该模型学会执行某项特定的任务。本申请实施例适用于但不限于以下一种或多种训练方法：监督学习、无监督学习、强化学习、和迁移学习等。有监督学习是利用一组具有已经打好正确标签的训练样本来训练。其中，已经打好正确标签是指每个样本有一个期望的输出值。与有监督学习不同，无监督学习是指一种方法，该方法没有给定事先标记过的训练样本，自动对输入的数据进行分类或分群。
4、推理
推理是指利用训练后的模型(训练后的模型可以称为推理模型)执行数据处理。将实际数据输入推理模型进行处理，得到对应的推理结果。推理还可以称为预测或决策，推理结果还可以称为预测结果、或决策结果等。
图2是本申请提供的混合模拟数字的大规模天线系统架构的一个示意图。
例如，如图2所示的天线系统架构配置于网络设备，该天线系统的前端配置有N个天线的均匀线性阵列(uniform linear array，ULA)，后端射频链路的个数R远小于天线个数N，网络设备通过该天线系统同时服务K个终端设备。在网络设备进行终端设备到网络设备的上行信道估计时，网络设备对天线阵列接收到的信号经模拟域、数字域处理后，得到处理后的接收信号Y为：
Figure PCTCN2021131856-appb-000001
其中，W BB为数字域组合矩阵，维度为R×R。W RF为模拟域组合矩阵，维度为R×N。N为加性高斯白噪声。p k为第k个终端设备发送的导频序列，维度为1×L p，L p为导频序列的长度。h k为第k个终端设备到网络设备的上行信道，维度为N×1。信道由多条路径叠加而成：
Figure PCTCN2021131856-appb-000002
其中，N p为路径数，α i为第i条路径的复增益，θ i为第i条路径到达网络设备的到达角。天线间隔为半波长时，ULA的阵列响应向量a(θ)＝[1 e jπsin(θ) … e jπsin(θ)(N-1)] T。
实际中，由于网络设备通常海拔较高、周围散射物较少，终端设备到网络设备的入射信号角度扩展(最大入射角与最小入射角之差的一半)较小，因此角度域信道信息具有结构稀疏特性。具体来说，角度域信道信息的非零元素数量远小于天线数，且集中在一个较小的范围内。利用这种结构稀疏特性，可以提升信道数据处理性能。例如，可以帮助提升信道估计性能、降低估计开销。角度域信道信息可由原信道经离散傅里叶变换(discrete  fourier transform，DFT)变换得到：
x＝B Hh,
其中x和h分别代表角度域信道信息与原信道，B为归一化DFT矩阵。此外，由于终端设备间导频序列相互正交，即满足当i≠j时，
Figure PCTCN2021131856-appb-000003
以及当i＝j时，
Figure PCTCN2021131856-appb-000004
因此可以提取出对应于第k个终端设备的归一化发送信号后的接收信号为：
Figure PCTCN2021131856-appb-000005
其中，Φ＝W BBW RFB称为测量矩阵，
Figure PCTCN2021131856-appb-000006
为等效噪声。
因此，在混合模拟数字架构下，大规模天线系统的信道估计问题即根据接收信号y k，设计合适的测量矩阵Φ与信道估计算法，估计角度域信道信息
Figure PCTCN2021131856-appb-000007
使其与真实的角度域信道信息x k之间的误差尽可能小。
本申请提出通过利用角度域信道信息的稀疏特性，信号接收设备的覆盖范围划分为多个覆盖区域，信号接收设备根据信号发送设备所处的覆盖区域，采用该覆盖区域相对应的人工智能推理模型，推理接收信号得到信道信息。能够提高接收信号的准确度。从而提高通信的可靠性。
图3是本申请实施例提供的通信方法300的一个示意性流程图。图3所示的通信方法由第一设备或配置与第一设备的装置执行，下文中以该通信方法由第一设备为例进行说明。第一设备为信号的接收设备，可以是网络设备也可以是终端设备，本申请对此不做限定。
S310，第一设备获取来自第二设备的第一信号。
例如，第二设备向第一设备发送参考信号(参考信号也可以称为导频信号)，第一设备可以通过阵列天线接收得到经历了信道传输后的该参考信号，该经历了信道传输后的参考信号即为第一信号。但本申请不限于此。
S320，第一设备根据第一模型，推理第一信号得到第一信道信息。
该第一设备的覆盖范围划分为多个覆盖区域，该多个覆盖区域中的每个覆盖区域对应一个信道信息推理模型。其中，第二设备处于第一设备的该多个覆盖区域中的第一覆盖区域。第一模型为该第一覆盖区域对应的信道信息推理模型。由于相近位置环境数据通常具有相关性，因此，根据不同覆盖区域采用相应的信道信息推理模型能够提高信道信息获取的准确度，从而较准确地还原接收信号。该多个覆盖区域可以是同一小区内的多个覆盖区域，也可以是多个小区内的覆盖区域，本申请对此不做限定。
作为示例非限定，该第一信道信息为角度域信道信息。
可选地，第一设备根据第一模型，推理所述第一信号得到第一信道信息，具体可以包括，第一设备根据该第一模型，推理得到测量信息，再根据测量信息处理第一信号得到第二信号。得到第二信号之后，第一设备再根据该第一模型，推理第二信号得到第一信道信息。
可选地，该第一模型可以是神经网络模型。
例如，该第一模型为神经网络模型中的自编码器模型，该自编码器模型包括编码器部分和译码器部分，具体地，第一设备可以通过第一模型的编码器部分推理得到测量矩阵Φ(即测量信息的一个示例)，并通过该测量矩阵Φ根据下式得到数字组合矩阵W BB，
Figure PCTCN2021131856-appb-000008
则该W BB为处理第一覆盖区域的信号对应的数字域组合矩阵。其中，
Figure PCTCN2021131856-appb-000009
表示W RFB的伪逆矩阵，例如可以是摩尔-潘罗斯(moore-penrose)伪逆矩阵。但本申请不限于此。由于模拟域组合矩阵W RF是由相移器相位决定的、且接收来自各个通信设备的信号的相移器是共用的。因此，当第一设备采用全向天线时，W RF可以是任意行满秩矩阵。或者，当第一设备非全向天线时，W RF可以是训练好的模拟域波束对应的模拟域组合矩阵。第一设备接收第一信号，通过相移器处理第一信号后，并采用通过测量矩阵Φ得到的W BB进行处理后得到第二信号。第一设备再根据该第一模型的译码器部分推理第二信号得到估计的角度域信道信息
Figure PCTCN2021131856-appb-000010
根据该方案，采用第一覆盖区域对应的模型推理接收到的第一信号，能够提高估计角度信道的准确性。根据该准确性较高的角度域信道信息对数据信号进行译码，可以减少译码失败的概率，提高通信的可靠性。
一种实施方式中，第一设备的多个覆盖区域对应的信道信息推理模型可以是预配置在该第一设备中的。
另一种实施方式中，在S320之前，第一设备采用训练信号对训练模型进行训练得到训练后的训练模型，该训练后的训练模型即为上述推理模型。
下面以第一覆盖区域的模型训练过程为例进行说明，其他覆盖区域对应的模型训练过程可以参考该第一覆盖区域的模型训练过程实施。在该实施方式中，首先，第一设备可以收集用于训练的第一训练信号集合，该第一训练信号集合包括多个训练信号，每个训练信号包括数据和标签，数据也可以称为样本，数据可以是带噪声的角度域信道信息，标签是角度域信道信息，用于计算训练模型输出的精度。该角度域信道信息可以是基于第一设备的实际接收信号计算得到的，该实际接收信号为接收到的来自第一覆盖区域的通信设备发送的信号，也可以是根据第一覆盖区域的信道模型由计算机模拟合成得到的信号。但本申请不限于此。
图4为本申请实施例提供的自编码模型训练过程的一个示意图。
如图4所示的自编码器模型中包括编码器部分和译码器部分，其中编码器部分可以包括卷积层1和卷积层2，译码器部分包括卷积层3、卷积层4、全连接层1和全连接层2。其中，编码器部分的卷积层1和卷积层2的输入维度均为2N×1，卷积核的维度(size)为N×1，卷积步长(strides)为N，其中N为天线数。卷积层1共包括R个卷积核(例如，卷积核可以由滤波器实现)，每个卷积核的参数可能不同，其中R为射频通道的个数。由以上的参数可知，训练信号中的数据输入编码器部分后，卷积层1的每个卷积核的输出维度为2，其中第一维度实现了测量矩阵第r行的实部与输入向量实部相乘，第二维度实现了测量矩阵第r行的实部与输入向量虚部相乘；卷积层2的每个卷积核的输出维度为2，其中第一维度实现了测量矩阵第r行的虚部与输入向量实部相乘，第二维度实现了测量矩阵第r行的虚部与输入向量虚部相乘且相乘结果取负值，其中r小于或等于R。卷积层1输出的第一维度的结果与卷积层2输出的第二维度的结果相加后得到测量后的数据y的实部；卷积层1输出的第二维度的结果与卷积层2输出的第一维度的结果相加后得到测量后的数据y的虚部。测量后的训练信号y输入译码器部分，译码器部分中的卷积层3的输入是2组维度为R×1的向量，相当于2个通道，每个通道都由32个维度为3×1的卷积核计算，2个通道的计算结果相加作为输出。卷积层4的输入是32组维度为R×1的向量，相当于32个通道，每个通道由32个维度为3×1的卷积核计算，32个通道的计算结果相加作为输出，输出维度为32组维度为R×1的向量，将该输出重构为一个一维的向量后送 入全连接层，由全连接层2输出根据测量后的数据y估计得到的角度域信道
Figure PCTCN2021131856-appb-000011
以估计得到的角度域信道信息
Figure PCTCN2021131856-appb-000012
与标签x之间的均方误差(mean square error，MSE)，
Figure PCTCN2021131856-appb-000013
其中，
Figure PCTCN2021131856-appb-000014
表示
Figure PCTCN2021131856-appb-000015
的范数。第一设备根据均方误差计算结果调整卷积层及全连接层的权重。以
Figure PCTCN2021131856-appb-000016
与x的均方误差MSE作为损失函数，采用随机梯度下降算法训练该自编码器模型，直至损失函数收敛，得到训练后的模型作为第一覆盖区域对应的角度域信道推理模型(即第一模型)。但本申请不限于此。
根据上述方案，第一设备基于信道环境中的训练数据对不同覆盖区域的信道信息模型进行训练，增加了模型与覆盖区域的匹配度，提高了模型进行信道估计的准确度。
在本申请中，第一设备的多个覆盖区域的划分可以包括但不限于以下方式。
方式一，该多个覆盖区域中的一个覆盖区域为以第一设备为圆心的一个圆心角覆盖的扇区。
例如，如图5所示为第一设备的多个覆盖区域的一个示意图。每个覆盖区域对应一个圆心角范围，圆心角还可以称为水平方位角，表1是多个覆盖区域的水平方位角的范围的一个示例。如表1所示，覆盖区域1包括的水平方位角范围为0≤θ＜θ 1、覆盖区域2包括的水平方位角范围为θ 1≤θ＜θ 2等。但本申请不限于此。可选地，多个覆盖区域中任意两个覆盖区域的圆心角大小可以相等，例如，每个覆盖区域的圆心角均为β，则第一设备的覆盖范围可以包括
Figure PCTCN2021131856-appb-000017
个覆盖区域。例如，β可以为5°、10°等，但本申请不限于此。多个覆盖区域中的两个覆盖区域的圆心角大小可以是不相等的，例如，第一设备可以根据信道特性将信道相关性较大的圆心角范围划分为同一覆盖区域，因此多个覆盖区域中的两个覆盖区域的覆盖圆心角大小可以不同。
表1
覆盖区域标识 水平方位角(θ)
1 0≤θ＜θ 1
2 θ 1≤θ＜θ 2
… …
方式二，该多个覆盖区域中的一个覆盖区域属于以第一设备为圆心的一个圆心角覆盖的扇区，其中，任意两个覆盖区域所属的扇区不同，或任意两个覆盖区域与所述第一设备的间距不同。
例如，图6为第一设备的多个覆盖区域的另一个示意图。如图6所示任意两个覆盖区域所属的扇区不同，或任意两个覆盖区域与所述第一设备的间距不同。例如图6中覆盖区域1与覆盖区域2属于同一扇区，但覆盖区域1至第一设备的间距与覆盖区域2至第一设备的间距不同。以及，覆盖区域2至第一设备的间距与覆盖区域N至第一设备的间距相同，但覆盖区域2和覆盖区域N属于不同的扇区。表2为覆盖区域的水平方位角θ的范围以及距离d的范围的一个示例。如表2所示，覆盖区域1包括与第一设备的距离的范围为0≤d＜d 1以及水平方位角的范围为0≤θ＜θ 1。覆盖区域2包括与第一设备的距离的范围为d 1≤d＜d 2以及水平方位角的范围为0≤θ＜θ 1。覆盖区域1包括与第一设备的距离的范围为0≤d＜d 1以及水平方位角的范围为θ 1≤θ＜θ 2等。但本申请不限于此。需要说明的是， 表2仅为一个示例，覆盖区域还可以通过如表3所示的距离范围表和如表4所示的水平方位角范围表两个表确定覆盖区域。本申请对此不做限定。
表2
Figure PCTCN2021131856-appb-000018
表3
覆盖区域标识 距离(d)
1 0≤d＜d 1
2 d 1≤d＜d 2
3 0≤d＜d 1
4 d 1≤d＜d 2
表4
Figure PCTCN2021131856-appb-000019
方式三，该多个覆盖区域中的一个覆盖区域为以第一设备为圆心的一个水平方位角和垂直方位角确定的区域。
例如，图7为一个覆盖区域的示意图。如图7所示的覆盖区域是通过水平方位角θ和垂直方位角
Figure PCTCN2021131856-appb-000020
确定的一个区域。覆盖区域的水平方位角
Figure PCTCN2021131856-appb-000021
的范围可以如表4所示，垂直方位角的范围可以如表5所示。例如，通过表4可以确定覆盖区域3的水平方位角θ的范围为θ 1≤θ＜θ 2，通过表5可以确定覆盖区域3的垂直方位角
Figure PCTCN2021131856-appb-000022
的范围为
Figure PCTCN2021131856-appb-000023
对于覆盖区域的垂直方位角，可以是将第一设备的垂直方位角0到180°范围划分出多个覆盖区域的垂直方位角范围，例如在具体实施中第一设备处于水平面时可以在垂直方位角0到180°范围划分出多个覆盖区域的垂直方位角范围。或者，当第一设备架高时，例如架设在海拔较高的山上或高层楼上，可以将垂直方位角0至360°范围划分出多个覆盖区域的垂直方位角范围。其中，垂直方位角在90°左右时可以是与飞机、卫星等设备通信的覆盖区 域。但本申请不限于此。另外，可以是均匀的划分为多个覆盖区域的垂直方位角，也可以是根据信道特性不同划分出多个覆盖区域的垂直方位角的大小可以不同，本申请对此不做限定。覆盖区域的水平方位角与方式一中类似，为了简要，在此不再赘述。
表5
Figure PCTCN2021131856-appb-000024
下面介绍第一设备确定第二设备处于第一覆盖区域的可选地实施方式。
实施方式一
图8是本申请提供的通信方法的另一个示意性流程图。
S810，第一设备向第二设备发送第一指示信息，该第一指示信息用于指示第一设备的多个覆盖区域，该多个覆盖区域与多个信道信息推理模型相对应。
相应地，第二设备获取来自第一设备的该第一指示信息。可选地，第一指示信息可以是广播信息。也就是说，该第一指示信息能够被多个与第一设备通信的设备接收到。
可选地，第一指示信息中包括该多个覆盖区域中的每个覆盖区域的边界的定位信息。
例如，定位信息可以是全球定位系统(global position system，GPS)信息，第一设备的多个覆盖区域可以如图5所示。第一指示信息包括每个覆盖区域的标识(identifier，ID)以及与该ID对应的该覆盖区域的两个边界的GPS信息。例如，第一指示信息包括{GPS0，GPS1，ID1，GPS2，GPS3，ID2}等，其中，ID的前两个GPS信息为该ID标识的覆盖区域的两个边界的GPS信息，其中ID1标识的覆盖区域1的两个边界的GPS信息分别为GPS0和GPS1，第二设备接收到该第一指示信息后可以根据GPS0和GPS1分别确定第一覆盖区域的边界的水平方位角，则第一覆盖区域的范围在GPS0和GPS1确定的水平方位角之间的范围。第二设备可以根据该方式基于第一指示信息确定每个覆盖区域。从而确定第二设备所处的覆盖区域，即第一覆盖区域。
可选地，第一指示信息还包括每个覆盖区域的标识信息和/或每个覆盖区域对应的模型的标识信息。
第二设备确定其处于的第一覆盖区域后，可以在S820中通过第二指示信息指示第一覆盖区域的标识信息，或指示第一覆盖区域对应的模型的标识信息。
S820，第二设备向第一设备发送第二指示信息，该第二指示信息用于指示第二设备处于第一覆盖区域或指示第一模型。
相应地，第一设备接收来自第二设备的第二指示信息。该第二设备在S810中确定其处于的第一覆盖区域后，可以在S820中通过第二指示信息通知第一设备第一覆盖区域的标识信息。使得第一设备根据第一覆盖区域的标识信息确定采用第一覆盖区域对应的第一模型处理获取到的第二设备的信号。或者该第二设备在S810中确定所处覆盖区域对应的 模型的标识信息，则第二设备可以通过第二指示信息通知第一设备第一模型的标识信息。第一设备根据第二指示信息中第一模型的标识可以确定采用第一模型处理获取到的第二设备的信号。
实施方式二
第一设备可以获取来自第二设备的位置信息，根据该位置信息确定第二设备处于第一覆盖区域。
基于本申请的方案，通过实验比较了本申请提供的方法与传统信道估计方法的性能。其中，实验参数包括：天线数N＝128，取一个覆盖区域的圆心角范围为β＝5°，信号噪声比(signal noise ratio，SNR)为20dB，射频通道数R＝32以及R＝16，实验结果为信道估计的归一化均方误差(normalized mean square error，NMSE)，表6中Φ A为传统的Zadoff-Chu测量矩阵，Φ B是通过第一模型推理得到的测量结果。实验中分别根据测量矩阵Φ A和Φ B，测量(或称为处理)接收信号，分别比较了通过稀疏贝叶斯学习(sparse bayesian learning，SBL)、正交匹配追踪(orthogonal matching pursuit，OMP)、变分贝叶斯推理(variational bayesian inference,VBI)算法对测量后的接收信号进行信道估计的NMSE结果。以及，通过本申请的第一模型进行信道对测量矩阵Φ A和Φ B分别得到的测量后的接收信号进行信道估计的NMSE。
由表6可以看出，比较表6中的同一行结果，采用相同的信道估计算法、相同射频通道数R时，相基于Φ B测量的信道估计NMSE结果均小于基于Φ A测量的信号的信道估计NMSE。可以看出应用本申请通过第一模型得到的测量矩阵Φ B时信道估计性能优于应用Zadoff-Chu测量矩阵时的信道估计性能。对于表6中的同一列结果，在采用同一测量矩阵时第一模型推理得到的信道信息的NMSE结果明显优于传统信道估计算法SBL、OMP、VBI的信道估计性能。而测量矩阵以及信道估计均采用本申请提供的覆盖区域对应的信道信息推理模型时，即通过表6中最后一行最后两列结果与其他结果相比较，本申请提供的方法进行信道估计的性能显著优于传统算法。
表6
Figure PCTCN2021131856-appb-000025
本申请提出通过利用角度域信道信息的稀疏特性，信号接收设备的覆盖范围划分为多个覆盖区域，信号接收设备根据信号发送设备所处的覆盖区域，采用该覆盖区域相对应的人工智能推理模型，推理接收信号得到信道信息。能够提高接收信号的准确度。从而提高通信的可靠性。
以上，结合图2至图8详细说明了本申请实施例提供的方法。以下详细说明本申请实施例提供的装置。为了实现上述本申请实施例提供的方法中的各功能，各网元可以包括硬件结构和/或软件模块，以硬件结构、软件模块、或硬件结构加软件模块的形式来实现上述 各功能。上述各功能中的某个功能以硬件结构、软件模块、还是硬件结构加软件模块的方式来执行，取决于技术方案的特定应用和设计约束条件。
图9是本申请实施例提供的通信装置的示意性框图。如图9所示，该通信装置900可以包括处理单元910和收发单元920。
在一种可能的设计中，该通信装置900可对应于上文方法实施例中的第一设备，或者配置于(或用于)第一设备中的芯片，或者是其他能够实现第一设备执行的方法的装置、模块、电路或单元等。
可选地，第一设备可以是网络设备或终端设备。应理解，该通信装置900可对应于本申请实施例的方法300、800中的第一设备，该通信装置900可以包括用于执行图3、图8中的方法300、800中第一设备执行的方法的单元。并且，该通信装置900中的各单元和上述其他操作和/或功能分别为了实现图3、图8中的方法300、800的相应流程。
当该通信装置900用于实现上述方法实施例中的第一设备执行的相应流程时，收发单元920，用于获取来自第二设备的第一信号；该处理单元910，用于根据第一模型，推理第一信号得到第一信道信息。
在某些可能的实现方式中，该处理单元910具体用于根据该第一模型，推理得到测量信息；根据该测量信息处理该第一信号，得到该第二信号；根据该第一模型，推理该第二信号得到该第一信道信息。
在某些可能的实现方式中，该处理单元910具体用于根据该编码器部分，推理得到测量信息，以及根据该译码器部分，推理该第二信号得到该第一信道信息。
在某些可能的实现方式中，在该收发单元920获取来自该第二设备的该第一信号之前，该收发单元920还用于获取第一覆盖区域的第一训练集，该第一训练集中包括多个训练信号；该处理单元910还用于基于该第一训练集对训练模型进行训练，得到该第一模型。
在某些可能的实现方式中，该训练模型包括编码器部分和译码器部分，该处理单元910具体用于将该第一训练集中的角度域信道信息样本输入该编码器部分得到目标编码信号，以及将该目标编码信号输入该译码器部分得到目标译码信号。该处理单元910再基于该角度域信道信息标签和该目标译码信号得到训练模型的参数信息，以及根据该参数信息，调节该编码器部分的参数和该译码器部分的参数。
在某些可能的实现方式中，该收发单元920还用于获取来自第二设备的位置信息；该处理单元910还用于根据该位置信息确定该第二设备处于该第一覆盖区域。
还应理解，该通信装置900为配置于(或用于)第一设备中的芯片时，该通信装置900中的收发单元920可以为芯片的输入/输出接口或电路，该通信装置900中的处理单元910可以为芯片中的逻辑电路。在另一种可能的设计中，该通信装置900可对应于上文方法实施例中的第二设备，例如，或者配置于(或用于)第二设备中的芯片，或者是其他能够实现第二设备执行的方法的装置、模块、电路或单元等。
该第二设备为与第一设备通过无线链路通信的设备，可选地，该第二设备可以是网络设备或终端设备。
应理解，该通信装置900可对应于根据本申请实施例的方法300、800中的第二设备，该通信装置900可以包括用于执行图3、图8中的方法300、800中第二设备执行的方法的单元。并且，该通信装置900中的各单元和上述其他操作和/或功能分别为了实现图3、图8中的方法300、800的相应流程。
当该通信装置900用于实现上述方法实施例中的第二设备执行的相应流程时，该收发单元920，用于获取第一指示信息；该处理单元910，用于根据该第一指示信息确定第一覆盖区域，该收发单元920，还用于发送第二指示信息。
还应理解，该通信装置900为配置于(或用于)第二设备中的芯片时，该通信装置900中的收发单元920可以为芯片的输入/输出接口或电路，该通信装置900中的处理单元910可以为芯片中的逻辑电路。可选地，通信装置900还可以包括存储单元930，该存储单元930可以用于存储指令或者数据，处理单元910可以执行该存储单元中存储的指令或者数据，以使该通信装置实现相应的操作。
应理解，该通信装置900中的收发单元920为可通过通信接口(如收发器或输入/输出接口)实现，例如可对应于图10中示出的通信设备1000中的收发器1010。该通信装置900中的处理单元910可通过至少一个处理器实现，例如可对应于图10中示出的通信设备1000中的处理器1020。该通信装置900中的处理单元910还可以通过至少一个逻辑电路实现。该通信装置900中的存储单元930可对应于图10中示出的通信设备1000中的存储器。
还应理解，各单元执行上述相应步骤的具体过程在上述方法实施例中已经详细说明，为了简洁，在此不再赘述。
图10是本申请实施例提供的终端设备1000的结构示意图。
该通信设备1000可对应于上述方法实施例中的第一设备，如图10所示，该第一设备1000包括处理器1020和收发器1010。可选地，该第一设备1000还包括存储器。其中，处理器1020、收发器1010和存储器之间可以通过内部连接通路互相通信，传递控制和/或数据信号。该存储器用于存储计算机程序，该处理器1020用于执行该存储器中的该计算机程序，以控制该收发器1010收发信号。
应理解，图10所示的通信设备1000能够实现图3、图8所示方法实施例中涉及第一设备的过程。第一设备1000中的各个模块的操作和/或功能，分别为了实现上述方法实施例中的相应流程。具体可参见上述方法实施例中的描述，为避免重复，此处适当省略详细描述。
该通信设备1000可对应于上述方法实施例中的第二设备，如图10所示，该第二设备1000包括处理器1020和收发器1010。可选地，该第二设备1000还包括存储器。其中，处理器1020、收发器1010和存储器之间可以通过内部连接通路互相通信，传递控制和/或数据信号。该存储器用于存储计算机程序，该处理器1020用于执行该存储器中的该计算机程序，以控制该收发器1010收发信号。
应理解，图10所示的通信设备1000能够实现图3、图8所示方法实施例中涉及第二设备的过程。第二设备1000中的各个模块的操作和/或功能，分别为了实现上述方法实施例中的相应流程。具体可参见上述方法实施例中的描述，为避免重复，此处适当省略详细描述。
其中，该第一设备可对应于如图1所示的系统中的终端设备，第二设备对应于如图1所示的系统中的网络设备。或者，该第一设备可对应于如图1所示的系统中的网络设备，第二设备对应于如图1所示的系统中的终端设备。
上述处理器1020可以和存储器可以合成一个处理装置，处理器1020用于执行存储器中存储的程序代码来实现上述功能。具体实现时，该存储器也可以集成在处理器1020中， 或者独立于处理器1020。该处理器1020可以与图9中的处理单元对应。
上述收发器1010可以与图9中的收发单元对应。收发器1010可以包括接收器(或称接收机、接收电路)和发射器(或称发射机、发射电路)。其中，接收器用于接收信号，发射器用于发射信号。
应理解，图10所示的通信设备1000能够实现图3、图8所示方法实施例中涉及终端设备的过程。终端设备1000中的各个模块的操作和/或功能，分别为了实现上述方法实施例中的相应流程。具体可参见上述方法实施例中的描述，为避免重复，此处适当省略详细描述。
本申请实施例还提供了一种处理装置，包括处理器和(通信)接口；所述处理器用于执行上述任一方法实施例中的方法。
应理解，上述处理装置可以是一个或多个芯片。例如，该处理装置可以是现场可编程门阵列(field programmable gate array，FPGA)，可以是专用集成芯片(application specific integrated circuit，ASIC)，还可以是系统芯片(system on chip，SoC)，还可以是中央处理器(central processor unit，CPU)，还可以是网络处理器(network processor，NP)，还可以是数字信号处理电路(digital signal processor，DSP)，还可以是微控制器(micro controller unit，MCU)，还可以是可编程控制器(programmable logic device，PLD)或其他集成芯片。
根据本申请实施例提供的方法，本申请还提供一种计算机程序产品，该计算机程序产品包括：计算机程序代码，当该计算机程序代码由一个或多个处理器执行时，使得包括该处理器的装置执行图3、图8所示实施例中的方法。
本申请实施例提供的技术方案可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本发明实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、网络设备、终端设备、核心网设备、机器学习设备或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line，DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机可以存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质(例如，数字视频光盘(digital video disc，DVD))、或者半导体介质等。
根据本申请实施例提供的方法，本申请还提供一种计算机可读存储介质，该计算机可读存储介质存储有程序代码，当该程序代码由一个或多个处理器运行时，使得包括该处理器的装置执行图3、图8所示实施例中的方法。
根据本申请实施例提供的方法，本申请还提供一种系统，其包括前述的一个或多个第一设备。还系统还可以进一步包括前述的一个或多个第二设备。
可选地，第一设备可以是网络设备或终端设备，第二设备可以是与第一设备通过无线链路进行通信的设备。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (26)

  1. 一种通信方法，其特征在于，包括：
    第一设备获取来自第二设备的第一信号；
    所述第一设备根据第一模型，推理所述第一信号得到第一信道信息，所述第二设备处于所述第一设备的多个覆盖区域中的第一覆盖区域，所述第一模型为所述第一覆盖区域对应的信道信息推理模型。
  2. 根据权利要求1所述的方法，其特征在于，所述第一信道信息为角度域信道信息。
  3. 根据权利要求1或2所述的方法，其特征在于，所述根据第一模型，推理所述第一信号得到第一信道信息，包括：
    根据所述第一模型，推理得到测量信息；
    根据所述测量信息处理所述第一信号，得到第二信号；
    根据所述第一模型，推理所述第二信号得到所述第一信道信息。
  4. 根据权利要求1至3中任一项所述的方法，其特征在于，在所述第一设备获取来自所述第二设备的所述第一信号之前，所述方法还包括：
    所述第一设备获取第一覆盖区域的第一训练集，所述第一训练集中包括多个训练信号；
    所述第一设备基于所述第一训练集对训练模型进行训练，得到所述第一模型。
  5. 根据权利要求1至4中任一项所述的方法，其特征在于，所述多个覆盖区域中的一个覆盖区域属于以所述第一设备为圆心的一个圆心角覆盖的扇区，所述第一覆盖区域属于第一圆心角覆盖的扇区，所述多个覆盖区域中的任意两个覆盖区域所属的扇区不同，或所述多个覆盖区域中的任意两个覆盖区域与所述第一设备的间距不同。
  6. 根据权利要求1至5中任一项所述的方法，其特征在于，所述方法还包括：
    所述第一设备获取来自所述第二设备的位置信息；
    所述第一设备根据所述位置信息确定所述第二设备处于所述第一覆盖区域。
  7. 根据权利要求6所述的方法，其特征在于，所述方法还包括：
    所述第一设备发送第一指示信息，所述第一指示信息用于指示所述多个覆盖区域，所述多个覆盖区域与多个信道信息推理模型对应；
    所述第一设备获取第二指示信息，所述第二指示信息用于指示所述第二设备处于所述第一覆盖区域或指示所述第一模型。
  8. 一种通信方法，其特征在于，所述方法还包括：
    第二设备获取第一指示信息，所述第一指示信息用于指示多个覆盖区域，所述多个覆盖区域与多个信道信息推理模型相对应；
    所述第二设备发送第二指示信息，所述第二指示信息用于指示所述第二设备处于所述第一覆盖区域或指示第一模型，所述第一覆盖区域为所述多个覆盖区域中的一个，所述第一模型为所述第一覆盖区域对应的模型。
  9. 根据权利要求8所述的方法，其特征在于，所述多个覆盖区域中的一个覆盖区域属于以第一设备为圆心的一个圆心角覆盖的扇区，所述第一覆盖区域属于第一圆心角覆盖的扇区，所述多个覆盖区域中的两个覆盖区域所属的扇区不同，或所述多个覆盖区域中的两个覆盖区域与所述第一设备的间距不同。
  10. 根据权利要求8或9所述的方法，其特征在于，所述第一指示信息包括以下一项或多项：
    所述多个覆盖区域中每个覆盖区域的参考位置信息、所述多个覆盖区域中每个覆盖区域的标识信息或所述多个信道信息推理模型中每个模型的标识信息。
  11. 根据权利要求8至10中任一项所述的方法，其特征在于，所述第二指示信息中包括所述第一覆盖区域的标识信息或所述第一模型的标识信息。
  12. 一种通信装置，其特征在于，所述通信装置配置于第一设备，包括：
    收发单元，用于获取来自第二设备的第一信号；
    处理单元，用于根据第一模型，推理第一信号得到第一信道信息，所述第二设备处于所述第一设备的多个覆盖区域中的第一覆盖区域，所述第一模型为所述第一覆盖区域对应的信道信息推理模型。
  13. 根据权利要求12所述的装置，其特征在于，所述第一信道信息为角度域信道信息。
  14. 根据权利要求12或13所述的装置，其特征在于，所述处理单元具体用于：
    根据所述第一模型，推理得到测量信息；
    根据所述测量信息处理所述第一信号，得到所述第二信号；
    根据所述第一模型，推理所述第二信号得到所述第一信道信息。
  15. 根据权利要求12至14中任一项所述的装置，其特征在于，在所述收发单元获取来自所述第二设备的所述第一信号之前，
    所述收发单元还用于获取第一覆盖区域的第一训练集，所述第一训练集中包括多个训练信号；
    所述处理单元还用于基于所述第一训练集对训练模型进行训练，得到所述第一模型。
  16. 根据权利要求12至15中任一项所述的装置，其特征在于，所述多个覆盖区域中的一个覆盖区域属于以所述第一设备为圆心的一个圆心角覆盖的扇区，所述第一覆盖区域属于第一圆心角覆盖的扇区，所述多个覆盖区域中的任意两个覆盖区域所属的扇区不同，或所述多个覆盖区域中的任意两个覆盖区域与所述第一设备的间距不同。
  17. 根据权利要求12至16中任一项所述的装置，其特征在于，
    所述收发单元还用于获取来自所述第二设备的位置信息；
    所述处理单元还用于根据所述位置信息确定所述第二设备处于所述第一覆盖区域。
  18. 一种通信装置，其特征在于，所述通信装置配置于第二设备，包括：
    收发单元，用于获取第一指示信息，所述第一指示信息用于指示多个覆盖区域，所述多个覆盖区域与多个信道信息推理模型相对应；
    处理单元，用于根据所述第一指示信息确定所述第二设备处于所述多个覆盖区域中的第一覆盖区域；
    所述收发单元，还用于发送第二指示信息，所述第二指示信息用于指示所述第二设备处于所述第一覆盖区域或指示第一模型，所述第一模型为所述第一覆盖区域对应的模型。
  19. 根据权利要求18所述的装置，其特征在于，所述多个覆盖区域中的一个覆盖区域属于以所述第一设备为圆心的一个圆心角覆盖的扇区，所述第一覆盖区域属于第一圆心角覆盖的扇区，所述多个覆盖区域中的两个覆盖区域所属的扇区不同，或所述多个覆盖区域中的两个覆盖区域与所述第一设备的间距不同。
  20. 根据权利要求18或19所述的装置，其特征在于，所述第一指示信息包括以下一项 或多项：
    所述多个覆盖区域中每个覆盖区域的参考位置信息、所述多个覆盖区域中每个覆盖区域的标识信息或所述多个信道信息推理模型中每个模型的标识信息。
  21. 根据权利要求18至20中任一项所述的装置，其特征在于，所述第二指示信息中包括所述第一覆盖区域的标识信息或所述第一模型的标识信息。
  22. 一种通信装置，其特征在于，包括至少一个处理器，与存储器耦合；
    所述存储器用于存储程序或指令；
    所述至少一个处理器用于执行所述程序或指令，以使所述装置实现如权利要求1至11中任一项所述的方法。
  23. 一种芯片，其特征在于，包括至少一个逻辑电路和输入输出接口；
    所述输入输出接口用于输入来自第二设备的第一信号；所述逻辑电路用于根据第一模型，推理所述第一信号得到第一信道信息，所述第二设备处于第一设备的多个覆盖区域中的第一覆盖区域，所述第一模型为所述第一覆盖区域对应的信道信息推理模型。
  24. 一种芯片，其特征在于，包括至少一个逻辑电路和输入输出接口；
    所述输入输出接口用于获取第一指示信息，所述第一指示信息用于指示多个覆盖区域，所述多个覆盖区域与多个信道信息推理模型相对应；
    所述逻辑电路用于根据所述第一指示信息确定第二设备处于所述多个覆盖区域中的第一覆盖区域；所述输入输出接口还用于发送第二指示信息，所述第二指示信息用于指示所述第二设备处于所述第一覆盖区域或指示第一模型，所述第一覆盖区域为所述多个覆盖区域中的一个，所述第一模型为所述第一覆盖区域对应的模型。
  25. 一种计算机可读存储介质，其特征在于，存储有指令，当所述指令在计算机上运行时，使得所述计算机执行如权利要求1至11中任一项所述的方法。
  26. 一种计算机程序产品，其特征在于，包括指令，当所述指令在计算机上运行时，使得计算机执行如权利要求1至11中任一项所述的方法。
PCT/CN2021/131856 2021-02-22 2021-11-19 基于空间划分的数据处理方法和通信装置 WO2022174642A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP21926356.3A EP4266603A4 (en) 2021-02-22 2021-11-19 DATA PROCESSING METHOD BASED ON SPATIAL DIVISION AND COMMUNICATION DEVICE

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110197721.6A CN114978382A (zh) 2021-02-22 2021-02-22 基于空间划分的数据处理方法和通信装置
CN202110197721.6 2021-02-22

Publications (1)

Publication Number Publication Date
WO2022174642A1 true WO2022174642A1 (zh) 2022-08-25

Family

ID=82930273

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/131856 WO2022174642A1 (zh) 2021-02-22 2021-11-19 基于空间划分的数据处理方法和通信装置

Country Status (3)

Country Link
EP (1) EP4266603A4 (zh)
CN (1) CN114978382A (zh)
WO (1) WO2022174642A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115460633A (zh) * 2022-09-29 2022-12-09 中电信数智科技有限公司 6g网络ai信号优化的方法、系统、设备和存储介质

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117786418A (zh) * 2022-09-29 2024-03-29 华为技术有限公司 模型匹配的方法和通信装置
WO2024138673A1 (zh) * 2022-12-30 2024-07-04 Oppo广东移动通信有限公司 无线通信的方法及通信设备

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014022964A1 (zh) * 2012-08-07 2014-02-13 华为技术有限公司 传输消息的方法、设备及终端
CN106302274A (zh) * 2016-08-26 2017-01-04 清华大学 一种大规模mimo系统多用户信道估计与跟踪方法
CN110035388A (zh) * 2019-03-21 2019-07-19 平安科技（深圳）有限公司 确定用户区域的方法、计算机可读存储介质及终端设备
CN110113088A (zh) * 2019-05-07 2019-08-09 东南大学 一种分离型数模混合天线系统波达角智能化估计方法
CN111478783A (zh) * 2019-01-23 2020-07-31 中国移动通信有限公司研究院 一种配置无线传输参数的方法和设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014022964A1 (zh) * 2012-08-07 2014-02-13 华为技术有限公司 传输消息的方法、设备及终端
CN106302274A (zh) * 2016-08-26 2017-01-04 清华大学 一种大规模mimo系统多用户信道估计与跟踪方法
CN111478783A (zh) * 2019-01-23 2020-07-31 中国移动通信有限公司研究院 一种配置无线传输参数的方法和设备
CN110035388A (zh) * 2019-03-21 2019-07-19 平安科技（深圳）有限公司 确定用户区域的方法、计算机可读存储介质及终端设备
CN110113088A (zh) * 2019-05-07 2019-08-09 东南大学 一种分离型数模混合天线系统波达角智能化估计方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4266603A4

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115460633A (zh) * 2022-09-29 2022-12-09 中电信数智科技有限公司 6g网络ai信号优化的方法、系统、设备和存储介质
CN115460633B (zh) * 2022-09-29 2023-10-27 中电信数智科技有限公司 6g网络ai信号优化的方法、系统、设备和存储介质

Also Published As

Publication number Publication date
EP4266603A4 (en) 2024-09-18
CN114978382A (zh) 2022-08-30
EP4266603A1 (en) 2023-10-25

Similar Documents

Publication Publication Date Title
WO2022174642A1 (zh) 基于空间划分的数据处理方法和通信装置
Gante et al. Deep learning architectures for accurate millimeter wave positioning in 5G
KR102154481B1 (ko) 딥러닝을 이용한 대규모 mimo 시스템의 빔포밍 장치 및 방법
CN108646213A (zh) 一种室内多径环境下直达波aoa判定方法
US20230362039A1 (en) Neural network-based channel estimation method and communication apparatus
CN110113088A (zh) 一种分离型数模混合天线系统波达角智能化估计方法
CN113595666A (zh) 大规模mimo系统基于时频联合相关性的信道预测方法和系统
CN113965881B (zh) 遮挡效应下的毫米波一体化通信与感知方法
US11962436B2 (en) User detection technique, and method and apparatus for channel estimation in wireless communication system supporting massive multiple-input multiple-output
Owfi et al. A Meta-learning based Generalizable Indoor Localization Model using Channel State Information
WO2023097636A1 (zh) 数据处理的方法和装置
Xiao et al. Lstm based multiple beamforming for 5g haps iot networks
WO2023070675A1 (zh) 数据处理的方法及装置
WO2024059969A1 (zh) 一种信道估计方法、装置及系统
Tang et al. Data-adaptive CNN used for multi-parameters estimation in high-precision indoor localization
CN113242072B (zh) 一种基于卷积神经网络的混合波束形成方法
WO2023169253A1 (zh) 一种通信方法及装置
WO2023160656A1 (zh) 一种通信方法及装置
WO2023179675A1 (zh) 信息处理方法和通信装置
WO2023088138A1 (zh) 信道状态信息处理方法及装置
WO2024138692A1 (zh) 空间滤波器预测方法、终端设备及网络设备
CN116367305B (zh) 一种基于掩码自编器和信号引导的无线定位方法及系统
WO2024169600A1 (zh) 一种通信方法及装置
WO2024103352A1 (zh) 一种通信方法、装置及系统
WO2023016503A1 (zh) 一种通信方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21926356

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021926356

Country of ref document: EP

Effective date: 20230719

NENP Non-entry into the national phase

Ref country code: DE