Beam prediction method and device, network equipment, camera and storage medium

Abstract

The invention provides a beam prediction method and device, network equipment, a camera and a storage medium, and relates to the technical field of wireless communication. The method comprises the following steps: acquiring user feature information and environment feature information of a target user; and according to the user feature information and the environment feature information, predicting a target beam sequence number sent to the target user by a network device and / or a probability that the target user is shielded. According to the scheme of the invention, beam prediction and / or shielding prediction are / is realized by using the environment feature information, user privacy and public security are protected, image data stream processing is not needed, and resource overhead is saved. The problems that in the prior art, image data collected through a third-party camera relates to user privacy and public safety, the storage space occupied by the image data is large, and processing of the image data needs a large number of computing resources are solved.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a beam prediction method, device, network equipment, camera and storage medium. Background Art

[0002] The large bandwidth support of millimeter waves and large-scale Multiple-Input Multiple-Output (MIMO) have the problem of unstable transmission links due to high path loss and high sensitivity to link obstruction. The occlusion prediction algorithm predicts the situation of link obstruction in the future time window, and switches the communication base station in time to establish a new line-of-sight link, which can solve the link instability problem caused by link obstruction. In the existing technology, environmental perception data (such as environmental image information data obtained by third-party cameras) is used to assist in realizing low-latency and highly reliable beam prediction and occlusion prediction methods. However, the image data collected by third-party cameras involves user privacy and public safety issues, the image data occupies a large storage space, and the processing of image data requires a lot of computing resources. Summary of the invention

[0003] The purpose of the present invention is to provide a beam prediction method, device, network equipment, camera and storage medium, which predicts the target beam sequence number sent by the network device to the target user and / or the probability of the target user being blocked by the user characteristic information and the environment characteristic information.

[0004] To achieve the above object, an embodiment of the present invention provides a beam prediction method, which is applied to a network device, including:

[0005] Obtain user characteristic information and environmental characteristic information of the target user;

[0006] According to the user characteristic information and the environment characteristic information, a target beam sequence number sent by a network device to the target user and / or a probability that the target user is blocked are predicted.

[0007] Furthermore, the method further comprises:

[0008] When the probability that the target user is blocked is greater than a preset threshold, determining the target network device according to the target beam sequence number;

[0009] A transmission link is established between the target user and the target network device.

[0010] Furthermore, the user characteristic information includes at least one of the following:

[0011] Location information;

[0012] A historical beam sequence number sent by the network device to the target user;

[0013] A pilot receiving signal;

[0014] The environmental characteristic information includes at least one of the following:

[0015] Building information;

[0016] Vehicle information;

[0017] Road information;

[0018] Lawn information.

[0019] Further, the user characteristic information of the target user includes: a user characteristic information set sent by at least one of the target users;

[0020] The environmental feature information includes: a set of environmental feature information sent by at least one camera;

[0021] The user characteristic information set includes at least one set of user characteristic information, and the environment characteristic information set includes at least one set of environment characteristic information.

[0022] Further, predicting the target beam sequence number sent by the network device to the target user and / or the probability that the target user is blocked according to the user characteristic information and the environment characteristic information includes:

[0023] Determine a target feature subset from the user feature information set and the environment feature information set by a target algorithm, wherein the target feature subset includes target environment feature information of the environment where the target user is located and / or target user feature information;

[0024] The target beam number and / or the probability that the target user is blocked are predicted according to the target feature subset.

[0025] Further, in the case of predicting the target beam sequence number, the target environment feature information includes: vehicle information and road information, and / or, the target feature information includes a historical beam sequence number sent by the network device to the target user;

[0026] In case of predicting the probability of the target user being blocked, the target environment feature information includes: vehicle information, and / or the target feature information includes location information of the target user.

[0027] Furthermore, the method further comprises:

[0028] A first prediction accuracy for predicting the target beam number and a second prediction accuracy for predicting the probability that the target user is blocked are determined according to the target feature subset.

[0029] Further, predicting the target beam sequence number sent by the network device to the target user according to the environmental feature information and the feature information includes:

[0030] Construct a beam prediction network;

[0031] Training the beam prediction network to obtain a target beam prediction network;

[0032] The target feature subset is used as the input of the target beam prediction network to obtain the target beam sequence number.

[0033] Further, predicting the probability of the target user being blocked according to the environmental feature information and the feature information includes:

[0034] Build an occlusion prediction network;

[0035] Training the occlusion prediction network to obtain a target occlusion prediction network;

[0036] The target feature subset is used as the input of the target occlusion prediction network to obtain the probability that the target user is occluded.

[0037] To achieve the above object, an embodiment of the present invention provides a beam prediction method, which is applied to a camera, comprising:

[0038] Collect image data stream;

[0039] Extracting environmental feature information from the image data stream;

[0040] The environmental characteristic information is sent to a network device.

[0041] Further, extracting environmental feature information from the image data stream includes:

[0042] Construct feature extraction network;

[0043] Training the feature extraction network to obtain a target feature extraction network;

[0044] Environmental feature information is extracted from the image data stream through the target feature extraction network.

[0045] To achieve the above object, an embodiment of the present invention provides a beam prediction device, which is applied to a network device, including:

[0046] An acquisition module is used to acquire user characteristic information and environment characteristic information of a target user;

[0047] A prediction module is used to predict the target beam sequence number sent by the network device to the target user and / or the probability that the target user is blocked according to the user characteristic information and the environment characteristic information.

[0048] To achieve the above object, an embodiment of the present invention provides a beam prediction device, which is applied to a camera, comprising:

[0049] An acquisition module, used for acquiring image data stream;

[0050] An extraction module, used for extracting environmental feature information from the image data stream;

[0051] The sending module is used to send the environmental characteristic information to the network device.

[0052] To achieve the above object, an embodiment of the present invention provides a network device, including: a transceiver and a processor;

[0053] The transceiver is used to obtain user characteristic information and environmental characteristic information of the target user;

[0054] The processor is used to predict, based on the user characteristic information and the environment characteristic information, a target beam sequence number sent by the network device to the target user and / or a probability that the target user is blocked.

[0055] To achieve the above object, an embodiment of the present invention provides a camera, including: a transceiver and a processor;

[0056] The transceiver is used to collect image data stream;

[0057] The processor is used to extract environmental feature information from the image data stream;

[0058] The transceiver is also used to send the environmental characteristic information to a network device.

[0059] To achieve the above-mentioned purpose, an embodiment of the present invention provides a terminal device, including a transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; when the processor executes the program or instruction, the beam prediction method as described above is implemented.

[0060] To achieve the above objective, an embodiment of the present invention provides a readable storage medium having a program or instruction stored thereon, and the program or instruction, when executed by a processor, implements the steps in the beam prediction method as described above.

[0061] The beneficial effects of the above technical solution of the present invention are as follows:

[0062] The beam prediction method of the embodiment of the present invention predicts the target beam sequence number sent by the network device to the target user and / or the probability of the target user being blocked by the acquired user characteristic information and environmental characteristic information of the target user. The beam prediction method of the embodiment of the present invention uses the environmental characteristic information to implement beam prediction and / or occlusion prediction, thereby protecting user privacy and public safety, and does not require processing of image data streams, saving resource overhead. It solves the problem that the image data collected by the prior art using a third-party camera involves user privacy and public safety, the image data occupies a large storage space, and the processing of the image data requires a large amount of computing resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 One of the flowcharts of the beam prediction method applied to a network device according to an embodiment of the present invention;

[0064] Figure 2 A schematic diagram of the structure of a beam prediction network and an occlusion prediction network applied to an embodiment of the present invention;

[0065] Figure 3 This is a second flowchart of a beam prediction method applied to a camera according to an embodiment of the present invention;

[0066] Figure 4 A logical flow diagram of a beam prediction method according to an embodiment of the present invention;

[0067] Figure 5 A schematic diagram of the structure of a beam prediction device applied to a network device according to an embodiment of the present invention;

[0068] Figure 6 It is a structural schematic diagram of a beam prediction device applied to a camera according to an embodiment of the present invention;

[0069] Figure 7 is a structural diagram of a network device according to an embodiment of the present invention;

[0070] Figure 8 is a structural diagram of a camera according to an embodiment of the present invention;

[0071] Fig. 9 A structural diagram of a terminal device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0072] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0073] It should be understood that the references to "one embodiment" or "an embodiment" throughout the specification mean that the specific features, structures, or characteristics associated with the embodiment are included in at least one embodiment of the present invention. Therefore, the references to "in one embodiment" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0074] In various embodiments of the present invention, it should be understood that the size of the serial numbers of the following processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0075] Additionally, the terms "system" and "network" are often used interchangeably herein.

[0076] In the embodiments provided in the present application, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined according to A. However, it should also be understood that determining B according to A does not mean determining B only according to A, and B can also be determined according to A and / or other information.

[0077] like Figure 1 As shown, an embodiment of the present invention provides a beam prediction method, which is applied to a network device and includes the following steps:

[0078] Step 101, obtaining user characteristic information and environment characteristic information of a target user;

[0079] Step 102: predicting the target beam number sent by the network device to the target user and / or the probability that the target user is blocked according to the user characteristic information and the environment characteristic information.

[0080] Optionally, the environmental feature information includes a target camera image or environmental feature information sent by a target base station.

[0081] It should be noted that the network equipment includes a base station and a network side equipment.

[0082] The beam prediction method of the embodiment of the present invention predicts the target beam sequence number sent by the network device to the target user and / or the probability of the target user being blocked by the acquired user characteristic information and environmental characteristic information of the target user. The beam prediction method of the embodiment of the present invention uses the environmental characteristic information to implement beam prediction and / or occlusion prediction, thereby protecting user privacy and public safety, and does not require processing of image data streams, saving resource overhead. It solves the problem that the image data collected by the prior art using a third-party camera involves user privacy and public safety, the image data occupies a large storage space, and the processing of the image data requires a large amount of computing resources.

[0083] Optionally, the method further comprises:

[0084] When the probability that the target user is blocked is greater than a preset threshold, determining the target network device according to the target beam sequence number;

[0085] A transmission link is established between the target user and the target network device.

[0086] It should be noted that the target network device and the network device are different network devices, and the target network device may be a base station.

[0087] The beam prediction method of an embodiment of the present invention determines the target network device according to the predicted target beam number of the target user when the target device is blocked, and establishes a transmission link between the target user and the target network device, that is, it can maintain stable communication quality by switching the communication base station.

[0088] In one embodiment of the present invention, the optimal beam direction and line-of-sight link condition are obtained by maximizing the transmission rate.

[0089] Optionally, the user characteristic information includes at least one of the following:

[0090] Location information;

[0091] A historical beam sequence number sent by the network device to the target user;

[0092] A pilot receiving signal;

[0093] The environmental characteristic information includes at least one of the following:

[0094] Building information;

[0095] Vehicle information;

[0096] Road information;

[0097] Lawn information.

[0098] The beam prediction method of the embodiment of the present invention predicts the target beam number by using the acquired environmental feature information, thereby eliminating surface features of buildings, vehicles or roads, and only retaining the category and layout information of environmental objects. It is more private than the original image information and does not require processing of image data, but only needs to process the environmental feature information, thereby reducing the computational overhead of the communication system.

[0099] Optionally, the user characteristic information of the target user includes: a user characteristic information set sent by at least one of the target users;

[0100] The environmental feature information includes: a set of environmental feature information sent by at least one camera;

[0101] The user characteristic information set includes at least one set of user characteristic information, and the environment characteristic information set includes at least one set of environment characteristic information.

[0102] It should be noted that the third-party camera end uploads the environmental characteristic information (or, target environmental characteristic information) corresponding to the user's nearby environment to the network device. If there are multiple target users in the environment at the same time, each of the target users uploads their user characteristic information. The beam prediction network and occlusion prediction network deployed by the network device end realize beam prediction and occlusion prediction based on the acquired environmental characteristic information (or, target environmental characteristic information) and the user characteristic information uploaded by each target user.

[0103] Optionally, predicting, according to the user characteristic information and the environment characteristic information, a target beam sequence number sent by the network device to the target user and / or a probability that the target user is blocked includes:

[0104] Determine a target feature subset from the user feature information set and the environment feature information set by a target algorithm, wherein the target feature subset includes target environment feature information of the environment where the target user is located and / or target user feature information;

[0105] The target beam number and / or the probability that the target user is blocked are predicted according to the target feature subset.

[0106] Optionally, the target algorithm is a feature selection algorithm.

[0107] In one embodiment of the present invention, the feature selection algorithm includes: a scoring sorting method and a sequential forward floating selection method.

[0108] It should be noted that the core of the target algorithm is to select the optimal feature subsets from the user feature information set and the environment feature information set to perform the number prediction and occlusion prediction;

[0109] The user characteristic information set and the environment characteristic information set can be expressed as:

[0110]

[0111] Assuming that the optimization goal of the algorithm is prediction accuracy A, initialize the feature subset X (0) and historical feature set Xh ist The algorithm is an empty set, and the number of iterations i←0 is initialized. The algorithm optimizes the feature subset by continuously iterating, mainly including two alternating iterations: one step is to add features, that is, to select the features that can best increase the prediction accuracy A from the unselected features:

[0112]

[0113] Then update:

[0114] X (i+1) ←X (i) +x +

[0115] Xh ist ←{Xh ist ,X (i)}

[0116] i←i+1

[0117] Then reduce the features, that is, if there is feature information in the selected feature information that will reduce the prediction accuracy A, then select the feature subset X (i) Delete the feature information that can most reduce the prediction accuracy A. Specifically, if:

[0118]

[0119] Then delete the feature information x - :

[0120]

[0121] Then update:

[0122] X (i+1) ←X (i) +x -

[0123] i←i+1

[0124] Replace the prediction accuracy A with the first prediction accuracy And the second prediction accuracy The target feature subset X corresponding to the target beam number can be predicted by the feature selection algorithm. bmThe target feature subset X corresponding to the predicted probability of the target user being blocked bl .

[0125] In one embodiment of the present invention, in the case of predicting the target beam number, the target feature subset is a vehicle and a road;

[0126] In case of predicting the probability that the target user is blocked, the target feature subset is a vehicle.

[0127] Optionally, in the case of predicting the target beam sequence number, the target environment feature information includes: vehicle information and road information, and / or the target feature information includes a historical beam sequence number sent by the network device to the target user;

[0128] In case of predicting the probability of the target user being blocked, the target environment feature information includes: vehicle information, and / or the target feature information includes location information of the target user.

[0129] The beam prediction method of the embodiment of the present invention predicts the beam and occlusion condition of the target user through the target feature information, without the need for pilot training and beam scanning, thereby achieving low-latency and highly reliable beam prediction and occlusion prediction.

[0130] Optionally, the method further comprises:

[0131] A first prediction accuracy for predicting the target beam number and a second prediction accuracy for predicting the probability that the target user is blocked are determined according to the target feature subset.

[0132] Optionally, the target feature subset is used as input, and the first prediction accuracy is calculated by the following function:

[0133]

[0134] The second prediction accuracy is calculated by the following function:

[0135]

[0136] in, is the first prediction accuracy; (n) is the labeled nth test sample; is the second prediction accuracy; 1(·) is the indicator function; is the target beam sequence number; is the output of the prediction network; is the output of the occlusion network; N tot is the total number of test samples.

[0137] The beam prediction method of the embodiment of the present invention can determine the accuracy of beam prediction and occlusion prediction by calculating the first prediction accuracy and the second prediction accuracy, thereby providing data support for base station switching (reconstruction of transmission links).

[0138] Optionally, predicting a target beam sequence number sent by a network device to the target user according to the environment feature information and the feature information includes:

[0139] Construct a beam prediction network;

[0140] Training the beam prediction network to obtain a target beam prediction network;

[0141] The target feature subset is used as the input of the target beam prediction network to obtain the target beam sequence number.

[0142] like Figure 2 As shown in FIG. 1 , the beam prediction network is composed of a convolution layer, a pooling layer, a fully connected layer, a Dropout layer, and a stack of nonlinear activation functions. The input is the target feature subset, and the output is the optimal beam number of the target user. The pre-training function used is as follows:

[0143]

[0144] Among them, Ω bm M is a trainable parameter of the wave number prediction network; bm is the size of the beam codebook, y bm is the output of the beam prediction network, and y{bm,lab} is the label sequence number of the beam.

[0145] Specifically, the model of the massive MIMO communication system is as follows:

[0146]

[0147] in, is the received signal on the kth carrier; is the transmitted signal on the kth carrier; ω is the beam vector; is the noise on the kth carrier, is the channel matrix on the kth carrier, which can be further expressed as:

[0148]

[0149] Among them, f D,k Downlink k-th carrier frequency; α l , and v l are the attenuation, phase shift and delay of the lth propagation path respectively; and It can be expressed as:

[0150]

[0151] in, d is the antenna spacing; c is the speed of light; are the corresponding azimuth and elevation angles. Assume that ω is selected from the codebook ω, M bm =|ω| is the codebook size, then the achievable transmission rate is:

[0152]

[0153] in, is the power of the transmitted signal; then the optimal beam vector can be expressed as:

[0154]

[0155] The optimal beam vector ω opt The serial number in the codebook ω can be used as the training label y{bm,lab} of the beam prediction network;

[0156] Then the target beam number is calculated by the following function:

[0157]

[0158] Optionally, predicting the probability of the target user being blocked according to the environmental feature information and the feature information includes:

[0159] Build an occlusion prediction network;

[0160] Training the occlusion prediction network to obtain a target occlusion prediction network;

[0161] The target feature subset is used as the input of the target occlusion prediction network to obtain the probability that the target user is occluded.

[0162] like Figure 2 As shown in FIG. 1 , the occlusion prediction network is composed of a convolution layer, a pooling layer, a fully connected layer, a Dropout layer, and a stack of nonlinear activation functions. The input is the target feature subset, and the output is the probability that the target user is occluded. The pre-training function used is as follows:

[0163] f Lbl (Ω bl )=y{bl,lab}logy bl +(1-y{bl,lab})log(1-y bl )

[0164] Among them, Ω bl is a trainable parameter of the occlusion prediction network; ybl is the output of the occlusion prediction network; y{bl,lab} is the label sequence number of the beam.

[0165] like Figure 3 As shown, an embodiment of the present invention provides a beam prediction method, which is applied to a camera and includes the following steps:

[0166] Step 301, collecting image data stream;

[0167] Step 302, extracting environmental feature information from the image data stream;

[0168] Step 303: Send the environmental characteristic information to the network device.

[0169] The beam prediction method of the embodiment of the present invention collects image data streams through a camera, and extracts environmental feature information on the camera side, thereby sending the environmental feature information to a network device. The image data stream is processed on the camera side, saving resource overhead of the network device.

[0170] Optionally, extracting environmental feature information from the image data stream includes:

[0171] Construct feature extraction network;

[0172] Training the feature extraction network to obtain a target feature extraction network;

[0173] Environmental feature information is extracted from the image data stream through the target feature extraction network.

[0174] Optionally, the feature extraction network is composed of a convolutional layer, a pooling layer, a fully connected layer, a signal discarding Dropout layer and a non-linear activation function stack, and the network input is an image data stream acquired by the camera.

[0175] It should be noted that the size of the input layer of the feature extraction network is consistent with the size of the input image, and the number of input channels is 3, and the size of the output layer of the feature extraction network is consistent with the size of the input image, and the number of output channels is 1.

[0176] Optionally, the feature extraction network is trained by the following function:

[0177]

[0178] Among them, Ω Ext are the trainable parameters of the environment image semantic extraction network, W and H are the width and height of the input image, M con is the total category of environmental characteristic information, is the semantic feature map output by the network, C∈RH×W is the training label of the environmental feature map, C(i,j) is the environmental feature classification value at the training label coordinate (i,j) of the feature map, C(i,j)=0,1,…M con The value of -1 is a non-negative integer.

[0179] like Figure 4 As shown, at least one group of image data streams collected by at least one camera is subjected to a feature extraction network to extract environmental feature information; at least one target user sends user feature information; a target feature subset is determined after selection by a feature selection algorithm; the target feature subset is used as an input to the beam prediction network to obtain a target beam sequence number of the target user; and the target feature subset is used as an input to the occlusion prediction network to obtain an occlusion state of the target user.

[0180] like Figure 5 As shown, an embodiment of the present invention provides a beam prediction device, which is applied to a network device 500, including:

[0181] The acquisition module 501 is used to acquire user characteristic information and environment characteristic information of the target user;

[0182] The prediction module 502 is used to predict the target beam sequence number sent by the network device to the target user and / or the probability that the target user is blocked according to the user characteristic information and the environment characteristic information.

[0183] The beam prediction device of the embodiment of the present invention predicts the target beam sequence number sent by the network device to the target user and / or the probability of the target user being blocked by the acquired user characteristic information and environmental characteristic information of the target user. The beam prediction method of the embodiment of the present invention uses the environmental characteristic information to implement beam prediction and / or occlusion prediction, thereby protecting user privacy and public safety, and does not require processing of image data streams, saving resource overhead. It solves the problem that the image data collected by the prior art using a third-party camera involves user privacy and public safety, the image data occupies a large storage space, and the processing of the image data requires a large amount of computing resources.

[0184] like Figure 6 As shown, an embodiment of the present invention provides a beam prediction device 600, which is applied to a camera, including:

[0185] The acquisition module 601 is used to acquire image data stream;

[0186] An extraction module 602, used to extract environmental feature information from the image data stream;

[0187] The sending module 603 is used to send the environmental characteristic information to the network device.

[0188] The beam prediction device of the embodiment of the present invention collects image data streams through a camera, and extracts environmental feature information on the camera side, thereby sending the environmental feature information to a network device. The image data stream is processed on the camera side, saving resource overhead of the network device.

[0189] like Figure 7 As shown, an embodiment of the present invention provides a network device 700, including: a transceiver 720 and a processor 710;

[0190] The transceiver is used to obtain user characteristic information and environmental characteristic information of the target user;

[0191] The processor is used to predict, based on the user characteristic information and the environment characteristic information, a target beam sequence number sent by the network device to the target user and / or a probability that the target user is blocked.

[0192] like Figure 8 As shown, an embodiment of the present invention provides a camera 800, including: a transceiver 820 and a processor 810;

[0193] The transceiver is used to collect image data stream;

[0194] The processor is used to extract environmental feature information from the image data stream;

[0195] The transceiver is also used to send the environmental characteristic information to a network device.

[0196] A terminal device according to another embodiment of the present invention, such as Fig. 9 As shown, it includes a transceiver 910, a processor 900, a memory 920, and a program or instruction stored in the memory 920 and executable on the processor 900; when the processor 900 executes the program or instruction, the above-mentioned method for beam prediction is implemented.

[0197] The transceiver 910 is used to receive and send data under the control of the processor 900 .

[0198] Among them, Fig. 9In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by processor 900 and various circuits of memory represented by memory 920 are linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The transceiver 910 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. For different user devices, the user interface 930 may also be an interface capable of externally and internally connecting required devices, and the connected devices include but are not limited to a keypad, a display, a speaker, a microphone, a joystick, and the like.

[0199] The processor 900 is responsible for managing the bus architecture and general processing, and the memory 920 can store data used by the processor 900 when performing operations.

[0200] A readable storage medium according to an embodiment of the present invention stores a program or instruction thereon. When the program or instruction is executed by a processor, the steps in the beam prediction method as described above are implemented and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0201] The processor is the processor in the terminal device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0202] It should be further explained that the terminals described in this specification include but are not limited to smart phones, tablet computers, etc., and many of the functional components described are called modules in order to more particularly emphasize the independence of their implementation methods.

[0203] In the embodiment of the present invention, module can be implemented with software so that it can be executed by various types of processors. For example, an executable code module of an identification can include one or more physical or logical blocks of computer instructions, for example, it can be constructed as an object, process or function. Nevertheless, the executable code of the identified module does not need to be physically located together, but can include different instructions stored in different positions, and when these instructions are logically combined together, it constitutes a module and realizes the specified purpose of the module.

[0204] In fact, executable code module can be a single instruction or many instructions, and can even be distributed on a plurality of different code segments, distributed among different programs, and distributed across a plurality of memory devices. Similarly, operating data can be identified in the module, and can be implemented and organized in the data structure of any appropriate type according to any appropriate form. The operating data can be collected as a single data set, or can be distributed in different locations (including on different storage devices), and can only be present on a system or network as an electronic signal at least in part.

[0205] When a module can be implemented by software, considering the level of existing hardware technology, a person skilled in the art can build a corresponding hardware circuit to implement the corresponding function of the module that can be implemented by software without considering the cost. The hardware circuit includes a conventional very large scale integration (VLSI) circuit or gate array and existing semiconductors such as logic chips, transistors, or other discrete components. The module can also be implemented by a programmable hardware device, such as a field programmable gate array, a programmable array logic, a programmable logic device, etc.

[0206] The above exemplary embodiments are described with reference to the accompanying drawings, and many different forms and embodiments are feasible without departing from the spirit and teachings of the present invention. Therefore, the present invention should not be constructed as a limitation of the exemplary embodiments proposed herein. More specifically, these exemplary embodiments are provided so that the present invention will be perfect and complete, and the scope of the present invention will be conveyed to those who are familiar with the technology. In these figures, the component sizes and relative sizes may be exaggerated for clarity. The terms used here are only based on the purpose of describing specific exemplary embodiments and are not intended to be limiting. As used herein, unless the text clearly indicates otherwise, the singular forms "one", "an" and "the" are intended to include these multiple forms. It will be further understood that the terms "including" and / or "comprising" when used in this specification indicate the presence of the features, integers, steps, operations, components and / or components, but do not exclude the presence or increase of one or more other features, integers, steps, operations, components, components and / or their groups. Unless otherwise indicated, when stated, a range of values ​​includes the upper and lower limits of that range and any subranges therebetween.

[0207] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A beam prediction method, applied to a network device, characterized in that: include: Obtain user characteristic information and environmental characteristic information of the target user; According to the user characteristic information and the environment characteristic information, a target beam sequence number sent by a network device to the target user and / or a probability that the target user is blocked are predicted.

2. The beam prediction method according to claim 1, characterized in that: The method further comprises: When the probability that the target user is blocked is greater than a preset threshold, determining the target network device according to the target beam sequence number; A transmission link is established between the target user and the target network device.

3. The beam prediction method according to claim 1, characterized in that: The user characteristic information includes at least one of the following: Location information; A historical beam sequence number sent by the network device to the target user; A pilot receiving signal; The environmental characteristic information includes at least one of the following: Building information; Vehicle information; Road information; Lawn information.

4. The beam prediction method according to claim 1, characterized in that: The user characteristic information of the target user includes: a user characteristic information set sent by at least one of the target users; The environmental feature information includes: a set of environmental feature information sent by at least one camera; The user characteristic information set includes at least one set of user characteristic information, and the environment characteristic information set includes at least one set of environment characteristic information.

5. The beam prediction method according to claim 4, characterized in that: The predicting, based on the user characteristic information and the environment characteristic information, a target beam sequence number sent by a network device to the target user and / or a probability that the target user is blocked includes: Determine a target feature subset from the user feature information set and the environment feature information set by a target algorithm, wherein the target feature subset includes target environment feature information of the environment where the target user is located and / or target user feature information; The target beam number and / or the probability that the target user is blocked are predicted according to the target feature subset.

6. The beam prediction method according to claim 5, characterized in that: In the case of predicting the target beam sequence number, the target environment feature information includes: vehicle information and road information, and / or the target feature information includes a historical beam sequence number sent by the network device to the target user; In case of predicting the probability of the target user being blocked, the target environment feature information includes: vehicle information, and / or the target feature information includes location information of the target user.

7. The beam prediction method according to claim 5, characterized in that: The method further comprises: A first prediction accuracy for predicting the target beam number and a second prediction accuracy for predicting the probability that the target user is blocked are determined according to the target feature subset.

8. The beam prediction method according to claim 5, characterized in that: The predicting, based on the environmental feature information and the feature information, a target beam sequence number sent by a network device to the target user comprises: Construct a beam prediction network; Training the beam prediction network to obtain a target beam prediction network; The target feature subset is used as the input of the target beam prediction network to obtain the target beam sequence number.

9. The beam prediction method according to claim 5, characterized in that: The predicting, according to the environmental feature information and the feature information, the probability that the target user is blocked includes: Build an occlusion prediction network; Training the occlusion prediction network to obtain a target occlusion prediction network; The target feature subset is used as the input of the target occlusion prediction network to obtain the probability that the target user is occluded.

10. A beam prediction method, applied to a camera, characterized in that: include: Collect image data stream; Extracting environmental feature information from the image data stream; The environmental characteristic information is sent to a network device.

11. The beam prediction method according to claim 10, characterized in that: The step of extracting environmental feature information from the image data stream comprises: Construct feature extraction network; Training the feature extraction network to obtain a target feature extraction network; Environmental feature information is extracted from the image data stream through the target feature extraction network.

12. A beam prediction device, applied to a network device, characterized in that: include: An acquisition module is used to acquire user characteristic information and environment characteristic information of a target user; A prediction module is used to predict the target beam sequence number sent by the network device to the target user and / or the probability that the target user is blocked according to the user characteristic information and the environment characteristic information.

13. A beam prediction device, applied to a camera, characterized in that: include: An acquisition module, used for acquiring image data stream; An extraction module, used for extracting environmental feature information from the image data stream; The sending module is used to send the environmental characteristic information to the network device.

14. A network device, characterized in that: include: transceivers and processors; The transceiver is used to obtain user characteristic information and environmental characteristic information of the target user; The processor is used to predict, based on the user characteristic information and the environment characteristic information, a target beam sequence number sent by the network device to the target user and / or a probability that the target user is blocked.

15. A camera, characterized in that: include: transceivers and processors; The transceiver is used to collect image data stream; The processor is used to extract environmental feature information from the image data stream; The transceiver is also used to send the environmental characteristic information to a network device.

16. A terminal device, comprising: A transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; characterized in that when the processor executes the program or instruction, the beam prediction method as described in any one of claims 1 to 9 is implemented; or, when the processor executes the program or instruction, the beam prediction method as described in claim 10 or 11 is implemented.

17. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by the processor, the steps in the beam prediction method as described in any one of claims 1 to 9 are implemented; or, when the program or instruction is executed by the processor, the beam prediction method as described in claim 10 or 11 is implemented.

Images