Search

CN-122027703-A - Data generation method, device and equipment for vision-aided wireless communication

CN122027703ACN 122027703 ACN122027703 ACN 122027703ACN-122027703-A

Abstract

The invention discloses a data generation method, device and equipment for vision-aided wireless communication. The method comprises the steps of constructing a task list comprising at least two communication links, for any communication link, carrying out ray tracing simulation in a pre-constructed three-dimensional city model based on three-dimensional geographic coordinates of a signal transmitter and a signal receiver, calculating all reachable paths of preset reflection times of the communication link, generating a channel data file comprising physical layer characteristics of each reachable path, generating a view angle control instruction based on the three-dimensional geographic coordinates of the signal transmitter and the signal receiver, sending the view angle control instruction to a geographic space data platform client, acquiring a scene two-dimensional image of a specific view angle, storing the scene two-dimensional image as an image file, and constructing a data sample for the communication link according to the channel data file and the image file. The embodiment of the invention can realize the alignment of the channel and the image source in the real scene.

Inventors

  • YIN SIXING
  • LIU ZHAOJUN

Assignees

  • 北京邮电大学

Dates

Publication Date
20260512
Application Date
20260126

Claims (10)

  1. 1. A method of data generation for vision-aided wireless communication, the method comprising: Constructing a task list comprising at least two communication links, for any communication link, carrying out ray tracing simulation in a pre-constructed three-dimensional city model based on three-dimensional geographic coordinates of a signal transmitter and a signal receiver of the communication link, calculating all reachable paths of preset reflection times of the communication link, and generating a channel data file comprising physical layer characteristics of each reachable path; Generating a view angle control instruction based on three-dimensional geographic coordinates of the signal transmitter and the signal receiver, and sending the view angle control instruction to a geospatial data platform client to drive the client to position a view to a specific view angle; And constructing a data sample for the communication link according to the channel data file and the image file.
  2. 2. The method of claim 1, wherein the open source map data comprises a two-dimensional outline polygon of a building, and geographic coordinates and heights of vertices of the two-dimensional outline polygon, the method further comprising: Stretching the two-dimensional outline polygon to a corresponding height along the vertical direction to generate a three-dimensional triangular mesh model, wherein the vertex coordinates of the three-dimensional triangular mesh model are determined by the geographic coordinates and the height of the vertices of the two-dimensional outline polygon; Performing geometric verification and repair on the three-dimensional triangular mesh model to eliminate model cracks and non-manifold edges and ensure that the model meets the geometric integrity required by ray tracing calculation; and configuring electromagnetic properties for the triangular patches of the repaired three-dimensional triangular mesh model to obtain the three-dimensional city model.
  3. 3. The method according to claim 2, wherein said configuring electromagnetic properties for the triangular patches of the repaired three-dimensional triangular mesh model to obtain the three-dimensional city model comprises: configuring material classification marks for triangular patches of the repaired three-dimensional triangular mesh model; And associating electromagnetic attribute parameters to the triangular patches with corresponding material classification identifications based on a predefined material electromagnetic attribute mapping library to form the three-dimensional city model, wherein the electromagnetic attribute parameters comprise relative dielectric constants and conductivities.
  4. 4. The method of claim 1, wherein the perspective instructions comprise an observation center point and a heading angle, wherein the generating perspective control instructions based on three-dimensional geographic coordinates of the signal transmitter and the signal receiver comprises: Taking the three-dimensional geographic coordinates of the signal receiver as an observation center point; calculating a direction vector from the signal receiver to the signal transmitter; and projecting the direction vector to a horizontal plane, and taking the direction indicated by the projected vector as the course angle, so that the signal transmitter is positioned right above the two-dimensional image of the scene.
  5. 5. The method of claim 1, wherein the data samples include a metadata file, a channel data file, and an image file, wherein the constructing data samples for the communication link based on the channel data file and the image file comprises: Forming a joint key by using the identification of the signal transmitter and the identification of the signal receiver of the communication link as a global unique index of the data sample; establishing an association relationship between a channel data file and an image file corresponding to the communication link based on the global unique index; generating a metadata file that encapsulates, in a structured format, the simulation parameters of the communication link, the globally unique index, and the storage paths of the associated channel data file and image file.
  6. 6. The method as recited in claim 1, further comprising: The communication links in the task list are distributed to a plurality of parallel execution computing units to synchronously execute ray tracing simulation generation channel data of the plurality of communication links and parallel generation image files.
  7. 7. A data generation apparatus for vision-aided wireless communication, the apparatus comprising: The system comprises a channel data generation module, a three-dimensional city model, a channel data generation module and a communication module, wherein the channel data generation module is used for constructing a task list comprising at least two communication links, for any communication link, carrying out ray tracing simulation in a pre-constructed three-dimensional city model based on three-dimensional geographic coordinates of a signal transmitter and a signal receiver of the communication link, calculating all reachable paths of preset reflection times of the communication link, and generating a channel data file comprising physical layer characteristics of each reachable path; The image data generation module is used for generating a view angle control instruction based on the three-dimensional geographic coordinates of the signal transmitter and the signal receiver, and sending the view angle control instruction to a geospatial data platform client to drive the client to position a view to a specific view angle; And the data sample generation module is used for constructing data samples for the communication link according to the channel data file and the image file.
  8. 8. An electronic device, the electronic device comprising: At least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the data generation method of vision-aided wireless communication of any one of claims 1-6.
  9. 9. A computer readable storage medium storing computer instructions for causing a processor to execute the data generation method of vision-aided wireless communication of any one of claims 1-6.
  10. 10. A computer program product comprising a computer program which, when executed by a processor, implements a data generation method of vision-aided wireless communication according to any one of claims 1-6.

Description

Data generation method, device and equipment for vision-aided wireless communication Technical Field The present invention relates to the field of wireless communications technologies, and in particular, to a method, an apparatus, and a device for generating data in visual auxiliary wireless communications. Background With the rapid development of 5G and 6G wireless communication technologies, channel modeling and channel estimation are key links restricting the performance of a communication system. However, multipath reflections, occlusions, and architectural layouts in complex urban environments have significant impact on wireless propagation, making it difficult for channel estimation based solely on mathematical models to accurately reflect the true propagation characteristics. In recent years, vision-aided wireless Communication (Vision-AIDED WIRELESS Communication) has become a research hotspot. The key idea is to use the environmental image information from the camera or other visual sensors to assist in predicting the channel characteristics, beam direction or shielding condition, thereby improving the sensing and self-adapting capability of the communication system. Research in this direction is driven by the development of deep learning and computer vision techniques, but practical applications still face a key bottleneck, the lack of large-scale, spatially consistent image-channel paired data sets. The current visual data set is mainly oriented to computer visual tasks such as target detection, semantic segmentation and scene understanding, focuses on semantic information and appearance characteristics of images, and lacks geometric and electromagnetic characteristic descriptions related to wireless propagation environments. On the other hand, the data set in the existing wireless communication field is mainly based on channel parameters, path loss, angles or spectrum characteristics, and the visual dimension and the spatial semantic structure of the environment are often ignored. In addition, these two types of data sets are generally heterogeneous in source, have no unified coordinate system, have inconsistent sampling angles, and are difficult to realize spatial alignment and semantic correspondence between visual and wireless modalities. Therefore, the existing data set of vision-aided wireless communication is based on an idealized geometric model or a simulated simplified building structure, can not embody the visual characteristics of buildings, streets and shields in a real environment, and is difficult to meet the requirements of reality, alignment and expandability in the vision-aided wireless communication research. What is needed is a data acquisition method that can automatically generate image-channel paired data samples in a real geographic scenario to provide high quality data support for subsequent deep learning based channel modeling and environmental awareness. Disclosure of Invention The invention provides a data generation method, device and equipment for visual auxiliary wireless communication, which are used for solving the problem that the existing simulation data are mostly based on an idealized geometric model or a simplified building structure and cannot reflect the visual characteristics of buildings, streets and shields in a real environment. According to an aspect of the present invention, there is provided a data generation method of vision-aided wireless communication, including: Constructing a task list comprising at least two communication links, for any communication link, carrying out ray tracing simulation in a pre-constructed three-dimensional city model based on three-dimensional geographic coordinates of a signal transmitter and a signal receiver of the communication link, calculating all reachable paths of preset reflection times of the communication link, and generating a channel data file comprising physical layer characteristics of each reachable path; Generating a view angle control instruction based on three-dimensional geographic coordinates of the signal transmitter and the signal receiver, and sending the view angle control instruction to a geospatial data platform client to drive the client to position a view to a specific view angle; And constructing a data sample for the communication link according to the channel data file and the image file. According to another aspect of the present invention, there is provided a data generating apparatus for vision-aided wireless communication, comprising: The system comprises a channel data generation module, a three-dimensional city model, a channel data generation module and a communication module, wherein the channel data generation module is used for constructing a task list comprising at least two communication links, for any communication link, carrying out ray tracing simulation in a pre-constructed three-dimensional city model based on three-dimensional geographic coordinates of a signal transmitte