Search

DE-102024210905-A1 - Methods for improving communication via a vehicle's wireless communication network

DE102024210905A1DE 102024210905 A1DE102024210905 A1DE 102024210905A1DE-102024210905-A1

Abstract

The invention relates to methods for improving communication via a wireless communication network (110, 210) of a vehicle (100), wherein the vehicle (100) comprises several components communicating via the wireless communication network (110, 210), the method comprising: providing (316) communication information (310), wherein the communication information (310) was obtained by analyzing protocol data (302) using a first machine learning model, preferably a large-language model (LLM), according to one or more predefined criteria, wherein the protocol data (302) was acquired from and/or generated by and/or related to the communication of the components via the wireless communication network (110, 210) in the vehicle (100); Generating (318) a digital twin (322) of the wireless communication network (110, 210) based on the communication information (310) using a second machine learning model (320), preferably a second LLM; analyzing (324) the digital twin (322) using an optimization algorithm (326) to obtain improvement information (328) for potential improvements to the operation of the vehicle (100); providing (330) the improvement information to adapt the configuration of the wireless communication network (110, 210) and/or at least some of the components according to the improvement information (328).

Inventors

  • Pedro Maia De Sant Ana

Assignees

  • Robert Bosch Gesellschaft mit beschränkter Haftung

Dates

Publication Date
20260513
Application Date
20241113

Claims (13)

  1. A method for improving communication over a wireless communication network (110, 210) of a vehicle (100), wherein the vehicle (100) comprises several components communicating over the wireless communication network (110, 210), the method comprising: Providing (316) communication information (310), wherein the communication information was obtained by analyzing protocol data (302) using a first machine learning model, preferably a first large-language model (LLM), according to one or more predefined criteria, wherein the protocol data was captured from, generated by, and/or related to the communication of the components over the wireless communication network (110, 210) in the vehicle (100); Generating (318) a digital twin (322) of the wireless communication network (110, 210) based on the communication information (310) using a second machine learning model (320), preferably a second LLM; Analyzing (324) the digital twin (322) using an optimization algorithm (326) to obtain improvement information (328) for potential improvements to the operation of the vehicle (100); Providing (330) the improvement information to adapt the configuration of the wireless communication network (110, 210) and/or at least some of the components according to the improvement information.
  2. Procedure according to Claim 1 , wherein the potential improvements to the operation of the vehicle (100) for which the digital twin (322) is analyzed include at least one of the following: improvement of the configuration of the wireless communication network (110, 210), reduction of energy consumption.
  3. Procedure according to Claim 1 or 2 , where the analysis of the digital twin (322) includes the potential improvements of the vehicle (100): predictions of various future scenarios based on the communication information (310).
  4. Method for improving communication over a wireless communication network (110, 210) of a vehicle (100), wherein the vehicle (100) comprises several components communicating over the wireless communication network (110, 210), the method comprising: providing (304) protocol data (302), wherein the protocol data was captured from and/or generated by and/or relates to the communication of the components over the wireless communication network (110, 210) in the vehicle (100); analyzing (306) the protocol data (302) using a first machine learning model (308), preferably a large-language model (LLM), according to one or more predefined criteria to obtain communication information (310); providing (312) the communication information (310); Receiving (334) improvement information (328), wherein the improvement information was obtained by analyzing a digital twin (322) of the wireless communication network (110, 210) using an optimization algorithm to obtain improvement information (328) on potential improvements to the operation of the vehicle (100), wherein the digital twin (322) was generated based on the communication information (310) using a second machine learning model, preferably a second LLM; and adapting (336) the configuration of the wireless communication network (110, 210) and/or at least part of the components according to the improvement information (328).
  5. Procedure according to Claim 4 , wherein the protocol data (302) are analyzed periodically or according to a predefined schedule using the first machine learning model, in particular according to predefined requirements of the multiple components or the wireless communication network (110, 210).
  6. Procedure according to Claim 4 , wherein the log data (302) are analyzed using the first machine learning model in response to a request triggered by an external command, in particular only obtaining a deviation of the log data (302) from previously obtained log data for analysis.
  7. Procedure according to one of the Claims 4 until 6 , wherein the improvement information (328) is obtained according to one of the procedures of Claims 1 until 3 to be won.
  8. Method according to any of the preceding claims, wherein the protocol data (324) includes one or more of the following data: channel status information, channel occupancy time, channel quality indicator, control unit performance metrics in terms of energy consumption, control unit performance metrics in terms of computing efficiency, network signal strength, device plug-in and/or unplug-in events, camera compression rates, system-level errors, warning messages.
  9. Method according to one of the preceding claims, wherein the wireless communication network (110, 210) is a vehicle-specific wireless communication network (120, 210) and/or comprises several subnetworks.
  10. Method according to one of the preceding claims, wherein the multiple components comprise at least two of the following components: sensors, actuators, cameras, control units.
  11. Computing device (120, 130) comprising a processor designed to perform the method(s) according to any one of the preceding claims.
  12. A computer program that includes instructions which, when executed by a computer, cause the computer to perform the procedure(s) according to one of the Claims 1 until 10 to execute.
  13. Computer-readable medium on which the computer program is stored. Claim 12 is stored.

Description

The present invention relates to methods for improving communication via a wireless communication network of a vehicle, wherein the vehicle comprises several components communicating via the wireless communication network, a computing device and a computer program for carrying out the method. State of the art Vehicles such as passenger cars (PKW) may be equipped with a wireless communication network, in particular a vehicle-integrated wireless communication network, e.g. for controlling the vehicle or vehicle parts and especially for communication between different components of the vehicle such as sensors, actuators, cameras and control units. Disclosure of the invention According to the invention, methods for improving communication via a wireless communication network, a computing device, and a computer program with the features of the main claims are proposed. Advantageous embodiments are described in the dependent claims and the following description. The invention relates generally to vehicles with a wireless communication network, in particular an in-vehicle wireless communication network, i.e., a wireless communication network used for communication between different vehicle components, e.g., sensors such as cameras, radar sensors, etc., actuators, and control units. However, the wireless communication network could also be used for communication between the vehicle or a vehicle part or component and an external device, e.g., another vehicle or a central server (via cellular communication; also referred to as the cloud). Different types of wireless communication can be used, e.g., WiFi, Bluetooth, mobile communication networks (cellular networks) such as 5G or 6G, or even satellite communication networks, e.g., GPS. The increasing complexity and growing demands placed on such (vehicle-integrated) wireless communication networks, coupled with the rapid advancements in vehicle technology and the enormous proliferation of connected devices in vehicles, are leading to increasingly complex wireless communication networks, which also include subnetworks. With the evolution of modern vehicles into highly networked ecosystems, improving or optimizing the allocation of network resources and the operation of vehicle functions is becoming increasingly important for ensuring efficiency, safety, and the best possible user experience. Within the present invention, a possibility for improving communication via a vehicle's wireless communication network is proposed. In one embodiment, communication information is provided, wherein the communication information was obtained by analyzing protocol data using a first machine learning model, preferably a large language model (LLM), according to one or more predefined criteria. The protocol data was captured from, generated by, and/or related to the communication of the components over the vehicle's wireless communication network. The protocol data may include communication protocol data, aggregate protocol data from which communication data is extracted, and other aspects such as which hardware components are involved. The criteria for extracting the data from the protocol data may include, for example, the current radio access technology (RAT), transmit power, handover events, frequency, GPS coordinates, and/or other essential network performance indicators. In one embodiment, the protocol data includes one or more of the following: channel status information, channel occupancy time, channel quality indicator, control unit performance metrics related to energy consumption, control unit performance metrics related to computing efficiency, network signal strength, device connection and/or disconnection events, camera compression rates, system-level errors, and warning messages. Such data is generated through ordinary communication over the wireless communication network and stored, for example, in a database within the vehicle. Energy consumption can relate to the energy consumption of one of several control units, specifically control units designed to communicate over the wireless communication network. The communication information can be acquired using a computing device in the vehicle and then sent, e.g. via a mobile communication network, to an external computing device (e.g. in the cloud). Based on the communication information, a second machine learning model, preferably also an LLM, generates a digital twin of the wireless communication network. The digital twin is then optimized using an algorithm to obtain information on potential improvements to vehicle operation, particularly regarding potential improvements to communication over the vehicle's wireless communication network. As a result, communication over the vehicle's wireless communication network is improved. In one embodiment, the potential improvements to the operation, especially communication over the wireless communication network, of the vehicle for which the digital twin is analyzed include improv