Early detection of cable failure in automotive networks

Patent metadata

Field	Value
Publication number	US-12237982-B2
Application number	US-202217824944-A
Country	US
Kind code	B2
Filing date	May 26, 2022
Priority date	May 28, 2021
Publication date	Feb 25, 2025
Grant date	Feb 25, 2025

Abstract

Methods and systems provide for early detection of failures in cables and/or physical layer devices (PHY) linked thereto, of a communications network in a vehicle. The methods and systems employ a computer trained model that analyzes changes in detected values obtained from a given vehicle within an operational time period, against a range of operational values, collected from different vehicles and/or the given vehicle, for the respective PHY parameter, to determine that the cable and/or the PHY linked thereto in the given vehicle may fail within a predetermined time period.

First claim

The invention claimed is:

1. A method for detecting failure in cables and/or physical layer devices (PHY) linked thereto, of a communications network in a vehicle, the method comprising:
obtaining values for one or more physical layer device (PHY) parameters from one or more cables and/or a PHY linked thereto, of a communications network of a vehicle;
analyzing the obtained values to determine a range of operational values for each of the one or more PHY parameters,
the operational values comprising, for each of the one or more PHY parameters, predetermined acceptable values for performance of a cable and/or a PHY linked thereto, and
the range of operational values, for each of the PHY parameters, being based on analyzing a history of the obtained values for each of the PHY parameters from the one or more cables and/or the PHY linked thereto;

detecting values for the one or more of the PHY parameters from the cable and/or the PHY linked thereto of a communications network of a vehicle during an operational time period when the vehicle is operating;
applying, to the detected values, a computer trained model that analyzes changes in the detected values within the operational time period against the range of operational values, for the respective PHY parameter, to determine that the cable and/or the PHY linked thereto may fail within a predetermined time period; and
providing an indication, based on detected changes in the values, that the cable and/or the PHY linked thereto may fail within the predetermined time period.

2. The method of claim 1 , wherein the range of operational values for each of the PHY parameters is variable, and is based on the values for the PHY parameters being obtained continuously from the one or more cables and/or the PHY linked thereto.

3. The method of claim 1 , wherein the computer model analyzes the obtained values to determine the range of operational values for each of the one or more PHY parameters.

4. The method of claim 1 , wherein the PHY parameters include one or more of: return echo signal, signal to noise ratio (SNR), insertion loss, electromagnetic interference, signal power, signal amplitude, signal magnitude, signal frequency, signal phase, signal duration, signal shape, signal polarization, signal modulation, signal level, signal irradiance, clocking, signal power spectrum density, PHY power voltage, PHY current level, PHY temperature, cable noise, cable, cable power transmission, and increased or new echo locations in the cable.

5. The method of claim 1 , wherein the one or more physical layer device (PHY) parameters from one or more cables and/or PHYs linked thereto includes a plurality of PHY parameters, and, the providing the indication that the cable and/or PHY linked thereto may fail is based on the detected values for at least a plurality of the PHY parameters being indicative that the cable and/or PHY linked thereto is failing.

6. The method of claim 1 , wherein the determining that the cable and/or the PHY linked thereto may fail includes the determined values for at least one of the PHY parameters for the operational time period being within the range of operational values and trending toward moving outside of the range of operational values for the one of the PHY parameters.

7. The method of claim 1 , wherein the determining that the cable and/or the PHY linked thereto may fail includes the determined values for at least one of the PHY parameters for the operational time period being within and/or outside of the range of operational values, including, trending toward moving outside of the range of operational values, for the one of the PHY parameters.

8. The method of claim 1 , wherein the computer trained model comprises a rules-based model.

9. The method of claim 1 , wherein the computer trained model comprises an Artificial Intelligence (AI) model or a machine learning (ML) model.

10. The method of claim 9 , wherein the AI or ML model is trained to:
identify the detected values in accordance with a corresponding PHY parameter from amongst the one or more PHY parameters; and
distinguish among a plurality of predetermined Physical Layer (PHY) parameters in cables and/or Physical Layer devices (PHYs) linked thereto, in the vehicular communication network, so as to determine whether the cable and/or PHY linked thereto, in the vehicle communication network, may fail within a predetermined time period, by analyzing changes in the detected values, against a range of operational values, for the respective PHY parameter, the range of operational values learned from computer analysis of a multiplicity of previous values corresponding to data, signals, and/or echoes, for the PHY parameter.

11. The method of claim 9 , wherein the AI or ML model is trained to determine whether a plurality of determined values obtained in the operative time period indicates a probability of failure in a predetermined time, of the cable and/or the PHY linked thereto.

12. The method of claim 1 , wherein the computer trained model additionally analyzes the detected values from within the operational time period, and determines whether certain of the one or more PHY parameters are to be determined within another operational time period of a predetermined length.

13. The method of claim 9 , wherein the AI or ML model includes a neural network model.

14. The method of claim 1 , wherein the obtaining values for one or more PHY parameters from the one or more cables and/or the PHY linked thereto, includes obtaining the values from the one or more cables and/or the PHY linked thereto of the vehicle and/or one or more different vehicles.

15. The method of claim 1 , wherein the range of acceptable operational values includes the most recently determined range of acceptable operational values.

16. A system for determining failures of cables and/or physical layer devices (PHYS) linked thereto of a vehicular communication network, the failure determination system comprising:
one or more processors; and,
a program memory storing executable instructions that, when executed by the one or more processors, cause the system to:
obtain a plurality of values corresponding to data, signals, and/or echoes for a physical layer device (PHY) parameter, the values of the plurality of values corresponding to data, signals, and/or echoes received from a cable and/or PHY linked thereto, of a vehicle communication network of a given vehicle;
apply, to the at least the plurality of values, a computer trained model that distinguishes among a plurality of predetermined Physical Layer (PHY) parameters in cables and/or Physical Layer devices (PHYs) linked thereto, in the vehicular communication network, so as to determine whether the cable and/or PHY linked thereto, in the vehicle communication network, may fail within a predetermined time period, by analyzing changes in the plurality of values obtained within the operational time period, against a range of operational values, for the respective PHY parameter, the range of operational values learned from computer analysis of a multiplicity of previous values corresponding to data, signals, and/or echoes, for the PHY parameter,
the operational values comprising, for each of the one or more PHY parameters, predetermined acceptable values for performance of a cable and/or a PHY linked thereto, and
the range of operational values, for each of the PHY parameters, being based on analyzing a history of the obtained values for each of the PHY parameters from the one or more cables and/or the PHY linked thereto; and

provide an indication, based on detected changes

Assignees

• Marvell Asia Pte Ltd

Inventors

• Razavi Majomard Seid Alireza
• Shen David

Classifications

• H04L67/12

  specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks · CPC title

• H04L43/0817

  by checking functioning · CPC title

• H04L41/16

  using machine learning or artificial intelligence · CPC title

• G06N3/08

  Learning methods · CPC title

• G06N20/00

  Machine learning · CPC title