Method and system for configuring look-up table for decentralized congestion control, decentralized congestion control method using such a look-up table

A method for configuring a look-up table for decentralized congestion control (DCC) of a network comprising at least two RSUs and, in the vicinity of each RSU, a plurality of TX UEs capable of communicating with respective RSU and using the look-up table to implement decentralize congestion control for sidelink communication of the TX UEs, and a parameter server (PS) connecting the RSUs, comprising:

• monitoring sidelink traffic conditions in the vicinity of each RSU;
• training a machine learning model for each RSU;
• aggregating the machine learning models;
• updating the machine learning models;
• and
• dynamically configuring a look-up table for a targeted TX UE in the TX UEs based on the monitored sidelink traffic conditions and the updated machine learning models,
• wherein the machine learning model is trained to estimate a number of TX UEs performing sidelink transmission in the neighborhood of the targeted TX UE,
• the neighborhood is an area in which the targeted TX UE is adapted to communicate with other TX UEs through sidelink communication, and
• the look-up table for the targeted TX UE is configured based on a limit of Channel Resource utilization (CRlimit) calculated as a function of the estimated number of TX UEs performing sidelink transmission in the neighborhood and a measured sidelink Channel Busy Ratio of the targeted TX UE.

A system for configuring a look-up table for decentralized congestion control (DCC) of a network comprising at least two RSUs and, in the vicinity of each RSU, a plurality of TX UEs capable of communicating with respective RSU and using the look-up table to implement decentralize congestion control, and a parameter server (PS) connecting the RSUs,

• wherein the RSUs are configured to:
• monitor sidelink traffic conditions in the vicinity of each RSU;
• train a machine learning model for each RSU;
• wherein the parameter server is configured to:
• aggregate the machine learning models;
• update the machine learning models;
• wherein the RSUs are further configured to:
• dynamically configure a look-up table for a targeted TX UE in the TX UEs based on the monitored sidelink traffic conditions and the updated machine learning models,
• wherein the machine learning model is trained to estimate a number of TX UEs performing sidelink transmission in the neighborhood of the targeted TX UE,
• the neighborhood is an area in which the targeted TX UE is adapted to communicate with other TX UEs through sidelink communication, and
• the look-up table for the targeted TX UE is configured based on a limit of Channel Resource utilization (CRlimit) calculated as a function of the estimated number of TX UEs performing sidelink transmission in the neighborhood and a measured sidelink Channel Busy Ratio of the targeted TX UE.