17816481. NODE SELECTION FOR RADIO FREQUENCY FINGERPRINT (RFFP) FEDERATED LEARNING simplified abstract (QUALCOMM Incorporated)

NODE SELECTION FOR RADIO FREQUENCY FINGERPRINT (RFFP) FEDERATED LEARNING

Organization Name

QUALCOMM Incorporated

Inventor(s)

Marwen Zorgui of San Diego CA (US)

Srinivas Yerramalli of San Diego CA (US)

Mohammed Ali Mohammed Hirzallah of San Diego CA (US)

Taesang Yoo of San Diego CA (US)

Xiaoxia Zhang of San Diego CA (US)

Mohammad Tarek Fahim of San Diego CA (US)

Rajat Prakash of San Diego CA (US)

Roohollah Amiri of San Diego CA (US)

NODE SELECTION FOR RADIO FREQUENCY FINGERPRINT (RFFP) FEDERATED LEARNING - A simplified explanation of the abstract

This abstract first appeared for US patent application 17816481 titled 'NODE SELECTION FOR RADIO FREQUENCY FINGERPRINT (RFFP) FEDERATED LEARNING

Simplified Explanation

The patent application describes techniques for training a machine learning model. Here is a simplified explanation of the abstract:

• A user equipment (UE) receives one or more selection criteria from a network entity.
• The selection criteria are used to determine whether the UE should participate in training the machine learning model.
• During a first period of time, the UE checks if it satisfies the selection criteria.
• After a second period of time, the UE transmits updated parameters for the machine learning model to the network entity.
• The machine learning model is updated during the second period of time if the UE satisfies the selection criteria.

Potential Applications:

• This technology can be applied in various fields where machine learning models need to be trained.
• It can be used in industries such as healthcare, finance, and transportation to improve predictive models and decision-making processes.
• It can also be utilized in smart devices and Internet of Things (IoT) applications to enhance their learning capabilities.

Problems Solved:

• The technique allows for selective participation of user equipment in training the machine learning model.
• It ensures that only UEs that satisfy specific selection criteria contribute to the model's training.
• This helps in improving the efficiency and accuracy of the training process.

Benefits:

• The technique optimizes the use of network resources by selectively involving UEs in training.
• It reduces the computational burden on UEs that do not meet the selection criteria.
• By updating the machine learning model based on satisfied selection criteria, the model's performance can be enhanced.

Original Abstract Submitted

Disclosed are techniques for training a machine learning model. In an aspect, a user equipment (UE) receives, from a network entity, one or more selection criteria for determining whether the UE is to participate in training the machine learning model, determines whether the UE satisfies the one or more selection criteria during a first period of time, and transmits, to the network entity, after a second period of time, updated parameters for the machine learning model, wherein the machine learning model is updated during the second period of time based on a determination that the UE satisfies the one or more selection criteria.