18004286. CONFIGURABLE METRICS FOR CHANNEL STATE COMPRESSION AND FEEDBACK simplified abstract (QUALCOMM Incorporated)

CONFIGURABLE METRICS FOR CHANNEL STATE COMPRESSION AND FEEDBACK

Organization Name

QUALCOMM Incorporated

Inventor(s)

Pavan Kumar Vitthaladevuni of San Diego CA (US)

Taesang Yoo of San Diego CA (US)

Naga Bhushan of San Diego CA (US)

June Namgoong of San Diego CA (US)

Bo Chen of Beijing (CN)

Ruifeng Ma of Beijing (CN)

Krishna Kiran Mukkavilli of San Diego CA (US)

Tingfang Ji of San Diego CA (US)

CONFIGURABLE METRICS FOR CHANNEL STATE COMPRESSION AND FEEDBACK - A simplified explanation of the abstract

This abstract first appeared for US patent application 18004286 titled 'CONFIGURABLE METRICS FOR CHANNEL STATE COMPRESSION AND FEEDBACK

Simplified Explanation

The patent application describes methods, systems, and devices for wireless communications that improve the reporting of channel state information (CSI) between a user equipment (UE) and a base station. Here are the key points:

• The techniques aim to efficiently report CSI to the base station with an appropriate level of accuracy.
• The base station can indicate the desired level of accuracy to the UE for reporting CSI.
• The UE encodes the CSI using a first neural network, while the base station decodes the CSI using a second neural network.
• The first and second neural networks form a neural network pair that can be trained by the UE based on the level of accuracy indicated by the base station.
• The base station can indicate a loss function that corresponds to the desired level of accuracy, and the UE can train the neural network pair using this loss function.

Potential Applications:

• This technology can be applied in wireless communication systems, such as cellular networks, to improve the accuracy and efficiency of reporting channel state information.
• It can enhance the performance of wireless communication devices, allowing for better signal quality and more reliable connections.

Problems Solved:

• The patent addresses the challenge of efficiently reporting channel state information in wireless communication systems.
• It provides a solution to ensure that the reported CSI has an appropriate level of accuracy, which can help optimize the performance of the communication system.

Benefits:

• By using neural networks to encode and decode CSI, the patent enables more efficient and accurate reporting of channel state information.
• The ability to train the neural network pair based on the indicated level of accuracy allows for customization and optimization of the reporting process.
• This technology can lead to improved signal quality, reduced interference, and enhanced overall performance in wireless communication systems.

Original Abstract Submitted

Methods, systems, and devices for wireless communications are described. Generally, the described techniques at a user equipment (UE) provide for efficiently reporting channel state information (CSI) to a base station with an appropriate level of accuracy. In particular, the base station may indicate a level of accuracy to the UE for reporting CSI. The UE may encode the CSI using a first neural network, and the base station may decode the CSI using a second neural network. The first and second neural networks may form a neural network pair, and the UE may train the neural network pair based on the level of accuracy indicated by the base station. For example, the base station may indicate a loss function corresponding to a level of accuracy with which CSI is to be reported by the UE, and the UE may train the neural network pair using the loss function.