Qualcomm incorporated (20240129008). NEURAL NETWORK BASED CHANNEL STATE INFORMATION FEEDBACK simplified abstract

NEURAL NETWORK BASED CHANNEL STATE INFORMATION FEEDBACK

Organization Name

qualcomm incorporated

Inventor(s)

Taesang Yoo of San Diego CA (US)

Weiliang Zeng of San Diego CA (US)

Naga Bhushan of San Diego CA (US)

Krishna Kiran Mukkavilli of San Diego CA (US)

Tingfang Ji of San Diego CA (US)

Yongbin Wei of La Jolla CA (US)

Sanaz Barghi of Carlsbad CA (US)

NEURAL NETWORK BASED CHANNEL STATE INFORMATION FEEDBACK - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240129008 titled 'NEURAL NETWORK BASED CHANNEL STATE INFORMATION FEEDBACK

Simplified Explanation

The present disclosure relates to neural network based channel state information (CSI) feedback in wireless communication systems.

• A device obtains CSI for a channel, creates a neural network model with a CSI encoder and decoder, and trains the model by encoding and decoding the CSI instance to minimize a loss function.
• The device obtains encoder and decoder weights based on training the neural network model.

Potential Applications

This technology can be applied in:

• Wireless communication systems
• 5G networks
• Internet of Things (IoT) devices

Problems Solved

This technology helps in:

• Improving the efficiency of CSI feedback
• Enhancing the performance of wireless networks
• Reducing latency in communication systems

Benefits

The benefits of this technology include:

• Faster and more accurate CSI feedback
• Enhanced reliability of wireless communication
• Improved overall network performance

Potential Commercial Applications

This technology can be used in:

• Telecommunication companies
• Network equipment manufacturers
• IoT device manufacturers

Possible Prior Art

One possible prior art could be the use of traditional methods for CSI feedback in wireless communication systems.

What are the limitations of this technology in real-world implementations?

There may be challenges in:

• Scalability of the neural network model
• Compatibility with existing network infrastructure

How does this technology compare to traditional methods of CSI feedback?

This technology offers:

• Potential for more efficient and accurate CSI feedback
• Improved performance in dynamic wireless environments

Original Abstract Submitted

various aspects of the present disclosure generally relate to neural network based channel state information (csi) feedback. in some aspects, a device may obtain a csi instance for a channel, determine a neural network model including a csi encoder and a csi decoder, and train the neural network model based at least in part on encoding the csi instance into encoded csi, decoding the encoded csi into decoded csi, and computing and minimizing a loss function by comparing the csi instance and the decoded csi. the device may obtain one or more encoder weights and one or more decoder weights based at least in part on training the neural network model. numerous other aspects are provided.