Samsung electronics co., ltd. (20240113759). METHOD AND APPARATUS FOR MATRIX ENHANCEMENT IN MIMO RECEIVER WITH DEEP LEARNING simplified abstract

From WikiPatents
Jump to navigation Jump to search

METHOD AND APPARATUS FOR MATRIX ENHANCEMENT IN MIMO RECEIVER WITH DEEP LEARNING

Organization Name

samsung electronics co., ltd.

Inventor(s)

Hyukjoon Kwon of San Diego CA (US)

Shailesh Chaudhari of San Diego CA (US)

Dongwoon Bai of San Diego CA (US)

METHOD AND APPARATUS FOR MATRIX ENHANCEMENT IN MIMO RECEIVER WITH DEEP LEARNING - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240113759 titled 'METHOD AND APPARATUS FOR MATRIX ENHANCEMENT IN MIMO RECEIVER WITH DEEP LEARNING

Simplified Explanation

The patent application describes a method and apparatus for enhancing received signals using a neural network and interference whitening techniques.

  • A vector is generated from a covariance matrix of a received signal.
  • The vector is input to a neural network to obtain an enhanced vector.
  • The enhanced vector is converted into an enhanced matrix.
  • Interference whitening is performed on the received signal using a whitening filter based on the enhanced matrix.

Potential Applications

This technology could be applied in various fields such as telecommunications, radar systems, and signal processing.

Problems Solved

1. Enhancing received signals to improve signal quality and reduce interference. 2. Efficiently processing signals using neural networks and whitening filters.

Benefits

1. Improved signal quality. 2. Enhanced performance of communication systems. 3. Reduction of interference in received signals.

Potential Commercial Applications

"Enhanced Signal Processing Technology for Communication Systems"

Possible Prior Art

There may be prior art related to signal processing techniques using neural networks and whitening filters in communication systems.

Unanswered Questions

How does the neural network optimize the enhanced vector?

The specific optimization process of the neural network in enhancing the vector is not detailed in the abstract.

What types of signals can benefit the most from this technology?

The abstract does not specify the types of signals that would see the most significant improvements using this technology.


Original Abstract Submitted

a method and an apparatus are provided in which a vector is generated from a covariance matrix of a received signal. the vector is input to a neural network to obtain an enhanced vector. the enhanced vector is converted into an enhanced matrix. interference whitening is performed on the received signal using a whitening filter based on the enhanced matrix.