Patent · US Active

Signal-multiplexing apparatus and method based on machine learning

US11050494B2 · kind B2 · utility

3Cited by
3References
16Claims
0Family size

Assignee

Inventors

Key dates

Filing dateAug 12, 2019
Grant dateJun 29, 2021
Priority date
Expiry dateAug 12, 2039

Classification

  • Technology area (CPC H)Electricity
  • CPC primaryH04L25/0254
  • WIPO fieldTelecommunications
  • WIPO sectorElectrical engineering

Abstract

Disclosed herein are a signal-multiplexing apparatus and method based on machine learning. The signal-multiplexing method based on machine learning, performed by a signal-multiplexing apparatus based on machine learning, including training a transmitted signal using a machine-learning technique, performing complex mapping to transmit the transmitted signal as multiple signals, receiving the multiple signals and regularizing the multiple signals based on a preset rule corresponding to the machine-learning technique, outputting an estimated value for the transmitted signal by training regularized multiple signals using the machine-learning technique, calculating a difference between the transmitted signal and the estimated value for the transmitted signal, and detecting the transmitted signal from an output value when the difference is less than or equal to a preset value.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.