- 专利标题: METHOD OF TRAINING AN ARTIFICIAL INTELLIGENCE CIRCUIT, MONITORING METHOD AND SYSTEM FOR MONITORING A RADIO FREQUENCY ENVIRONMENT
-
申请号: US18192372申请日: 2023-03-29
-
公开(公告)号: US20230394360A1公开(公告)日: 2023-12-07
- 发明人: Kok Meng Wong , Peter John Green , Wee Chong Pay , Rajashekar Durai , Cyril Lee
- 申请人: Rohde & Schwarz GmbH & Co. KG
- 申请人地址: DE Muenchen
- 专利权人: Rohde & Schwarz GmbH & Co. KG
- 当前专利权人: Rohde & Schwarz GmbH & Co. KG
- 当前专利权人地址: DE Muenchen
- 优先权: EP 177192.6 2022.06.03
- 主分类号: G06N20/00
- IPC分类号: G06N20/00 ; H04W24/08
摘要:
A method of training an artificial intelligence circuit is provided. The method includes the steps of providing a training data set, the training data set encompassing training data of a frequency spectrum including a guard band and/or training data concerning a radio frequency power in the guard band, and feeding the training data set into the artificial intelligence circuit to be trained, which processes the training data set in order to learn a standard frequency spectrum and/or a standard radio frequency power in the guard band, thereby enabling the artificial intelligence circuit, when trained, to determine a deviation from the standard frequency spectrum and/or to determine a deviation from the standard radio frequency power in the guard band. In addition, a monitoring method for identifying an anomaly in a radio frequency spectrum as well as a system for monitoring a radio frequency environment are described.
信息查询