US 12,328,602 B2
ML model management in O-RAN
Dawei Ying, Hillsboro, OR (US); Jaemin Han, Portland, OR (US); Leifeng RUan, Beijing (CN); Qian Li, Beaverton, OR (US); and Geng Wu, Portland, OR (US)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Sep. 24, 2021, as Appl. No. 17/484,074.
Prior Publication US 2022/0014942 A1, Jan. 13, 2022
Int. Cl. H04W 24/02 (2009.01); G06N 5/022 (2023.01); H04L 67/02 (2022.01); H04W 16/14 (2009.01)
CPC H04W 24/02 (2013.01) [G06N 5/022 (2013.01); H04L 67/02 (2013.01); H04W 16/14 (2013.01)] 17 Claims
OG exemplary drawing
 
1. An apparatus for a Non real-time (Non-RT) radio access network intelligence controller (RIC) (Non-RT RIC) in an open radio access network (O-RAN), the apparatus comprising: memory; and, processing circuitry coupled to the memory, the processing circuitry configured to:
receive, over an AI interface, a register machine learning (ML) model request from an AI-ML consumer of a Near-RT RIC, the register ML model request comprising a ML model instance to be registered with an AI-ML producer of the Non-RT RIC, the ML model instance comprising an indication of supported monitoring types; and
send a response, over the AI interface, to the AI-ML consumer, the response comprising a registered model instance, the registered model instance comprising an indication of monitoring types to be used and an indication of a periodicity of update messages to be sent by the AI-ML consumer, wherein the register ML model request comprises a hypertext transfer protocol (HTTP) put request and the response comprises an HTTP post.