Abstract
Techniques are presented herein that employ a HyperLedger (i.e., a private blockchain) to allow authenticated machine learning (ML) components to interact with each other and sign messages (having pre-processed data) which are exchanged between the components to provide message authenticity. Along with authenticity, aspects of the presented techniques provide for the traceability and accountability of ML components. Aspects of the presented techniques may be employed across different operators and domains wherein a HyperLedger provides for the authenticity of the ML components. Further aspects of the presented techniques may be employed within an operator or within an instance (such as an Access and Mobility Management Function (AMF), a Session Management Function (SMF), etc.) wherein a centralized ML system or software can act as central repositories for all of the authenticated ML components along with their corresponding public key (which may be used for verification of the signed messages that are exchanged between the ML components). In such a case, the functionality of the HyperLedger may be performed by a centralized ML system or software.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
M M, Niranjan and Kenchaiah, Nagaraj, "AUTHENTICATED MACHINE LEARNING IN 5G NETWORK DEPLOYMENTS", Technical Disclosure Commons, (May 08, 2022)
https://www.tdcommons.org/dpubs_series/5121