Using machine learning (ML) to make observations of network operations is faced with many constraints, including collection constraints, storage constraints, and processing constraints. Additionally, in many instances, data collected from a network may be unusable and will incur collection, storage, and processing costs with potentially limited return. Presented herein are techniques through which pre-filtering tasks can be distributed to wireless access points (APs) to highlight valuable metrics and learn from network deployments.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.