System and method for machine learning applied to in-band to out-of-band communications management

Abstract

An information handling system operating a low power communications engine comprising a wireless adapter for communicating on a low power communication technology network for receiving low power communication technology data traffic for at least one always-on remote management service for the information handling system, a controller receiving a location status of the information handling system via the low power communication technology network indicating a location or network, where the controller executes code instructions for a low power communications engine to assess a location trust level from an environment characteristics analysis engine to determine whether the location status is a trusted zone location or an untrusted zone location utilizing binary classification machine learning based on input variables including data relating to history of activity at the location or on the network learned by the environment characteristics analysis engine from reported operational or network activity, and the controller to trigger an embedded controller to wake a BIOS of the information handling system and forward the incoming low power communication technology data traffic to in-band applications on the information handling system if the information handling system location status is determined to be in one trusted zone location and the controller to ignore the received low power communication technology data traffic if the information handling system location status is in one untrusted zone location.