Classification problem with 2 level features

Consider an automated house, where we can collect router data every minute. The objective is to predict router fail/no fail (1/0) in a future time window.

The router sends data at two different levels every minute:

1. Router properties of the device: Memory, Temperature, Usage (router level)
2. Data collected from N different IoT devices, received rates, signals (IoT device level)

Since the target is defined at the router level, how could the IoT individual data be incorporated into the classifier?