If it will never be 0 for an extended period of time, its best not to use that as a metric for triggering the event.
We use a baseline because we know over a period of time we could have… a single file get processed. In your case you may want to set it to 500kbps (or whatever your measurement is) over a small period of time. This way even with minimal “noise” on the line from other tasks, you can still get the alerts.
If somehow that won’t work you are likely going to have to start parsing the traffic to filter out the data you don’t care about.
One last option that works for us because we use influx, is Telegraf has a ping plugin, you could just ping a public source every so often and when it starts failing you can use that metric to trigger alerts on.