Flow metering tools for link-layer traffic on: Bluetooth, Zigbee, WiFi
HI! The most recent version(s) of this tool are available at this repo:
- Auto-detection of the protocol within the frame is defaulted to packet 0. In the event of malformed (or incomplete) packets, this can create a problem where it is not detected and not produce the table. Planned Fix: Find non-malformed packet to detect, or override via flag to force protocol choice.
All datasets are already in the CSV flow formats.
TODO: Relabel all CSV files into the abstract classes except for the ones under "ProtoFlow Originals", which are complete.
Below is an example of device-to-class mappings for classification purposes.
| Device | Class |
|---|---|
| ASUS Router RT-N12 | router |
| ASUS RT-AC1200GE | router |
| August Smart Lock | smart_lock |
| Barnes & Noble Nook | ereader |
| Withings Blood Pressure monitor | smart_bp_monitor |
| Bose Home Speaker 300 | smart_speaker |
| C by GE 3-Wire On/Off Toggle | smart_switch |
| Echo W. Hub | smart_assistant |
| Fitbit 4 Health & Fitness Tracker | fitness_tracker |
| Galaxy A21 | smartphone |
| Garmin Index S2 Smart Scale | smart_scale |
| iRobot Roomba | smart_vacuum |
| Kindle | ereader |
| Kinsa Quickcare Smart Thermometer | smart_thermometer |
| PETKIT WiFi Feeder | smart_ped_feeder |
| Phillips Hue Bridge | smart_bridge |
| Pixel 4a | smartphone |
| Samsung Galaxy Watch Active | fitness_tracker |
Notes: Some of them are odd ducks (e.g., thermometer, bp monitor) as they don't have a greater class to belong to. In general, these abstractions can be anything as long as we have a clear umbrella for a large amount of these. For instance, we have multiple smartphones, routers, fitness trackers, etc.