Indoor Beacon Tracking system based on Bluetooth Low Energy and iBeacon technology. The tracking client can be installed on any Linux based micro-computer such as Raspberry Pi. It identifies each UUID of the beacons and utilizes Python multi-threading to record the entry time, exit time, duration in the particular area with corresponding event. In the meanwhile, the client sends real-time event data to IoT cloud by TCP. Further big data analytics and beacon visualization will be processed on cloud.

For indoor positioning and triangulation, multiple clients (hardware, at least 3+) should be placed in correct location according to reasonable coordination.

Finally, it's the time to upgrade the system and move forward! (Coming in April 2021) 😎

1. Replace old Python 2.7.x to Python 3.8.x in order to support more fancy Python functions and modules.

2. Integrate MQTT Pub/Sub capability. The tracking client will publish system and beacon events to IoT cloud via MQTT 3.1.1 or MQTT 5.0. Besides, the client will subscribe to a command topic to receive service commands for real-time configuration updating or further operations such as restart. All messages are in JSON format! The topic structure:

 Event:   service/${region}/event/${client-uuid}
 Command: service/${region}/cmd/${client-uuid}
 Data:    service/${region}/data/${client-uuid}

3. Integrate sliding window algorithm to reduce the side effect of multi-path, and send better average RSSI values for each supervised beacon.

4. Based on MQTT subscription, setup dynamic RSSI fence/threshold, for easy installation and debugging.

5. Add SQLite3 for better beacon UUID management on edge-devices.

6. Upgrade beacon tracking logic on client side.

Some Linux libraries and Python packages are required:

sudo apt install libbluetooth-dev
sudo apt install libboost-python-dev
sudo apt install libboost-thread-dev
sudo pip3 install gattlib
sudo pip3 install pybluez
sudo pip3 install paho-mqtt
sudo pip3 install sqlalchemy

Or pip install based on requirements.txt

Check config/app-config.ini file, in mqtt section, you may update to your own MQTT broker and corresponding settings.

Query Client Machine Status

{
   "command": "status",
   "trace": "test-trace-uid"
}

Shutdown the Beacon Tracking Client

{
   "command": "shutdown",
   "trace": "test-trace-uid"
}

Publish one-time beacon scan command (duration is optional)

{
   "command": "scan",
   "duration": 5, 
   "trace": "test-trace-uid"
}

Response

{
  "data_type" : "JSONArray",
  "action" : "data",
  "message" : "OK",
  "timestamp" : "2021-04-04 23:50:03 +0800",
  "data" : [ {
    "uuid" : "06f62a6d-f66f-88bd-b66e-2b12fd6e96aa",
    "major" : 0,
    "minor" : 512,
    "power" : 171,
    "rssi" : -43,
    "address" : "34:D6:66:E7:34:88"
  }, {
    "uuid" : "06f6213d-f88e-aabc-a36d-2b49fd6c36ad",
    "major" : 0,
    "minor" : 0,
    "power" : 191,
    "rssi" : -47,
    "address" : "28:E4:C2:E6:A1:D2"
  } ]
}

Change Scan Mode to Auto (duration, period are optional)

{
   "command": "change-scan-mode",
   "scan-mode": "auto",
   "duration": 3, 
   "period": 8,
   "trace": "test-trace-uid"
}

Server side code based on Spring Boot 2.4.x and ReactJS framework, old project has been moved to branch v2015-1-3

You are welcome to ask any questions here!