LoRA on AWS Driver

Loads live information from MQTT and LoRA devices via Amazon Web Services IoT Core.

This driver has specific extensions to handle LoRA payload decoding and disconnected-device detection.

Include Application Name in Device Name means that the addresses you bind to will be in the form Application.Device instead of just Device. Useful when you have a large number of devices or multiple locations/sites.

Set Offline Data To Blank

This option is used to set all related values to blank when a device times out. See the later section on Detecting Offline Devices.

When pasting in the content of your certificates and keys, rememeber to include the —–BEGIN and —–END lines.

MQTT is extremely powerful when used for slower data, but encoding and decoding MQTT payloads can become very inefficient when dealing with fast-update information.

If you are sending out very large volumes of data, please consider alternatives such as REDIS, which may perform better at scale.

The point name is made up of both Device ID and Value Key. For example, Device.Temperature.

You can add either a model number or decoding string to process the LoRA message payload. This will extract values such as Temperature, Humidity, Vibration etc. from the encoded data send by the sensor.

You can use models or strings usable with the lorapayload library.

Note that you only have to specify the decoding string or model number on one of the bound points. The behaviour of the system when you have multiple different decoders or model numbers is not defined.

You can alert when a sensor has not send a signal in a fixed amount of time.

To do this, you can bind an alert to DeviceName or Application.DeviceName with the option '#' + the maximum number of seconds between updates.

For example, if you have a device called 'MachineSensor' that should send updates every 10 minutes, you could bind an alert to MachineSensor with the option #600.

After 10 minutes with no new updates from the device, the value will skip to '1'.

In this example, here's our MQTT payload…

{
   "sequence": 293834,
   "content": {
       "sensors": {
             "temperature": 25.3,
             "humidity": 43
       },
    },
    "deviceName": "MyDevice",
    "applicationName": "MyApp",
    "dt": "2012-20-20 10:10:00"
}

MyApp.MyDevice.content.sensors.humidity would get the current humidity value, and MyApp.MyDevice.content.sensors.temperature would get the temp.

In this version, we have the following MQTT payload that contains encoded LoRA data…

{
  "end_device_ids": {
    "device_id": "sensor",
    "application_ids": {
      "application_id": "myapp"
    },
    "dev_eui": "xxx",
    "join_eui": "xxx",
    "dev_addr": "xxx"
  },  
  "received_at": "2025-07-10T01:21:50.080004435Z",
  "uplink_message": {
    "session_key_id": "xxx",
    "f_port": 99,
    "f_cnt": 28,
    "frm_payload": "AZIBMyUAAAAAAA==",
    "rx_metadata": [
      {
        "gateway_ids": {
          "gateway_id": "elpro-pkct",
          "eui": "xxx"
        },
        "timestamp": 3036917906,
        "channel_index": 7,
        "received_at": "2025-07-10T01:21:49.834375860Z"
      }
    ],
    "settings": {
      "data_rate": {
        "lora": {
          "bandwidth": 125000,
          "spreading_factor": 7,
          "coding_rate": "4/5"
        }
      },
      "frequency": "918200000",
      "timestamp": 3036917906
    },
    "received_at": "2025-07-10T01:21:49.847558325Z",
    "confirmed": True,
 
  }
}

The address myapp.sensor.Temperature will find the value, but the temperature is encoded.

To acces it, you'll need to set the optional text to 'R718CT' (the model number of the appropriate sensor, or '[1][x92][1](Battery:B:*0.1)(Temperature:<H:*0.1)' which are the instructions on how to split the binary paylod into usable pieces.