This is an example project for synchronizing clocks of end devices using class C.
In LoRaWAN, there are at least three ways for synchronizing end device clocks:
- Using application layer clock synchronization. This requires request/response messages per end device and do not guarantee exact time synchronization between end devices;
- Using class B beacons. This requires gateways with GPS sync and the end device to be aware of time zone and leap seconds;
- Using class C messages. This requires the end device only to be aware of time zone.
This is an example application for method 3: scheduling a class C message at a fixed time, with the absolute time as message payload. Since this is downlink only, this method can be used for unicast and multicast sessions.
- A LoRaWAN network powered by The Things Network Stack V3 for LoRaWAN. If you do not have a network available, please see the Getting Started.
- A LoRa® end device capable of running the Arm Mbed OS, for example the ST B-L072Z-LRWAN1 LoRa®Discovery kit or L-TEK FF1705.
- A gateway that has either GPS lock or provides round-trip times (RTTs). When relying on RTTs, gateways running LoRa Basics™ Station or Semtech UDP Packet Forwarder 4.0 or higher (sending
TX_ACK
) can be used.
- Provision an end device in the LoRaWAN stack supporting class C, for example:
$ ttn-lw-cli end-devices create app1 dev1 \ --dev-eui 0004A30B001C0530 \ --app-eui 800000000000000C \ --frequency-plan-id EU_863_870 \ --root-keys.app-key.key 752BAEC23EAE7964AF27C325F4C23C9A \ --lorawan-version 1.0.3 \ --lorawan-phy-version 1.0.3-a --supports-class-c true
- Configure, compile and run the device application
- Configure, compile and run the controller
Controller output:
sync clock dev1: now = 2019-05-10T12:13:33.275Z; target = 2019-05-10T12:13:43.000Z; unix time = 1557490423
Device output:
Received message on port 2 (4 bytes): f7 6a d5 5c
Synchronized clock to Fri May 10 12:13:43 2019