Cost, power, IP, and latency issues are driving IoT designers away from current standards like Zigbee and toward an emerging set of long-range wireless solutions. LoRa, an abbreviation of “Long Range”, is one such technology that is gaining traction.
LoRaWAN™ (Long Range WAN) is an emerging LPWAN specification developed by the LoRa® Alliance (www.lora-alliance.org). Like all LPWAN solutions, LoRaWAN is designed to operate at low power and low bit rates to maximize battery life. End-devices may run for years off a small battery. Fixed sensors can be powered indefinitely with a small solar panel. In addition, LoRaWAN is designed for simplified installation and long-range bidirectional communication with end-devices. End-devices may be located up to 10 miles from a gateway, with a clear line of sight. Each end-device connects wirelessly with one or more gateways in a star-of-stars topology. Gateways act as transparent bridges relaying traffic to and from a central network server.
The network server manages the data rate and transmission frequency of each end-device to maximize battery life and gateway capacity. LoRaWAN data rates range from 0.3 kbps to 50 kbps, and spread-spectrum chirp transmission allows signals with different data rates to act as virtual channels on the same frequency. Data security is managed through separate layers of encryption at the network, application, and end-device levels. The spread-spectrum approach does increase the spectrum width compared to a narrowband technology such as SigFox.
End-devices in LoRaWAN installations are classified according to communication method and power consumption.
Class A devices have the lowest power consumption, by means of scheduled chirp transmission windows each followed by two brief receive windows. Transmission is scheduled by the end-device based on its own data needs. Class A capability must be implemented in all LoRaWAN compliant end-devices.
Class B end-devices use the same transmit-receive system as Class A devices, but they schedule additional receive windows, coordinated through time-synchronized beacon signals from the gateway. This pulls more power from the end-device’s battery, but allows better coordination through the network server.
Class C devices offer a nearly continuous receive window, closed only during the device’s scheduled transmit windows. This draws the most power, but provides the most granular control.
Some of the considerations for a good LoRaWAN use case include:
Simple sensors that need to transmit infrequently
Tolerance for relatively high signal latency
Little need to control the end-device
No need to update end-device firmware over-the-air
End-devices deployed in the dozens to hundreds in one local network
Can deploy several gateways to cover every node
Use cases exploring LoRaWAN technology include:
Metering power, water flow, or similar quantitative monitoring.
Tracking systems for shipping containers, packages, industrial equipment, golf carts, city vehicles, patients, children, or pets.
Selective control of lighting, climate, etc. in different rooms of a building in response to changing occupancy, outside conditions, or other variables.
Control of low-flow irrigation systems in response to soil-moisture sensors.
Monitoring occupancy of parking spaces in a structure. Structure owners can identify cars that have exceeded their allotted time, and subscribed users can be notified of available spaces, or when their cars are approaching a violation.
Low-cost private networks as an alternative to commercial networks with paid access.
Semtech LoRa chips do have a second source, HopeRF, but it is unclear that they are a viable supplier. LoRaWAN compliant modules and modems are produced by a number of vendors:
ARM based https://developer.mbed.org/components/SX1276MB1xAS/
Using leading-edge technologies can quickly become “bleeding edge” because often documentation and prototype designs may not live up to the marketing promises. LoRa is an emerging technology and not a mature product. We have an interesting article on handling this type of risk “Voler Provides Risk Management For Leading Edge Components“.
Limitations for LoRaWAN include network capacity and signal interference as the number of end-devices in the network increases. Because of the pseudo-random ALOHA algorithm used to schedule transmissions, inevitably some signals will collide as the number of end-devices on any given gateway rises into the thousands. See Adelantado, et al for an in-depth discussion.
Want to learn more
Here are a couple of resources to learn more about LoRaWAN
Senet is a nationwide public Low Power Wide Area Network (LPWAN) network service provider
Orange Labs in San Francisco provides a LoRa sandbox for developers
Open Sensors’ IoT University is a great free online 5-day “Introduction to IoT”