LoRa and LoRaWAN – Part 2: LoRaWAN Device Classes: Review and Comparison of Class A, B, and C Leave a comment

LoRaWAN Device Classes: Review and Comparison of Class A, B, and C

In the previous article about What is LoRa and LoRaWAN? | Differences, Applications, and Advantages in IoT, we discussed the basics. In this article, we will talk about the classes of LoRaWAN devices. The LoRaWAN standard defines three types of devices: Class A, Class B, and Class C. All LoRaWAN devices must support Class A, while Class B and Class C are essentially extensions or add-ons built on the Class A specifications.

All three classes enable bidirectional communication — meaning they can send data to the network (uplink) and receive from the network (downlink). However, the main difference lies in when and how the device listens for network messages.

An important point is that when performing remote software updates (FUOTA) on devices, Class A alone is not sufficient. In such cases, the device must switch to Class B or Class C to enable more reliable and timely message reception.

LoRaWAN Device Classes: Class A Devices

All LoRaWAN devices are required to support Class A. In this mode, the device can send an uplink message (data to the network) at any time. Immediately after the uplink transmission ends, the device opens two short intervals to receive messages from the network. These intervals are called Receive Windows.

There is a specific delay between the end of the uplink transmission and the start of each receive window, known as RX1 Delay and RX2 Delay. If the network server cannot respond to the device within these two windows, the downlink message is postponed and sent at the first opportunity after the next uplink.

In simple terms, Class A offers the greatest energy savings because the device only activates when sending a message or listening immediately afterward, spending the rest of the time in sleep mode. This makes it ideal for sensors that transmit data only occasionally (such as temperature, humidity, or location).

The network server can deliver the response in one of the two receive windows; either in the first window (RX1) or in the second window (RX2), but it never uses both simultaneously. Therefore, when discussing downlink messages, three different scenarios arise:

1. The server delivers the message to the device in the same first window (RX1).
2. If no response is given in RX1, it can respond in the second window (RX2).
3. If neither of these two windows is used, the downlink message is deferred to the next uplink and sent right after it.

In this way, the receive timing in Class A always remains predictable, because the device only listens briefly twice and then returns to sleep mode.

Class A devices have extremely low energy consumption, enabling them to operate on battery for years. They spend most of their time in sleep mode and only wake up occasionally to send data. The interval between uplink messages is typically long. An important point to note is that Class A cannot receive a downlink message until it sends an uplink. This results in relatively high latency for receiving messages in this class.

These characteristics make Class A the best choice for scenarios requiring very low energy consumption and where data is sent only occasionally. Common applications of Class A include: environmental monitoring, animal tracking, forest fire detection, water leak detection, smart parking management, asset tracking, and waste management.

LoRaWAN Device Classes: Class B Devices

Class B devices actually extend the capabilities of Class A. The main difference is that these devices periodically open windows called Ping Slots to receive downlink messages. To ensure synchronization, the network periodically broadcasts a synchronization signal called a Beacon through gateways (either individually or in groups). End devices receive these beacons and synchronize their internal clocks with the network.

This time synchronization allows the network server to know exactly when to send a downlink message to a specific device or even a group of devices. The time interval between two beacons is called the Beacon Period.

Of course, just like Class A, after each uplink, two short windows RX1 and RX2 also open. But the advantage of Class B is that, in addition to those, it has regularly scheduled times for listening to network messages, which significantly reduces message reception delay compared to Class A.

Class B devices have much lower delay in receiving downlink messages compared to Class A, because they regularly open Ping Slots. However, they are still far behind Class C devices in terms of reaction speed. Most Class B devices also operate on battery. Of course, their battery life is shorter than Class A, because they spend more time in active mode — both to receive beacons and due to the open Ping Slots.

This reduction in battery life comes with a significant advantage: the device can receive control messages with much less delay. For this reason, Class B is typically used in applications that require relatively fast response, but not instantaneous — such as smart meters (electricity, water, and gas) or street lighting systems.

An interesting point is that Class B devices can operate like Class A if needed. This flexibility allows prioritizing either energy consumption or response speed depending on the scenario.

LoRaWAN Device Classes: Class C Devices

Class C devices also extend the capabilities of Class A, with the key difference that their receive windows are always open — except when sending an uplink message. As a result, these devices can receive downlink messages almost at any moment, which makes message reception latency very low. This capability is extremely useful for commands like dimming street lights or activating a water shut-off valve in smart meters.

In Class C, just like Class A, there are two windows RX1 and RX2, but the major difference is that RX2 remains almost always open until the next uplink is sent. The operation works as follows: the device sends an uplink message, then a short RX2 opens, followed by a short RX1. After that, RX2 opens again and stays continuously open until the next uplink. A new uplink is only sent when no downlink message is being received.

Due to this architecture, Class C is the best option for scenarios where commands need to be executed with minimal delay — where response time is much more important than energy consumption.

Class C devices have the lowest latency compared to Class A and Class B, because they are almost always ready to receive downlink messages. However, this very feature causes high energy consumption — since their receive window must remain continuously open. For this reason, these devices typically cannot operate on battery for long periods and are usually connected to mains power.

Class C is primarily used in applications where commands must be executed quickly and without delay. Of course, just like Class B, Class C devices can also operate in Class A mode if needed to reduce energy consumption.