How LoRaWAN Technology works
With its star topology and cleverly implemented signal transmission technology, LoRaWAN technology is specifically designed for the energy-efficiency and secure networking of devices in the Internet of Things. We can explain how the technology works.
The Internet of things imposes many requirements on the network technologies used. What is needed is an architecture that is designed for thousands of nodes that can be far from populated areas and in hard-to-reach places – from sensors that monitor water flow and pollution in rivers and canals to consumption meters in the basement.
The architecture must also safely support battery-powered sensor nodes while simplifying installation and maintenance. That speaks for radio operation. Network technology must take into account the strict power consumption requirements for end nodes, many of which are to be operated with a single battery for decades. High security is essential to prevent eavesdropping and to ward off hackers.
The design of such a network technology begins on the physical level. Similar to a number of other radio protocols that are used for IoT applications, LoRaWAN technology uses the spread spectrum modulation. An essential difference between LoRaWAN and other protocols is the use of an adaptive technique based on chirp signals – and not on conventional DSSS (direct sequence spread spectrum signaling). This approach offers a compromise between reception sensitivity and maximum data rate, which supports this adaptation node by node thanks to the modulation configuration.
With DSSS, the phase of the carrier is dynamically shifted according to a precalculated code sequence. A number of successive codes are applied to each bit to be transmitted. This sequence of phase shifts for each bit produces a signal that changes much faster than the carrier, thus spreading the data over a wide frequency band. The higher the number of code pulses (chips) per bit, the higher the scatter factor. This spread makes the signal less susceptible to interference, but reduces the effective data rate and increases the power consumption per bit transmitted. Because the transmitter is more resistant to interference, it can reduce the overall power level. DSSS, therefore, offers lower power consumption with the same bit error rate. DSSS causes electricity and investment costs, which limits the application in IoT nodes.
The accurate reference clock is important for LoRaWAN technology
To ensure that the receiver can process the incoming code chips and convert the stream back into data, DSSS relies on an exact reference clock on the circuit board. Such clock sources are rather expensive and the increasing accuracy of the clocking also increases power consumption. The CSS technology used by LoRaWAN technology (chirp spread spectrum) can be implemented more cost-effectively because it does not rely on a precise clock source. A chirp signal is a signal whose frequency varies over time.
In the case of LoRaWAN technology network, the frequency of the signal increases over the length of the code chips of the respective data bit group. To improve reliability, LoRaWAN adds error correction information to the data stream. In addition to the immunity of systems with a spread spectrum, CSS offers a high level of immunity to multipath distortion and fading, which is problematic in urban environments – just like Doppler shifts: overlays change the frequency. The CSS technique is more robust because Doppler shifts cause only a small change in the time axis of the baseband signal.
More range or higher data rate
Like DSSS, LoRa can vary the number of code chips per bit. The standard defines six different scattering factors (SF). With a higher SF, the range of a network can be increased – but with more performance per bit and a lower overall data rate. With SF7, the maximum data rate is approximately 5.4 kbit / s and the signal can be considered strong enough at a distance of 2 km – although this distance depends on the terrain. With SF10, the estimated range increases to 8 km with a data rate of slightly less than 1 kbit / s. This is the highest SF in an uplink: a transmission from the node to the base station. A downlink can use two even larger SF. The SFs are orthogonal. This allows different nodes to use different channel configurations without influencing each other. In addition to the physical level that prepares data for CSS modulation and transmission, LoRaWAN defines two logical layers that correspond to levels 2 and 3 of the layered OSI network model (Open Systems Interconnection).
• Level 2 is the LoRa data connection level. It offers fundamental protection of message integrity based on cyclical redundancy checks. LoRaWAN establishes basic point-to-point communication.
• Level 3 adds the network protocol feature. The LoRaWAN protocol offers nodes the opportunity to signal each other or to send data to the cloud via the Internet – using a concentrator or a gateway.
LoRaWAN technology uses a star topology: All leaf nodes communicate via the most suitable gateway. The gateways take over the routing and, if more than one gateway is within range of a leaf node and the local network is overloaded, can redirect the communication to an alternative. Some IoT protocols use mesh networks to increase the maximum distance of a leaf node from a gateway. The consequence is a higher energy requirement of the nodes for the forwarding of messages to and from the gateways, as well as for an unpredictable shortening of the battery life.
The LoRaWAN architecture ensures that the battery of each IoT node can be dimensioned appropriately and predictably for the application. The gateway acts as a bridge between simpler protocols, which are better suited for resource-restricted leaf nodes, and the Internet Protocol (IP), which is used to provide IoT services. LoRaWAN technology also takes into account the different functions and energy profiles of the end devices by supporting three different access classes. All devices must be able to support class A. This is the easiest mode that helps maximize battery life. This class uses the widely used Aloha protocol.
Automatic collision avoidance integrated
A device can send an uplink message to the gateway at any time: The protocol has built-in collision avoidance when two or more devices try to send at the same time. Once a transmission is complete, the end node waits for a downlink message that must arrive within one of two available time slots. Once the response is received, the end node can go to sleep, which maximizes battery life.
A LoRaWAN gateway cannot activate a class A end node if it is in the idle state. He has to wake up by himself. This is due to local timers or an event-controlled activation, which is triggered by an event at a local sensor input. Actuators such as valves in a fluid control system must be able to receive commands sent by a network application – even if they have no local data for processing and communication. These devices use Class B or C modes.
With class B, each device is assigned a time window within which it must activate its recipient in order to search for downlink messages. The node can remain in sleep mode between these time windows. Uplink messages can be sent if the device is not waiting for a downlink message. Class B is used when the latency of up to several minutes can be tolerated. Class C supports significantly lower latency times for downlink messages since the receiver front end remains almost constantly active. A class C device is not in receive mode only if it sends its own uplink messages. This class is used by network powered end nodes.
Continuous encryption of the transmitted user data
In contrast to other protocols proposed for the IoT, LoRaWAN offers end-to-end encryption of the application data – right down to the cloud servers that are used to manage and provide the services. In addition to end-to-end encryption, LoRaWAN technology ensures that every device connected to the network has the required credentials and lets IoT nodes check whether they are not connecting to a gateway with a false identity. To ensure the required level of authentication, each LoRaWAN device is programmed during production with a unique key, which is referred to in the protocol as an AppKey.
The device also has a unique identifier worldwide. To make it easier for devices to identify their gateway connections, each network has its own identifier in a list managed by the LoRa Alliance. Computers that are identified as join servers are used to authenticate the AppKey of any device that wants to join the network. Once the join server has authenticated the AppKey, it creates a pair of session keys that are used for subsequent transactions. The NwkSKey is used to encrypt messages that are used to control changes at the network level, e.g. to set up a device on a specific gateway. The second key (AppSKey) encrypts all data at the application level. This separation ensures that the user’s messages cannot be intercepted and decrypted by a third network operator.
Another level of security is achieved through the use of secure counters that are integrated into the message protocol. This feature prevents packet playback attacks in which a hacker intercepts packets and manipulates them before feeding them back into the data stream. All security mechanisms are implemented via AES encryption, which has been proven to guarantee a high level of security. Due to its nationwide supply, energy efficiency and security, LoRaWAN technology is suitable for many applications as a protocol for setting up IoT networks.