Within the Internet of Things, connectivity between objects (Things) and the network (Internet) is clearly needed. For applications that require small territorial coverage, we are familiar with the Bluetooth Low Energy (BLE) and ZigBee standards, which fall within the broader category of Personal Area Networks (PAN). For applications that instead have the need to cover larger areas, it is necessary to use a Wide Area Network (WAN) or better, within the IoT, a Low Power WAN (LPWAN). LPWAN technology is very suitable for connecting devices that need to send small amounts of data over long distances while allowing a long battery life.

The number of IoT (Internet of Things) devices is expected to grow continuously on a global scale: the most recent estimates assume that by 2020 there will be over 30 billion connected devices. A certainly noteworthy figure, which requires design choices to which the developers of the devices and the infrastructure for communication must adapt. These choices impact on the scalability and energy efficiency of the devices, with reflections on the type of technology used to achieve connectivity towards the IoT network concentrator (or gateway).

The new LPWAN technologies operate at frequencies included in the ISM (Industrial, Scientific, Medical) license-free bands. Unlike mobile network operators, therefore, LPWAN network operators do not have to buy expensive licenses for the assignment of radio spectrum bands. LoRaWAN is a LPWAN specification that allows battery-powered devices to connect to an IoT network over a long-range, using low bandwidth, in a regional, national or global network.

LPWAN

The Low-Power Wide-Area Network (LPWAN) is a type of wireless telecommunications network designed to allow long-range communications with low bit rates. LPWAN technology is designed for machine-to-machine (M2M) network environments. With a decrease in power requirements, a greater range and a lower cost compared to a mobile network, LPWAN networks are designed to allow a more extensive range of M2M applications and the Internet of Things, which are often constrained by budget and power problems.

The most well-known LPWAN is LoRaWAN for wireless connection of objects via a regional, national, or global network. The LoRaWAN specification provides interoperability between intelligent objects, without the need for complex local installations. It also offers freedom for the user, developers, and companies to develop the Internet of Things.

Connectivity for IoT

LoRA is a wireless technology for long-distance communication, emerging as one of the leading solutions to realize the IoT infrastructure. Distinguished by reduced power absorption and low data transfer rates, LoRA is used by IoT devices connected to network end-points to communicate with internet-connected gateways. The gateways behave like real bridges, allowing full interaction between the end-point devices and a central network server.

LoRA technology uses two distinct layers:

  • a physical layer, which uses the Chirp Spread Spectrum (CSS) radio modulation technique;
  • a MAC protocol layer (the LoRaWAN mentioned above).

The LoRA physical layer, developed by the company Semtech, allows long-range communications using low-power devices and requiring very low bandwidth. LoRaWAN is instead a protocol that allows multiple IoT endpoints to communicate with a gateway using LoRA technology. While the LoRA modulation technique is proprietary, LoRaWAN is an open standard, developed by the LoRa Alliance itself. The typical structure of a LoRA network is based on a star topology, with the presence of three different types of devices, as shown in Figure 1.

Figure 1: a typical example of a LoRA network

The LoRA physical level

Although proprietary, LoRA uses a chirp spread spectrum (CSS) modulation, based on the use of chirp (signals whose frequency increases or decreases over time) for information encoding. By the high linearity of the chirp pulses, the frequency offsets between the receiver and the transmitter are equivalent to temporal offsets, easily eliminated in the decoding circuit. This aspect also makes the modulation immune to the Doppler effect, corresponding to a frequency offset. Figure 2 shows the typical waveform of a chirp. On the axis of the abscissas is indicating the time, while on the axis of the ordinates the amplitude. Note that the frequency is not constant but varies over time.

Figure 2: the typical aspect of a chirp

The parameters that affect the LoRA modulation are as follows:

  • bandwidth;
  • spreading factor (SF): identified by the base 2 logarithm of the number of chirps per symbol: in practice, it indicates the duration of each chirp. LoRA uses the SFs from 7 to 12: SF7 corresponds to the chirp of shorter duration, SF12 the longer duration;
  • code rate (CR), related to the forward error correction (FEC) mechanism included in LoRA.

The LoRA frame primarily involves a preamble, which begins with a constant sequence of chirps with increasing frequency covering the entire frequency range. The last two chirps identify the sync word, a single byte used to differentiate LoRA networks that use the same frequency bands. An end-point device configured with a specific sync word will ignore any message received if the associated sync word does not match the configured one. The sync word is followed by 2.25 chirp. The total duration of the preamble can be configured between 10.25 and 65539.25 symbols. After the preamble, the header is transmitted (optionally): when present, the header has a code rate of 4/8. The payload (with a maximum length of 255 bytes) is transmitted after the header and is followed by a CRC (optional).

The LoRAWAN protocol

The LoRaWAN network has a star-shaped, or rather, star-like architecture. Unlike a mesh network (mesh network) in which individual nodes receive and retransmit information from adjacent nodes in order to increase the coverage area of the network (with the disadvantage of keeping all network components awake in order to handle information that is in most cases irrelevant to themselves), in a LoRa network the terminal nodes are most often “sleeping”.

LoRaWAN is a MAC protocol based on the physical level offered by LoRA and designed to support sensor networks. The main components of a LoRaWAN network are as follows:

  • low-power end-point devices that communicate with the gateways via LoRA;
  • gateways: intermediate devices that forward packets received from end-points to a server using a broadband network;
  • network server: decodes the packets received from the sensors and prepares any reply.

LoRa Gateways are the interface to endpoints. The messages received from the endpoints are in turn transferred to another unit that has the function of Network Server. In general, the connection to the Network Server is made via a standard IP connection, and public or private networks can be used. The Gateway will send the messages to the endpoints if the server has something to say to the remote devices.

LoRa Network Server is the unit that manages the network. The network server fulfills all the tasks foreseen by the LoRaWAN protocol, such as that of eliminating duplicate packets (the messages of an endpoint can be picked up by several gateways simultaneously), adapting the data rate between the various nodes, send the data to the relevant applications (Figure 3).

Figure 3: SAM R34 Xplained Pro development kit of Microchip for LoRa design

Please visit EDN Asia for the complete article