Wired & Short Range wireless networks
The first option is to wire the sensors. Or use short-range wireless networks. But of course, it does not offer a global scope, and is therefore very restrictive. In the context of wireless networks, we can mention:
- WIFI: big consumer of energy, it does not allow a realistic operation on battery, and is not well adapted to an industrial environment.
- Bluetooth, ZigBee, Z-Wave
M2M – 2G, 3G, 4G & 5G networks
Until now, when it comes to connecting an object (a machine, a vehicle …) to know its status, its location, or interact with it, the only global solution – a global geographical coverage, national, even international – consisted of equipping it with a SIM card, and using the GPRS / 3G network of a telecom operator. The problems of this technology, commonly called M2M (Machine to Machine), are known:
- Expensive equipment: difficult to find a 3G modem less than 20-30 Euro …
- Very high energy consumption, requiring either a permanent power source (and therefore a wiring) or a powerful battery, making the size of the sensor incompatible with many use cases
- Large footprint
- A subscription cost, for connectivity, not insignificant, related to the need for a radio network (antennas operators) dense and expensive.
It is therefore difficult, with the M2M, to connect small objects, or not having an electric source. Or to accept that the sensor costs more than the connected object itself … Not to mention that it is undesirable for health, given the power of emissions, to carry a connected object in 3G, 24 hours a day.
LPWAN – Low Power Wide Area Networks
It is in this context that LPWAN – Low Power Wide Area Network – technologies have emerged. The objectives of this technology are to propose sensors
- Cheap – in the end only a few euros …
- Small sizes – a few centimeters away, or less …
- Can operate on battery, with great autonomy (5-10years!)
- Optimized to communicate for very small data rates, for a few Euro cents a month
If the LPWAN technology is not recent, with already implemented in the 90’s (for example the Alarm Net network in the United States), it will be necessary to wait for the creation of SIGFOX (2009) and the launch of its first network in 2012, in France, so that it begins to be structured and to become popularized.
The Main Players in the LPWAN – Low Power Wide Area Network – for IOT projects are: SigFox and LoRaWAN. Other technologies, such as RPMA (Ingenu), Weightless, Strij (Russia), aren’t so popular yet. NB-IOT is also appearing…
SigFox & LoRaWAN
The two most used technology players today are SigFox and LoRa. If there are different techniques still both solutions (SigFox is the Ultra Narrow Band), from a technical point of view, the two technologies are relatively equivalent.
SigFox and LoRaWAN operate on a free ISM (Industrial, Scientific and Medical) frequency band, so there is no need for a license. If this frequency band is free, it nevertheless requires compliance with a rule, which is not to use more than 1% of the bandwidth, which explains the limitation of the number of data that can be transferred …
The principle of operation is that the sensor (device) is in sleep mode most of the time (to save the battery), and wakes up at regular intervals – usually, at most every 10 minutes – to transmit the measured data . 12-byte data packets that are broadcast and retrieved by all visible Gateways. This data is then transmitted over the Internet to an IOT platform that will process the data.
Uplink (data from the device to the network: 140 messages of 12 bytes / day
Downlink (data from the network to the device): 4 messages / day
Key Strengths: a global network (with roaming “by design”)
Uplink (data from the device to the network: 140 messages of 12 bytes / day
Downlink: more flexible
Key Strengths: Possibility to build its own LoRaWAN private networks. Tens of Thousands of networks in the world
NB-IOT (Narrow Band IOT)
NB-IOT, for “Narrow Band IOT”, is the answer of traditional GSM operators to also offer a LPWAN network. This is a recent specification, an evolution of 4G (LTE) that offers a low power consumption solution, with limited throughput, based on an update of the current 4G network.
If this solution offers the advantage of global country coverage (since it is only an update of the current network), it is likely that the energy consumption will not be as low, and that the cost of the sensors will be higher. Moreover, it is impossible to build a private network based on NB-IOT. It is in any case an interesting alternative, but far from perfect.
