IoT is on the threshold of the next great format war. Will the winner take all?
Estimates for the number of IoT devices that will be connected by 2020 keep rising. At the moment it’s in the tens of billions. Google and Amazon are competing hard to win the voice-controlled smart home market. But what’s the case for scenarios like smart agriculture and smart cities that require long-range, low-power transmission?
The most promising technologies are probably LoRa and NB-IoT (sorry Sigfox), each with their own plusses and minuses. It is important to consider the relative strengths and weaknesses of each technology, but it’s also important to consider that “superior” technology doesn’t necessarily correlate with success. The Sony Betamax vs. VHS war is a classic example of how the outcomes of format wars aren’t always easy to predict.
The example is well known, but not everyone knows the details. The short version is that Sony did not effectively take into account what consumers wanted. They expected that consumers would be willing to pay more for superior video quality, when in reality the factors that mattered were longer recording time, cost, and compatibility with other systems — VHS quickly became the standard in many homes with many manufacturers of compatible VCR machines versus only Sony for the Betamax.
What lies in store for IoT, however, remains unwritten. Let’s discuss a comparison of LoRa and NB-IoT and see what conclusions, if any, we can draw.
LoRa was also developed originally in France but was later purchased by the US company Semtech in 2015. LoRa has been standardized via the LoRa Alliance and is currently deployed in 42 countries and being rolled out further with the help of various network operators. NB-IoT was developed in China by Huawei and is a technology based on narrow band technology. It’s still being tested, and is currently rolling out worldwide, expanding primarily via mobile network operators such as Vodafone and T-Mobile.
Factors to consider
LoRa uses unlicensed spectrum — this means that the volume and frequency of traffic are reduced at a significant cost advantage. Because the spectrum is unlicensed, there is also a greater risk of interference. NB-IoT uses licensed spectrum, this improves things like speed, volume and bandwidth, and increases the type of traffic allowed on the network at a significant increase in relative price. A bit of research has shown LoRaWAN subscription fees to be around 1 USD/year and NB-IoT to be around 6 USD/year, though they will certainly differ based on the provider.
LoRa and NB-IoT are in sleep mode most of the time, which improves battery life. However, NB-IoT utilizes synchronous communication as opposed to asynchronous with LoRa, which reduces latency at the cost of reduced battery life. If the battery is a primary concern, then LoRa is the best candidate, however, where latency is an issue, NB-IoT has a clear advantage.
NB-IoT offers much better scalability versus LoRa, basically 100k devices for NB-IoT versus 50k for LoRa per cell. NB-IoT also has the advantage of offering larger payloads at 1600 bytes maximum versus 243 for LoRa.
Network coverage and range
LoRa has a range of about 20 km in real-world conditions. By contrast, NB-IoT has a range of closer to 10 km. NB-IoT utilizes LTE networks, meaning that it’s not suitable in places where mobile network availability is lacking, such as in remote areas. However, because it is being deployed by mobile network operators, there is likely to be wide network availability in areas with LTE coverage. Because NB-IoT is being installed by mobile network operators, the rollout is expected to be better than LoRa, however, LoRa benefits from reduced costs and wider area per base station.
Take a look at this graphical representation of the differences between NB-IoT and LoRa — notice that their relative strengths and weaknesses are on relatively opposite ends of the spectrum:
Both NB-IoT and LoRa have respective advantages to one another. LoRa offers longer range and better battery life at thecost of network speeds and frequency of data. Because NB-IoT is being rolled out by traditional telecom companies, it has the benefits of coverage in those areas where LTE service is already being offered in the near future, if not now. LoRa networks are being deployed by different organizations worldwide, and the coverage will depend on networks being set up by members of LoRa Alliance. LoRa is definitely cheaper to use than NB-IoT as well as a much more “open” standard. Despite the many advantages of LoRa over NB-IoT, you might still prefer to utilize NB-IoT because of higher speeds or network availability.
At Stack7, we design our equipment with a modular approach. What this means is that our sensors can be fitted with different radios based on client requests. Rather than building sensors from the ground up everytime a different radio is required, we can simply swap out the module and adjust the firmware to make sure that devices are working as they should. In this way, we keep speed to deployment as fast as possible and don’t need to decide on one particular standard versus another. Flexibility is the key.
What is in store for the future? If we wanted to predict the outcome according to VHS vs. Betamax, it would seem that the cheaper, more open standard will be a winner. However, a lot has changed since the 70s, and at this point, it’s anyone’s game. We suspect that both networks will exist side-by-side for a while, each benefiting from relative strengths depending on the use-case. Still, LoRaWAN gives the most freedom and flexibility to developers and therefore is the most likely to succeed in the longer run.