The IoT is diverse, with a broad range of devices and applications each with a cellular connectivity solution to match. By weighing up a range of factors, product designers can ensure they make the best connectivity choices to fit the IoT solution. That way, they optimise the in-life operation of devices once they have been deployed. Flexible connectivity is ideal for many installations and in this the eSIM, where it is appropriate, can help keep products agile, says Paul Bullock, head of MVNO, Wireless Logic.
What to consider for cellular connectivity
When designing for the IoT, consider:
- Device: Device and deployment characteristics will influence the choice of cellular connectivity. Considerations include the lifetime of the device and how much power it needs to stay deployed. Mass deployments in remote, global or expansive locations need consistent power as they are likely to stay in position for an extended time.
- Data: What type and volume of data will the device need to exchange? It may send only small data packets, or it may need higher bandwidth to exchange video. It’s a good idea to consider how the device application may develop over time, for example from sending audio files to adding video. Whatever the data volume and type, connectivity must be secure because mass IoT deployments present a large attack surface that is at constant risk. Therefore, consider data value and multi-network resilience.
- Distribution: Distribution is a key consideration because network options and commercial arrangements will differ for national, regional and international deployments. It is always preferable to have a single stock keeping unit (SKU) for devices, but that may not be possible if devices destined for different markets have different SIMs. In that case, region-specific SIMs may have to be added when devices reach their destinations for late-stage connectivity.
- Coverage: This concerns the wireless technology used to connect devices. Low power wide area network (LPWAN) (NB-IoT and LTE-M) are in vogue, but their global coverage is found wanting and even at a national level there are gaps. Be certain your device will definitely have coverage and there is no chance you’ll need to deploy to countries with no LPWAN at all. Notoriously, the UK offers LTE-M on the east side and NB-IoT on the west. There are many similar coverage gaps around the world, even a well-developed country like Brazil cannot offer consistent LPWAN coverage.
The phasing out of 2G and 3G
A large number of IoT applications were designed to connect over 2G and 3G. These networks are now being phased out, in the case of 3G the switch off has already happened in many regions or is being planned to do so in the very near future. The main reason for this is that mobile network operators (MNOs) want to free up spectrum for next generation, cost efficient and greater revenue generating technologies.
Most companies with IoT deployments connecting via 3G will be far into their migration plans. 2G is slightly different in that it is a technology that has been entrenched in large deployments of machine-to-machine (M2M) and IoT connected devices especially across Europe and therefore there is indication that in many countries, 2G won’t be sunsetting until at least the end of the decade. Whilst that may seem far off, forward planning is critical for devices deployed for many years.
There are a range of 2G/3G connectivity alternatives, each with their own features. These should be weighed up to assess their suitability for a new IoT device in design:
This is best suited to stationary IoT devices that send small amounts of non-real-time data, are battery or solar powered and are in places where other technologies would struggle to get a signal. NB-IoT offers low hardware and operating costs, so makes long-term, mass deployments viable. It is battery efficient and can therefore support devices that stay in the field for a long time, such as sensors which have low and intermittent data throughput. NB-IoT has a strong signal penetration, in fact the deepest reach of any low LPWAN, so it can cope with basements or underground car parks if sensors or water meters, for example, are deployed below street level.
- LTE-M (Cat-M1): This delivers the highest throughput speed and bandwidth of any LPWAN technology so it can cope with over-the-air (OTA) updates to, for example, firmware or network operator profile. It delivers latency and speeds broadly equivalent to 3G, is battery efficient and cost effective.
LTE-M (Cat-M1) can support a wide range of IoT applications, but is ideally suited to low power devices that require higher speed or two-way data transfers, such as those supporting voice or SMS services. It can be used by mobile, as well as stationary, devices as it supports cell to cell roaming.
However, some deployments would struggle with either NB-IoT or LTE-M for the coverage reasons already mentioned. Neither NB-IoT nor LTE-M is available on 4G and NB-IoT is not currently compatible with eSIM.
Where these limitations make LPWAN unsuitable, companies can consider:
- LTE Cat-1 and Cat-1 BIS: LTE Cat-1 (Cat-1 BIS being the single antenna version) is a mature technology. Hardware costs and power consumption are slightly higher than with LTE-M and NB-IoT but for some deployments the benefits will outweigh that. It has global support, because Cat-1 is a standard 4G technology, and existing roaming agreements mean global network access is possible with a single SIM SKU. It is suitable for mobile applications and is compatible with eSIM.
Lower latency and higher bandwidth make Cat-1 a strong alternative for 2G/3G and it can support a wide range of IoT applications. It should be considered for low power IoT devices that require higher speed, two-way data transfers or mobility, and can support a three-to-five-year battery life or applications that use rechargeable batteries.
eSIM for flexible connectivity
eSIM is an ideal technology for flexibility because it supports OTA provisioning of network operator credentials. That means the same SIM can be inserted into each device, regardless of where they are going, as connectivity can be provisioned later. That delivers the single SKU that is so important for operational simplicity, particularly for large international IoT deployments. This makes for simpler, more streamlined manufacturing processes and connectivity that uses local networks at local rates.
Plus, the eSIM makes in life network operator changes possible without the need to physically swap out SIM cards. That way, companies can take advantage of new commercial arrangements and attractive deals.
There are a range of considerations to take into account when designing IoT devices with cellular connectivity in mind. The devices themselves, the data they will exchange and the geographies they are destined for, will all play a part in determining the right cellular connectivity choice.
Network technologies are not forever, as companies who have completed, or are in the process of, migrations off 2G and 3G will know, therefore the lifetime of the technology standard is another factor to bear in mind. The benefit of some technologies is how flexible they are – the eSIM, where it is suitable, brings agility to IoT deployments. In all, an IoT connectivity platform provider will be able to advise on optimal cellular connectivity for each particular IoT use case.
The author is Paul Bullock, head of MVNO, Wireless Logic.