LPWA: a market in the making
Whenever a new concept like IoT comes along, there is always speculation and conjecture over how the market for it will develop, particularly when there are competing technologies involved and LPWA is proving to be no exception. At present, how the market for IoT systems and devices evolves and what its eventual shape will be depends, as ever, on who‘s doing the talking. It is already possible however, even though IoT is still a relatively nascent concept, to discern the dynamics at work as the industry moves ahead, writes Peter Dykes.
From the communications perspective, IoT is all about connecting hundreds of millions of diverse, battery-driven devices which require secure bidirectional communication, mobility and localisation services over wireless networks. IDC forecasting is predicting 30 billion connected objects by 2020, with the sector worth US$2.75 trillion by that time and competition for a slice of the potentially lucrative LPWA market has, unsurprisingly, been hotting up for some time.
Essentially, players fall into two camps. On the one hand there are proprietary technologies such as SigFox and Ingenu, and on the other there are the standards-based technologies such as LoRa, backed by the LoRa Alliance, and LTE-based systems developed by the 3GPP group. Experience teaches that ultimately, the standards-based technologies will win out because they are vendor and technology agnostic, and while LoRa is arguably behind LTEbased standards in terms of large-scale infrastructure availability, it is beginning to overtake the likes of SigFox and Ingenu in market positioning.
Indeed, the lack of in-situ infrastructure is holding back the deployment of LoRa at present. Exponentially rising demand for LoRa hardware and gateways is swamping vendors with the knock-on effect of delaying network rollouts, according to Nigel Chadwick, the chief executive of Stream Technologies. He says, “This can be frustrating for platform vendors who, while they may be receiving orders for large-scale networks, are having difficulties getting the hardware from the vendors.” Chadwick adds however that the situation will be resolved in 2017 when production volumes come into line with demand.
Infrastructure availability is not such a big problem for the 3GPP systems as they are primarily aimed at cellular operators and consist of little more than enhancements to existing networks. That said, end point hardware in the form of LTE chipsets only began shipping at the beginning of September 2016, following Release 13 earlier in the year, but it does mean that commercial cellular-based IoT solutions will probably be first to market ahead of LoRa networks, given the reduced infrastructure rollout requirement.
Once both technologies have begun to roll out, likely market shares going forward will become easier to predict. This is because the true cost of IoT modules will become clear and it will be far easier for end users to calculate the TCO for an LPWA network. Currently however, there is some dispute over the price of modules. Shane Rooney, the executive director of IoT at the GSMA, says, “A lot of attention has been focused on the module costs and some of the others such as LoRa and SigFox have been saying sensors will only cost one or two dollars each, but this is simply not true. Module costs are actually a lot higher than that and realistically, they are all going to be a similar price, so the actual cost ultimately depends on scale.”
Rooney believes that module costs for cellularbased systems will be far lower because operators will be able to access much greater economies of scale than will buyers of LoRabased networks.
Customers for LoRa-based systems do have an option to reduce network TCO by becoming operators themselves however. Most LoRa deployments are expected to be in large-scale industries, agriculture, utilities and to some extent smart buildings and smart cities. Among the larger private utilities companies who have the ability to roll out a network for their own use, LoRa network suppliers are reporting that these customers are realising they can sell access to their networks to third parties and thus recoup their initial investment. As a result, it becomes far easier for a large private organisation to roll out a nationwide network and reduce the total cost of ownership (TCO) of hardware and network management platforms. Then they can calculate for example, the cost of millions of sensors over a fixed period and from that work out how much they can charge for access to each one. In addition, deployments in towns and small areas which have poor cellular connectivity but which require M2M and sensor connectivity could also enable enterprises to make a similar offering.
Rooney admits that the argument is valid, but has reservations. He says, “Monetising third party access over what is essentially a private network definitely has potential for recouping the original investment, but we’ve yet to see it happen. It will all depend on the application and what else is being offered because it won’t be achieved on connectivity alone.“
That IoT is a global phenomenon is beyond doubt, with demand coming from most parts of the world for both LoRa and LTE-based LPWA networks, however LoRa is likely to face stiff competition from 3GPP vendors who feel they have all the bases covered with three standards ranging from EC-GSM IoT, which is aimed at cellular operators running 2G in developing markets such as Africa, India and South America where infrastructure rollout costs are high, through to Cat-M1 for operators who are LTEready. The intermediate standard, NB-IoT is probably the most ubiquitous of the three and could prove to be a disruptive influence, as it has advantages over the other 3GPP standards and meets the challenge of LoRa head on. That said, in markets such as North America and Europe, where 2G networks have been or are in the process of being rolled up, LTE-based solutions are likely to dominate.
The weight of argument seems to be that the cellular-based standards will dominate in the medium to long term, with technologies such as LoRa confined largely to private installations such as industrial and agricultural complexes, however the market will remain fragmented for the foreseeable future and there should be room for all. It’s just that some LPWA technologies might not get as big a slice of the market than they are currently predicting.