Within one week of one another, the Netherlands and South Korea both announced that they had completed nationwide deployments of LoRa networks. LoRa is a low-power wide area network (LPWAN) technology that is catching on very quickly around the world. In both cases, national telecoms were responsible for deploying the networks — KPN led the charge in the Netherlands and SK Telecom rolled out the network in South Korea. Both telecom operators aim to leverage these new networks to develop more business into IoT use cases as LoRa is ideal for sensors and machine-to-machine (M2M) communications. Both networks will offer low price points: SK Telecom claims its “plans are highly affordable — costing merely one-tenth of SK Telecom’s LTE-based IoT services — and thus are expected to support active development and provisioning of diverse IoT services by easing the cost burden of startups and SMEs.” A more detailed SK Telecom pricing plan can be found on the official press release.
To learn more about the KPN deployment, we caught up recently with Simon Philipsen, Commercial Product Manager for KPN LoRa, who explained that KPN has officially covered the entire nation in LoRa, and that it plans to deploy more base stations to increase coverage density. By the end of 2016, KPN expects it will have deployed about 1,100 base station gateways, offering complete indoor and outdoor coverage, regardless of the location in the Netherlands. Key technology partners for the deployment include Actility (software for infrastructure and gateways) and Kerlink (gateway hardware). Simon said that initial tests achieved transmission distances between base stations and endpoints of about 30 kilometers, but once KPN has rolled out all the base stations, signals will not have to travel more than a few kilometers. He said that this high density of base stations will enable KPN to provide optimal service-level agreements (SLAs).
He said that pricing for KPN LoRa connectivity will cost between €4 and €14 (~$4.50 and ~$15.40) per device per year, depending on the usage of the device and network. KPN and its partners are underway with pilots for a variety of use cases including dike monitoring, groundwater monitoring, urban lighting and waste management. Philipsen explained that LoRa modules will cost about €5 in high volumes; total device cost depends on different factors, but the simplest integrated devices can be as cheap as €10. Several tests from partners have demonstrated devices lasting many years on a single charge (in some cases over 10 years on two AA batteries), depending on the use case. KPN does not expect that its LoRa network will cannibalize much of its existing 3G and 4G M2M business; rather, it expects that the LoRa network will create new business opportunities for completely new IoT solutions. Before deciding to go with LoRa, KPN assessed competitive technologies including the emerging NB-LTE-IoT standard and Sigfox. Compared with NB-LTE-IoT, KPN chose LoRa based on quality, technology readiness and industry adoption. Compared to Sigfox, LoRa offers more bandwidth and control over the infrastructure. Sigfox has already covered large portions of Europe (including almost all of the Netherlands), making it the most direct competitor to KPN LoRa today.
We reached out to contacts at Sigfox for a comment — Thomas Nicholls (EVP, Communications) offered the following quote: “LoRa is a proprietary technology sold by the chip vendor Semtech. It can be used as the radio technology for someone who wants to install and operate a managed network. Semtech’s approach is to let anyone deploy networks and then sell transceivers (radio components) or license them to other chip vendors. Compared to Sigfox’s managed global network, the major differences in the two approaches is that Sigfox has no vendor-specific tie-ins and therefore provides an open model, which ensures low pricing and access to the existing install-base of transceivers from the likes of Texas Instruments, Atmel and Silicon Labs. Also, Sigfox is based on an approach with one single global network, where customers can perform one single integration and sign one single subscription contract for global connectivity. In the LoRa model, you have the vendor tie-in and you’ll need to sign contracts with the individual operators of managed networks, with individual SLAs. Today there’s no managed LoRa-based network that has any major coverage area, compared to Sigfox’s 1.3 million km2 of coverage in over 20 countries. Also, the major operators evaluating LoRa have openly stated that they do not believe in the model where you can end up with multiple networks in a given zone, since LoRa is very vulnerable to interference, which would cause issues for operators wanting to provide industrial-grade service-levels.”
It’s true that until recently Semtech had been the sole manufacturer of LoRa chipsets, but recently the LoRa Alliance has begun to license the IP to other vendors. LoRa is a major threat to Sigfox, so Sigfox likes to focus on this element in its messaging about competition with LoRa. Proponents of LoRa, however, view the LoRa ecosystem as being more accommodating to open standards and open-source innovation than Sigfox. The LoRa standard is governed by an alliance group and LoRa operators can manage the LoRa infrastructure and back end; meanwhile Sigfox is a singular private company and always maintains the infrastructure and backend for Sigfox networks. In a recent interview with Lux, Will Yapp (VP, Business Development), from U.S.-based LoRa network operator Senet, offered the following quote regarding the Netherlands and South Korea networks: “Both of these deployments further exemplify how fast open standards drive adoption of new technology. It also shows that major global carriers recognize the fact that one technology alone will not address all use cases, so by supporting multiple, open standards-based solutions, they can provide a full service offering to their clients based on the requirements of the use case, the needed ROI and the time to market. That is why 3GPP, Wi-Fi and LoRa will lead the way in wireless networking, each addressing a wide variety of use cases and in some cases being used together in a single use case to offer the full solution.”
It will be interesting to watch LoRa and Sigfox battle for bandwidth in the coming years. Our expectation is that LoRa’s ecosystem will prove to be more open than Sigfox’s, and that this will foster more development for LoRa at a global level. Additionally, Sigfox’s narrower bandwidth range and limited number of messages will relegate it to low-bandwidth, low-frequency applications. That’s not to say Sigfox won’t still be very successful — the company is growing rapidly and will likely achieve the IPO it is targeting for 2017 or 2018. But LoRa will most likely emerge as the dominant LPWAN solution for a broader range of use cases. One thing is clear, regardless of which solution ends up winning most of the pie: LPWAN is past the introduction phase and is ready for primetime. Those developing connected products and services should investigate LPWAN solutions as a key enabler; those working on internal enterprise IoT initiatives should do the same.
All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.