May 07, 2017I recently attended the LoRa Alliance conference in London and got to see the latest and greatest in low-power wireless access (LWPA) and LoRa-enabled technologies. Never heard of LPWA or LoRa, and not sure why they matter? In addition to sharing my perspective on what's to come in the IoT connectivity arena, I want to take a step back and discuss some important questions, such as: Why do we need a special way to connect the billions of smart devices expected in the coming years? What exactly is LPWA, and why should we care about LoRa? Why is a connectivity standard for the Internet of Things so important?
The Importance of Standards
First, let's discuss what a standard does. Have you ever noticed, when traveling between countries, how your mobile phone can connect? It connects whether you are in Nigeria, the Philippines or Canada. Yet, you cannot simply plug in your charger to a wall plug in each of these places without cumbersome adapters. Well, that is exactly why a standard is needed.
Standards also create a global market, enable massive economies of scale, drive down cost, and attract more investments and continued innovation. So, in order to arrive at the 50 bio devices connected, as touted by some industry commentators, we need a standard for the IoT.
In order to deliver on the expectations for connectivity ubiquity, coverage, battery life, OPEX and ease of provisioning for the IoT, current solutions do not suffice. At present, there are basically two connectivity standard camps: players supporting the licensed band and players supporting the unlicensed band. In simple terms, it is a bit like Wi-Fi vs 3G and 4G. One technology can be deployed by anybody anywhere (your Wi-Fi router at home or in the office), while the other is "owned" by a licensed operator who has paid a 10- or 20-year concession fee to a government-regulated entity to use a certain spectrum.
Why Other IoT Connectivity Standards Failed
We have seen a range of unsuccessful attempts to create a new IoT connectivity standard from Bluetooth, DECT derivatives, ZigBee, UNB, Weightless and others. Most of them have failed due to a lack of industry ecosystem support, the technology's complexity and basic bad timing, but they have also failed due to the unclear end-user need. Who needs a connected toothbrush if it comes with a premium price and no intelligence or consumer value? Then there is, of course, Wi-Fi, GSM, 3G and LTE, which all are successful standards, but not for the gap that LoRa is now filling. In a previous post, I wrote about trade-offs between latency and bandwidth being the key determinants for applications, which makes this so significant.
Where LPWA Fits Into the IoT Standard Landscape
LPWA covers a set of technologies, such as weightless, ultra-narrow-band (UNB) technology, random-phase multiple-access (RPMA), LoRa, narrowband IoT (NB-IoT), long-term evolution for machines (LTE-M) and others that, in principle, should offer the features required for rapid mass adoption of IoT deployment.
It is starting to look like LoRa is gaining traction for a variety of use cases serving the IoT within the unlicensed spectrum. LoRa is the alternative IoT connectivity network to traditional licensed spectrum solutions such as NB-IoT and LTE-M. Buying expensive licenses to run your own network is unnecessary, as LoRa operates in frequency bands that are "unlicensed." This enables practically anybody to set up a few gateways (that are type-approved, of course), switch them on and wirelessly connect end devices.
From toothbrushes and dog collars to thermostats and more, this could add a wide variety of devices to the wireless grid. For example, a Dutch company that I met at the event, Xignal, is literally building a better mousetrap—yes, a wirelessly connected mousetrap. For large industrial sites, it is important to understand whether or not your trap been activated, and where it is located. LoRa Alliance technology can enable this.
The LoRa Alliance is an open, non-profit association of members collaborating to drive an open global standard for secure IoT connectivity. At the event, it was encouraging to see so many IoT players coming together—hundreds of C-level decision-makers in IoT, from both large corporations and smaller startups, who are actively building on LoRa. The whole value chain was present, from sensor makers, chip manufacturers and connectivity-layer providers to gateway manufactures and all the way to end-to-end service providers. But why the huge interest in this seemingly marginal technology? The answer is actually quite simple:
1. Low power: LoRa end devices can operate for many years on a single AA battery, meaning no fixed power source is necessary.
2. Wide area: LoRa has a long range (up to 15 kilometers in rural environments) and strong propagation, allowing urban and underground use cases.
3. Low cost: LoRa chipsets and subscriptions are low-cost, which even allows use cases in which devices are disposable. LoRa technologies are significantly cheaper than traditional cellular options, both in terms of CAPEX and OPEX.
4. Simple provisioning: LoRa devices are designed so that management and provisioning are simple and flexible.
What You Need to Know
So what are the key takeaways from the LoRa conference that everyone should know? My top three:
1. The time for LPWA is now, but cost slows down adoption:
LoRa networks are being rolled out rapidly, but demand is still lagging behind due to unclear use-case scenarios. Although the number of LPWA connections is still small, I expect that they will increase rapidly during the next few years, and that most of those initial deployments will either be proprietary or LoRa-based. The LoRa ecosystem has matured to a point at which the market is ready to take off. Still, device and connectivity cost needs to come down before widespread adoption is possible. The largest issue right now for the alternative LPWA technologies is that device costs are not yet well understood, while many industry players are making inflated, low-cost marketing claims. Ultimately, however, device cost will be a big driver (or barrier) for uptake. We expect 3GPP-based NB-IoT and LTE-M will come onstream soon as well; however, for the time being, LoRA has the upper hand in actual deployments.
2. Unclear business model and regulatory issues create additional hurdles for some verticals:
Different LPWA-spectrum allocations and regulatory approaches are likely to drive industry fragmentation. The current LPWA device deployments have been limited to primarily private deployments, delaying the growth of public network connections. The LoRa use cases with the most potential and actual deployments are smart metering, smart buildings, supply chain, asset tracking and agriculture.
Mission-critical applications will require QoS and high availability, and industrial use cases will have the most stringent requirements. The current regulated spectrum solutions generally allow better QoS and latency control, though they themselves are also subject to statistical assumptions and the whimsical behaviors of radio wave propagation (yours truly graduated with a master's degree in radio channel equalization techniques).
On the telecom operator side, it seems that all operators want to play a part, but the revenues are minuscule compared to cellular. Most operators have the IoT on their strategic agenda, but the existing business model does not allow them to reap much value. It is still unclear what the role of operators will be. We are, however, seeing early signs of consolidation by operators to accrue more value in end-to-end solutions—for instance, Vodafone acquiring Cobra Automotive.
3. The connectivity market will remain small:
The LPWA connectivity market will be small compared to cellular connectivity for mobile phones. The majority of the value will be in applications and analytics, simply because the IoT stack (comprising chips, modules, end devices, gateways, network servers, provisioning, analytics and so forth) is generally simpler and optimized for high unit volumes, narrow bandwidth and low-cost connections.
The LoRa ecosystem needs to make it easier for end customers to understand and buy solutions. End-to-end offerings and out-of-the-box connectivity will help customers in non-technical industries more readily adopt the IoT. Service providers are already quoting prices in the range of $0.50 to $1.00 per connection annually. This is expected to come down as competition increases.
All in all, it's a very exciting time to be an IoT investor, not only because of LPWA or 5G, but primarily due to the massively important enterprise and consumer solutions these technologies enable.
Bo Ilsoe has more than two decades of experience in venture capital, organizational and strategic development, and sales and marketing. Bo began his career in the technology business with Alcatel and Nokia, where he launched some of the world's first GSM networks. Since 2002, he has been an investor and has created in excess of a billion dollars in shareholder value with successful exits such as CapitalIQ, Heptagon and Fyber (formerly known as Sponsorpay). Having lived in seven different countries, including six years in Singapore, and having invested in more than ten countries, he comes with a global view to business. He graduated with honors with a Masters in Electronics Engineering degree from Aalborg University, Denmark. Bo is based in Europe.
