After you’ve been in this industry for a while, you start to learn to trust your nose when it comes to new technologies, writes IoT Now magazine’s editor, Alun Lewis. Exposed to the usual hype, disinformation and vapourware, it’s all too easy to become sceptical – even cynical – about many claims that are made.
Once every so often however, something comes along that does represent a real potential game-changer – and that something, this time around, does seem to be Low Power Wide Area (LPWA) networking.
Out at a LoRa Alliance meeting in Rotterdam in early November, there was a palpable buzz in the air, huddles of folk in intense conversation and, from the stage, real life stories of deployments, devices and successful roaming and interworking. And that’s only one part of the LPWA story, where a mix of different technologies are under discussion: some part of the 3GPP standards bodies, aimed at licenced spectrum; others looking to exploit unlicensed spectrum via currently unstandardised, semi-proprietary solutions.
There’s certainly been a clear need in the market for some alternative to cellular or satellite communications for some time now. With our obsession for pushing for ever higher bandwidths to satisfy humanity’s seemingly insatiable need for audio and video content to stimulate increasingly jaded palates, the much more humble and basic requirements of ‘things’ have been largely forgotten.
Coincidentally, these needs were actually identified around 25 years ago in a spoof paper “Gigabit Network Economics and Paradigm Shifts” that proposed that there were certain applications that didn’t need high bandwidths, such as tracking glaciers and the US postal service. Another, subsequent paper even suggested ultra-secure, ultra-low speed comms using neutrinos, and military grade comms using tanks with zeros and one painted on the sides.
Putting the humour aside though, the roles and characteristics of probably the majority of connected devices and sensors that will drive the next phase of our migration to a ‘smart’ environment have specific demands. Battery life is critical, so devices using these new technologies – LTE-M, LoRa, SIGFOX, Ingenu’s RPMA, GSM-EC, Weightless, NB-IoT – are intended to keep running for between five and ten years with no human intervention. Other issues involve the ability of radio waves to penetrate buildings easily and, as always, the cost and availability of the spectrum itself.
If there’s one infallible guide to a new hot area, it’s the number of analyst reports that get released on the topic concerned and the last few months have seen a flurry from various quarters – all with roughly similar bullish conclusions and the usual caveats.
Beecham Research is just one of these, predicting that LPWA could provide as much as 26% of the total IoT connectivity market by 2020 – adding up to around 345 million connections worldwide. According to David Parker, senior analyst at Beecham, “LPWA represent the most dynamic and potentially game changing development in the M2M/IoT market. The lower speeds of LPWA are the trade-off for longer range, offering networks optimised for machine connectivity with much lower deployment costs than traditional cellular networks. LPWA will both compete and collaborate with cellular and other network technologies to stimulate market growth with more connectivity options for end-users”.
The report also warns of the hype around ‘big data’ applications where everything is discussed in terms of the 3Vs – velocity, volume and variety. “Our look at LPWA highlighted that there are many applications that are not big data and not necessarily real-time, interactive or immersive,” echoes Robin Duke-Woolley, CEO of Beecham. “So, from a connectivity point of view, the market will move towards 4G-5G for satisfying big data IoT, while on the other side LPWA and equivalent networks will address the low data IoT requirement.”
The Beecham Research report additionally investigated the increasing number of LPWA technologies including SIGFOX and companies in the LoRa Alliance, which are currently leading the LPWA field in terms of network deployment, industry support, investment and customers. Most of these LPWA solutions use the ISM (Industrial Scientific and Medical) bands better known for use by short range wireless technologies like Zigbee, WiFi and 6LoWPAN. However, recent advances have enabled LPWAs to be established using the ISM bands over longer distances, up to 50km in rural areas and 5- 10km in urban areas. Another LPWA technology with long-range, low-power characteristics is known as TV White Spaces (TVWS), which uses the gaps in between VHF/UHF parts of the spectrum, previously used for TV broadcasting. TVWS promises connectivity over distances of 10Km and with superior in-building penetration when compared to 3G or 4G.
“New entrants working in the ISM and TVWS bands are promoting overall market growth and providing a spur to action within the GSMA world,” concludes Parker. “Developing standards for the cellular operating community is a slower process, but the emergence of LTE-M and Narrowband IoT (NB-IOT) will allow cellular operators to compete with these new entrants on a level playing field of range, battery life and costs.”
These developments – just like the arrival of affordable and mass market cellular communications in the 1990s – are inevitably going to have major and inevitably unpredictable second order effects. This is especially true if you consider their impacts on the evolution of smart city concepts. If we can take analyst reports as one important metric of a hype wave, then the large number of smart city conferences being held must also represent another signal. Indeed, a number of LPWA-based smart city projects are already underway. Interestingly, India is very active in both these areas at the moment, driven in part by Narendra Modi’s announcement that India will create 100 smart cities over the next few years and complementary announcements from Wipro and TATA that they are exploring LPWA applications and running trials.
India’s Tech Mahindra is also active in this market as Prajakt Deotale – head of their Europe Consulting Services operation – explains:
“TechM is supporting end-to-end enablement of several IoT solutions on LoRA. It’s important however for organisations to have end-to-end services that span across IoT devices such as vendor management, network consultancy, data storage, OSS/BSS integration and applications development which are all specific to LoRA. Tech Mahindra has been researching the subject of LPWA extensively internally; with comparative analysis of the several solutions available in the market like SIGFOX, LoRa and others taking place in order to decide the applicability of solutions on these networks.”
He continues, “Our research so far indicates that solutions which have low to moderate data requirement like smart lighting, smart bins, smart parking and so on are best suited for LPWA networks. Solutions with high data requirements like connected cars, are not suitable. It will be interesting to see how these technologies evolve further. In particular, how might mobile service providers use these LTE IoT variants to compete against companies, operators and service providers using LPWA technologies? LTE IoT variants are definitely late entrants in the optimised IoT network space and would need to play catch-up with LPWA for some time. However, LTE IoT can overcome one of the biggest disadvantages of LPWA technology – that is, network throughput. This advantage could make it a preferred solution going forward. Also, one major consideration here would be that both LPWA and LTE IoT are device dependent i.e. there are no commercial gateways available that will work with both. Therefore, if someone wants to move from LPWA to LTE IoT, they would have to undergo an additional infrastructure investment to replace the existing devices with the ones compliant with the newly chosen technology.”
The issue of how mobile service providers are going to react to new and potentially subversive technologies impacting one of their historic markets is provoking a variety of responses. At the previously mentioned LoRa Alliance event in Rotterdam in early November, Belgium’s Proximus and the Netherland’s KPN demonstrated interworking and roaming on their two deployed LoRa network infrastructures, supported by Actility.
One perspective comes from Arnaud Vamparys, senior VP for Seamless Wireless Strategy at Orange: “As part of our strategic plan Essentials 2020, IoT/M2M is a key diversification domain for Orange. LPWA is an important network segment of IoT and it is also the segment for which current cellular networks require the most important evolutions. This IoT market segment needs long-range networks for ubiquitous connectivity associated with low cost and long battery life connectivity modules, which is typically something that existing mobile networks are not currently capable of.”
He adds, “In order to quickly address this new IoT market, Orange is deploying a national LoRa network in France. At the same time, Orange Labs is pushing international standards to address these requirements with a software upgrade of current mobile networks. This second phase is set to start at the end of 2016 and will further accelerate IoT business for Orange. We believe that the most appropriate option for European and African countries is to upgrade 2G (GSM) networks with a new standard called Enhanced Coverage GSM (EC-GSM), as in a vast majority of European and African countries, 2G territory coverage will remain significantly better than 4G coverage until 2020. Territorial coverage – and not just population – is a key promise for the LPWA market with objects located virtually anywhere and not only in most populated areas.”
Vamparys concludes: “Aside from EC-GSM, the 4G (LTE) standard will also evolve to enable LPWA business with an option called Narrow Band IoT (NB-IoT). This option still needs some debate to decide its radio structure and Orange is participating in these discussions to reduce potential worldwide fragmentation. Depending on what the final standard(s) will look like, Orange could consider these option(s) for some markets in Europe, where 4G territory coverage can compete with 2G in the short term.”
This apparent proliferation of standards within standards is also commented on by Paul Egan, IoTUK principal consultant at the UK’s Digital Catapult organisation: “It is accurate to say that mobile operators globally have been slow to appreciate the opportunity in LPWA and to make any large investments. This is partly due to a lack of clear market pull from their customers, but more importantly because no standards-based low power mobile technology exists that could be deployed within their existing networks until now. In September ETSI/3GPP approved the use of the Narrow Band Cellular IoT (NB-CIoT) proposal from a consortium led by Huawei/Neul, Qualcomm and many mobile operators. Interestingly, this standard will now be integrated into Release 13 in December 2015. It is a surprisingly rapid development, adoption and approval of a new standard which could be seen as a response to the LPWA players.”
Egan adds, “Early entrants in the LPWA space included SIGFOX, OnRamp, NWave and Neul. These new entrants all proposed the use of license-exempt spectrum for their solutions, but had differing business models. Neul was acquired by Huawei in September 2014 and since then has been developing the NB-CIoT system to work within existing sub 1GHz licensed mobile networks and discontinued the work in licenseexempt bands. There are many differences between the competing ecosystems and the technologies – but there are many similarities as well. Both approaches aim to deliver scalable solutions that solve problems associated with existing cellular systems. These include: significant improvements in link budgets – in some cases an extra +20dB: a massive reduction in power by re-architecting the air interface to allow end-point radios to be switched off almost completely allowing for many years of battery life from simple cells; and low cost infrastructure deployments.
He concludes: “If we look at link budgets, most of these gains are achieved through trading bandwidth for sensitivity and coding. This allows for data rates to be in the Kbps range and below and still offer a high probability of the uplink signal getting through even in very challenging RF conditions and locations. Costs are also saved in the license-exempt systems by using commoditised standard product silicon radios. In the NB-CIoT, much of the system has been designed to remove much of the standard essential Intellectual Property (IP), so allowing chip vendors to produce devices at similar costs to the licenseexempt systems. This is a big change to the current cellular eco-system where a small number of players collect significant IP royalties.”
These key – and seemingly eternal – issues of standardisation paths are also raised by Olivier Beaujard, VP market development at Sierra Wireless, “There are many different LPWA technologies being discussed right now, all in varying stages of development, standardisation, and commercial availability, which is causing some confusion for customers. They can be split most simply between proprietary and standardised technologies. Standardised solutions are more viable long term than proprietary solutions because longevity is a function of ecosystem support much more than technology. We strongly believe that LTE-M will emerge as the preferred LPWA technology as it is a standardised solution that will re-use the existing cellular networks and not compete with them.
“LTE-M is the continuity candidate for LPWA because of how it builds on the success of cellular networks in enabling IoT applications and solutions. Taking a divergent path on LPWA risks too much as IoT enters the mainstream and scales from today’s levels to greater ubiquity over the next decade or so. LTE-M delivers a stable future with a combined low-power and low-cost solution which uses the strong, field-proven security mechanisms now expected by all IoT industries. LTE-M can be quickly deployed in global markets thanks to the re-use of 4G/LTE network infrastructure and we can expect it to hit the market in 2H 2017.”
So, it looks like another level of etheric hum is about to be added to our already crowded airwaves – only this time it’s going to be composed of machines and things talking to one another. Just as almost every household already has a usually unnoticed background symphony of whirs, clicks and buzzes taking place as thermostats switch on and off and central heating and refrigerator pumps burst into sudden life, we’ll be similarly unaware of the secret life of machines and their private discussions …