Now Reading
Specific apps need LTE chipsets optimised for IoT

Specific apps need LTE chipsets optimised for IoT

Posted by IoT Now MagazineJanuary 16, 2017

Eran Eshed is the vice president of worldwide marketing and sales at Altair Semiconductor. Here, he tells Christopher Taylor, the director of analyst firm Strategy Analytics’ RF and Wireless Components advisory service, how the company is progressing following its acquisition by Sony earlier in the year and how he views the prospects for single-mode LTE

Christopher Taylor: How is Altair progressing under Sony, and in single-mode LTE as a whole?

Eran Eshed: For Altair, it has been about eight months since the acquisition concluded and we are past the integration challenges. We keep investing in our target products and markets as before, and in some cases have even stepped up the investment which I believe provides our customers with a level of confidence that we are financially solid and strategically focused. One of the major advantages of becoming part of a larger company is the ability to bundle products and offer a platform-level solution rather than a discrete modem.

We’ve been progressing very nicely. We have increased our focus on the new R13 technologies – Cat-M1 and Cat-NB1 – and have accelerated our development. A highlight of this is that we demonstrated the first live connection on an LTE Cat-M1 network. We did this with AT&T and Ericsson infrastructure equipment using our ALT- 1210 chipset in the IoT devices.

This followed a somewhat lengthy process of interoperability testing and functional testing, and we now have our chipsets up and running over the air with a live connection, while some of our competitors have announced similar achievements after us. We wouldn’t consider the technology mature yet and nationwide coverage will take some time, but I think this is an important milestone for the 3GPP and cellular industry in the light of all the LPWA wars – the proprietary camp versus the LTE Cat-M camp and the LTE NB1 camp. LTE Cat-M is real and no longer just the subject of 3GPP meetings and white papers.

CT: Will we see Altair LTE chips in a Sony smartwatch, for example?

EE: Conceivably we could see this sort of development, and although some customers have raised some concerns that Altair would become a captive, internal supplier to Sony, this is not the case, we are operating as a catalogue chipset provider to whoever is willing to buy our chipsets. The flipside is that there is no guarantee that we will get business with Sony; we have to win those sockets on our chipsets’ merits and we cannot take the business for granted.

CT: Given the industry’s progress with Cat-M, where does this leave Cat-1 and Cat-0?

EE: LTE Cat-0 has been dead for a while. There was little motivation to deploy it from the outset and now it is even less appealing.

LTE Cat-1 is still gaining momentum. Fast progression between standards probably has hurt Cat-1, but if you want to deploy IoT within the next 12 months using LTE and not 2G or a proprietary technology such as LPWA, Cat-1 is the best and only real option today. Nobody will be able to deploy LTE Cat-M in a commercial service in the next six to nine months because of coverage and other issues. It’s important to consider that the design cycles are very long in the IoT market, sometimes the time taken from starting design to market introduction can easily be two years, so deployments that we will see next quarter will be design-ins that we had more than a year ago in many cases. Altair has quite a few LTE Cat-1 customers in the pipeline, some of them Tier 1 module and device makers, so we think that 2017 will be a good year for Cat-1. The market probably will still not reach tens of millions of units in 2017, but based on customer design-in activity, we expect to see real deployments happening.

On top of that, Cat-1 offers more bandwidth than Cat-M, so it may not be the best option for smart city sensors or wearable devices, but for applications that require more bandwidth but don’t need 100 Mbps or 150 Mbps, then Cat-1 is very suitable and mature and offers VoLTE capability. LTE Cat-M will eventually offer VoLTE too, but this will take time so there is a market window for LTE Cat-1. I think that, even beyond this window, Cat-1 will continue to have a role in supporting applications that need bandwidth that’s a bit higher.

CT: Which markets or applications are leading the way for single mode LTE?

EE: First of all, wireless broadband, such as fixed wireless applications, CPE/LTE routers, gateways, Mi-Fi devices, USB dongles and others represent a solid market, but not one that is growing at a tremendous rate. You will find these applications and devices mostly in emerging LTE countries, but also in the US, Japan and other developed countries for fixed line replacement for voice and broadband access.

IoT is obviously a very diverse market, and has tens of different vertical markets and applications. The ones that we see driving volumes today are:

  1. Smart metering for electricity, gas, and water.
  2. Telematics, mostly in tracking and UBI (Usage Based Insurance). Telematics includes stolen vehicle recovery, ‘Buy Here, Pay Here’ offerings and leased vehicle management. For example, car dealers install millions of trackers every year in the US alone for managing buyers with suboptimal credit history so that the cars can be turned off and recovered if the buyer defaults.
  3. Security alarm panels, which until recently used GSM or CDMA connections, and now need to have these older technologies replaced. These are all established markets with established business cases, what we call M2M today, and are migrating to new technology.

These three are in my view the most significant in terms of volume and adoption of LTE, but there are several other areas such as various sensors, propane gas level measurement, bridge oscillation monitoring, garbage collection … it’s endless, and these will grow significantly.

CT: Beyond M2M, will single mode LTE handsets be viable as a replacement for today’s ultra lowcost 2G and 3G handsets?

EE: Altair offers single-mode LTE Cat-4, Cat-6 and Cat-10 chipsets today, but we’re not focused on handsets.

CT: Qualcomm recently announced partnerships and design wins for itss MDM9206 and MDM9207 LTE chipsets for LTE Cat-M and NB1. What does Qualcomm’s entry into the market mean for LTE chip specialists such as Altair and its competitors?

Eran Eshed:
If you build a solution optimised for a specific use case, there’s an opportunity to differentiate

EE: One perspective is that Qualcomm’s entry means that this is now a significant market, not that we needed this indication, but it means that the market is real and maturing. Altair, Sequans and even Marvell cannot invent new markets; if companies like Qualcomm, MediaTek, Hisilicon, Intel and others are not in a space, then it probably means the space is not real, so this is a good sign.

Qualcomm became so dominant in phones because that market is so consolidated in terms of its’ requirements. Regardless of whether a phone is a featurephone or smartphone, the volume driver is essentially one application and Qualcomm did a great job of having a strong grip over intellectual property rights and scaling to huge volumes. This has worked well for them, but the cellular world is changing.

It is no longer just about phones, but it is about 100 different end markets with different needs. Of course the wireless modem is a common requirement, but the diversity of requirements is unlike anyone has known before, which creates opportunities for smaller players. As long as we differentiate ourselves, for example with extremely low power consumption, then, while some designers of devices with mains power such as electric meters won’t care, in perhaps ten other spaces people will care a lot about this capability.

If we focus on some of these areas, it gives us the opportunity to dominate each of them and the volumes – even in IoT terms – will be huge. You can extend this to other areas if you understand the use case of the application such as the tracking device, the health device or the wearable device. If you build a solution optimised for that, then there’s an opportunity to differentiate. This differs from before, where someone had all the variants, sizes, functions and integration options covered. In addition, integration is not necessarily an advantage here; the applications processors Qualcomm and others offer are not an advantage for IoT, and neither are the peripheral functions such as Bluetooth and Wi-Fi in most cases.

Qualcomm has a broad portfolio, great technology and it is a fierce competitor. However, we have built a sustainable business in its shadow. In contrast to larger players, Altair can take advantage of its flexibility and speed of execution to give smaller – relative to the handset market – customers intimate relationships with more accommodation and attention. Customers know that loss of a one million unit order is not even a rounding error for Qualcomm, but for us it is a big, big deal, and we are extremely excited to win this kind of business.

Overall, the entry of more suppliers into the LTE IoT chip market is a positive, and forces us to be more competitive and aggressive, and to continue to design better products. We will continue to focus on what we do well and keep increasing our business, in the process showing that experience in IoT and customer engagements count. There is definitely room for the LTE IoT chip specialist.

About The Author
IoT Now Magazine

Leave a Response