Creative Connectivity - the site for eHealth, telematics and wireless

Zombie Wireless Standards

September 3rd, 2013 |  Published in Wireless Connectivity  |  27 Comments

Back in 2010, Mark Thomas, the head of PA Consulting’s Strategy and Market practice published a book called The Zombie Economy.  In it he defined a Zombie company as one which is generating just about enough cash to service its debt, so the bank is not obliged to pull the plug on the loan.  The issue with such companies is that they can limp along, and just about survive, but as they don’t have enough money to invest, they fall over once the economy picks up, as they become uncompetitive.  The problem they pose is that by continuing to exist in this Zombie state they threaten the development of other companies, acting as a damper to more sustainable businesses.

It struck me that there’s a close analogy in the area of wireless standards where we have what are effectively Zombie wireless standards.  There’s not necessarily anything fundamentally wrong with these individual standards, other than that they have failed to get traction and so limp along.  Here, the problem is that they tend to jealously claim a particular application sector or market segment, blocking other more successful standards from entering.  That has a damping effect on product development, creating silos which keep putting off innovation in the hope that one day the standard will gain traction, constantly delaying growth and interoperability.  Because they’re not being incorporated into enough products, they have effectively lost their ability to function and have become half-dead, half-alive ‘Zombies’.

I think it’s time to recognise the damage that this is doing.  Rather than pursuing multiple parallel paths, the industry needs to concentrate on a far smaller number of short range wireless standards. They in turn need to embrace the requirements of a wider range of sectors.

Let’s start with the wireless standards that are firmly in the land of the living.  There are three of them – Wi-Fi, Bluetooth and Bluetooth Smart.  (If you think the last two are the same, they’re not – have a look at this article.)  There are two reasons for singling these three out, which is the installed base and the rate at which they are growing.  Each of these three is already incorporated into hundreds of millions of products and that number is growing year on year.   These two points bestow a lot of advantages to the standards.  The most important one is that it provides critical mass.  It allows enough companies to make profits from supporting the necessary elements of the standard, whether that’s designing chips, writing protocol stacks, developing applications or providing testing and certification services.  That ensures that there is a vibrant ecosystem of people with a vested interest in working to make the standard cheaper and easier to use, and also to evolve and improve the standard.  It’s not cheap to develop a global wireless standard – it needs the industry to contribute hundreds of millions of dollars, which only happens if the standard is being used.

When a wireless standard becomes established, it begins to invest in a different level of detail.  As more companies introduce it into their products, interoperability, security and robustness become more important to make sure that it works in the real world, rather than just in the lab.  These are the much less glamorous aspects of standards development, but they’re the ones that increase confidence in it, generating a virtuous circle of development.

In contrast, the Zombie standards tend to concentrate their limited development resource into adding bells and whistles to meet the perceived requirements of niche market after niche market. But they rarely finish the work, instead devoting time to aggressive PR to conceal these deficiencies.  The problem here is that without the resource to solve interoperability, backwards compatibility and certification issues, each niche market will waste time flirting with these standards before discarding them.  Having been bitten, they then lose confidence in wireless, putting back the evolution of that particular sector.

Zombie standards can generally be spotted because they advertise their merit either with a single technical feature, or a single, highly vertical market.  As I’ve said before, they’re not necessarily flawed or inadequate, but they fail to achieve the scale they need to become robust and then evolve with their markets.  However, their continued existence threatens much larger market segments who mistakenly back them, particularly when the relationship results in the manufacturers within that segment losing faith in wireless.

It’s time to name them, and there is something of an irony that two at the top of the list coincidentally start with “z” and end with “e” – ZigBee and Z-Wave.  ZigBee has always sold itself on its mesh capability, although its most significant deployments in smart metering have generally failed to use that feature.  It follows the pattern of trying to change its spots to court any potential suitor.  Hence we’ve had versions for health devices, home automation, TV remote controls, lighting and smart metering.  Within the smart metering sector alone there are now five different versions of the specification with limited to no interoperability between them.  That leaves product manufacturers with a real concern, as they have very little confidence that a product they bring to market will work with other products, or that the standard will remain backwardly compatible with a future version in a few years’ time.

Z-Wave has not been through as many versions, and offers a correspondingly better level of interoperability.  But it shares the issue of other Zombie standards, which is that it does not, nor is ever likely to exist within smartphones or tablets.  That leads to a double whammy that is fatal to most Zombie wireless standards.  The first issue is that in order to make their product usable, manufacturers have to supply both ends of the wireless link, which costs more.  The second is that because their devices can’t be controlled from a smartphone or tablet, they don’t have access to the millions of apps developers who are constantly upping the pace of development and expectation of user experience.  That means that many of the Zombie wireless products start to look outdated, often before they even come to market.

ANT+ is another contender for a Zombie standard.  It has done well in fitness devices, and has even made it into a few smartphones, but there’s a growing trend for companies that used ANT+ in their products to jump to Bluetooth Smart, led by some of the leading fitness brands.  The option of connecting to an iPhone or iPad is just too persuasive.  In response, ANT is starting to promote itself as a contender for smart homes, following the well trodden path of other Zombie standards of constantly looking for another niche where they can pretend to be a big fish in a small pond.

You can level some of the blame for creating Zombie standards at the doors of Bluetooth and Wi-Fi.  The Zombies have survived because the larger standards never showed any great commitment to niches such as sports and fitness or smart energy.  If they had, these standards would never have achieved the limited traction they have.  But that has led to a classic Catch-22 situation.  Bluetooth and Wi-Fi ignored these markets, allowing the Zombie standards to claim ownership of them.  When these markets started to take off, manufacturers using these Zombie standards found that customers couldn’t connect the products to their smartphones and tablets, limiting the market size and blocking any disruptive newcomers.  This meant there was little market growth, so the established standards continued to ignore the markets.

There’s a further problem arising from the lack of growth.  Without ongoing funding from market usage, standards tend to atrophy.  That’s a problem for commercial applications, as the environment in which a standard has to operate is not static.  New standards and increased deployment of competing standards clog the spectrum, making coexistence more of a problem.  Unless a standard can track this and evolve, it risks becoming less and less effective in an ever more congested spectrum.

Zombie standards don’t just distort the market, they distort the way we train engineers, not least because they seem to be happy for academia to abuse and misuse then.  Rather than training young engineers and designers to use mainstream wireless standards, too many universities concentrate on these Zombie standards, presumably because they feel that they can change and tinker with them without anybody getting upset.  ZigBee is a classic offender, happy to lend its name to anything a student builds using a 802.15.4 radio. This cavalier attitude to the principles of a standard results in a steady stream of graduates who need several years of retraining before they’re of use to companies  developing commercial products.

The Zombie effect doesn’t apply to proprietary wireless, which still has advantages.  If you want lowest power, greatest range, or if there is any other parameter where you can’t compromise performance, proprietary remains a good choice.  The only issue is that it throws away interoperability.  But if you can own the entirety of an application, proprietary wireless has a lot going for it.  Many of the Zombie standards could have a future as a proprietary solution if they ever get over losing their desire to be a brand.  The only issue in this case is who then owns and develops them, particularly who owns the IP.  One of the rarely spoken facts about Zombie standards is that using them may mean you’re infringing patents, because they don’t embrace enough member companies to give them sufficient IP coverage.

So there are two wake-up calls.  The first is for Bluetooth and Wi-Fi to recognise and engage with the markets that the Zombie standards are now stifling, bringing relevant companies into their standards processes and giving them the features they need.  The second is for manufacturers to realise which way the wind is blowing and move their efforts to these mainstream standards.  With these in place, silicon companies, who are currently burdened with the overhead of supporting multiple protocols, can concentrate their resources into the ones that will make a difference.

As with the Zombie companies, it’s not necessarily anyone’s fault that they’ve failed, but the markets need consolidation.  They won’t be the first wireless standards to fail and they won’t be the last.  But if we want growth, we need to clear the field for innovation that can succeed at volume, not just in a range of niche products.  Because it’s in everyone interest to turn niches into vibrant ecosystems.

27 comments ↓

#1 Harald Naumann on 09.13.13 at 5:24 am

Nice article, but I miss 6LoWPAN. 6LoWPAN we have on 2400 MHz and on 868/915 MHz. We also have it on PowerLine Communication. It is already on IPv6.
ZigBee has interoperability problems. Two software layers called ZigBee profiles developed by two different companies can create profiles that interfere with different versions of ZigBee profiles. Worst case two devices that have ZigBee chips might not interoperate. And even two of the same chips can’t communicate to each other if the run different profiles like Smart Energy and Home Automation. To cover my home I need the mentioned profiles and if I get older than there is a need for Health profile. On top of this chaos a ZigBee router on Smart Energy can’t route Home Automation or Health profile. This will force three different ZigBee routers on several places in my home. ZigBee routers are always on. That means the three ZigBee routers will charge the triple of energy. Is a system that waste 300 % of energy is still a low power standard? Correct me if I am wrong.
6LoWPAN by ARM
Sensinode is today ARM. With Sensinode ARM is able to offer the IoT from one hand (wireless hardware, software stacks and services). The stack has a need for 32K of flash only, offers extreme robustness on optional frequency-hopping MAC and support for both 2.4 GHz and sub-GHz bands such as 868/915 MHz. While 6LoWPAN was originally designed to support the IEEE 802.15.4 radio, their stack can be ported to a variety of different MAC/PHY solutions.
With Sensinode ARM got a service platform through powerful yet simple HTTP REST APIs. They also provide a Java SDK complete with commercial source code. Their device library provides automatic registration and maintenance, CoRE resource discovery, a CoAP server API, a CoAP client API along with eventing. The library is available for C, Java SE, and Android. Sensinode is actively developing and promoting open standards at the IETF, ZigBee, ETSI and the IPSO Alliance. And ARM Cores is already included with a lot of MCUs and SoCs (e.g. Freescale, STM, NXP, ATMEL).
ARM acquires Internet of Things Startup Sensinode
http://techcrunch.com/2013/08/27/arm-acquires-internet-of-things-startup-sensinode-to-move-beyond-tablets-and-phones/
ZigBee wants to be the Bluetooth of the internet of things. Too bad everyone hates it
http://gigaom.com/2013/08/30/zigbee-wants-to-be-the-bluetooth-of-the-internet-of-things-too-bad-everyone-hates-it/

#2 Nick on 09.26.13 at 5:05 pm

Harald,

The reason 6LoWPAN is not there is that this is an article about interoperable wireless standards, which I interpret to mean interoperability at a user level. 6LoWPAN does not meet that definition – it’s a building block for one part of the protocol stack, which is used by a number of other standards.

6LoWPAN advocates do a lot of damage by promoting it as if it were an application level protocol. It’s not. Yet I’ve seen it listed in the specifications in consumer product literature. This is madness. I do blame many academics and hobbyists for promoting this idea, presumably due to a lack of understanding about what interoperability really is and assuming that picking on a small part of a bigger entity makes the whole thing work. This is very sloppy thinking and does no service to growing the market for any wireless standard.

#3 Harald Naumann on 09.26.13 at 8:00 pm

> I do blame many academics and hobbyists for promoting this idea, presumably due to a lack of understanding about what interoperability really is and assuming that picking on a small part of a bigger entity makes the whole thing work.
HN: Nick, ARM bought a 6LoWPAN company. Other companies have have done the same before. The wish of the academics and hobbyists will come true.
Anyhow, we have more Wireless-M-Bus devices that ZigBee in my country. W-MBus contains no Z.
> So there are two wake-up calls. The first is for Bluetooth and Wi-Fi to …
Sorry, but I do not agree.
> ANT+ is another contender for a Zombie standard. It has done well in fitness devices, and has even made it into a few smartphones, but there’s a growing trend for companies that used ANT+ in their products to jump to Bluetooth Smart, led by some of the leading fitness brands.
HN: Are you sure? It jumped in an Android device last days.

#4 Nick on 09.26.13 at 9:32 pm

Don’t assume that every company purchase is for their product, rather than the expertise of their engineering staff. As I said in the previous reply, 6LoWPAN is ONLY a component part. I expect to see it it other wireless standards within the next few years.

If you’re aware of a successful wireless standard that has come out of academia, let me know. To the best of my knowledge, GSM, DECT, Wi-Fi and Bluetooth did not. That’s partly because the major part of a SUCCESSFUL standard is the hard work on interoperability and testing, which I’m afraid are lacking in most academic standards. For more on that, see my article on interoperability.

I also avoided wireless-MBUS, because of its limited distribution. I’m interested in the Internet of things, not the subset of the Internet of a few Things in Germany.

And I’m pretty sure about ANT. One Android device does not make a successful standard.

I’m sorry if this sounds negative, but the whole point of the article was to concentrate on the winners and put the losers out of their misery. Sometimes you need to be cruel to be kind.

#5 Harald Naumann on 09.27.13 at 4:16 pm

OMS does not contain a Z:
http://www.oms-group.org/en_index.html
OMS is compatible with a widely used KNX Standard (ISO/IEC 14543-3 = EN 50090).
OMS is an open communication standard based on European norms and is not a proprietary solution of a single manufacturer. The Open Metering System is therefore best suited to be used as a long-term, stable interface in all congeneric products which access consumption data – for example in intelligent homes or buildings.

An interface is made available for data visualisation (consumption display), integration of the building automation for the end customer as well as for future services such as tariff or load management. This interface is fully compatible with the widely used KNX standard (ISO/IEC 14543-3 = EN 50090).
Members of OMS-Group:
figawa
KNX
Plus:
Adeunis RF
Adunos GmbH
Allmess GmbH
AMBER wireless GmbH
Apator Metrix
Aquametro AG
BRUNATA Wärmemesser GmbH & Co. KG
BTR NETCOM GmbH
Deutsche Telekom Technischer Service GmbH
devolo AG
DIEHL Metering
Elster GmbH
Elster Messtechnik GmbH
Flonidan DC A/S
GOK GmbH
GÖRLITZ AG
GWF MessSysteme AG
Hager Electro GmbH & Co KG
Heinz Lackmann GmbH & Co. KG
iAd GmbH
IMST Design und Systems International GmbH
Ing. Büro Lertes GmbH & Co. KG
Iskraemeco
ITRON GmbH
Kamstrup A/S, Mettmann
Landis + Gyr GmbH
Michael Rac GmbH
NZR GmbH & Co. KG
Power Plus Communications AG
Radiocrafts AS
Relay Reinecke Elektronikentwicklung und Layout GmbH
Rossweiner Armaturen und Messgeräte GmbH & Co. OHG
Steinbeis Transferzentrum Embedded Design und Networking
Sensus GmbH
Siemens AG
SMARVIS GmbH
Solvimus GmbH
Sontex SA
SSV Software Systeme GmbH
Techem Energy Services GmbH
Tixi.Com GmbH
Webolution GmbH
WEPTECH elektronik GmbH
Wien Energie Gasnetz GmbH
Wikon Kommunikationstechnik GmbH
ZENNER INT. GmbH & Co. KG
Germany is one of the mayor countries in EU.

> I’m interested in the Internet of things, not the subset of the Internet of a few Things in Germany.
The few things in a mayor country in Europe? If you really would like to cover Europe, then you have to accept, that we have often the same currency, but not the same wireless standards in EU.

#6 Nick on 09.27.13 at 4:32 pm

Let me point out the difference between M2M, which is what I think you’re caught up with, and the Internet of Things.

M2M is largely about silos. It may use proprietary, semi-proprietary or global standards, but it typically has two limitations. One is high cost, because the many niche players don’t achieve economy of scale. And the other is that data is generally not shared, but resides within the company running the M2M application.

In contrast, the Internet of Things needs cheap ubiquitous sensors, which come with volume. They will grow out of wireless standards that achieve global, not local volume. It also needs data to be shared so that it has aggregators who can take data from multiple sources and apply analytics to produce valuable and compelling insight. We haven’t really got either of those today, but we’re beginning to see the building blocks appearing.

I would classify all of the companies you list as M2M companies, not IoT companies. I’d also class OMS and KNX as M2M standards. Lots of companies play a game of getting their proprietary standard released as an IEC or IEEE or ANSI standard, putting together a special interest group and claiming it’s a global standard. It isn’t. It’s just a proprietary standard with good PR and in most cases is what I’ve described as a Zombie standard, getting in the way of volume sales and delaying the arrival of the real Internet of Things.

Nor does it need to have a Z to be a Zombie standard. That was a joke, born out of an amusing conincidence. KNX can apply to join the ranks of Zombies with neither Z nor E.

#7 Harald Naumann on 09.27.13 at 8:23 pm

Since 25 years I am forwarding data wireless from point A to point B. We had no name for this task. Nobody named it M2M or IoT.
W-M-Bus and OMS are in use. ZigBee and Z-Wave are not important. If you want to enter the German / EU market then you have no option – you have to integrate standards like W-M-Bus and OMS. Just count the companies and the meters they already control. Be aware there there are EU security levels behind. All this are Zombies?
You also have to accept Enocean and KNX. Sorry, but the world is not IP only.

#8 Nick on 09.28.13 at 7:13 am

I think what you are saying proves my point. There will always be an M2M market, but it’s not the same as the Internet of Things. Traditionally M2M solutions have often been local and will remain so. And when the IoT arrives, they will become less attractive.

I’ve never believed in IP to every device – it’s not consistent with low power, but it is part of the equation. The important point about Zombie standards, and I include KNX and EnOcean in that category is that by pushing short term opportunities, they delay the bigger one.

#9 Zia on 10.04.13 at 3:41 pm

Nick,

For home automation, how do we solve the range issues with Bluetooth Smart if the Smart Ready internet gateway is on the home router (not smartphone)?

(Yes, a small niche yet, but Z-wave and Zigbee may have an edge here?)

#10 Nick on 10.05.13 at 8:26 am

There is currently work going on within the Bluetooth SIG to address the issue of range. For many homes, this isn’t a problem. I’ve been using Bluetooth for personal projects in my home for many years and it covers all three floors of the building. Most portable credit card terminals (Chip & PIN) use a Bluetooth connection to an access point, as the industry has found it is the most robust wireless standard. Having said which, you do need to concentrate on doing a decent RF design for the product. Many issues of Bluetooth range have been down to poor RF design within a product, not the chips or the spec.

#11 Zia on 10.05.13 at 12:29 pm

Nick, thank you for your reply.

May I ask if you are referring to “Bluetooth Low Energy” for credit card terminals? I do not see any qualified credit card terminal listed at: http://www.bluetooth.com/Pages/Bluetooth-Smart-Devices-List.aspx. Paypal Here (https://www.paypal.com/uk/webapps/mpp/how-to-use-paypal-here#overview) mentions Bluetooth but not “Low Energy”. In any case, Paypal Here is likely to be used in close proximity to a smartphone.

It would be interesting to read about your home projects using Bluetooth Low Energy (blog post idea? if there is already one on this site, I will find it :) Also, releasing these projects (with kit info for purchase) may be nice supplemental material for your book readers.

[My primary interest is in low power sensing / monitoring devices, which imply BT LE usage … just reiterating for emphasis :)]

#12 Harald Naumann on 10.05.13 at 6:05 pm

Q: For home automation, how do we solve the range issues with Bluetooth Smart if the Smart Ready internet gateway is on the home router (not smartphone)?
A: Bluetooth Low Energy offers a simple star network only. ZigBee offers a meshed network. Anyhow, any embedded PC on Linux or Android can run as a gateway. Behind my TV is a beta version of a Google TV box with an app for home automation. If I upgrade it with a Bluetooth Low Energy USB dongle, then living room and terrace will be covered. The corridor will be covered as well. The kitchen is maybe already to far away.
The second floor, the third floor and the basement are not covered for sure. I need a meshed technology or a technology that offers hops like ANT. With ANT I already made tests to cover all floors.

#13 Nick on 10.05.13 at 6:59 pm

You’re right – Credit card readers are classic Bluetooth, but low energy has slightly better range, due to the higher modulation index, so I felt the comparison was valid.

I’ve been meaning to write up my home projects – one is control for a garden watering system, but haven’t found time. I’m currently planning to update them with Bluetooth low energy and I’ve promised myself to write those up as I go along. So keep watching.

#14 Zia on 10.05.13 at 7:42 pm

Harald, Nick: thank you for your responses.

Let me make it a little more concrete. My mom lives with my sister in the suburbs (US east coast; 2500 sq feet covered area; 1 acre lot)

Wired internet and computers are at one end of the house. Even Wi-Fi signal can be very weak in the bedrooms at the other end of the house.

Assume that the internet router is Smart Ready. If there is health monitoring device w/ BLE, what is the best way to ensure internet connectivity? For most of the day, there are no smartphones at home. Harald’s idea for using a PC / laptop is neat, but assume that is not available either.

Perhaps if there was a low-cost seamless “range extender”, BLE over Smart Ready router may become viable.

On a related note: for BLE health devices, a Smart Ready router may be a good network redundancy provider when smartphone is not available.

#15 Harald Naumann on 10.05.13 at 8:54 pm

Zia my idea for my parents is public here:
Home appliance as a wireless hub / gateway
http://www.gsm-modem.de/M2M/m2m-faq/home-appliance-as-a-wireless-hub-gateway/
Right now, I am not sure which technology I will use. With Bluetooth LE I can cover floors, but not the whole house.

#16 Nick on 10.08.13 at 12:59 pm

I wonder whether we are behaving too much like technology geeks? For most people the first question is what is the compelling application, not the range. Unless we can get people really interested, range may be academic.

#17 Harald Naumann on 10.08.13 at 5:41 pm

Nick, I do not know how many BLE proximity beacons you own. I own 3 and get number 4 soon.
BLE on alone is useless for AAL. BLE offers no tree of meshed topology and BLE does not cover my small house with 120 square meters on 3 floors plus basement (further 50 square meters). The range is much too small.
On 868/915 MHz I can cover the house and garden on a star topology. On same band we get a licence free technology called 6LoWPAN. 6LoWPAN we get on 868/915 and 2400 MHz plus Power Line Communication.
I do not understand why you hype BLE previously called Bluetooth Ultra Low Power, and in early 2001 as named WiBree. “Wi” for wireless, and “Bree” for Bree in the Lord of the Rings. Bree was the town where the two major roads, the Great East Road and Greenway, crossed.
From the idea to come real we needed more than 10 years.
6LoWPAN is moving much faster, covers more bearers and offers a bigger range. Bigger range means less cost for hardware.
Nick, I am neutral. I have access to all mayor wireless technologies. Nevertheless my latest projects in ISM are based on SubGHz. I gave my customers the freedom on all wireless technologies. Why nobody of them selected BLE?

#18 Zia on 10.08.13 at 9:20 pm

Nick, thanks for your reply. Home applications that I have in mind are the usual suspects – security sensors, lighting / temperature control, health monitoring etc.

Granted that these aren’t big markets but I think that was the point of your post (Bluetooth ignores them and we have the various “Z” tech filling the void)

IMO, this is not academic … range issues will have to be solved for home applications to come out (this isn’t chicken and egg – range comes first IMO)

#19 Nick on 10.08.13 at 10:21 pm

I think we’ll see innovation first. Have a look at the Canary campaign on Indiegogo. They’ve just raised almost $1,959,906 for a Wi-Fi based security system, not because of range or technical features, but because it’s disruptive.

In contrast, eSensor’s campaign for a Bluetooth low energy gateway, which was purely technical, only raised $133.

It’s not necessarily the technology – it’s how you tell the story.

#20 Nick on 10.09.13 at 7:58 am

Why do I support BLE? Because I think it will become a dominant standard. I was asked the same question when I started supporting Bluetooth in 2000. Everyone told me I was mad because HomeRF was the better standard. Today HomeRF is dead and two billion Bluetooth chips are sold each year.

The technology is only one factor in making a standard successful. Applications and interoperability are arguably more important. As the old proverb says “if you don’t use it, you lose it”. So getting products bought and used by customers is key. That’s not just true in wireless standards. If consumers bought purely on technical merit, the iPhone would have been a commercial failure and we’d all have gone down the Betamax route.

BLE is not perfect. But if it starts selling hundreds of millions of chips and getting apps developers excited, then money pours into the ecosystem. That means more engineers will have the resources to extend the specification with longer range, higher speed and mesh. Success breeds success. We’ve seen Bluetooth and Wi-Fi succeed with that virtuous circle, whilst other standards have struggled and died. My prediction is that BLE will do the same. Let’s make a date to come back in ten years to see who is right? I’d be happy to wager a case of good German wine on the outcome.

#21 Harald Naumann on 10.09.13 at 9:09 am

BLE is perfect. It is perfect for a lot of solutions close to your body, around your smartphone and for something with small to medium range. I already have more than 100 requests for locating on BLE or ANT.
To cover my house and garden the 2400 MHz technologies are wrong. Even on free space 2400 MHz has a higher loss. This is just physics. On 868 MHz a simple star network will cover our whole house. By using of 6LoWPAN I can select on wireless or wired protocol. 6LoWPAN is already on power line communication.
Since more than 25 years I am in wireless applications. Today I am the FAE for the third biggest distributor in the world. Nick, I will ask to get the power line communication on 6LoWPAN in portfolio as well. I can´t split technologies in wireless and wired anymore. My business card is already renamed from FAE Wireless to FAE Connectivity.
I assume that we will make big steps next 5 years. We have not to wait 10 years. Anyhow, nobody will know 100% what will happen.
Just to let you know that I will stay from 10th. of Nov to 15th. of Nov in London. Is this a good time to drink a glass of wine together?

#22 Zia on 10.09.13 at 1:28 pm

Nick, thanks again. I appreciate your point about applications and interoperability.

Let me try to clarify my question(s) one more time: take the perspective of a product manager (PM), who has defined applications to develop. In choosing among competing technologies to undergird these applications, PM must choose on “technical” dimensions (in addition to scale, interoperability, etc).

I have looked through the Bluetooth Shanghai conference material. The refrain was to build applications using Bluetooth Smart. If Bluetooth Smart cannot cover a suburban home, I fail to see why it can be used for this application category. Then Z-wave, Zigbee, Insteon etc should survive purely because of range.

From a user perspective: If I have to go to the far corner of the house with my mobile phone in hand to get in range … and check if the window is open, I’d rather do it manually (thanks but no thanks to home automation tech).

Few polemical points (I am old enough to remember the crazy 1990s):

– Story is important but you can always over promise and under deliver
– Indiegogo may have a selection problem – consumers see a use for Canary and hence the pre-orders … enterprises / firms / product developers likely don’t go on Indiegogo to pre-order
– (somewhat redundantly) investment / pre-order is not a convincing argument for me … (again ref to 1990 tech bubble)

[I feel sympathetic to Harald’s points on sub-GHz wireless and power line tech b/c of range]

#23 Nick on 10.09.13 at 2:03 pm

Let’s start with range. 2.4GHz won’t work in every house, but it will work in most. In the UK there’s been a lot of work done looking at the in-home range of smart meters. The resulting assumption is it will work in over 70% of homes. That’s very conservative, as the 30% where it might not work include tower blocks, meters in cellars and meters buried underground in the garden. That is for UK homes. It will vary elsewhere. In the US, where there are fewer thick, brick walls, the success rates is reckoned to be above 90%. If the house is bigger than that, it is probably a more valuable property with more affluent owners who will have a higher end system fitted.

Sub GHz will be better. As Harald says, that’s physics. But whereas a 2.4GHz product can ship globally, sub-GHz ones can’t as there’s no global harmonisation. So manufacturers need to make and certify multiple different products for different territories. An added problem is that the amount of sub-GHz spectrum is limited, so the access rules are far more draconian than they are at 2.4GHz, which makes some applications difficult. And the longer range can result in greater interference. We’re starting to see that in some UK smart metering trials.

Even with a sub-GHz radio, it’s good to have frequency hopping. A number of reports have now confirmed the issues of single channel radios in real homes. That’s equally true at sub-GHz. However, because of the spectrum access issues, there’s not a lot of tried and tested frequency hopping options for these on the market. And no interoperable ones that I’m aware of. (Let me be clear here – if the standard does not have an independent certification scheme it does not meet my definition of being interoperable.)

If you want to connect the device directly to a phone or tablet, then you’re stuck with 2.4GHz, as that’s the only short range radio they use. You say “with my phone in hand…” If that’s part of your solution, you’ve committed yourself to Bluetooth or Wi-Fi.

Some routers support sub-GHz, but starting in 2014, it looks as if Wi-Fi and BLE will become the standard combination for domestic routers. If you don’t take advantage of these, you need to provide your own router or bridge, which adds cost to the overall solution.

In time BLE range will increase – there is work going on at the moment, but it’s a few years away. There are a lot of chip companies making BLE chips now and they all want to be able to own these smart home applications, so you have a lot of vested interest in making it fit this market. I’ve seen demos that give over 1km open field range.

If you have to design a product now, it’s a hard choice. I’d recommend making sure you fully understand your target market segment. If you want to sell to mansions, BLE is probably the wrong choice. If you’re targeting apartments, it’s a good choice. In an ideal world we’d have had a harmonised sub-GHz spectrum with plenty of bandwidth, so that we never needed to use the 2.4GHz spectrum. But real life’s not that easy. The industry only got access to 2.4GHz because regulators thought it was difficult to use, no-one else wanted it and they didn’t see any prospect of charging for it. The industry’s not made a bad job of proving the regulators wrong.

#24 Zia on 10.09.13 at 3:24 pm

Nick, thanks for your reply. To clarify, my focus here is to screen a single tech (Bluetooth v4.0) for a single app category (Smart Home). I am not committed rather screening Bluetooth v4.0. [yes, I meant BLE “phone in hand” as I don’t see anything WiFi based without main powers]

I should also add that it would be better to focus the discussion on BLE rather than any 2.4 GHz. The “zombie” techs in 2.4 GHz may use more power and have meshing to increase their range. Does BLE have meshing in the works? Will higher range BLE increase power consumption?

As your book reader, may I request if you could give any references to the work being done on BLE range? I have been searching around but not finding much that is useful. Specifically:
– 70% of UK homes (is this BLE or some other 2.4 GHz tech)
– BLE in home routhers?
– 1 km line-of-sight range

Finally, as WiFi moves to 5 GHz (11n, 11ac) over next few years, do you see frequency hopping in 2.4 GHz to remain such a big deal? I am assuming (without ready evidence) that most of 2.4 GHz ISM spectrum usage is due to WiFi.

#25 Zia on 10.09.13 at 5:07 pm

A quick question: is a video recording of your “Smart Home” presentation on Bluetooth World (or similar content elsewhere) available publicly? Generally, the entire content is relevant to me but in particular, the following two slides:
*** Owning the Gateway
*** Who will Provide the Gateway

{Even an audio podcast (voice over for slides) would do …}

[I have searched on videos.google.com and seen some of your other presentations with overlapping content … YouTube channel “BluetoothTech” has mostly Robin Heydon’s presentations from Bluetooth World]

#26 Zia on 10.10.13 at 7:19 pm

Interesting launch: http://www.engadget.com/2013/10/10/dropcam-pro/?ncid=rss_truncated

Bluetooth Low Energy (or whatever the purist will call) is incorporated into Dropcam Pro so that sensors could use Dropcam Pro as an internet hub / gateway.

Nest seems to have built-in Zigbee but kept it disabled but may eventually enable it.

#27 Nick on 10.11.13 at 1:08 pm

I don’t believe that the recordings are generally available, as the Day 2 presentations contained some confidential information about Bluetooth strategy.

However, I am planning to write a couple of articles about Gateways over the next month, so keep an eye on the blog.

Leave a Comment

About Creative Connectivity

Creative Connectivity is Nick Hunn's blog on aspects and applications of wireless connectivity. Having worked with wireless for over twenty years I've seen the best and worst of it and despair at how little of its potential is exploited.

I hope that's about to change, as the demands of healthcare, energy and transport apply pressure to use wireless more intelligently for consumer health devices, smart metering and telematics. These are my views on the subject - please let me know yours.

You can Subscribe via RSS »