Choosing a Wireless Standard

The first question that most designers ask when adding wireless to their product is “which wireless standard to use?”  In some cases, where it is connecting to an existing product, that’s easy to answer.  If it’s not, it’s a lot more difficult.  It’s one reason I wrote a book about it – to try and help designers answer that question.  But another part of the same question is how well the different standards promote themselves as a solution? 

This year has seen some major changes within some of those wireless standards.  The ZigBee Alliance has lost Benno Ritter – for many years the global marketing face of ZigBee.  And the Bluetooth SIG has replaced its Chairman, Mike Foley, as well as its CTO, Andy Glass.  Both are interesting moves, as each of these standards is still evolving.  ZigBee is taking on smart lighting, home automation and smart metering, whilst Bluetooth is finally seeing Bluetooth Smart appearing in the mainstream.  In a recent issue of Incisor magazine, Vince Holton wrote about the loss of passion within the Bluetooth SIG – a sentiment that I’d echo and also extend to some of the other wireless standards.  But that’s an opinion formed from being close to these groups.  A few years ago I ran to survey to try and see what the general engineering opinion was of the different wireless standards. Prompted by Vince’s article, I thought it would be useful to run the survey again to see what designers think as we approach the end of 2012.

The last time I ran the survey I made the mistake of sending it to too many people within the wireless industry, which made the results rather skewed. This time around I tried to concentrate on responses from product designers.  Where I spotted an answer from someone involved in wireless standards or silicon design I’ve excluded it, so I hope this is a bit more representative of the actual level of knowledge.

To start with, I wanted to gauge how well designers think that Bluetooth, Wi-Fi, ZigBee and ANT evangelise their respective standard.  These four standards differ from traditional standards organisations in that they are driven by industry alliances rather than the more traditional standards bodies.  That has a couple of implications.  In general it means that they get their standards written faster, as well as getting products to market faster.  (That’s based on fact, not PR).  Because these groups are interested in customer experience as well as raw technology, they also have a greater emphasis on interoperability and certification regimes, albeit at different levels of competence.  And to help momentum, they have marketing departments that go round and attempt to promote their specific standard above all others.  

So how well do they do?  I know that a lot of engineers also get information from the silicon companies who sell chips, but I wanted to look at the performance of the standards groups themselves.  Not least because their competence in marketing affects the number of silicon vendors who support each standard.

Here the volume standards – Bluetooth and Wi-Fi were well ahead, with engineers feeling they had a good knowledge of them.   It was interesting that some designers felt they knew nothing about Bluetooth, whilst a fair percentage had similarly little exposure to ANT.  None had failed to hear of Wi-Fi or ZigBee, although they felt that ZigBee could do better in saying what they do.

The answer on whether they made enough information available showed them as much better matched.  ANT was the exception, falling a fair way behind, but given it’s the youngest, that’s not necessarily surprising. 

When asked how much respondents felt they knew about each standard, the main three fared well, with most feeling they had a thorough or sufficient knowledge.  This is where ANT suffered – only a small percentage felt they were familiar with it.  I suspect the real level of knowledge about the differences between them is somewhat lower, but I didn’t ask that question, so I can only surmise.

To try and see how recent that knowledge was I asked whether people were aware of the latest versions of each of these standards and how they thought that each differed from the previous version.  I chose Bluetooth Low Energy / Smart, Wi-Fi Direct and ZigBee RF4CE.  These are the same three I enquired about last time.  Whilst there has been a lot of underlying progress over the last two years, but I didn’t feel that there had been much information disseminated about more recent ZigBee profiles or Wi-Fi initiatives, so I decided to repeat the previous question.  There are still a lot of people who don’t know about Wi-Fi direct or RF4CE and a fair few who are unaware of Bluetooth Smart.  The good news is that the message about Bluetooth Smart being a distinct, new standard does appear to be getting across.  But from this I was not convinced that any of these three bodies have really explained the story about their new offerings.

The next question was to try and gauge whether designers thought that particular standards were more or less suited to their industry.  The surprise winner is ZigBee, followed by Bluetooth, with ANT drawing with Wi-Fi.  I’m not quite sure what to say about that, other than that it has little to do with the target markets which the various groups are attempting to win.  For reference, the respondents who listed an industry area were predominantly from Consumer Electronics, Industrial, Automation, Smart Energy and Medical.  Automotive and Defence were noticeably absent.  Around half of those who replied were from North America, and the other half split between Europe and Asia.

I then tried to see which standards people would be designing with, asking which were relevant to their particular industry.  At this point I included “classic” versions of Wi-Fi, Bluetooth and ZigBee to see whether that would have an effect.  It didn’t – almost all were equally liked.  The three laggards were ANT, RF4CE and classic Bluetooth (which ironically ships most chips).  There weren’t enough responses to try and break this down against each of the different industry areas, but looking at individual responses, the correlation was fairly random.

Finally I tried asking the old chestnut of where they went for information.  I don’t think this offered any surprises – the web is now dominant, with everything else decreasing until we reach exhibitions as the least useful source of information.  Sadly blogs don’t do much better, but I’m not giving up on the basis of one survey result.

What this says to me is that most designers consider the different wireless standards to be on a fairly level playing ground.  That ought to be a major wake-up call for Bluetooth Smart and Wi-Fi, who are still losing engineering mindset to ZigBee and ANT.  They’re also starting to come under fire from easier to use chipsets running in the sub-GHz band, which are poised to take away a lot of the potential low cost Internet of Devices and wireless sensor market (more of which in a future blog). 

Which brings me back to the Incisor article about the lack of passion.  In recent years I’d argue that the major wireless standards have become complacent and coasted.  Bluetooth allowed its high speed ambitions to be trumped by Wi-Fi and still doesn’t appear to grasp the importance of Bluetooth Smart as the bedrock of its future.  Wi-Fi in turn has become obsessed with the lure of go-faster stripes, when there is much more opportunity to embrace the IoT world with lower power.  And ZigBee is still desperately looking for a market that will result in it shipping in volume.  That means it’s falling into the trap of spinning ever more variants of its profiles rather than concentrating on robustness and interoperability.  But I also think that the ZigBee Alliance is beginning to accept the fact that their future may just be a set of IP application profiles which sit on top of Wi-Fi, rather than continuing to be a wireless standard

Bluetooth and Wi-Fi need to up their game.  Just because they’re in every laptop and phone does not guarantee that they will be used or have a future.  There is a new challenge coming from the sub-GHz chip vendors, who are transferring the skills they learnt with these standards at 2.4GHz into the lower frequency bands.  That’s where I see the passion in wireless today.  And that’s the story they’re pushing out to the start-ups in the Internet of Things space.  If you look on Kickstarter, or at IoT incubator events, most of the new products are running proprietary protocols at 868MHz or 910MHz.  They may not offer interoperability, and we’re certainly not going to see them integrated in phones, phablets or PCs in the near future.  But they’re easier to design with and cheaper to get to market.  Unless today’s standards organisations recognise the challenge and realise that the playing field has changed, they’ll grow old in their incumbent slots, whilst a new generation of designers ignore standard “brands” and simply use wireless.