DASH7 Wireless Sensor Networking Blog

interesting piece from Network World that basically says zigbee is a no-go on the mobile handset and that LTE and WiFi will dominate, and perhaps NFC, too.  I’d basically agree, with the caveat that DASH7 co-occupies the NFC space as I’ve argued here before.  I don’t agree, as you’ll see in my comment, with the idea that WiFi is a panacea for everything and can magically become “low power”, nor do I think WiFi is going to displace Bluetooth anytime soon.   Note that WiFi adoption on smartphones in China is lagging … so it’s no slam dunk in the world’s largest market.   I look at GPS more as a sensor than as a radio, btw.

So smartphones will have four radios in the end, in my view:

1.  LTE for high bandwidth, very long range voice and data comms

2. WiFi for high bandwidth, campus-area voice and data comms

3.  Bluetooth for cable replacement

4.  NFC/DASH7 for mobile payments, social networking, location based services, mobile advertising, sensor networking, and more.

If you’ve been paying attention to the buzz around Google Android recently, you’ve probably noted the tsunami of discussion around augmented reality or “AR”.  You can check out discussions on AR here and here, but in short, the idea is that using the GPS coordinates on your smartphone combined with the onboard compass (so your phone knows roughly which direction you are pointing at) can allow an application to overlay “meta data” atop an image you may be looking at through your phone’s camera lens.  Lots of applications, but the most cliche analog I’ve been hearing lately that unfortunately will probably stick is this:  think of the first Terminator movie when Ahhnuld gets a bunch of data about a guy he’s about to pummel in order to score the jacket he wears in the movie.Putting image recognition aside for a moment, the way we will acquire data about an indoor environments like this in the short term is via active RFID like DASH7.  I see the smartphone being the nexus for all this data acquisition, so the wireless transport we choose is imperative.  NFC – only works within 10 centimeters or less of the thing being tracked/read, not secure.   Passive RFID – kills the battery, won’t support sensors, short range, not secure.  WiFi, Bluetooth — also kill the battery, can’t track things that move, can’t penetrate difficult substances, short range.

DASH7, on the other hand, tracks moving things, uses just a fraction of the power of the next alternative, can talk through concrete walls and water, supports PKI, can co-exist nicely with both NFC and WiFi (other technologies at 2.45GHz will be drowned out by 802.11n — the wireless sensor networking “elephant in the room” that few are brave enough to acknowledge … except the brave souls here at the DASH7 Alliance!), and is very affordable both from a device and a TCO standpoint.

If you are contemplating smartphone-based AR investments, you’ll want to consider both indoor environments as well as moving objects and moving smartphones.  We think that DASH7 has a central role to play here … I’ll blog with some more app opportunities shortly.

BTW – James Cameron did Terminator about 20 years ago and I guess we are just about there with at least some of the innovations he and the creators of that incredible flick envisioned.  I saw “Avatar” and thought there was a bunch of tech that was cool but which was already being deployed in the military or that was not far off … perhaps Cameron has an eye for deploying tech in his movies which is not sheer fantasy … though the cloning/consciousness transplantation concept in Avatar is beyond 20 years either Scalzi fans or Ray Kurzweil devotees may disagree … discuss …

Between iPhone and Android, 2010 should bring multiple announcements relating to RFID integration with smartphones.     The conventional wisdom is that NFC will be the first RFID technology to see broad adoption, targeting mobile payments.  But NFC has a range limit of just a few centimeters  so I expect NFC to be a niche feature and a more fully-featured flavor of RFID will ultimately be what excites developers.  For them, the current RF options on a smartphone — GPRS, WiFi, Bluetooth, GPS — are simply non-starters for the vast majority of RFID applications.

Much of the discussion about the capabilities of RFID is vague and in some cases, incorrect.  One invaluable resource we can tap to avoid making poor long-term RFID platform decision is the past 10-20 years of RFID experience gained by enterprise customers like Walmart, Target, Metro, DoD, and others.  Billions of dollars and lots of hype later, RFID’s success in the enterprise is mixed.  Some customers, like the US Department of Defense, have reaped massive benefits, while others, like the legions of WalMart suppliers who were given an offer they couldn’t refuse around passive RFID, have mostly not been happy with the return on their RFID investments.

Many of the business reasons for RFID’s success or failure in the enterprise may not read directly on the applications for RFID and smartphones — how Walmart mandated that its suppliers tag pallets of pantyhose with passive RFID tags, for instance –  there are nonetheless some fundamental lessons about RFID that are directly applicable to how RFID will be applied to smartphones.

First, a quick reading of the history of automatic identification technology suggests that enterprise will indeed show consumer markets the way for RFID similar to the way enterprise drove adoption of other ID technologies like barcodes.  We’re printing around 10 trillion barcodes annually so there’s no debating the success of the technology.  Similar analogs can be drawn from adoption of fax, email, and mobile telephony itself.

Second, the billions of dollars deployed against RFID projects in the enterprise over the past decade or so provide us with a rich portfolio of “lessons learned” about RFID that are ready to be applied to smartphones.  My top five lessons learned as they apply to integrating RFID with smartphones are:

1.  Buying, installing, and maintaining fixed RFID “reader” infrastructure can be expensive. 

RFID “readers” — roughly similar in concept to a wifi access point or a cellular base station — acquire data from the RFID “tags” nearby.  RFID tags can be passive (one-way, no battery, short range) or active (two-way comms capability, has a battery, long range, supports public key encryption and sensors).  Historically, nearly all RFID readers were plugged into AC outlets and had ethernet cables for purposes of backhaul.  Smartphones, however, have an opportunity to turn this infrastructure paradigm on its head and virtually eliminate the need for an AC outlet or an ethernet cable. (See this post for more background).  Approaching RFID opportunities with this in mind should be at the forefront of any smartphone developer plans for RFID.

2.  Open loop networks require global standards and low cost devices.

Mobile phones are part of open loop networks, in effect.  Anyone with access to a phone can call someone with a mobile phone provided you have their number.  Ditto internet email, SMS messaging, etc.  All of these are based on global standards which allow someone in Japan to send an SMS message to someone in Brazil without the need for intermediary gateways to translate one protocol into another.  Closed loop networks, on the other hand, are typically proprietary and only accessible to “members” of the network.  Classified military communications networks, corporate VPN’s, and CompuServe are examples of closed loop networks.  Basically, anyone can participate in an open loop network whereas closed loop networks are accessible to only some of us.

Enterprise experience with both passive RFID (e.g. NFC, EPC Gen2) and active RFID (DASH7) standards provides the smartphone community with global RFID standards from which to choose.   For passive RFID, the market may have already spoken, as NFC deployments are well underway while operating with EPC Gen2 requires too much power for a smartphone battery to realistically handle.  Since mobile payments apps will fail if NFC read ranges increase (due to heightened security/privacy fears), NFC appears locked into its current short-range paradigm for the foreseeable future.

So eliminating EPC Gen2 from contention and isolating NFC to extremely short range apps, the only serious RFID alternative for smartphones is active RFID.  The global standard for active RFID is ISO 18000-7, aka “DASH7″, which provides the smartphone community with a robust two-way wireless platform for RFID applications that offers both long range (2 km+) and minimal power draw from the host batttery.  (DASH7 battery life is routinely 5-10 years (!) in the enterprise, a huge attraction for developers.)  Since DASH7 operates at 433 MHz, it can be deployed using the same 13.56 MHz silicon as NFC with minimal impact to the bill of materials.

Speaking of frequency, a corollary lesson to that of standards is the frequency itself.  There is a great deal happening in the world of radio spectrum allocation and it’s not uniform from country to country or region to regionn.  Some RFID technologies (and the distant cousins like ANT) selected 2.45 GHz as their “favorite” frequency because it’s a slice of spectrum that can accommodate high bandwidth apps like video, the spectrum does not require an FCC license, and 2.45GHz is globally available.  The only problem with using “junk” bands like 2.45 GHz is that you are not the only one using it — 2.45GHz is a victim of something economists refer to as the “tragedy of the commons”, which means if the government provides free and unfettered public access to a valuable asset, eventually self-interested individuals will overwhelm the asset the point of making the asset useless.  In pre-industrial revolution days, this resulted in the decimation of public lands by cattle owners who over-grazed on public lands, but today this results in the decimation of 2.45GHz by all sorts of applications including garage door openers, microwave ovens, WiFi, cordless phones, bluetooth, and more.  2.45GHz is full of interference and is really only safe for WiFi now.  433 MHz is the only globally available choice and besides, its signal propagation properties are simply terrific.

3.  Attempts to re-purpose wireless broadband technologies like WiFi and Bluetooth for RFID have mostly gone nowhere.  This will continue to be the case for smartphones.

Intuitively, we would all prefer to use “what we’ve got” rather than add yet another RF component to a smartphone.  WiFi and Bluetooth are already baked into many smartphones (except in China, where WiFi integration into smartphones is stalled) but for purposes of RFID applications, neither technology has ever been taken terribly seriously by the enterprise.  Why?  First, both technologies are power hogs.  If you have a Jabra bluetooth headset or have ever tried to maximize the hours you get out of laptop battery while surfing the net using wifi, you know just how quickly those batteries burn out.  Second, both wifi and bluetooth are session-based protocols, designed from the ground up to replace a wireline session — like a “session” of streamed YouTube videos or a session of streamed Led Zeppelin MP3’s.  This is of course great if you are streaming large files but if you are doing anything remotely similar to RFID – reporting the location and/or condition of an object, these technologies are overkill.  Not only do they drain your battery faster, but they won’t work if the “tagged” object is in motion or if the smartphone itself is moving — WiFi and Bluetooth are “high latency” technologies that fail when you ask them to connect to things that move.  RFID is about short, bursty messages that are often only a few bytes in length.  DASH7= squirt gun.  WiFi, Bluetooth = firehose.

Note:  the bluetooth folks are promoting a new flavor of bluetooth called “low energy bluetooth” which is still some years away from widespread adoption but regardless is still designed for high-bandwidth transactions at short ranges.  Lower power than today’s bluetooth for sure, but still using 2.45 GHz and seeking to replace headset and mouse cables.  Bluetooth for cable replacement – sure.  Bluetooth for RFID and sensor networks — no so much.

4.  Passive RFID tags are like a cheap hotel on a Friday night.

Passive RFID reminds me of the cheap “hotels” you stay in while backpacking around Europe.  You get a room for the night with a bed, but because the walls are so thin, your neighbors can hear everything you say and do and vice versa … and the sleep is often not restful.  With passive RFID, you get a cheap way of acquiring an ID wirelessly, but with the risk that with some modest effort any amateur hacker/voyeur could intercept any of your communications.

This is a principle challenge for NFC adoption today vis-a-vis mobile payments.   Even though an NFC-enabled phone needs to be within just a few centimeters of the NFC reader panel, end users still remain concerned that their banking information is being transported over public airwaves, reminiscent of the early days of Amazon.com when many were loathe to share credit card information online.  Given the short range of NFC, I think their privacy issue is ultimately solvable and I look forward to ditching my credit cards in favor of an NFC-enabled phone.

However, for longer range passive RFID — usually discussed in the context of EPC Gen2 tags — the privacy matter is more serious.   While enterprise customers have found it easy to rationalize this concern away, consumer markets can’t be waved off so easily.   In fact, in many of my conversations with analysts, end users, and others about the vast potential for the “internet of things”, privacy is one of the top questions I get, irrespective of income, education, industry expertise, or experience with cheap hotels.  The concerns about surveillance, particularly in the U.S., are strongly felt and the smartphone RFID innovators ignore this issue at their peril.

If you don’t know the encryption business as it relates to RFID, a one-way wireless technology like passive RFID can theoretically encrypt its ID’s using a “private” key approach.  However, enterprise RFID companies have learned the hard way that for customers who care about encryption, private key-based approaches (e.g. DES or Triple DES) are insufficient and unacceptable.  Especially in open loop networks.   If a bad guy gets his hands on your private key via your smartphone, your bank account could be toast.  For reasons I won’t get into here, your one-way passive RFID tag with private key encryption can still be “hacked”, as it were.  So in sum, if encryption and, by extension, privacy is as big a deal as I believe it already is — passive RFID is a non-starter.

For open loop environments like smartphones, two-way active RFID is the only serious RFID option where security and privacy is a concern.  Technologies like DASH7, for instance, support the exchange of keys under public key encryption.   Note:  enterprise requests for public-key encryption via RFID have been limited to-date and I don’t have much to share in terms of  reference implementations.  There are reasons for the slow uptake of public key encryption in the enterprise that I can get into in a different post.  Regardless, the implementation of RFID in open loop networks combined with the existing concerns from the privacy community will require us to take this issue more seriously in the near future if RFID is to be successfully integrated with smartphones.

5.  The enterprise is deploying sensors and smartphones will need to talk to them.

A huge driver of RFID and smartphone integration now and in the future will be the proliferation of very low cost MEMS sensing devices that are coupled with sub-1GHz radios like DASH7.  Thus a batteryless passive RFID tag cannot support transmitting sensor data (the sensor needs a battery, after all), so all passive RFID technologies are in effect eliminated from serious consideration for integration into mobile handsets as readers.  Mobile handsets can still function as passive RFID tags — as they do with NFC — but as readers, there are too many negatives to envision passive RFID reader capability getting serious traction on the handset.

I’ll write more in the future about the potential of smartphones to serve as the “go to” acquisition tool for wireless sensor networks everywhere, but hopefully if you are still reading this windy post, you agree that sensors will play a central role in the internet of things.  There will be so many different implementations of sensors — traffic sensors, mildew sensors in your building, power meters, radiation in your warehouse, moisture sensors in your lawn — that the vast majority of sensors will be  deployed not on smartphones themselves but as “tags” that a smartphone “reads”.  Enterprise customers have been deploying sensors with RFID for years (e.g. the US DoD) and with the current and coming cost reductions in both MEMS devices and DASH7 systems-on-a-chip, nearly every semiconductor company I speak with sees an obvious path for accelerated sensor/RFID adoption.

I only spent a single day at CES this year, but:

- 802.11n WiFi is clearly on the rise. This presents serious challenges to HAN technologies (802.15.4, ZigBee) deploying at 2.4GHz. An engineer from a well-known company in the smartgrid space expressed complete ignorance about the issue, which was either a headfake or something the industry needs to get real about and quickly.

- GPS is extremely mainstream now, though I saw virtually no mention of issues around accuracy or lack of indoor location capability.

- Lucent, in collaboration with Georgia Tech, had a terrific augmented reality demo that was extremely well-informed. Many academics out there have been working for 10 years or more on AR and they are finally getting their 15 minutes.

- Z-wave had what I thought was an impressive display of force. Though this is obviously no substitute for real market traction, the sheer number of devices deployed using Z-wave is considerable. It’s a proprietary standard, which limits its ultimate appeal, unfortunately.

- Tons of home automation companies, demonstrating the lack of maturity in the marketplace despite years of effort. I did not walk away from CES with the sense that a catalyzing force was afoot in the home automation space that was going to drive adoption. No one I spoke with felt that the utility companies represented a serious near-term opportunity to penetrate home automation, which was telling.

- We had dinner at Emeril “turn it up another notch” Lagasse’s “Lagasse Stadium” in the Palazzo. Next time I come to Vegas to bet on NCAA games, this is where I’ll hang out.

Now back to the Bay Area with the rest of the “anti-fun” crowd leaving Vegas on a Friday night …

2009 was an important year in the history of wireless sensor networks, but not for all the “usual” reasons.  2009 was less about flamboyant breakthroughs than about quiet but hyper-important infrastructure and investment decisions whose effects that will be felt for many years to come.   Here are my top ten events of 2009 that will most impact the adoption of wireless sensor networking in the months and years ahead as well as my (related) predictions for 2010.  Enjoy and feel free to comment!

1.      DoD RFID III DASH7 Award

In January 2009 DoD awarded a contract for $429 million in DASH7 infrastructure.  Significant because the US Dept of Defense operates the largest and most complex supply chain in the world and both allied defense forces as well as suppliers of many types, including CPG companies, will eventually be required to deploy DASH7 infrastructure.  Given the global footprint of DoD’s supplier network, this is the most profound development of year in terms of global adoption of a wireless sensor networking (WSN) technology based on a single global standard.  That DoD is standardizing internally and externally with  DASH7 is a major event that most outside the supply chain sector missed but will have ramifications for years to come in both defense and commercial sectors similar to DoD’s pivotal impact on barcode and fax adoption.

2010 prediction:  accelerated adoption of DASH7 by defense forces  in Korea, Japan, Taiwan, Singapore, and India.

2.    DoE SmartGrid Stimulus Awards

Federal funding for new investments in measurement and control infrastructure for the nation’s electricity grid will have downstream ramifications for hundreds of applications which may not even be energy related. While this year’s stimulus spending mostly subsidizes  large utilities who are using the technology for demand planning purposes, psychologically the “battle” for control of home automation networks began in earnest in 2009.  Large amounts of government money are enabling a wide array of companies, most of whom are unconnected to utilities, dip their toe into this space and offer a variety of solutions for home healthcare, home multimedia, environmental command and control, and much more.  Many wireless sensor networking protocols — DASH7, WiFi, Bluetooth, ZWave, Zigbee, and others — are taking unique approaches to this market and the smartest companies are digging in for a long ramp up as despite the sugar high of stimulus money, few see utility companies as the innovators that will ultimately drive this market.  2009 was the year of  important investments by component suppliers, device integrators, software vendors, systems integrators, utilities, telco’s, and many that will prove pivotal in raising awareness around opportunities in home/building automation.

2010 prediction:  focus on mobile handsets as the command and control center for home automation displaces set top box/desktop-centric visions.

3.    Formation of the DASH7 Alliance

A small event to some, but evolving the ISO 18000-7 standard from a defense-centric technology to one that is global and interoperable across geographies, industries and applications, will eventually mean access to a single global, interoperable standard for extremely low power wireless sensor data acquisition that is simply not attainable with competing technologies. With the catalyst of the defense sector driving initial adoption (see #1 above), 2009 was the year DASH7 entered the arena as a serious wireless sensor networking alternative and in the eyes of many, the technology that is hands-down superior to the next best alternative.  If WSN is important to your business, put the DASH7 Alliance on your to-do list for 2010.

2010 prediction:  DASH7 Alliance aligns more closely with complementary technologies like cellular, passive RFID, WiFi, and 2D barcodes.

4.    Chinese Political Leadership Embraces Wireless Sensor Networking

When Chinese premier Wen Jiabao pronounced the internet of things as a national imperative for China, it got surprisingly little play in North America, but in Asia it was widely advertised and for those who study China and its increasing sway over technology adoption, this was an important moment as it signaled that not only do government leaders there realize that WSN is critical to China’s future as a manufacturing power, but the internet of things will permeate many other industries where China is or hopes to be a global leader.  As the hub of the world’s supply chain, China’s acknowledgement of the importance of the IOT should strike everyone as pivotal.

2010 prediction:  New announcements from top three Chinese mobile carriers around environmental sensing businesses/pilots.

5.     Twitter:  Human sensor network.

The explosion of Twitter, a service that only a couple of years ago people openly ridiculed, demonstrates the power of a “human sensor network”.  The ability to add non-human sensors to Twitter and other social networking applications is a natural evolution of the current application that will present great opportunities and challenges, but would have been delayed were it not for the “teachable moment” that Twitter provided the wireless sensor networking community this year.

2010 prediction:  presentation of sensor data in 2D environments like Twitter continues to prove disappointing and sensor data over Twitter remains a novelty until 2011/2012.

6.    NTT DoCoMo environmental sensing announcement.

DoCoMo’s announcement of a pilot in Tokyo whereby environmental sensor data would be acquired by mobile phones was a watershed announcement in that it signals how wireless carriers will play a pivotal role in providing a data acquisition interface for unstructured environments like a city.  As the role of sensors and location markers becomes clearer through the growth of augmented reality applications, carriers will see the opportunity to drive new commerce
and advertising revenues as well as leverage GPRS/CDMA backhaul for transporting sensor data as well as for executing transactions resulting from access to sensor data.

2010 prediction:  Regulators in Japan increasingly recommend 433 MHz as the frequency in Japan for a range of logistics and sensor networking applications, to the exclusion of technologies operating at 2.45GHz, which present a range of interference and performance problems there and elsewhere.

7.    The Wikitude Browser and Augmented Reality

The Wikitude browser demonstrated on a basic level how we will integrate masses of sensor, location, and metadata from the physical world around us.  Demonstrated on the Apple iPhone, it relies on GPS (primarily) for deriving location, which limits the application to outdoor environments.  But future implementations will augment GPS data with location and sensor data from DASH7-enabled tags that can be updated by merchants or municipalities simply and cheaply, with minimal maintenance of the tag.

2010 prediction:  Several major CPG companies launch pilots using augmented reality with DASH7.  Augmented reality will become a huge theme in 2010.

8.    IBM’s SmarterPlanet advertising campaign

Regardless of your opinion of IBM, their Smarter Planet campaign this year was terrific.  Suuure it’s nice to have big marketing budgets, a cynic would say, but this campaign articulated opportunities in WSN that raised end user awareness and that will provide benefits to the entire industry.  Sam Palmisano has done a terrific job of aligning what seems like the entire company around this theme and many of us noticed.

2010 prediction:  IBM’s focus on Smart Water is particularly prescient … expect lots of announcements around a “Smart Water Grid” that augments the electrical one that made headlines in 2009.

9.    Wi-Fi Alliance ad-hoc networking announcement

This announcement signalled WiFi’s intent to capture more of the home automation market that is coveted by the Bluetooth/Zigbee/ZWave crowd.  Home automation is already shaping up to be a very heterogeneous landscape with many options, but this ultimately serves to create still further fragmentation, at least for the near future.  I don’t see the same downside for bluetooth that some see with this – I like my bluetooth headset and wireless keyboards – but clearly the WiFi folks fired a warning shot to the rest of the home automation crowd.

2010 prediction:  integration of high-bandwidth/high power WiFi connectivity @ 2.4GHz with low bandwidth/low power connectivity @ 433 MHz (DASH7).

10.      H1N1 “Swine Flu” Virus.

What does swine flu have to do with WSN, you say?  This virus scare/pandemic has illustrated for many just how interdependent we’ve all become and how our massively complex and high-volume commercial relationships with people from across the globe can not only benefit us economically, but also threaten us physically (and economically, if enough are stricken).  The need for better pandemic preparedness and cold chain management will be solved in part by the availability of a wireless sensor networking standard that can allow anyone to operate or interoperate with another network similarly to the way we roam from wifi hotspot to wifi hot spot.  Use of a low cost, reliable WSN technology that we can either re-use or dispose of at the end of the journey is a major advancement that will not only help us mitigate the imapct of pandemics like H1N1, but also — don’t laugh — make us more competitive.  Counter-intuitive, I know, but the smarter we make our supply chains, the more competitive we become.

2010 prediction:  major regulatory breakthrough in the way RF devices are allowed/disallowed on aircraft for purposes of cold chain/perishables management.

What do you think?  Comments are enabled below … have a great 2010 everyone!

We have been heads down lately for a number of upcoming announcements, hence the light blogging. If you are a prospective DASH7 developer, I think you will like what we have coming up.

A big topic of conversation in October in many corners was the WiFi Direct announcement that landed smack dab in the middle of the annual ZigBee Alliance meeting in Australia. Sometimes the guys who really do kick butt feel no need to be subtle in their signals, and this announcement was no showcase of subtlety. In the spirit of Halloween, here’s a scary forecast for the ZigBee folks: there will be three ways of connecting our physical environment to the internet in the near future: Wireline Ethernet, WiFi, and DASH7. Read more…