Smart Energy Profile 2.0 – a case of too much PAP?
- in Smart Energy
Just when you though Smart Energy was settling down, and it was going to be smart meters all round, the smart grid movement is getting its knickers in a twist. It’s not a problem about what needs to be done, but about the standards, what goes into the standards and who is responsible for them? In the past few weeks both NIST and the ZigBee Alliance have had some major tantrums, which raises questions about the speed and degree of technology push that is being forced on the industry.
There is no lack of agreement about the need to improve the grid and the way that we consume energy. Growing demand, political concern over the stability of supplies, climate change worries, new challenges in the form of electric vehicles and decades of underinvestment in generating capacity and the grid have persuaded Governments around the world to support and mandate investment in new “smart” technology from smart meters in homes to intelligence in the grid. The last time the world saw a similar level of stimulus was in the 1930s, during the great depression. So this really is likely to be a once in a lifetime event. The political will is there, the question is who decides how it is going to be done? Groups like NIST in the US are pushing hard to put things in place, but are groups like this too academic in their approach? Over the last year they’ve set up eighteen Priority Action Plans or PAPs to oversee development. (A potentially unfortunate acronym as my dictionary defines pap as “worthless or oversimplified ideas”). And according to a recent pronouncement they obviously don’t think the industry is doing enough to meet the challenge. But before we look at that, let me share a quote with you:
“I hate those guys. I hate those legislators and politicians – not because they restrict business and screw up the markets, even though they do and it does. I hate governments because I know those guys. I went to school with them. And let me tell you, the weakest, most ignorant, most drunken incompetents work for the US government. And the bottom of the barrel, know-nothing dicks design the regulations for a market they know nothing about. Why should we look at the regulations they’ve put in place by committee and go “Yeah, you suck at your jobs, fine, we’ll ignore that and suck at ours too?”
Not my words, but those of Lucy Prebble from her brilliant play “Enron“. It’s a diatribe that she gives to Jeffrey Skilling – Enron’s President, as his empire starts to crumble. Strangely, from a character that has little to commend himself throughout the rest of the play, it’s a dramatic moment where you suddenly start to feel sympathy with him, particularly if you’ve ever worked in a regulated industry. Of course, that speech is just fiction and has nothing to do with the current situation…
The current spat is about whether to mandate IP over the entirety of the smart grid. (IP as in Internet Protocol, not Intellectual Property, although the legislators ought to be a bit more aware of some of the implications of the latter than they appear to be.) IP is the foundation of the Internet as we know it today and is generally a good thing, as long as you aren’t limited by power consumption or spectral bandwidth. Or to put it another way, it’s fine as long as you’ve got a cable and a power supply. It’s an obvious choice for most of the grid infrastructure and connectivity down to each house, but when you get to individual, battery powered devices within the home it start to become less than ideal. I’ve written abut the reasons for this before, so won’t rehash them here. Except to reiterate the fact that we need more experts from the embedded world to get involved in this debate.
This is all getting acrimonious because smart meters which are currently being deployed don’t use IP. They mostly use ZigBee Smart Energy Profile 1.0, which uses an optimised protocol that was designed for low power wireless communications. And they work well. The only limitation is that they need a gateway device to convert the ZigBee protocol to IP when they want to communicate with the outside world over the Internet. But that’s not a problem, as they’re always going to need a gateway because ZigBee in any incarnation is only designed to have a range that’s largely limited to the home.
The IP lobby, which is almost religiously fundamentalist in their views, don’t like this. They want the smart metering world to convert to IP as potentially promised in the ZigBee Smart Energy Profile 2.0 as soon as possible. Preferably yesterday. And last week they formed a new PAP group – PAP18 – the nineteenth one, whose mandate is essentially to tell ZigBee what to do about making the transition from SEP1.0 to SEP2.0 painless.
The timing was interesting. A few days before there was a schism within the ZigBee Alliance, where it has been reported that the working groups failed to approve the decision to use IP as a transport within Smart Energy Profile 2.0. The deliberations of the working groups are confidential, but elements of that debate have been reported by EE Times and Smart Grid Today. In the light of this, the NIST decision appears to be an attempt to exert pressure on the ZigBee Alliance to conform to their crusade for IP.
The ZigBee Alliance, like other similar industry groups, develops its standards through a process of consensus. That bestows a couple of very valuable advantages: it means that the standards process is less influenced by vested interests and it also gives the standard time to develop through debate, both of which are a good way of getting a better standard. Each person eligible to vote will have their own reasons for accepting or rejecting a proposal, and those reasons are private. But I suspect that one of the reasons for the recent rejection has been a reaction to the level of external pressure which has been applied to the membership. That will be further informed by a different level of technical understanding of what is being asked, as within the ZigBee working group there are far more members who understand the implications and potential folly of forcing IP into small, battery power devices. ZigBee is not alone in this tackling this dilemma – other low power wireless groups are having the same debate over taking IP to resource constrained devices. There is not an easy answer, which is why this debate needs time, whatever wireless technology is being considered.
It is important not to be deflected by these current squabbles, and to understand that the industry does not need IP to connect to each individual device. We do not need to make this decision now. Smart meter manufacturers and utilities in regions that are leading deployment are working on a roadmap for the SEP 1.0 standard. SEP 1.1 is already being tested by multiple manufacturers prior to formal certification and working groups are adding new functionality into a future SEP 1.2. There is a valid roadmap for SEP 1.x that has no need of IP, as long as a gateway is installed in the home, which is what is happening.
There has been immense pressure to include IP. The effect of that pressure, largely from NIST within the US has been the recent setback in ZigBee Alliance voting. At best it will reduce confidence amongst those utilities embarking on their smart metering strategies. At worst it will add a year or more of delays while they wait to see what happens? For those utilities who have chosen to take the SEP 1.x route this isn’t a problem – they can just carry on with what they’re doing. However, those who believed the roadmap for SEP 2.0 and decided to wait for it may find themselves waiting a lot, lot longer.
I’ll stress again, whether or not we take IP to meters and IHDs within the home is irrelevant. It is perfectly possible to access them using the non-IP approach of ZigBee SEP 1.0. Which takes me back to those words penned by Lucy Prebble in “Enron”. If you get a chance to see the play, it is well worth it. If not, go out and buy a copy and read it. You’ll have plenty of reading time while you wait to find out what SEP 2.0 will be. And if you know someone in NIST, lend them your copy when you’ve finished.
1) You are right about smartphones and regarding the concern for battery-powered devices. However, your original assertion re. cables and power supplies vindicates the general impression that many people seem to have that wireless and IP simply don’t work. I don’t agree with that. The wavefront of IP extending into the Internet of Things is happening now and you will find IP on devices powered by batteries lasting for many years (you may wish to look at Prof. Adam Dunkels’ presentation here: http://www.iab.org/about/workshops/smartobjects/tutorial/Dunkels.pdf)
2) You may be right, but the draft originates from Nokia, one of the driving forces behind Bluetooth. I don’t think the IETF needs to proselytise. Nor does there really need to be a debate about correctness. The ubiquity of the Internet Protocol speaks for itself.
Let me try and answer these:
1) IP works over your phone, because you plug it in to charge every day – that’s not a low power embedded device. What worries the people who are working on low power, wireless devices is what effect IP has when you’re running off primary cells or energy harvesting. That’s the subject of a lot of debate amongst people designing these products, whether they’re ZigBee, Bluetooth, Wi-Fi or proprietary wireless. This isn’t academic – it’s about whether devices such as IHDs will have the battery life that comsumers want. The opinion of many of us who make these devices is that taking IP to the device solution will not allow us to meet consumer expectations at a practical price point. I beleive we should not be telling consumers that they need to charge these products every day, rather than designing protocols that support the maximum possible battery life.
2) I am not aware of anyone within the BTLE community who considers 6LoWPAN over BTLE as an acceptable proposal. The 6LoWPAN IETF group would like it to happen, but that’s very different from implying that the Bluetooth community does. The reason I deliberately chose religious terminology in this article is to reflect to proselytising approach that the IP camp takes. But just because they worship gourds or sandals and are vociferous about their chosen cause doesn’t mean that it’s correct.
3) As you say, I cannot breach any confidentiality. However, I’m sure you will agree that the level of internal technical debate that this has generated is very positive. It shows a standards group working through constructive dialogue, which is exactly how standards groups should work.
1) “It’s fine as long as you’ve got a cable and a power supply”. WiFi/802.11 on my smartphone? IP works fine over that now.
2) Interesting that there is a Internet Draft to put 6lowpan (in other words, IPv6 datagrams) over BTLE, which I believe is your favoured low power wireless protocol (http://tools.ietf.org/html/draft-ietf-6lowpan-btle-00).
3) I can’t go into detail for confidentiality reasons but your article is incorrect in many areas. However, I will point out that the statement “A few days before there was a schism within the ZigBee Alliance, where it has been reported that the working groups failed to approve the decision to use IP as a transport within Smart Energy Profile 2.0.” is fundamentally incorrect.