Mar 24, 2014

IoT Update: I Give Up

posted by Bryon Moyer

Last year I proposed an overall architecture for the Internet of Things (IoT). The goal was to clarify the many different pieces required to make this work. And, in particular, to clarify which companies do which parts of the IoT.

There are so many companies that say the “enable the IoT.” But what does that mean? Last year, it could have meant many things, and so I tried to make some sense out of it. My intent was to come back and revise and refresh that effort.

That’s what I started to do recently – until throwing my hands up in dismay. There are so many companies claiming to participate in this business, and there’s typically not enough information available to place them properly in the various categories I set up. I have updated the table below, but only to the point where I surrendered.

You could argue that, as a journalist, I should be digging into each and every one of these companies to ferret out the truth. Up to a point, I agree; that’s what I did before. But after a while, I realized that I was turning into an industry analyst.

In reality, it would keep me from doing anything else for a while. Truly fleshing things out now would be something of a full-time job for a while.

Meanwhile, the number and range of companies tying their pitches to the IoT has ballooned. I could probably tie sneakers to it… let’s see… the first commercial application of a special new rubber in the soles, the volume sales of which will provide the revenues necessary to research new elastomers in home widgets that can be connected to the IoT! Boom! “New Footwear Supports the IoT”

<sigh>

So I’m going to keep watching for and covering interesting IoT technology and companies doing new, unique things that can clearly demonstrate a substantial IoT connection. (Like today’s M2M discussion of DDS.) But for the moment, characterizing all the companies claiming an IoT connection feels a tad too quixotic. I hate embarking on something and then backing off… but… there you have it.

Figure.png

Tags :    0 comments  
Mar 20, 2014

More Common-Process MEMS

posted by Bryon Moyer

Last year we took a look at a couple of proposals for universal processes from Teledyne/DALSA and CEA-Leti that could be used to make many different MEMS elements, trying to move past the “one product, one process” trap. We’ve also reported on the AMFitzgerald/Silex modular approach and their first device.

Well, the first design using CEA-Leti’s M&NEMS process has rolled out: a single MEMS chip with three accelerometers and three gyroscopes designed and implemented by Tronics. They’re not quite the smallest of the 6-DOF sensors, but they claim that, with more optimization, they will be. Right now their die size is 4 mm2. And they say that all main parameters are on track with their simulation models.

But this is just the first functional version; they’re going back to work some more while, at the same time, giving it a companion ASIC, releasing them at the end of this year.

They’re also using the same process to create a 9-DOF sensor set, with all of the sensors on a single MEMS chip. Also for release at the end of the year. And, the idea is, that, if they wanted to, they could also include a pressure sensor and a microphone, since they can all presumably be made on this same process. Yeah, you might wonder whether integrating a microphone with those other sensors has value; even if it doesn’t, being able to make it separately using the same process as the n-DOF chip still brings a huge level of manufacturing simplification.

These efforts, if successful, could represent a fresh breath of efficiency for some of the oldest sensors in the MEMS world. The industry also has new MEMS elements in the works, like gas sensors and such. If a standard process like this could be used for new sensors as well, then at some point new sensors could launch on standard processes rather than having to do the “one process” thing first like accelerometers and their ilk have done.

There are those who believe that these standard processes are too restrictive to allow the design of sensors with arbitrary characteristics. We’ll continue to keep an eye on this stuff to see whether these common-process skeptics can eventually be appeased or whether they’ll be proven correct.

Check out the details in Tronics’s release.

Tags :    0 comments  
Mar 20, 2014

Model-based Product Line Engineering

posted by Dick Selwood

OK –not the most attention grabbing headline, but it is what Atego are calling the total revamp of their modelling tools. System modelling is one of those techniques which have been around for a long time, but outside a group of high-end companies, mainly in aerospace and automotive, has never really taken off. Atego’ is hoping to change this. Atego was formed by the merger of Artisan and Aonix, both system development tools suppliers. Since the merger the company has made a number of acquisitions of specialist tools. Today they are launching Vantage, which is not just a relabeling of their tools, but a complete rework to provide a portfolio for the complete lifecycle of complex products, such as cars or trains.

There are three threads to the approach Model-based Systems and Software Engineering (MBSE), Asset-based Modular Design (SoS/CBD/SOA) and variable Product Line Engineering (PLE). Atego claims that the combination of these three proven approaches into Model–based Product Line Engineering (MB-PLE) can reduce development costs by 62% and bring 23% more projects in on time.

This is an implementation of ISO 26550-2013 ‘Software and systems engineering - Reference model for product line engineering and management’ and ISO 15288 ‘Systems and Software Engineering – System Lifecycle Processes’ standards.

The detail of each of these three threads is a bit of an alphabet soup. But simplifying to the point of caricature, what modelling now provides is better reuse, a system that provides for a range of modules within the model, each of which may be one of a number of variants. One set of variants might be a cars different engine options, another set  the gearbox options. The power train model is a combination of engine and gearbox, but certain combinations will be illegal. The modelling also provides a requirements capability and the ability to match the requirements of specific standards, in automotive again this is ISO 26262. 

This is not modelling for the masses, but for people working on complex systems, with variants, and with a long product life, should certainly look closely at this approach. Everyone should take time to look at the new video on the Atego web site http://www.atego.com/downloads/videos/introducing-ategos-mb-ple.mp4

Tags :    0 comments  
Get this feed  

Login Required

In order to view this resource, you must log in to our site. Please sign in now.

If you don't already have an acount with us, registering is free and quick. Register now.

Sign In    Register