White Paper: Control Networks and the Internet

lts for ""
Below is a cache of http://www.echelon.com/solutions/opensystems/papers/Control_Internet.pdf. It's a snapshot of the page taken as our search engine crawled the Web.
The web site itself may have changed. You can check the current page or check for previous versions at the Internet Archive. Yahoo! is not affiliated with the authors of this page or responsible for its content.
White Paper: Control Networks and the Internet Control Networks and the Internet
Revision 2.0
Reza S. Raji
reza@echelon.com
http://www.echelon.com
 Control Networks and the Internet
Rev 2.0
2
Introduction
There has been quite a lot of talk about the emergence of the Internet and its applicability to
nearly everything around us. Indeed the power of the Internet is not really in its sheer technical
innovation but in the way it cuts across traditional communication boundaries and permits new
possibilities for the average non-technical user.
The technology for the Internet has been around for over twenty-five years. What then has
prompted its explosive expansion in the last few years? The answer to this question is
important in understanding where the Internet is going and what things it will affect on its path
in the next few years. Many believe that the Internets potential has only been minimally
tapped so far. The true impact of the Internet and the many killer apps it will spawn are yet
to come.
The Internet and the World Wide Web have in essence allowed people to easily and
economically reach other people. Electronic mail, arguably the most popular application of the
Internet, has become an indispensable tool for many businesses and families. The number of
new businesses and individuals connected to the Internet is growing at an astonishing rate. In
short, people can now get in touch and stay in touch with others more easily than ever before,
thanks to the Internet.
The concept of the Internet appliance has been gaining a lot of press lately. The idea being that
small, task-specific, appliances could be allowed to interface to the Internet and perform very
specific functions (e-mail appliance, kitchen appliance, etc.). This, in essence, is the same
concept as accessing the Internet from the PC- connecting people to people - but at a much
lower hardware scale, and thus lower end-user cost. The Network Computer (NC) is another
idea with similar, yet more generic, applicability.
What if the power of the Internet could be extended beyond just connecting people? What if
the Internet was allowed to go beyond connecting desktops and laptops (or kitchen counter
tops) and could somehow be tied to the devices around us? The devices we come in contact
with everyday, lights, switches, thermostats, TVs, air-conditioners, security systems, fire
alarms, can all be thought of as Internet clients.
The idea proposed above can be implemented in a brute force way by simply assigning unique
IP addresses to all of these intelligent devices and placing a tiny web server in each one. Any
browser on the web could then monitor and control parameters on the embedded devices. In
reality, however, this idea has many practical limits. The Internets technology was not
designed to link small, embedded devices together. Small, low cost, control devices have their
own unique set of characteristics, which are very different from those of the computer world.
There is a need for a different kind of networking architecture at the device (or appliance)
level.
That is where control networks come in to the picture. Control networks link intelligent
devices together to permit a distributed and live network to reside in a home, building, or
factory by providing a low-cost, reliable, and flexible networking platform optimized for needs
of control. The network is the computer , the slogan popularized by Sun Microsystems a
few years ago can now be adapted to this new paradigm. The network is the control system.
It would only seem natural to extend the networking paradigm into control devices by allowing
the different networks to join and form a homogenous networking fabric. In the same way that Control Networks and the Internet
Rev 2.0
3
the intranets became an extension of the Internet, the local operating control networks, or
Infranets, could be linked to the Internet and intranets where information (data and control)
could flow from anywhere to anywhere, from anybody to anything. People could now reach
things as well as other people. Figure 1 shows the completed information infrastructure.
Internet
Intranet
Infranet
Figure 1. The Completed Information Infrastructure
The idea that everyday devices could be linked to the Internet certainly seems logical. We
have appliances and gadgets all around us, and given the current growth rate of the number of
Internet users, marrying those devices to the Internet would seem like a natural fit. Well, it is,
but a few hurdles have slowed down the fusion of the two concepts in the past.
The Internet and intranets finally began to enjoy wide spread acceptance and growth when
there was an easy-to-use universal standard for using them. The World Wide Web and the
ubiquitous web browser provided the impetus for the average person to bypass the technical
obstacles and start taking advantage of the Internets benefits. In turn, its growth resulted in
more content (more web sites), which further promoted its use.
In contrast, control networks are relatively new. More importantly, the need for
standardization in the way everyday devices and appliances communicate has never been more
prevalent. The World Wide Web standard needed the TCP/IP protocol, the HTTP protocol, and
the HTML markup language, all working in perfect harmony, to blossom. The control world is
establishing similar standards, protocols, and technologies so that it can fully take off and join
forces with the Internet.
This paper will attempt to bridge the gap between the Internet and the control world by
quantifying the requirements for their successful marriage. It is hoped that this paper will also
provide much-needed clarification of the confusing messages that have been hyped in the press
in the recent past.
The Embedded World
Embedded systems have traditionally been used in applications such as discrete and process
control automation in factories where a central source of computing power is required to run a
process or operation. In most cases the computing requirements, such as the graphical user
interface (GUI) or real-time performance of the system, are specific enough to rule out the use
of a generic desktop PC.
Despite
the relative lack of press coverage, the embedded systems world has been quietly
growing in the shadows of the PC world for quite some time. Major chip manufacturers have Control Networks and the Internet
Rev 2.0
4
devoted entire product families to microcontroller chips, the central processing core of
embedded systems. The growth of the microcontroller market has been significantly faster
than that of the traditional microprocessor - the central processing engine of personal
computers and its projected volume easily dwarfs that of the microprocessor. From a
business point of view, applications for embedded systems usually fall in one of two
categories: commercial or consumer.
Commercial embedded systems are used in such applications as building automation, process
control, and factory automation. Traditionally, such systems provide a central processing
station, which is wired to a multitude of physical sensors and actuators (e.g. temperature,
humidity, motion, etc.) throughout the facility. The embedded market for such commercial
applications has become fairly mature in the past decade or two. A wide range of
microcontrollers covering different computing needs (8, 16 and 32-bit) and offering different
features (choice of memory, I/O, timers, etc.) is available to system manufacturers. The
complexity of such systems, together with the reliability and real-time operational
requirements has also created a market for embedded operating systems. Real-time operating
systems (RTOS) are available from many manufacturers.
More recently, the consumer market has been taking advantage of the capabilities of embedded
systems. The surge in consumer devices has dramatically lowered the cost and size of the low-
end embedded system. Low to medium power (4, 8, and 16 bit) microcontrollers have quietly
found their way into microwave ovens, thermostats, stereos, and many other consumer devices.
Unlike the commercial embedded market, the software development process for these
consumer devices is not as formalized. The concept of the operating system does not really
exist for consumer embedded applications due to the insatiable consumer appetite for smaller
and cheaper products.
From a technical point of view, embedded applications can be categorized into two groups:
standalone or distributed.
Standalone embedded applications are those which are generally contained within a relatively
small self-contained enclosure. The application does not intelligently communicate with