Intelligence at the End of the Wire

by Michael Schwartz

In today’s world, computers are everywhere and in everything, as well as at the heart of our test equipment. If you would have told me 25 years ago that my oscilloscope would come with Windows® already installed, I would have called you crazy! But today it is hard to find that dividing line between test equipment and computer, especially in the high end instruments.

Now we all know that inside of every piece of digital test equipment there is a small embedded microcontroller responsible for converting analog to digital and back again. These microprocessors have traditionally been inexpensive, lower end chips, solely dedicated to the tasks required by the hardware. But as the world of test equipment continues to merge with computers, these embedded micro-controllers are becoming more powerful. As their computing power increase so does our test equipments’ ability to take on more complex measurements and tasks.

With the increase of computing power, test equipment becomes smarter—intelligent, if you will. And with intelligent test equipment at the end of the wire, we can now change how we think of measurements and instrumentation control.
No longer do we have to think of automation control in terms of sending low-level, bit wise configuration commands. We are now able to move into the next generation of automation control where our test equipment becomes part of a distributed computing system. Now our test equipment can be an intelligent team player in our local lab network all the way up to an enterprise-wide measurement system.

As we move into the next generations of automation control, the paradigm in how computers communicate with test equipment will change. More intelligent instruments, the ones with full blown operating systems like Windows®, will communicate and behave on the network more like a computing system than an instrument. Some of them already have HTTP web servers allowing users to interact with them using nothing more than a web browser. But this is just the start of computer to computer interaction with test equipment.

Screenshot of a Windows® embedded spectrum analyzer.
Screenshot of a Windows® embedded spectrum analyzer.

As we continue morph test equipment with computers, so too will we treat them as servers. Only instead or rendering up web pages, they will render our measurements. Many of them, I believe, will start to incorporate to Representational State Transfer (REST) calls as a standard means of communication.

For many of you, this may be the first time you have heard of REST communications. But it has been around for many years, and is at the heart of how servers on the internet communicate with each other. So it is inevitable that test equipment start to implement it in one form or another.

Like the evolution of HP-IB to GPIB to VISA to LXI, and all the other standards we have been through over the past 30 years, REST calls, as it relates to test equipment, will also evolve. But more system—like Metrology.NET™—will emerge, relying on REST communications as the back bone for their communications.

Its biggest advantage to REST is its scalability and flexibility. It allows engineers to design systems without relying on the specifics of the implementation. They don’t care if a server is running Windows® or UNIX; only that it complies with the message format. For metrology, it will allow us to design measurement systems with multiple implementations not based on specific commands to the hardware. REST communications, with intelligent test equipment, will allow us to define our measurement requirements and pass them to the measurement hardware in the REST call. Then the measurement hardware will perform the task as defined in the REST call and return the results. This will all function regardless of the specific hardware or version used to make the measurement.

We must embrace the fact our test equipment and computer systems are now one in the same; change how we think about the equipment we are trying to control and realize there is now more intelligence at the end of the wire. In doing, so we can focus on automation designs from a measurement definition approach and leave the specific implementation to the smart measurement systems.