26 Sep 2013
Challenges of Designing RF Test Equipment
Our editor Ian Poole talks to Jin Bains of National Instruments about the challenges facing the development of test instruments for the RF domain.
RF testing is an essential element in the development and manufacture of any system containing radio frequency or RF circuits.
With the rapid growth of radio technology in the past twenty years, RF testing is becoming far more important and the requirements for the test equipment are becoming more exacting.
This has been driven by the increase in use of cellular telecommunications as well as other wireless systems like Wi-Fi, Zigbee, NFC, WiMAX, and others. In addition to this there are very many other proprietary formats that all use RF technology and all need RF test.
All these systems need to be tested effectively and accurately. The requirements for many tests are much higher than they were several years ago because the performance requirements for these wireless and RF systems have all increased significantly.
This all makes the test equipment more challenging to design. It is necessary to provide equipment with a higher specification than the device being tested. Also the test equipment must be developed ahead of the need for the testing itself.
The main challenges
In order to understand what the main challenges are we talked to Jin Bains, Vice President of R&D, RF Products at National Instruments.
From the discussions it soon became clear that the main challenges appear to fit into a number of main categories:
Continually developing or evolving standards
One of the key points that Jin Bains made was regarding the various wireless and radio standards: “I think a lot of it comes down to the fact that there are so many emerging standards out there.”
These all need to be measured and they have many different requirements.
There are familiar radio communications standards like LTE, 802.11ac and many of the other new communications standards that are being deployed around the globe. Additionally there are many legacy standards that are still in use from 802.11n (and other earlier ones) to those like UMTS / WCDMA, GSM and others.
Apart from the industry standard systems, there are also very many proprietary ones, especially within the military and aerospace arena. Again these all have their own requirements for test.
Whatever they are, all these standards need to be measured in an efficient and flexible way. They have a wide variety of different requirements ranging from the frequencies used to modulation formats, power levels and many other aspects of the signals.
Said Bains: “ . . so we are looking at a fairly nimble adaptable test environment.”
With many emerging standards, any test system needs to be flexible. This is a key element, but not always easy to achieve. To enable a test instrument to fit many applications can mean it having far more capability in many areas than is needed for a given requirement and this can considerably increase the cost unless means are taken to ensure the system is flexible without adding cost.
As Bains commented: “The modular nature of PXI-based instruments enables the development of a flexible test system. From the NI standpoint we also use the SDR (software defined radio) model within test instruments. You can make a very flexible system like this and we have been using SDR in this way for a long time. It comes down to reconfigurability, primarily provided by the FPGA.”
PXIe Chassis with Modules
By enabling the FPGA within the system to be configurable, it is possible to adapt the system to exactly what is needed. This is a technique that has been pioneered by National Instruments, but being a very beneficial approach it is also starting to be emulated by other manufacturers in a number of different ways.
Again, Bains commented on the cost of test: “ . . another concern is the on-going pressure on the cost of test. And reducing the overall cost is being driven by the customer needs.
Bains further commented: “Driving the concept of Moore’s Law into the instrumentation world guides us into smaller form factors and pushes us towards the development of lower cost test products. One way this enables us to get to lower cost is that it forces more integration. On top of this, it requires very careful design in all aspects – electrical; mechanical; software; and so forth.”
Another area that helps is the modular approach adopted. Bains said: “Modularity allows you measure MIMO systems without having to build up a whole rack of equipment so you can build out a 2x2 or 4x4 MIMO system into one or possibly two chassis. That knocks down the cost. Consider the number of boxes you would need to do this with bench equipment.”
Another challenge for the developers of test equipment is size. No longer are development teams and managers satisfied with large stacks of equipment. This takes too much time to assemble and space is often at a premium. Many people require much smaller systems that are able to provide considerably higher levels of performance and more functionality.
Jin Bains stated: “Also size is a challenge: space and power are issues, because to gain the performance often requires high current levels in the RF amplifiers. You end up having to be creative on a variety of issues from frequency planning, to the mechanical solutions.”
Often high frequency components, especially those used for microwave frequencies can take up more space. Bains stated: “Larger components do present a challenge, but equipment like the VST (Vector Signal Transceiver) uses an interesting architecture that tends to avoid a lot of these components.”
NI PXIe-5644R VST Vector Signal Transceiver
Despite aspects such as configurability, it is not possible to get away from having a good basic instrument as the foundation, and this presents very many issues and challenges in itself.
Bains said: “In our case our front ends end up having to be pretty wide open, and that is a challenge. We end up having to use technologies that use pre-selection and filtering on both the RX and TX side to ensure the responses and spurious emissions are within the requirements.”
In fact the performance of the test instrument has to be much better than the unit under test to be able to measure the performance of the unit under test properly.
With performance being such an issue it is necessary to ensure the best techniques are adopted and this requires considerable thought, especially in the RF arena. Techniques such as using direct conversion receivers, IQ modulators and demodulators are used by all, but what then differentiates the different offerings.
Bains said: “We did a lot of algorithm development to enable a direct conversion system to provide the right level of performance. We did a lot of DSP work which is patented so there is some very good IP there.”
“We do corrections for the imperfections in the hardware performance within the DSP. These compensate for some of the hardware imperfections. Flatness, IQ correction, imbalance and image rejection,” he stated were all involved.
In fact one element of performance is the speed of test. This is becoming more important as often new devices or products need to be characterised. By ensuring that the performance and flexibility are maintained, it is possible to return overall test speeds many times that which would normally be possible.
“Speed,” Bains said, “is always a differentiator. Fast buses are one way that we can provide a differentiator. PCI Express Gen 3.0 is an advantage there. The other advantage is through a reconfigurable FPGA architecture that allows a lot of processing in real time which takes us into software-designed instrumentation. Enhanced optimised algorithms running on the FPGA allow real time in-line processing of data with results pushed to the host.”
The front-end components themselves within the instruments also play their part. He continued: “Data converters are a key speed enabler. We need to use the latest ones, and we leverage our relationships with the chip vendors to ensure we get access to the latest chips to get more bandwidth and data.”
RF test equipment has come a long way in recent years. Although some of the basic parameters such as noise floor and the like are difficult to improve because many instruments were approaching the theoretical limits, there have been many other ways in which equipment has been improved. Flexibility and speed of test are two.
In these ways, the equipment is now able to provide considerable improvements over older equipment. For example, tests to characterise new equipment would previously only have been able to analyse spot frequencies. Now with tests speeds, sometimes ten or a hundred times faster, full characterisation is possible, and many manufacturers have developed far more effective products because of this feature.
It is also likely that RF will push the frequency limits far more. With the VHF and UHF spectrum virtually full, new 5G technologies may well use millimetre waves as part of the standard offering. This will mean that testing at these frequencies will need to become more into the basic testing scene.
In many ways, RF testing provides challenges for the equipment developer, but some significant improvements have been seen which have altered and improved the ways manufacturers develop their equipment. For the future, the technology is likely to develop even faster as the needs become more acute.
Page 1 of 1
About the author
Ian Poole is the editor of Radio-Electronics.com. Having studied at University College London to gain his degree he went on to undertake a career in electronic development working for companies including Racal. He became the hardware development manager at Racal Instruments where he was in charge of the hardware development activities within the company. Later moving in to freelance work as a consultant he also developed Radio-Electronics.com to become one of the leading publications for professional electronics engineers. He is also a Fellow of the Institution of Engineering and Technology and is the author of over 20 books.
Most popular articles in Test & measurement
Share this page
Want more like this? Register for our newsletter