Kevin BiskingFor 2010, the engineers at National Instruments have structured the Automated Test Outlook into five sections. In each of these categories, we highlight a major trend that we believe will significantly influence automated test in the coming one to three years. The second trend is Multichannel RF test. The other four trends have been (previous post here) or will be featured in upcoming posts on this blog.

Multichannel RF Test
To some it may appear that the wireless revolution has already made its full impact on the test and measurement industry. After all, RF measurements topped the list of technologies test engineers were required to learn about in the past two years, according to the Test & Measurement World 2009 salary survey.

Until now, many companies have been able to incrementally evolve their test architectures to absorb new RF measurement requirements. However, two critical technology trends will change this approach. First, the emergence of multiple-input multipleoutput (MIMO) wireless technology, which offers significant increases in data throughput and link range without additional bandwidth or transmit power, will change test processes. Emerging communication standards based on MIMO technology include IEEE 802.11n, Mobile WiMAX Wave 2 and 3GPP Long Term Evolution (LTE). Second, the convergence of multiple wireless radios, such as GPS and WLAN, into a single system on a chip (SOC) will create new measurement requirements.

While the addition of each new wireless standard delivers benefits to consumers, it creates challenges for today’s test engineers. Added complexity results in longer test times and cost overruns, forcing test engineers to evaluate alternative approaches. This trend has created a new demand in the marketplace for multichannel RF testing configuration. A multichannel RF test architecture enables parallel test – that is, testing multiple wireless enabled devices in parallel and/or testing multiple communication standards, such was Bluetooth and 3G, on the same device in parallel.

MIMO uses multiple antennas at both the transmitter and receiver. For example, a 2x2 system contains two transmitters and two receivers. A multichannel RF instrument architecture is required when fully characterizing a MIMO device during validation/verification or when implementing MIMO technology for nonproduction applications such as RADAR and beamforming. Today’s 2x2 MIMO system will be an 8x8 MIMO system tomorrow, making scalability a key requirement for next-generation RF test systems. With advances in multiradio SOCs, design engineers can pack additional wireless technologies, such MIMO, into already multifaceted devices such as the next-generation smartphones. Testing a MIMO radio in production does not typically require a multichannel architecture because full spectral characterization is not required. It will, however, be one more radio that requires testing.

To implement a parallel test architecture for multiradio devices, engineers will need RF instrumentation that can economically scale as they require more channels but is flexible enough to test multiple frequencies. This market requirement creates a need for a new class of application specific RF instrumentation with a parallel hardware and software architecture that includes advanced synchronization capabilities.

For example, the typical general-purpose vector signal analyzer hardware architecture is based on a three-stage superheterodyne downconversion process architecture that yields many benefits such as intermediate frequency (IF) image rejection that enables wideband acquisition on a single channel. For multichannel applications, the new class of RF instrumentation is based on simplified architectures such as signal stage downconversion and direct digital downconversion to baseband. These modern architectures and the commercial availability of low-cost, high-performance semiconductor components – such as analog-to-digital converters (ADCs)/digital-to-analog converters (DACs), field-programmable gate arrays, amplifiers, and attenuators – will reduce the cost for these instruments while preserving measurement fidelity.

In addition, the new RF instruments must be “MIMO ready,” offering a new level of synchronization that goes beyond sharing signals such as the reference clock (usually 10 MHz) and the occasional start trigger. This traditional approach of synchronizing multiple RF instruments is sufficient to guarantee simultaneous signal acquisition, but it does not guarantee true phase coherency. As a result, a multichannel RF acquisition system with only a shared 10 MHz reference is characterized by substantial channel-to-channel phase skew. Achieving true phase coherency between multiple channels of RF signal acquisition requires the synchronization of all synthesized local oscillators (LOs), ADC sample clocks, and start triggers directly between each RF instrument. Instruments with this capability can achieve better than 0.1 degree channel-to-channel skew at a 1 GHz carrier frequency.

The software component of the architecture is even more important because processing a multistandard configuration is computationally intensive. The modern software architecture enables parallel data streams where one or more processing units are dedicated to each RF channel. Common parallel processing architectures found in the marketplace today include multiprocessor, hyperthreading, multicore, and FPGA. There are still additional technologies on the horizon, such as the Intel Turbo Boost technology, which is featured in the latest-generation Intel microarchitecture codenamed Nehalem. It automatically allows processor cores to run faster than the base operating frequency if it is operating below power, current, and temperature specification limits.

To fully use these processor technologies, engineers need to apply parallel programming techniques such as task parallelism, data parallelism, and pipelining at both the algorithm and application software levels.
The multichannel test architecture reduces aggregate test times, increases test throughput, and improves instrument usage. But the flexibility of the architecture is just as important. For example, a MIMO configuration is typically dynamic in nature, whereas the manipulation of the phase and amplitude of each transmitter can optimize sign al performance and direction. With each additional MIMO transmitter, the software complexity increases exponentially.

In the same way that emerging wireless technologies such as MIMO antenna systems have profoundly influenced transceiver designs, they have left their mark on RF instrumentation. The multichannel wireless systems of the future will be based on a low-cost architecture that is parallel from signal to software.