Advances in Analog and Digital Technologies Evolve Modern Analyzers
By Ben Zarlingo, Agilent Technologies

In stark contrast to the exponential pace of growth set by digital circuits, improvements in their analog counterparts have come slowly and at considerable cost. While today’s RF engineers correctly understand the application of Moore’s Law to digital circuits (e.g., processors and memory), they understandably wish it applied more directly to analog circuits. RF circuits are more linear and therefore not as amenable to the advances characterized by Moore’s Law. Consequently, while digital circuits continue to experience a veritable growth explosion, advancements on the RF side are hard won.

RF engineers have enjoyed some very substantial gains in recent years though, particularly with regard to the speed, resolution and fidelity of analog-to-digital converters (ADCs) and digital-to-analog converters (DACs). By combining these advances with the DSP capabilities made possible by improved processors and memory, advanced spectrum/signal analyzer designs have resulted that now combine improved performance and functionality with an evolutionary migration path from existing high-performance spectrum analyzers. Featuring all-digital IF sections with improved resolution bandwidth (RBW) filters, relatively wide analysis bandwidths and advanced processing capabilities like digital modulation analysis, such instruments are critical to meeting the challenges of characterizing signals in today’s aerospace/defense and commercial communications applications (Figure 1).

Advancing Modern Analyzers
What innovations are enabling evolutionary improvements over earlier analyzers? To answer this question, first consider that spectrum analyzer performance is fundamentally defined by a number of parameters, including:

• Analyzer noise floor, described in terms of displayed average noise level (DANL) or noise figure
• Third-order distortion, also known as third order intercept (TOI)
• Phase noise, also known as analyzer phase noise floor
• Amplitude accuracy, both single-frequency and integrated band power

One of the most critical of these parameters is noise floor—a key figure of merit that describes the analyzer’s own noise contribution to all spectrum and power measurements. Analyzer noise figure can be a limiting factor in several types of measurements. On small signals, for example, noise limits the low end of dynamic range as well as measurement accuracy. When large signals are present, analyzer attenuation is commonly increased to limit the distortion due to overload; however, this brings small signals closer to the analyzer’s own noise, and thus the analyzer noise forces a compromise in the amount of attenuation. Noise can also have a large impact on measurement speed, potentially requiring narrower RBWs for measurements like spur searches and resulting in slower sweeps due to increased settling time.

Thus, a key innovation for both low-level and high-level measurements is reducing the analyzer’s effective noise floor. In addition to improvements from components and hardware design, it is now becoming practical to precisely measure and model analyzer noise contribution and then to use fast DSP to dramatically reduce it. An example of this noise floor extension technology is shown in Figure 2.

An additional benefit of reducing analyzer noise floor is that it may allow low-level measurements to be made without preamplifiers, which can reduce the TOI performance or other elements of dynamic range.

For microwave measurements, analyzer noise figure can be degraded by signal routing or switching, which results in signal path loss. The loss goes up significantly with frequency and, while it can be characterized and corrected to preserve amplitude accuracy, it still affects the noise floor of the analyzer and thus, the accuracy of low-level signal measurements. Preamplifiers can compensate for this loss and improve signal/noise for small signals, but they can also cause distortion from larger signals present in the measurement. One innovative way to address this problem is to use an alternate low-noise input path. This low-noise path (LNP) allows lossy elements normally found in the RF input chain to be completely bypassed, resulting in the highest sensitivity possible without a preamplifier.

Innovations for Compatibility
A key concern for engineers who are looking to migrate to a new analyzer is compatibility. They are often torn between the guaranteed compatibility assured by using, and finding support for, their existing solutions and the benefits of changing to new, more supportable and more capable (e.g., in terms of performance, speed, functionality and interfaces) solutions. Further complicating matters, ATE engineers require a multifaceted level of compatibility that is difficult to achieve when migrating to a new solution.

A number of innovations that extend beyond conventional code and performance compatibility can make this migration much easier and less risky. Fast switching between compatibility and native modes, for example, provides ready access to both high compatibility and important new features. Error logs itemize unrecognized commands so engineers can determine their importance to the application at hand. Additionally, an attenuator offset mode can take advantage of dynamic range improvements by adding transparent attenuation to any manually- or automatically-selected input attenuation, allowing the analyzer to respond to overloads in a manner similar to that of an analyzer with an analog IF.

Other compatibility innovations include:

• A flexible compatibility mode with customizable settings: while code compatibility with previous-generation spectrum/signal analyzers is essential when replacing legacy instruments, a more sophisticated compatibility mode allows parameters to be individually set to compatible values or to values optimized for new measurement capabilities.

• Sweep time minimum limits: in an ATE system, faster is not always better and, in some cases, speed improvements may impair compatibility with measurement programs. In such instances, the ability to set the analyzer to automatically limit sweep speed can be extremely useful.

• Multiple, configurable IF outputs and video outputs: in some applications, compatibility depends on specific analyzer outputs, and a user-programmable IF output frequency can closely approximate multiple special IF options. Other outputs can include a log video output with fast rise time and wide bandwidth, and a traditional narrowband video output.

In addition to compatibility, information security is another key area of concern. Modern analyzers have a large amount of nonvolatile memory that can be a problem, especially if the analyzer is moved between secure environments or from secure to unsecure environments for use or for calibration. A removable hard disk drive provides an easy and dependable way to “sanitize” the analyzer, and extra/spare disk drives allow fast swapping with non-classified hard disk units when needed. An equivalent solid-state hard disk can handle applications with more stringent environmental requirements.

Conclusion
Meeting the demands of engineers working on current and future aerospace/defense and commercial communications applications requires an evolution in spectrum/signal analysis. In pursuit of that goal, analog and digital technologies are now being combined to provide tangible improvements in key spectrum analyzer performance. These include reductions in effective noise floor and wider measurement bandwidths with low distortion. Combining these improvements with extensive compatibility features produces evolutionary new analyzers that can effectively replace legacy systems and still provide engineers with the performance and functionality they need to maintain a competitive edge in today’s highly volatile marketplace.

About the Author
Ben Zarlingo is a product manager for communications test with Agilent Technologies’ Microwave & Communications Division. He received a BS in Electrical Engineering from Colorado State University in 1980 and has worked for Hewlett-Packard /Agilent Technologies in the areas of spectrum, network and vector signal analysis, with a primary focus on techniques for the design and troubleshooting of emerging communications technologies.

Agilent Technologies
www.home.agilent.com
TXTLINX.COM93
Email this article to a friend!