Home >> May 2011 Edition >> Realistic Testing Of Satellite-Based Comms Systems
Realistic Testing Of Satellite-Based Comms Systems
By Beate Hoehne, Agilent Technologies

Providers of high-end and satellite-based communications systems seem to face never-ending demand for new services, new capabilities and greater data capacity. As a result, system designers are pushing communications hardware to operate with wider bandwidths, at higher carrier frequencies, or both.

For years, a 1 GHz bandwidth provided sufficient headroom for a wide range of services. Today, however, the need to move more data in less time is driving modulation bandwidths to 2 GHz or, in some cases, 5 GHz. In addition, these bandwidths must be available at increasingly higher carrier frequencies.

These changes present challenges to designers, and they also have implications for the testing and analysis of communication systems, subsystems and components. As described below, the necessary test equipment must provide sufficient frequency coverage and bandwidth as well as the appropriate modulation and demodulation capabilities.

HoenheFig1 Picturing The Challenges
The globalization of “information anywhere, anytime” can create challenging, but common, scenarios. For example, it’s easy to imagine a video transmission originating from a mobile phone in Japan and being received by a viewer in Europe or North America. The challenges aren’t limited to commercial applications: Military communication systems are expected to handle more information in less time, and must do so in a highly reliable and secure manner.

Across these scenarios, the systems may use standards-based modulation (e.g., WCDMA, WiMAX), specialized variants based on those standards, or fully proprietary modulation schemes. With digital modulation techniques, wideband communication systems can provide better security and improved immunity from interference. These developments further intensify the challenges in testing and analysis.

Satellite-Based Systems Focus
Because so many satellites are already in orbit, the most cost-effective approach is to modify the existing infrastructure. This can be done with new modulation techniques that enable higher data rates and expand overall system capacity. As an example, using 16-QAM and 1 Gsymbol/s can provide a data rate of 4 Gb/s.

The alternative is launching one or more new satellites. Getting a new bird in the air presents another set of challenges, in particular the need to ensure interoperability between new and existing communication links. Such links may be required between terrestrial and space-based communications, or between satellites and multiple types of military radios.

HoenheFig2 Defining A Viable Test System
The preceding scenarios begin to outline the challenges in testing and analysis. Further complications come from the nature of wideband signals: They contain a significant amount of distortion, which makes it difficult to make valid measurements.

Four essential pieces of test equipment can address these challenges (Figure 1). On the receiver side, the first item is an arbitrary waveform generator (AWG), which can simulate the necessary modulated signals. Next is an upconverter, which translates the modulated signal up to the required RF frequency.

Measurements on the transmitter side benefit from two devices, a signal or spectrum analyzer and a wideband oscilloscope. Either of these may be enhanced with vector signal analysis (VSA) software that provides the necessary demodulation capabilities and essential measurements such as error vector magnitude (EVM).

Making The Essential Measurements
Many communications standards use EVM as a key indicator of system performance (Figure 2). As a result, the test setup itself must have a very low EVM. This will improve the likelihood of detecting subtle problems within the device under test (DUT).

HoenheFig3 Another proven test is a measurement of noise power ratio (NPR) using a multi-tone stimulus. Even though this technique has been around since the 1950s, it is an informative measurement that can substitute for measurements of intermodulation distortion (IMD).

An IMD measurement focuses on the middle of the signal band. In contrast, the multi-tone NPR test can cover the entire band with a single measurement. It also creates large signal peaks that stress the communication channel more than, for example, a two-tone test. The resulting measurement provides an at-a-glance view of the noise and distortion characteristics of a communication link.
Multi-tone has another advantage: The greater the number of tones, the greater the power in the band and the more accurate the test results. Ensuring a steady measurement result requires the use of hundreds or thousands of tones, and all must stand out against any distortion present in the link (Figure 3). This technique uses a notch filter, and the center frequency and width of the notch can be easily controlled. In effect, the scope or analyzer measures the “quietness” of the NPR in the DUT.

As final advantage, the multi-tone NPR test setup is simple and repeatable. This enables meaningful comparisons of results from before and after changes to a design, component or algorithm.

Another key measurement is the frequency response of the DUT. Getting a precise result depends on consistent amplitude flatness across all tones in the stimulus signal. This technique becomes less appealing if it is necessary to write error-correction routines, a process that can be difficult, complex and time consuming (Figure 4).

Fortunately, amplitude flatness can be measured and corrected using the signal or spectrum analyzer. This is done by reading every tone from the multi-tone signal, calculating the required pre-distortion, and generating a modified multi-tone signal that provides the necessary amplitude correction.

HoenheFig4 As shown in Figure 5, this produces an extremely flat multi-tone signal. The only downside is a decrease in spurious-free dynamic range (SFDR). As a result, the AWG used to generate the multi-tone signal must have sufficient resolution (i.e., have enough bits) to provide SFDR of 65 to 80 dB.

Selecting A Suitable AWG
In communication systems, nonlinear distortion is a key characteristic. Components of second-order distortion fall outside of the signal band but third-order distortion falls within the band. At today’s typical frequencies, an AWG must be capable of 12 GSa/s to simulate third-order distortion.

Further, a wide-bandwidth AWG makes it possible to create modulation that is wider than some of today’s bands. For example, 5 GHz of analog modulation bandwidth provides up to 10 GHz of modulation bandwidth. In addition, a wideband AWG also allows easy frequency hopping across one or more bands.

Accumulating all of the foregoing requirements, the key characteristics of a suitable AWG are as follows:

HoenheFig5 High SFDR: This ensures that tones will clearly stand out from distortion. It also provides sufficient margin to enable amplitude-corrected measurements.

Flat amplitude: This enables highly precise measurements of the DUT’s frequency response.

Wide bandwidth: This makes it possible to simulate third-order distortion and test the signal band.

These capabilities are embodied in the Agilent M8190A AWG, which provides excellent signal fidelity with 14-bit resolution at 8 GSa/s or 12-bit resolution at 12 GSa/s. Compared to other commercially available AWGs, the M8190A is uniquely capable of producing these levels of high resolution and wide bandwidth simultaneously. As a result, this AWG can create signal scenarios that push communication designs to the limit and provide deeper insights into system performance.

When operating in 14-bit mode, the M8190A provides SFDR performance of up to 80 dB; with 5 GHz analog bandwidth, it offers ample headroom for common test scenarios. It also includes 2 GSa of onboard memory for storage of multiple test scenarios and advanced sequencing capabilities that enable creation of highly realistic signal scenarios. When these capabilities are used in concert, the M8190A makes testing faster and more flexible.

It seems reasonable to expect the demand for information anywhere, anytime to continue growing unabated in commercial and military applications around the world. Looking to the future, the watchword is “flexibility” with regard to communication systems and the equipment used to test those systems.

HoenheFig6 In testing, using the versatility of a high-performance AWG is an essential step toward ensuring present and future flexibility in the test system. Further, an AWG equipped with ample onboard memory and advanced sequencing capabilities makes it possible to create highly realistic signal scenarios that provide thorough and detailed testing of communications systems during development, during system validation, and before real-world deployment.

Note: The examples in this article are based on MATLAB scripts that are available online at

About the author
Beate Hoehne designs and implements marketing strategies for the most comprehensive pulse and data generator product portfolio for Agilent Technologies’ Digital Verification Solutions Division. Beate is responsible for the marketing activities along the entire product life cycle. Beate joined Agilent Technologies (formerly Hewlett-Packard) in 1989 as a consultant in the computer business.