Real-Time Situational Awareness In Test And Evaluation
By Mark Elo, Tektronix
Verifying systems performance for a sensor, such as a radar or an electronic counter measure (ECM) system, under multiple operational conditions, is a difficult and demanding task. It includes extensive testing on the bench, in a lab environment, and out on the range before the sensor can be certified for combat use.
The risks involved in any program’s test and evaluation (T&E) phase are high, especially when testing involves scheduling large assets — such as aircraft, ships, and missiles — operating in a complex electromagnetic environment, with multiple types of emitters. Some emitters’ performance will be understood, while other emitters’ behavior may not be under operator control.
To minimize risk, success often hinges on real-time situational awareness and a thorough understanding of the radio spectrum at any given moment. What’s happening in one instant can be entirely different the next. Without real-time situational awareness, problems such as incorrect signal levels, over-ranged instrumentation, or unanticipated behavior of the system under test can lead to missing or bad data acquisitions, ultimately resulting in significant time and cost overruns.
Addressing these challenges involves the use of measurement equipment that can record large portions of the spectrum throughout the test, while simultaneously giving operators the ability to monitor the operational environment. After the test, the ability to scan the acquired data quickly to determine if the events recorded are valid ensures quality before passing the results on for full analysis and post-processing.
Understanding The Signals Under Test
To determine the best measurement device, it is important to understand the signals involved. A sensor (radar) or an ECM is best characterized as a time domain phenomenon. Fundamentally, a quantity of electromagnetic energy is transmitted that illuminates a target, and the time it takes for the reflected residual energy to arrive back at the receiver is the core of the test. The signal could be a continuous wave or a sequence of pulses with a specific mission goal. However, pulse rise and fall times, type of modulation, behavior of the transmitter amplifier, and — most importantly — transmission frequency can create a broad range of responses that need to be considered.
Time domain measurements traditionally are performed with oscilloscopes, while spectrum analyzers are best suited for frequency domain measurements. This presents a unique challenge, in that there are time domain behaviors to observe, but they are exhibited in the frequency domain. While swept spectrum analyzers offer wide frequency and dynamic ranges, their ability to characterize time domain data is limited. Oscilloscopes offer excellent time domain analysis, but lack in dynamic range, especially at high frequencies.
Advancements in analog-to-digital converter technologies, and in measurement instrumentation architectures — such as FFT-based analyzers or vector signal analyzers (the instrument captures both the phase and amplitude components of the signal, or its vector) — now allow for wide, instantaneous bandwidth captures of high dynamic range time domain data in the frequency domain. This is key when dealing with a transient/pulse-based time domain system with transmission frequencies in the gigahertz ranges, such as a radar or ECM.
Real-time spectrum analysis drives the next level of insight, beyond the ability to capture transient events in the frequency domain. Real-time measurement capability offers real-time multi-domain triggering, time-selective spectrum analysis, continuous wide bandwidth waveform storage, and high-quality persistent displays that provide much higher levels of measurement capability and insight when performing a T&E measurement campaign.
All architectures use a super-heterodyne process to convert high-frequency signals to a lower frequency for analysis (Fig. 1). The swept spectrum analyzer only measures the power in a filter (resolution bandwidth filter) at a specific frequency point derived from the frequency of the local oscillator (LO). It can display a range of power versus frequency by “sweeping” the LO and the x-axis of display. This allows for very broad frequency displays with excellent dynamic range, but the time domain acquisition bandwidth is limited to that of the resolution bandwidth filter.
Fig. 1 — Alternate instrumentation block diagrams
A vector signal analyzer (VSA) tunes the LO to a desired fixed frequency and has a much larger IF filter, versus the RBW filter bank in the spectrum analyzer. This allows large amounts of time domain data to be acquired and either displayed in the time domain, or translated to the frequency domain utilizing a fast Fourier transform (FFT).
A real-time spectrum analyzer (RTSA) is conceptually the same as a VSA, but more processing is performed earlier in the signal chain, allowing for faster processing of FFTs and triggers, thus providing the enhanced measurement capability required for a successful test.
What this all means is that the RTSA architecture allows observation of time varying behaviors in the electromagnetic spectrum and behavior of the system under test, as well as analysis in the time and frequency domain, all while simultaneously recording the events to a hard disk array.
Observing Time-Varying Behaviors In The Frequency Domain
As the RTSA architecture acquires many time domain records and converts them to frequency in real time, time and frequency can be simultaneously observed. One way to display this is with a spectrogram. A spectrogram adds the dimension of time while still allowing observation of frequency and amplitude. The x axis represents frequency, the y axis represents time, and amplitude is represented by color — usually, red for high power and green for low power.
An alternate method is to visualize this measurement by emulating the anomalies of a cathode ray tube (CRT), commonly found as the display on a spectrum analyzer before the turn of the century. A CRT utilizes a magnetically-controlled electron gun that fires electrons on a phosphor-coated screen. The screen illuminates and then slowly decays over time.
Fig. 2 shows a typical display using a phosphor emulation technique. Without phosphor emulation, the screen would just show the large LFM signal, with the CW signal “popping out” the top on the left. But with phosphor emulation, a second, lower-power LFM is shown overlapped in frequency. In addition, several single-frequency pulsed carriers and two continuous wave (CW) interferers can be observed. Of course, the bandwidth and acquisition speeds are much faster than a turn-of-the-century spectrum analyzer.
Fig. 2 — Multiple chirps in one band, shown with a phosphor emulation technique
Observing Time-Varying Behaviors In The Time Domain And Frequency Domain
As discussed in the introduction, an electromagnetic pulse of energy will vary in its relative position in time, frequency, and power level, either due to the target moving or specific countermeasure techniques being utilized. Examining multiple instances of a time-varying event is a great application for real-time analysis. As the instrument can trigger on time domain events observed in the frequency or time domain, the device under test (DUT) can be validated while demonstrating its expected behaviors.
In Fig. 3, the first two displays are time and frequency “waterfall” displays that show, over time, how the pulse and time frequency move relative to the reference pulse, demonstrating both range and doppler. A pulse table also provides the ability to classify each pulse to positively identify it is the signal of interest for the measurement.
Fig. 3 — Time vs. time, frequency vs. time pulse table, and time domain phosphor display emulation
Capturing Test Data
Characterization of system capabilities across operational conditions requires the ability to record and analyze minutes or even hours of wide bandwidth data. Equipment to meet these requirements can be very expensive, often requiring piecemeal solutions from multiple vendors or limited use, purpose-built equipment.
Lacking a commercial off-the-shelf (COTS) alternative, some radar test teams have opted to stream test data to an inherently non-deterministic computing device, such as a PC or laptop. The inevitable result of such a set-up is lost or missing data, and no ability to verify that data is being recorded as expected. When transient events of interest can be as short a few microseconds, it’s vital to have assurance that the data capture is 100-percent complete.
To address this requirement, COTS solutions are available that can stream an analyzer’s full real-time bandwidth to a redundant array of independent disks (RAID) memory. In addition, all other analysis (real-time spectrum analysis, modulation analysis, etc.) is available simultaneously during streaming. As noted, this ability to analyze while streaming ensures the quality of data collection, avoids re-runs, and ultimately saves time and reduces costs.
Putting It All Together
Fig. 4 shows a five-second recording being made by a wideband spectrum analyzer while a digital phosphor display provides real-time monitoring of the 800 MHz acquisition. The file size, available disk space, recording progress, and number of files recorded all are reported. Indicators of dropped frames and input overload also are presented in the same control screen.
Fig. 4 — Control screen for a data capture showing real-time situational awareness.
Once the spectrum data has been successfully captured, it’s critical to have software applications that can manage large files and display color-graded spectrums, spectrograms, and amplitude vs. time of files of unlimited length. Streamed recordings can easily reach tens of terabytes in size. Typical pulse analysis can include start/stop time, average/peak power, pulse duration, pulse repetition interval (PRI), and start/stop frequencies.
Test with Confidence
Understanding the radio frequency environment in real-time is not only important, it can make or break the success of an operation. Complexities in the electromagnetic spectrum add risk to any T&E measurement campaign. Problems such as incorrect signal levels, over-ranged instrumentation, or unanticipated behavior of the system under test can cause major time and cost overruns. Giving operators the ability to view the events while in data transfer mode and make real-time corrections, or halt the test and reconfigure, is key to the success of a T&E measurement campaign.
About The Author
Mark Elo is a Senior Technical Marketing Manager for Tektronix – he has over 25 years of experience in RF and microwave measurements, specializing in Communications, Radar and Electronic Warfare applications.