Labs: Hands-on Exercises with an RTSA

Lab 1: First Capture - Power-on Through a Wi-Fi Spectrogram

Procedure

1. Connect the SPECTRAN to the host via USB-C. Power on. Confirm the device appears in RTSA Suite PRO.
2. Connect a 2.4 GHz antenna to the RF input.
3. In RTSA Suite PRO, set center frequency to 2.437 GHz (Wi-Fi channel 6) and span to 30 MHz.
4. Set RBW to 100 kHz, FFT length to 4096, window to Hann.
5. Enable the waterfall display.
6. Watch as Wi-Fi traffic paints the waterfall. Observe beacons every 102.4 ms and data bursts in between.

Expected Result

A clear waterfall image showing Wi-Fi beacons as periodic horizontal stripes and data activity in between. The 22 MHz Wi-Fi channel is clearly visible as a wide block of energy.

Discussion

• What is the average duty cycle of a Wi-Fi network you observe?
• How do beacons differ visually from data frames?
• What happens to the waterfall when you stream a video over the Wi-Fi?

Lab 2: Finding a Transient with Persistence Display

Procedure

1. Tune the SPECTRAN to the band of interest (2.4 GHz for Bluetooth, 2.45 GHz for microwave leakage, 315 or 433 MHz for garage doors).
2. Set span and RBW appropriately.
3. Enable the persistence display. Set integration time to 5 seconds.
4. Trigger the device intermittently. Watch the persistence display accumulate.

Expected Result

Rare bursts that are nearly invisible on the live trace become bright dots on the persistence display. The same view shows continuous signals as bright bands and noise as a diffuse cloud.

Discussion

• Why does persistence reveal signals that single FFTs miss?
• What is the difference between persistence and a max-hold trace?
• When would you choose persistence over a waterfall?

Lab 3: Building a Frequency Mask Trigger for an ISM Band

Procedure

1. Set center 2.45 GHz, span 100 MHz.
2. With no nearby active transmitters, observe the noise floor for one minute. This is your baseline.
3. Open the FMT editor. Define a mask 10 dB above the noise floor across the entire span.
4. Arm the trigger. Walk away.
5. Have a friend turn on a Bluetooth device or other 2.4 GHz transmitter. The FMT should fire and capture the event.
6. Replay the captured I/Q to confirm the event was captured.

Expected Result

The mask trigger fires within milliseconds of any signal exceeding the threshold. The recorded I/Q shows pre-trigger samples (from the rolling buffer), the trigger event, and post-trigger samples.

Discussion

• How would you tighten the mask if the noise floor varies with frequency?
• Could you use FMT as a regulatory enforcement tool?
• What is the latency between event and trigger fire? Why does it matter?

Lab 4: Reading Raw I/Q and Plotting in Python

Procedure

1. Capture a 100 millisecond I/Q recording at 2.45 GHz, span 80 MHz. Save as SigMF.
2. In Python, install the sigmf package: pip install sigmf.
3. Run this code:

from sigmf import sigmffile
import numpy as np
import matplotlib.pyplot as plt
from scipy.signal import spectrogram

handle = sigmffile.fromfile('capture.sigmf-meta')
fs = handle.get_global_field('core:sample_rate')
iq = handle.read_samples()

print(f"Loaded {len(iq)} samples at {fs/1e6:.1f} MHz")

f, t, Sxx = spectrogram(iq, fs=fs, nperseg=1024, return_onesided=False)
plt.pcolormesh(t * 1e3, np.fft.fftshift(f) / 1e6,
               10*np.log10(np.fft.fftshift(Sxx, axes=0)), shading='auto')
plt.xlabel('time (ms)')
plt.ylabel('frequency (MHz from carrier)')
plt.colorbar(label='dB')
plt.title('Wi-Fi capture (offline analysis)')
plt.show()

Expected Result

A Matplotlib spectrogram nearly identical to the RTSA Suite PRO waterfall.

Discussion

• What advantages does offline Python analysis have over the built-in display?
• How would you compute the average power in a specific channel programmatically?
• Could you use this workflow for automated production testing?

Lab 5: Windowing Comparison (Hann vs Flat-top)

Procedure

1. Connect a tone source generating a clean sinusoid at 2.401 GHz at -30 dBm.
2. Tune the SPECTRAN to 2.4 GHz center, 2 MHz span.
3. With the tone exactly on a bin center, measure the amplitude. Record.
4. Shift the source frequency by half a bin (e.g., to 2.401001 GHz). Measure again. Record.
5. Switch the window from Hann to Flat-top. Repeat steps 3 and 4.

Expected Result

With Hann, the off-bin reading is 1 to 2 dB lower than the on-bin reading. With Flat-top, both readings are within 0.05 dB of each other.

Discussion

• Why is Flat-top the right choice for amplitude calibration?
• Why is Hann the better choice for general spectrum monitoring?
• When would you use Blackman-Harris instead of either?

Lab 6: EVM on a QPSK Signal

Procedure

1. Configure the source to transmit QPSK at 1 Msps with raised-cosine pulse shaping (rolloff 0.35), -30 dBm output.
2. In RTSA Suite PRO, load the QPSK modulation analysis preset.
3. Tune to the source's center frequency, set span to 5 MHz.
4. Observe the constellation diagram and the running EVM value.
5. Record the EVM in percent and dB.
6. Increase the source power to drive the analyzer's front-end into compression. Observe what happens to outer constellation points.

Expected Result

Clean constellation with EVM in the 1 to 3 percent range. As compression sets in, outer points scatter while inner points remain tight.

Discussion

• What is the EVM contribution of the analyzer itself?
• Could you separate transmitter EVM from receiver EVM?
• What other distortions would you expect to see in real transmitters?

Lab 7: Pulse Descriptor Word Extraction from a Radar Capture

Procedure

1. Set up the source to generate 1 microsecond pulses at 1 kHz PRF, center 1 GHz, output -30 dBm.
2. In RTSA Suite PRO, load the PDW extraction preset.
3. Tune to 1 GHz center, span 10 MHz.
4. Run for 5 seconds. Save the PDW table.
5. Open the table in a spreadsheet. Verify pulse width measurements, PRI calculations, and amplitude readings.

Expected Result

PDW table with 5000 rows, each containing TOA (with nanosecond precision), pulse width close to 1 microsecond, and amplitude near -30 dBm. Computed PRI distribution is centered on 1 ms with sub-microsecond variance.

Discussion

• How would the PDW extraction handle a frequency-hopping radar?
• What additional fields would you want in the PDW for EW analysis?
• How would deinterleaving work if you had two radars at different PRIs?

Lab 8: Direction Finding with IsoLOG 3D DF

Procedure

1. Mount the IsoLOG 3D DF antenna at a known surveyed position.
2. Connect to the SPECTRAN via the appropriate RF and control connections.
3. In RTSA Suite PRO, enable the DF mode and the 3D map view.
4. Place a transmitter (any 2.4 GHz signal) at a measured distance and bearing from the antenna.
5. Observe the bearing and elevation reported by the system.
6. Compare to the actual position. Compute error in degrees.

Expected Result

Bearing within 1 to 3 degrees of true position. Elevation correctly distinguishes ground-level vs elevated transmitters.

Discussion

• How does multipath affect the bearing accuracy?
• What is the optimal range for accurate DF with this antenna?
• How would you set up a two-site cross-bearing test?

Lab 9: 24/7 Recording and Search

Procedure

1. Configure RTSA Suite PRO to stream I/Q at 80 MHz bandwidth, center 2.45 GHz, to the external storage.
2. Start recording. Let it run for at least 1 hour.
3. During the recording, intentionally trigger interference: turn on a microwave oven, run a wireless camera, etc.
4. Stop recording. Note the file size (should be approximately 1.4 GB per minute, or 84 GB per hour).
5. Use RTSA Suite PRO's search function to find time periods of high power. Replay the matching segments.

Expected Result

The recording captures every event during the hour. The search function locates them by power threshold or pattern matching, and replay shows them as if live.

Discussion

• How would you scale this to multi-day recordings?
• What's the right compression strategy for long-term archives?
• How would you trigger automatic alerts during the recording?

Lab 10: Cascading Two SPECTRANs for Doubled RTBW

Procedure

1. Connect the two units via Ethernet to the same network and via the 10 MHz reference cable to ensure phase coherence.
2. In RTSA Suite PRO, configure the units for cascaded operation. Each unit covers a different center frequency, with the channels adjacent.
3. With a signal source generating a wideband test signal (an OFDM modulator with wide bandwidth), capture using both units simultaneously.
4. Verify that the stitched spectrum from both units shows a continuous wideband signal.

Expected Result

The two captures combine into a single coherent spectrum twice as wide as either unit alone. Phase relationships across the boundary are smooth.

Discussion

• What synchronization is essential for cascading?
• How would the math change for four cascaded units?
• What applications justify the added cost and complexity?