.svg){kind=small}
Cyntony recently attended Tough Stump Rodeo 2025, a test event geared towards pushing the limits of tactical communications, drones, TAK, and robotics. Industry participants included leading companies such as Boston Dynamics, Skydio, and Red Cat. Observers attended from military, search and rescue, law enforcement, and homeland security.
Benchmarking
In the final days of the Rodeo, we used the open terrain in the Ruby River valley to test the ground level limits of our long-range mesh network performance. We set up a Cyntony XTND-DIRECT base station on an ST-R 6-meter mast at the OPS center area, aiming straight down the valley (as shown below on the viewshed map generated with Radio Mobile), and deployed a wearable Doodle Labs radio, outfitted with Trival AD-62/DB-915-2450 antennas on a 3-meter mast, at test points 2 and 4 miles away.
Our goal was to evaluate real-world TCP and UDP throughput using the iPerf3 package for radios operating at both 2.45 GHz and 915 MHz, under close-to-the-ground conditions. We were particularly interested to determine peak performance from narrow-channel links (3 MHz) in a realistic field deployments.
Results and Discussion
At 4 miles, we were able to maintain a relatively stable link, achieving about 300 to 500 kbps throughput for both TCP and UDP tests in both bands. This performance exceeded the approximately 2 miles prediction of the Doodle Labs Throughput Estimation Tool for these test conditions.
While 4 miles may not sound extreme on paper, it is quite significant in practice. Operating near ground level introduces major challenges due to Fresnel zone interference. The first Fresnel zone, which needs to stay at least 60% clear to avoid destructive interference, is larger at lower frequencies. This means that at 915 MHz, the required vertical clearance is much higher than at 2.4 GHz.
Because of this, even though 915 MHz typically propagates farther in free space, its performance at ground level can degrade due to Fresnel obstruction and ground diffraction. That’s exactly what we observed: both 2.4 GHz and 915 MHz performed similarly at long range, likely because 915 MHz’s wider Fresnel zone was more compromised in this terrain.
At the 2-mile mark, Fresnel clearance was more manageable and free space RSSI higher, performance improved dramatically. On a 3 MHz channel, we consistently pulled ~3 Mbps throughput, which is excellent considering the constraints. The image below shows the Fresnel zone analysis of the 4 mile link at 915 MHz from Radio Mobile.
Conclusion
This test underscored how gain, antenna height, channel width, and Fresnel zone clearance all interact in a real deployment. It’s not enough to look at free-space path loss on paper. Terrain, diffraction, and vertical clearance play just as large a role in determining whether a link is viable. And in this case, we were able to stretch the system’s capabilities significantly farther than predicted thanks to a directional antenna and favorable terrain along the valley floor.
Bottom line: long-range, ground-level mesh networking is very possible, even low-SWaP configurations, if you understand the RF environment and design for it accordingly.