Real-World Bluetooth® LE Testing
June 20, 2025
Sponsored Blog
In our world of embedded systems and wireless connectivity, there’s no shortage of synthetic tests and perfectly controlled lab scenarios. But as any engineer worth their oscilloscope knows, the real challenge lies in the real world. It’s the real world where unexpected conditions arise, and where products earn their stripes, or show their flaws. At Infineon, we’ve always leaned into practical evaluation, and our latest deep dive into Bluetooth® Low Energy (LE) throughput with the AIROC™ CYW20829 Bluetooth LE MCU follows that same hands-on philosophy.
The idea for this evaluation came about during a break in our daily grind, as the team tossed around ideas for the next “testing adventure.” We’ve explored range. We’ve investigated coexistence. But this time, the focus shifted to throughput, specifically over-the-air throughput in a noisy environment. We decided our office (in Irvine, California) qualified as pretty noisy; it’s a live, chaotic, interference-rich office, not talking about some RF-shielded, noise-free cleanroom.
Why Throughput Matters
Before diving into the test setup and results, let’s clarify the metric at hand. Throughput, measured in bits per second, represents the actual rate of successful data transfer over a communication link. This isn’t the same as bandwidth, which is a theoretical maximum. Throughput reflects reality, a measurement of what gets through after all the noise, interference, and packet loss are accounted for.
When it comes to Bluetooth® LE, understanding real-world throughput is vital for applications that rely on streaming, data logging, or bulk transfers. And because Bluetooth operates in the crowded 2.4-GHz band, with everything from Wi-Fi to microwave ovens fighting for spectrum, it’s not just a nice-to-have test, it’s a necessity.
Our Irvine office is a pretty good proxy for a noisy, cluttered RF environment. Half of it is filled with cubicles, each home to at least a mouse, a keyboard, and a headset, all connected via Bluetooth. Add in at least two smartphones per employee (work and personal), plus several active Wi-Fi access points and printers, and you get an assortment of RF congestion.
The other half of the office? That’s our RF lab space, a borderline warzone of Bluetooth signals. Test boards are running, analyzers are scanning, and radios are chirping around the clock. It’s the sort of place where, if a Bluetooth device can survive, it can survive almost anywhere.
To get a clearer view of the RF activity, we ran a sweep of the 2.4-GHz band using a spectrum analyzer. Unsurprisingly, the results showed a sea of red, which represents a visually crowded band that would test the mettle of any Bluetooth® LE stack.
Test Setup: Mobile Meets Stationary
We conducted the test using two CYW20829 evaluation kits. One remained fixed in the center of the office as our reference node. The second became our mobile unit, moved from point to point throughout the office space. At each location, we initiated a connection to the stationary unit and conducted four throughput measurements. The average of those four readings was used to represent that location.
Data collection was handled using the AIROC™ Bluetooth® Test and Debug Tool. This tool lets us establish a connection using HCI commands, push a data payload, and measure the time taken for transfer, yielding the effective throughput in real time. The results were visualized through heatmaps that paint a clear picture of how performance varies throughout the office, depending on distance, interference, and environmental obstructions.
The Results Are In
Let’s start with the LE 1M Bluetooth® PHY. This mode emphasizes range and robustness over speed, which proved helpful in reaching the far corners of the lab. Even with active RF devices operating all around, the CYW20829 pushed through, delivering solid results. The highest throughput observed was 758 kbits/s, which is pretty close to the theoretical ceiling of about 800 kbits/s. That’s quite good considering the RF battlefield it was navigating.
Now for LE 2M Bluetooth® mode. As expected, we traded off a bit of range for speed, but the results were impressive nonetheless. We observed a peak throughput of 1348 kbits/s, which is just shy of the 1400-kbit/s ceiling. While this mode didn’t quite punch through to the farthest reaches of the lab space, it still maintained a connection into our lobby area with full file transfer integrity. That’s more than acceptable for many real-world use cases, particularly when you consider how bandwidth-hungry some modern applications have become.
Across both modes, the takeaway was consistent: the CYW20829 not only maintained connections under pressure, but did so with throughput figures that align with, and in some cases exceed, expectations given the PHY limitations.
What We Learned
This exercise reaffirmed several key points about designing and deploying Bluetooth® LE solutions:
- Real-world performance depends heavily on the operating environment. RF congestion, physical obstructions, and competing protocols all play a role.
- Throughput isn’t just a number—it’s a dynamic metric that reveals how a device will behave under practical use conditions.
- The CYW20829 is more than capable in the face of harsh, unpredictable RF environments, making it a compelling choice for Bluetooth LE applications where reliability and performance are non-negotiable.
One individual from our testing team logged far more steps than the rest of us. He returned from his testing rounds with a treasure trove of data, and even a newfound appreciation for automation. But he wasn’t done. His next suggestion? Benchmark the CYW20829 against competing Bluetooth® MCUs. That serves as a great reminder: testing doesn’t end with validating your own hardware. It continues by understanding how you stack up to everyone else.
What’s Next?
Stay tuned. We’re lining up a head-to-head comparison that puts the CYW20829 toe-to-toe with its industry peers. Because at the end of the day, it’s not just about being good. It’s about being the best.
