Sometimes the tiniest tweak can trigger an avalanche of regression test costs. It may be very tempting to upgrade to the latest technology from a vendor or the latest algorithm from the PhD’s in the signal processing group. But what’s the risk of that upgrade vs. the actual benefit? Do you dare push forward with that upgrade without commissioning a new set of costly and time consuming field tests for full validation? Is there a reasonable middle ground?
The answer is yes, there is a middle ground. An RF Recorder/Player coupled with a mature library of RF Recordings can help you find it.
In our experience, our customers conduct regression tests for at least one of these three reasons:
- To make sure nothing has broken.
- Ensure that performance hasn’t unexpectedly degraded.
- Quantify performance improvements.
Streamline Regression Tests
Before your initial deployment, you probably performed field tests. You may have had the foresight to make RF Recordings (since your engineers were out suffering in the field anyway). If so, then you should have a library of RF Recordings on the shelf ready to validate your latest release.
In many cases, this library will be all you need to validate backwards compatibility with existing signals and ensure that nothing major becomes “broken” in the latest revision. It’s important to determine that each release is actually stepping forward and not backwards. In an ideal world, regression tests clearly illustrate continuous receiver improvements. If the receiver is on a trajectory of continuous improvement it will more easily handle heavily impaired signals.
Deploy Releases without the Risk
The ability to quickly regression test in the lab—instead of the field—helps you make that risk vs. benefit upgrade decision faster, with more confidence, and less cost.
Consider equipping your engineering team with an RF recorder before sending them out to the field again. You’ll reduce your regression test cost, gain a library of recordings, and your team will definitely thank you for it.