Stabilograms – A Visual Tool for Testbed Performance

Stabilograms – A Visual Tool for Testbed Performance


Credit: ©italianestro/Shutterstock



NIST’s Testbeds offer real-world environments for conducting rigorous and replicable research.  Reliability of the testbed environment is imperative. Thanks to a new technology created by NIST researchers the ability to convert testbed variations into a visual representation to help scientists and engineers analyze their progress is now available.  


Testbed stability should be monitored or assessed as part of the execution and analysis of any experiment. Researchers at the National Institute of Standards and Technology have devised a simple graphical device, called a stabilogram, to provide an easy visual check for potential instability in a testbed’s operation.  


These testbeds often involve a multitude of measurement equipment and devices, running very complex execution protocols over long periods. Changes in the testbed environment, changes in a protocol during an experiment, or “drift” in the measurement devices can all introduce testbed instabilities that may interfere with the experiment’s goals. Characterizing the stability of an experimental testbed and its operation is vital in assessing a researcher’s hypothesis. 


This novel Stabilogram tool converts testbed variations into a visual representation to help scientists and engineers analyze their progress in achieving an experiment’s goals. In addition, this tool is intrinsic to understanding enabling a method to illustrate challenges to a greater diverse audience (i.e., we want the black box to be bigger than the red box). Understanding why portions of the testbed weren’t repeatable or the cause of systematic variations is critical for producing high-quality scientific research interpretations and conclusions. 


The squares in a Stabilogram represent the different sources of variability in the experiment response: a black square for natural random measurement variation, a blue square for the signal of interest, and red squares for interfering proxy variables in the testbed ..

Support the originator by clicking the read the rest link below.