Using commercially available technology and innovative methods, researchers at NBI have pushed the limits of how fast you can detect changes in the sensitive quantum states in the qubit. Their work allows researchers to follow rapid changes in qubit performance that were previously invisible.
The workhorse of any quantum-based application aimed at the coveted, but not yet fully realized quantum computer is the qubit. It is, however, a rather fragile workhorse. Qubits, and quantum processors in general, are highly sensitive to their environment. Typically the materials in which they are embedded contain microscopic defects that are still not fully understood. These defects can spatially fluctuate extremely fast, sometimes hundreds of times per second. As they fluctuate, the rate at which a qubit loses energy, and therefore useful quantum information, also changes.
Until now, standard characterization routines, which can take up to a minute, simply couldn’t “catch” these fast fluctuations. As a result, researchers could only measure an average energy-loss rate, which often gave an incomplete picture of the qubit’s true performance.
It is a bit unfair, really – you have a trusted workhorse, it is trying to pull the plow through the field and then you scatter moving branches and stones in its path/environment at such a high speed that the plowman can’t avoid them. Not good for the crop and final harvest!
Now, a team of researchers at the Niels Bohr Institute from Center for Quantum Devices and Novo Nordisk Foundation Quantum Computing Programme, led by Dr. Fabrizio Berritta, postdoctoral researcher, has implemented a real-time adaptive measurement approach that tracks the fluctuations in the qubit energy-loss (relaxation) rate as they happen. The work is part of an international collaboration with researchers from the Norwegian University of Science and Technology, Leiden University, and Chalmers University.
Using a fast classical controller, the method continuously updates its estimate of the qubit’s relaxation rate within just a few milliseconds, close to the intrinsic timescale of the fluctuations themselves, instead of seconds or minutes as the previous approaches.
The team achieved these speeds by using a specialized controller known as a Field-Programmable Gate Array (FPGA) — a form of classical processor that can operate extremely fast. By running the experiment directly on the FPGA, they could form a “best guess” of how quickly the qubit would lose its energy based on only a handful of measurements, without resorting to the “slow” roundtrips to a normal computer.
That speed comes with a challenge: FPGAs can be exceptionally complex to program for specific tasks. Even so, Fabrizio and the rest of the team succeeded in making the controller update its internal “knowledge” — a Bayesian model — after every single qubit measurement. This allowed the system to continuously adapt how it learned about the qubit’s state as efficiently as possible.
As a result, the FPGA controller and the qubit’s environment now evolve on roughly the same timescale, with measurements and detection happening accordingly — about a hundred times faster than has ever been demonstrated before.
On top of that, nobody knew before just how fast the fluctuations happen in superconducting qubits. But now we do, thanks to Fabrizio’s work at NBI.
FPGAs have been around for a while and have found use in other scientific fields. But the commercially available controller with FPGA used here, provided by Quantum Machines in the OPX1000, has shown remarkable results – it is programmable in a coding language comparable to Python, the coding software typically applied by physicists and so it is accessible for physicists around the world.
The use of the FPGA-powered controller from Quantum Machines on state-of-the-art quantum hardware is the result of a close collaboration between the research group at the Niels Bohr Institute, led by Associate Professor Morten Kjaergaard, and Chalmers University, where the quantum processing unit was designed and fabricated “The controller enables very tight integration between logic, measurements and feedforward: these components made our experiment possible”, says Morten Kjærgaard.
The promises of quantum technologies are many. But the story has been, and in some cases still is, a tale of many birds on the roof and the occasional visit of some of them in our hands. But progress is being made all the time, and sometimes it happens in leaps.
By exposing these previously inaccessible dynamics, these results redefine the timescales relevant for the characterization and calibration of superconducting quantum processors. With today’s materials and fabrication techniques, moving toward real-time calibration and monitoring appears to be a key step forward. Research at NBI continues to push in this direction. The present progress here shows the value of collaboration between research and industry and the innovative use of untraditional means.
“Nowadays, in quantum processing units in general, the overall performance is not determined by the best qubits, but by the worst ones: those are the ones we need to focus on. The surprise from our work is that a ’good` qubit can turn into a ’bad` one in fractions of a second, rather than minutes or hours.
With our algorithm, the fast control hardware can pinpoint which qubit is ’good’ or ’bad’ basically in real time. We can also gather useful statistics on the ’bad` qubits in seconds instead of hours or days.
We still cannot explain a large fraction of the fluctuations we observe. Understanding and controlling the physics behind such fluctuations in qubit properties will be necessary for scaling quantum processors to a useful size”, Fabrizio says.
Physical Review
Data/statistical analysis
Real-Time Adaptive Tracking of Fluctuating Relaxation Rates in Superconducting Qubits
13-Feb-2026