Model-based qubit noise analysis improves accuracy and speed

Researchers at the Johns Hopkins University Applied Physics Laboratory have developed model-based qubit noise spectroscopy (QNS) approaches inspired by classical signal processing techniques. The work addresses limitations of existing QNS methods, which estimate noise power at set points, by using structure in the noise spectrum to improve statistical accuracy and potentially achieve improvements. These new estimation approaches, detailed in a recent paper, maintain computational benefits while offering a path toward more effective qubit control and adaptive feedback for quantum systems. The team anticipates the flexibility of these models will also prove useful in quantum sensing. Classical signal processing techniques are now being directly applied to the analysis of qubit noise, with the Schrödinger wave autoregressive moving-average (SchWARMA) model leading this effort. Unlike existing qubit noise spectroscopy (QNS) methods that estimate noise power at specific points, these new approaches assume underlying structure within the noise spectrum to enhance accuracy. This adaptation allows for the…