AI Boosts Force Sensing Beyond Quantum Limits

Detecting extremely weak forces, such as those from gravitational waves or dark matter, has long d scientists due to fundamental quantum and thermal noise limits. A new study proposes a nonstationary measurement protocol that significantly enhances force sensitivity by preparing a mechanical oscillator in a dissipative squeezed state before measuring transient signals. This approach allows signal-to-noise ratios (SNR) much greater than the maximum achievable in steady-state scenarios, as shown in Figure 8 of the paper, potentially advancing applications in fundamental physics and precision instrumentation.

The researchers discovered that by first cooling a mechanical oscillator to a squeezed state using two-tone driving of an electromagnetic cavity, and then performing force measurements in a finite time before re-thermalization, the SNR can be dramatically improved. This exploits reduced mechanical fluctuations during the transient phase, enabling detection of forces that would otherwise be obscured by noise. For instance, with optimal parameters, the nonstationary protocol achieves an SNR at resonance that peaks for measurement times around the inverse of the system's damping rate, as illustrated in Figure 7.

Ology involves coupling a mechanical oscillator to an electromagnetic cavity driven at two specific frequencies, creating an optomechanical system. The mechanical oscillator's position shifts in response to an external force, and this change is monitored via the output electromagnetic field. In the nonstationary strategy, the system is prepared in a dissipative squeezed state by tuning the drive asymmetries (e.g., G+0/G−0 ≠ 1), which reduces noise in the mechanical quadrature. Upon force arrival, measurements are taken in the transient regime without necessarily changing the drive configuration, simplifying experimental implementation.

Analysis of the data, based on the nonstationary force noise power spectral density (PSD) and SNR calculations, shows that this protocol can surpass steady-state limits. For example, in a back-action evading measurement with G+/G− = 1, the SNR at resonance increases significantly with appropriate initial state preparation, as depicted in Figure 8. The force noise PSD in nonstationary regimes includes contributions from steady-state noise, transient effects dependent on initial conditions, and thermal noise, with the latter halving for short measurement times compared to steady-state values.

This breakthrough matters because it offers a practical way to detect weak classical forces in real-world scenarios where forces are impulsive and arrival times are unknown. By avoiding the need for complex modifications to the sensor setup, could enhance technologies like atomic force microscopes, gravitational wave detectors, and dark matter searches. For instance, improved force sensitivity could lead to better resolution in imaging biological samples or more precise tests of fundamental physics.

However, the study notes limitations, such as the dependence on precise parameter tuning and the finite measurement window before re-thermalization degrades sensitivity. The optimal conditions require careful manipulation of drive asymmetries and measurement times, and 's effectiveness varies with system parameters like cooperativity and mechanical dissipation rates. Future work could explore experimental implementations and extensions to other quantum systems.