Gaussian Fisher Information Is Superadditive

In the world of quantum physics, scientists often treat tiny particles as independent messengers. If you want to measure a property of a system, such as its temperature, the standard wisdom has long been to send in one probe at a time. Each probe is measured separately, and the results are simply added together. This approach relies on a principle called additivity, which suggests that the total information gathered is just the sum of what each individual probe can tell you. For decades, this rule held firm for almost every type of measurement, leading researchers to believe that the most efficient way to gather data was to keep their sensors isolated from one another. However, this assumption rests on the idea that the tools used to read the data are infinitely flexible and powerful. In the real world, our instruments are not perfect; they are limited by the laws of physics and the specific technology we have built. When we restrict our tools to the linear, standard methods used in modern optics and microwave circuits, a surprising exception to the rule of independence begins to emerge.

A team of researchers at Beihang University has discovered that under these realistic constraints, two independent probes are actually better when measured together than when measured apart. They focused on a specific type of measurement common in optical and microwave experiments, known as Gaussian measurement. This method uses standard equipment like beam splitters and detectors that measure the width of a wave's oscillation. The researchers found that when two independent modes of light or microwave radiation carry information about a parameter like temperature, combining them in a specific way allows a detector to extract more information than the best possible separate measurements could ever achieve. This phenomenon, which they call superadditivity, means that the whole becomes greater than the sum of its parts, but only when the probes are processed jointly through a collective measurement.

The key to this advantage lies in how quantum mechanics handles uncertainty. When a detector tries to measure a wave, it cannot know everything about it at once. To measure one aspect of the wave, it must sacrifice knowledge of another. In a standard setup where a single probe is measured alone, the detector often has to leave a "port" open to the vacuum of empty space, which introduces a unavoidable unit of noise. This noise acts like static on a radio line, blurring the signal. The researchers showed that by bringing a second independent probe into that empty port, the noise is no longer just static; it becomes a carrier of information. The second probe fills the void, and because the two probes respond differently to the parameter being measured, the noise that would have been wasted is transformed into a useful signal. This process is akin to a balanced beam splitter, a device that mixes two light beams, followed by two detectors that read the combined output.

The study proves that this joint measurement, which they call Bell homodyne detection, consistently beats the best separate readout for a wide range of conditions. The advantage is not infinite, however. The researchers calculated a strict upper limit to how much better the joint method can be. They found that the gain in precision never exceeds 17.157 percent for any possible scenario. For the specific and common case of thermal modes, which are systems in thermal equilibrium like a hot object emitting radiation, the maximum gain is even more precise: 12.699 percent. This peak performance occurs when the two probes have a specific frequency ratio of about 3.318 to one. The researchers did not just suggest this possibility; they provided a rigorous mathematical proof that no other Gaussian measurement can do better than this specific setup under these conditions.

This finding overturns a long-standing conjecture that had suggested such a gain was impossible for thermal systems. Previous theories argued that if two systems share the same temperature, no joint measurement could outperform separate ones. The new work demonstrates that this is only true if the systems are identical in every way. If the two probes have different frequencies, even if they are at the same temperature, they respond to changes in that temperature at different rates. This mismatch is the resource that the joint measurement exploits. The researchers showed that any difference in frequency opens a window of opportunity where the collective measurement wins. They verified this using interval arithmetic, a method that checks every possible value within a range to ensure the result holds true without exception.

The practical implications are significant because the hardware required to achieve this gain already exists. The setup does not require exotic, unproven technology. It can be realized using two coupled resonators, which are simple devices that trap electromagnetic waves, measured by standard homodyne detectors. Alternatively, it can be achieved with a phase-preserving amplifier, a common component in microwave circuits, where the second probe is fed into the amplifier's "idler" band. The researchers demonstrated that standard components, such as coupled resonators or amplifiers, can reach these theoretical limits. In one specific configuration involving a traveling-wave amplifier, they showed that the gain could be as high as 10 percent in ideal conditions, and still remain significant even with the losses found in real-world equipment.

The study also clarifies the boundaries of this advantage. It is not a universal rule that applies to all quantum measurements. If the probes are identical in every way, or if the measurement is allowed to be infinitely complex and non-linear, the advantage disappears. The gain is strictly a property of the limitations imposed by Gaussian measurements. Furthermore, the researchers proved that adding more probes does not create a runaway effect; the extra benefit is always pairwise. Even with many probes, the best strategy is to pair them up and measure each pair together, rather than trying to measure them all in one giant collective bundle. This means the total gain for a large system is simply the sum of the gains from these individual pairs, and it will never exceed the 17.157 percent limit found for a single pair.

Ultimately, this work redefines how we think about the efficiency of quantum sensors. It shows that the architecture of the optimal protocol depends entirely on the tools available to the experimenter. When we are limited to the linear, standard tools of optics and microwave engineering, the old rule of measuring one by one is no longer the best path. By understanding how to fill the empty ports of our detectors with a second probe, we can turn the unavoidable noise of the quantum world into a signal, squeezing a measurable improvement in precision out of systems that were previously thought to be at their limit. The result is a clear, proven path to better thermometry and sensing, grounded in the reality of the hardware we can build today.