Qubits to qudits: Using quantum mechanics to transmit information more securely

In the realm of encrypted quantum information distribution, sending a signal from point A to point B is like a baseball pitcher relaying a secret pitch call to the catcher. The pitcher has to disguise the signal from the opposing team and coaches, base runners, and even onlookers in the stands so no one else cracks the code. The catcher can’t just stay in one spot or rely on the same finger pattern every time, because savvy opponents are constantly working to decipher any predictable sequence. If the signs are intercepted or misread, the batter gains an advantage, and the entire inning can unravel for the pitcher.

But what if there was a way for pitchers to bolster their signals by adding extra layers of “dimensionality” to each call, effectively increasing the chances of delivering it correctly to the catcher no matter how many eyes are watching? What if by incorporating more nuanced gestures—a subtle shift in glove position, a specific tap on the mound—the pitcher could craftily conceal their intentions? With this higher-dimensional approach, the coded message could zip right past the other team’s efforts to steal the signs and land reliably in the catcher’s mitt.

Now, research led by Liang Feng of the University of Pennsylvania and collaborator Li Ge at the City University of New York and published last week in Physical Review X has resulted in a newly developed compact microlaser that transmits coded information more securely. Instead of the conventional two-dimensional quantum bits (qubits) that encode quantum information within a two-level system, the microlaser uses a higher-dimensional alternative, known as qudits, vastly expanding the capacity and resilience of quantum communication—akin to a pitcher layering multiple subtle gestures to avoid tipping pitches. This innovation allows quantum messages to carry more information while being less prone to interference, laying the groundwork for the future of secure, high-dimensional quantum networks.

“What we did, essentially, was shrink the gigantic optical setup that researchers typically use to create quantum signals onto a small laser chip,” says Feng, professor of electrical and systems engineering at the School of Engineering and Applied Science. “In this case, the corresponding energy consumption is quite low compared to what is being done today, but also the signal is far more robust.”

Close-up revealing how two red lens modules align with a glass plate, carefully tuning the path of the laser. The metal and plastic mounts allow researchers to fine-tune the angles for optimal quantum signal measurements. (Image: Courtesy of Yichi Zhang)

Given its compact, sleek design, Feng notes that the device can be made portable, which means a user, say a banker on Wall Street receiving encrypted tokens, can walk around with it without the noise of busy New York streets interfering with the signals being sent and received.

First author Yichi Zhang, a Ph.D. candidate at Penn Engineering, explains that quantum keys are ways to encode and decode real information. “So, imagine that every time you log onto a webpage for your bank, you need a one-time code. With this sort of quantum code, it would be theoretically impossible to crack due to the high level of specificity in the signal.” 

Zhang explains that their quantum key distribution (QKD) device generates spin-orbit photonic qudits, which refers to a special way of encoding information in light by manipulating both its shape (orbital angular momentum) and the way it “twists” as it travels through space (polarization). In other words, rather than encoding quantum information in just one property of a photon, the team’s system can multitask, allowing for more complex, high-dimensional encoding.

“Our microlaser system fundamentally improves upon existing QKD methods, which often rely on large, delicate free-space optical setups,” Zhang explains. “Before, these kinds of quantum signals required an entire optical table filled with precise, bulky equipment. With our microlaser, we’ve condensed all of that into a compact chip, which can be incorporated into real-world networking applications.”

Peeking under the hood

A key driver behind the team’s breakthrough was the use of non-Hermitian physics, which guides how energy and information may flow through a system. Unlike traditional Hermitian systems, where energy and information behave regularly and rigidly—like a perfectly balanced scale—non-Hermitian systems introduce new degrees of control, allowing energy exchange to be dynamically and conveniently fine-tuned. This flexibility enables the real-time generation and manipulation of high-dimensional spin-orbit qudit states using the extreme compact microlaser. As a result, emitted light—as the qudit carrier—can be precisely controlled, ensuring stable quantum key transmission with greater efficiency.

“We designed a microlaser that can emit four distinct quantum states with perfect spatial and temporal uniformity,” Feng explains. “This means we don’t have to worry about dephasing effects—which is just light losing synchronicity—or signal loss due to environmental fluctuations.”

The team’s experiments demonstrated that their system could reliably transmit quantum keys over simulated long-distance conditions, maintaining signal integrity across distances equivalent to over 100 kilometers in atmospheric transmission.

Artistic illustration of a quantum photonic chip generating four different spin-orbit states (shown on the two Bloch spheres). The spiraling paths on the chip guide and manipulate light, enabling secure, higher-dimensional quantum communication. (Image: Courtesy of Yichi Zhang)

“And our calculations suggest that, with further optimization—like replacing low-efficiency single-photon avalanche diodes with superconducting nanowire single-photon detectors—the system could push past 500 kilometers, making ground-to-satellite quantum communication a tangible reality,” Zhang says.

The researchers also address a major weakness in QKD known as multiphoton pulses in weak coherent state propagation, or simply put, sophisticated eavesdropping.

“In a perfect QKD system, every transmitted pulse contains just one photon, ensuring that only the intended recipient can receive it,” Feng explains. “But in practical implementations, laser-based systems sometimes generate multiphoton pulses, meaning more than one photon is emitted per signal pulse.” 

This means an eavesdropper could intercept one photon from a pulse, measure it, and send the remaining photons forward undisturbed—essentially copying the message without diminishing it and being detected.

To prevent this, the team implemented a decoy that introduces random variations in pulse intensity, so some pulses contain the expected number of photons, while others are deliberately weaker or just empty.

“By randomly altering the intensities of transmitted pulses, we can trick an eavesdropper into revealing themselves,” Zhang explains. “If someone tries to measure the quantum keys, they won’t be able to tell the difference between a real signal and a decoy, and we’ll be able to detect their interference.”

Looking ahead

On a device level, notes Feng, the team is trying to see if they can further increase the dimensionality of the system such that they can encode more quantum information and make its transmission more robust and more resilient.

“I guess next step would be, first, we want to really test this in a practical environment, such as, for example, a fiber network,” he says. “Second, beyond just QKD, can we see the application of this data for this small chip applied in a range of scenarios? For instance, can they serve as very important node in a quantum network?” 

Liang Feng is a professor of materials science and engineering and electrical and systems engineering in the School of Engineering at the University of Pennsylvania.

Yichi Zhang is a Ph.D. candidate in the Feng Group at Penn Engineering.

Other authors include Haoqi Zhao, Zihe Gao, and Tianwei Wu of Penn Engineering and Li Ge of the College of Staten Island, CUNY, with The Graduate Center, CUNY.

This research was supported by the Gordon and Betty Moore Foundation (GBMF12960 and grant DOI 10.37807); the National Science Foundation (ECCS-1846766), the Defense Advanced Research Projects Agency (W911NF-21-1-0340), and the Office of Naval Research (N00014-23-1-2882).