A new optical communications strategy is turning one of light’s most intricate properties into a potential engine for faster, more secure data networks. In a study published in Light: Science & Applications, researchers Zhou, Li, Yang and colleagues describe a “hybrid intelligent” approach designed to control high-dimensional encrypted orbital angular momentum, or OAM, combs and distribute multiple optical channels simultaneously. The work points toward a future in which a single beam of light could carry a vast number of independent data streams while also resisting interception and signal degradation.
The central idea relies on the structure of light itself. In addition to intensity, wavelength, polarization and phase, a light wave can possess orbital angular momentum. OAM occurs when the wavefront twists around the direction of propagation, forming a spiral pattern. Different twists are associated with different OAM modes, each of which can theoretically serve as an independent communication channel. Unlike conventional systems that primarily increase capacity by adding wavelengths or using more advanced modulation, OAM multiplexing adds another dimension to the information carried by light.
The researchers focus on an OAM “comb,” a collection of many distinct OAM modes arranged as a coordinated group. The concept resembles a frequency comb, in which precisely spaced optical frequencies provide a large set of stable channels. In an OAM comb, the channels are defined by different angular momentum states. If those modes can be generated, separated and detected with sufficient accuracy, they could dramatically increase the information density of free-space or fiber-based optical links. The challenge is that high-dimensional systems become difficult to manage as the number of channels grows.
Small imperfections can cause neighboring OAM modes to overlap, creating crosstalk in which information from one channel leaks into another. Atmospheric turbulence can distort beams traveling through open air, while optical components, alignment errors and device limitations can introduce additional noise. At the receiver, the system must identify the intended mode, reconstruct the transmitted information and correct errors before the data becomes usable. Managing these processes across many channels requires fast decisions and precise control—conditions that are increasingly suited to artificial intelligence.
The study’s hybrid intelligent strategy is designed to combine machine-learning capabilities with physical models and signal-processing methods. Rather than relying solely on a neural network or solely on conventional optical algorithms, a hybrid architecture can use knowledge of the communication system to guide the learning process. This may allow the system to distinguish genuine signal changes from distortions caused by turbulence, crosstalk or equipment imperfections. In practical terms, intelligence could be used to select modes, optimize transmission parameters, recognize degraded channels and improve decoding in real time.
Encryption adds another layer to the proposed architecture. Optical signals can be protected by transforming data across multiple dimensions, including OAM states, phase, amplitude and wavelength. A high-dimensional signal is more difficult to interpret without knowledge of the encoding rules, because an interceptor would need to determine not only the correct optical channel but also the relationships among many modes. The study’s emphasis on encrypted OAM comb multicasting suggests a system capable of sending protected information from one transmitter to multiple authorized receivers.
Multicasting is especially important for future networks. Conventional point-to-point links send information from one source to one destination, while multicasting distributes the same content to several endpoints. In an OAM-based architecture, a transmitter could potentially direct different encoded mode combinations toward multiple receivers, allowing shared information, private streams or dynamically allocated services to travel through the same optical infrastructure. Such functionality could support data centers, high-capacity wireless backhaul, satellite links, immersive media and other applications that demand both speed and flexible connectivity.
The significance of the work lies in bringing several demanding technologies together rather than treating them as isolated advances. OAM multiplexing can expand capacity, encryption can strengthen confidentiality, multicasting can improve network efficiency and intelligent processing can help stabilize a complex link. Combining them, however, also multiplies the engineering difficulties. A system must generate clean optical modes, preserve their identities during propagation, decode them rapidly and ensure that machine-learning decisions remain reliable under changing conditions.
For optical networks, the most important question is whether such laboratory-scale capabilities can be translated into robust, affordable hardware. Future deployments will require compact mode generators and receivers, efficient photodetectors, high-speed processors and standardized protocols capable of working with existing network equipment. Security will also need to be evaluated beyond the complexity of the optical signal itself, because every control system—including an intelligent one—can introduce new vulnerabilities. Long-term performance under atmospheric fluctuations, component aging and changing traffic demands will be equally important.
Even with those challenges, the research reflects a broader shift in communications engineering: the search for capacity is moving from simply increasing power or adding more conventional channels toward exploiting the full physical structure of light. By treating OAM modes as controllable information resources and using intelligent algorithms to coordinate them, the researchers propose a pathway toward optical networks that are denser, more adaptive and more secure. If the approach proves scalable, twisted light could become more than a striking laboratory phenomenon—it could form part of the hidden architecture carrying the world’s next generation of data.
Subject of Research: High-dimensional encrypted orbital angular momentum comb multicasting for optical data-transmission networks
Article Title: Hybrid intelligent strategy driving high-dimensional encrypted orbital angular momentum comb multicasting towards optical data-transmission networks
Article References: Zhou, S., Li, L., Yang, J. et al. Hybrid intelligent strategy driving high-dimensional encrypted orbital angular momentum comb multicasting towards optical data-transmission networks. Light Sci Appl 15, 339 (2026). https://doi.org/10.1038/s41377-026-02386-3
Image Credits: AI Generated
DOI: 10.1038/s41377-026-02386-3
Keywords: Orbital angular momentum, optical communications, high-dimensional encoding, encrypted transmission, multicasting, machine learning, optical networks, twisted light
