Holograms and AI create an uncrackable optical encryption system

Transmitting optical information through highly distorting media has long been a challenge. Atmospheric turbulence, fog, and underwater environments distort light signals, making communication difficult. As the need for secure and efficient data transmission grows, researchers are turning to machine learning and high-power lasers to overcome these obstacles.

Traditional free-space optical communication systems struggle with unpredictable distortions. When a laser beam travels through the Earth's atmosphere or water, it experiences absorption, scattering, and turbulence. These effects cause the beam to spread, fluctuate, and lose its original shape. This limits the amount of data that can be transmitted reliably.

Machine learning has emerged as a powerful tool in photonics. Researchers have used neural networks to analyze chaotic dynamics, predict nonlinear behaviors, and enhance optical imaging techniques. These methods have proven useful for reconstructing images distorted by scattering and turbulence.

One promising approach involves using multimode fibers (MMFs). These optical fibers support multiple light paths, but signals become scrambled as they propagate. Neural networks can decode these complex speckle patterns, enabling image transmission through MMFs and other distorting environments. However, overcoming distortions in free-space communication requires a different strategy.

High-power femtosecond laser beams offer a novel solution. When these beams travel through air or liquid, they form filaments—self-guided channels created by a balance between Kerr self-focusing and ionization defocusing. This process, known as laser filamentation, has applications in remote sensing, guiding electric discharges, and opening clear optical paths in foggy conditions.

A key advantage of laser filaments is their ability to maintain a structured beam over long distances. Researchers have explored whether these filaments can be used not just to clear a path for optical signals but to directly encode and transmit information. However, nonlinear propagation introduces significant distortions, making traditional decoding methods ineffective.

Instead of relying on conventional beam reconstruction techniques, scientists have turned to artificial intelligence. Neural networks can analyze the complex transformations that laser filaments undergo and reconstruct the original encoded information.

A research team led by Stelios Tzortzakis at the Institute of Electronic Structure and Laser in Greece has developed a new optical encryption system. This system encodes information as holograms within a high-power laser beam. As the beam passes through a nonlinear medium, the information becomes completely scrambled, making it impossible to reconstruct using physical calculations.

“From rapidly evolving digital currencies to governance, healthcare, communications, and social networks, the demand for robust protection systems to combat digital fraud continues to grow,” said Tzortzakis. “Our new system achieves an exceptional level of encryption by utilizing a neural network to generate the decryption key, which can only be created by the owner of the encryption system.”

The team published their findings in Optica, demonstrating that trained neural networks can successfully retrieve spatial information from scrambled holograms. By recognizing subtle patterns in the distorted light, the neural networks reconstruct the original images with remarkable accuracy.

To test the system, the researchers encoded thousands of handwritten digits, animals, and everyday objects into laser beams. They then passed these beams through a cuvette filled with ethanol, a liquid that induces strong nonlinear effects and thermal turbulence. This setup mimicked the chaotic behavior of optical propagation through the atmosphere.

After optimizing their experimental procedure, the team trained neural networks to decode the distorted images. The results were impressive—accuracy rates reached 90–95%, with the potential for further improvement through additional training.

“Our study provides a strong foundation for many applications, especially cryptography and secure wireless optical communication, paving the way for next-generation telecommunication technologies,” said Tzortzakis. “The method we developed is highly reliable even in harsh and unpredictable conditions, addressing real-world challenges like tough weather that often limit the performance of free-space optical systems.”

The researchers plan to enhance their encryption method by incorporating multi-factor authentication. They are also exploring cost-effective alternatives to high-power laser systems to make the technology more accessible.

As digital security becomes increasingly critical, innovations like this could revolutionize secure communication, protecting sensitive information from cyber threats.

Note: Materials provided above by The Brighter Side of News. Content may be edited for style and length.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.