Academic Jobs Logo

Japan Achieves World Record Ultra-High-Speed Data Transmission: 1.02 Petabits per Second Over 1,808 km

NICT's Breakthrough in Multi-Core Optical Fiber Technology

Be the first to comment on this article!

You

Please keep comments respectful and on-topic.

A man standing in front of a bridge with a hat on
Photo by Annie Spratt on Unsplash

Promote Your Research… Share it Worldwide

Have a story or a research paper to share? Become a contributor and publish your work on AcademicJobs.com.

Submit your Research - Make it Global News

The Groundbreaking Achievement by NICT Researchers

In a monumental leap for optical communications, researchers from Japan's National Institute of Information and Communications Technology (NICT) have shattered world records by transmitting data at 1.02 petabits per second (Pb/s) over a staggering distance of 1,808.1 kilometers. This ultra-high-speed data transmission record equates to over 1 million gigabytes per second, capable of downloading the entire Netflix catalog in mere moments. The feat was accomplished using a novel 19-core randomly-coupled multi-core fiber (MCF) with a standard cladding diameter of 0.125 mm, ensuring compatibility with existing infrastructure.

Announced on May 29, 2025, this breakthrough was detailed in a post-deadline paper at the Optical Fiber Communication Conference (OFC 2025), titled "1.02 Petabit/s Transmission Over 1,808.1 km in a 19-Core Randomly-Coupled Multicore Fiber." It marks the first petabit-class transmission exceeding 1,000 km in standard 19-core fiber, addressing the explosive data demands of the post-5G era.

Cross-section of the 19-core optical fiber with standard cladding diameter used in NICT's record-breaking transmission

Evolution of Optical Fiber Technology: From Single-Core to Multi-Core

Optical fiber communication has evolved dramatically since the 1970s when single-mode fibers first enabled gigabit speeds. Traditional single-core fibers, with one light path (core) surrounded by cladding, hit capacity limits around 100 terabits per second (Tb/s) due to the Shannon limit and nonlinear effects. To scale, researchers turned to space-division multiplexing (SDM), packing multiple cores or modes into one fiber.

Multi-core fibers (MCFs) emerged in the 2010s, initially uncoupled (isolated cores) then coupled for uniform performance. NICT's 19-core MCF uses random coupling, where light mixes between cores, simplifying manufacturing and enabling MIMO (multiple-input multiple-output) digital signal processing (DSP) to decode signals, akin to Wi-Fi MIMO. This design maintains low loss across C-band (1,530-1,565 nm) and L-band (1,565-1,625 nm), critical for long-haul transmission.

Step-by-Step Breakdown of the Record-Setting Experiment

The setup involved a recirculating loop system simulating real-world deployment:

  • Signal Generation: 180 wavelengths across C+L bands modulated with polarization-multiplexed 16QAM (quadrature amplitude modulation), branched into 19 cores with deliberate delays to simulate distinct paths.
  • Transmission Loop: Each 86.1 km loop of 19-core fiber, repeated 21 times for 1,808 km total. Loops included multiplexers/demultiplexers and amplifiers.
  • Amplification: Custom 19-core amplifiers for C and L bands separately, compensating losses (~0.18 dB/km).
  • Reception and DSP: Coherent detection per core, followed by offline MIMO-DSP to mitigate inter-core crosstalk, yielding 1.02 Pb/s post-FEC (forward error correction).

This closed-loop approach validated stability over ultra-long distances.OFC 2025 paper abstract

Schematic of the recirculating loop transmission system used by NICT researchers

Key Innovations: Random Coupling and MIMO-DSP

The randomly-coupled MCF avoids precise core spacing, reducing fabrication complexity. Light scrambling between cores creates uniform propagation, unlike uncoupled MCFs needing S-band amplifiers (not commercial). MIMO-DSP, processing 19x19 matrices, handles interference like massive MIMO in wireless.

Sumitomo Electric optimized core pitch and trench-assisted design for low loss and crosstalk. Amplifiers used multicore EDFAs (erbium-doped fiber amplifiers), a first for 19 cores over such distance.

Comparison to Previous Records and NICT's Track Record

RecordSpeedDistanceFiber TypeDate
NICT 2025 (New)1.02 Pb/s1,808 km19-core coupled MCFMay 2025
NICT Previous0.273 Pb/s1,001 km15-mode fiber~2024
NICT Short-range1.7 Pb/s63.5 km19-core MCFPrior
NICT Standard SMF402 Tb/s50 kmSingle-modeJun 2024

Capacity-distance product: 1.86 Eb/s·km, highest for standard fibers.

The Research Team and International Collaboration

Lead authors include Ruben S. Luis, Georg Rademacher, Benjamin J. Puttnam from NICT, with contributions from Yuki Goto (NICT), Takuji Nagashima, Taiji Hayashi (Sumitomo), and internationals like Martijn van der Hout (Eindhoven Tech), Stefano Gaiani (Politecnico di Milano). Hideyuki Furukawa (NICT) oversaw.

This showcases Japan's photonics prowess, bolstered by global academia.

Implications for 6G, Data Centers, and Submarine Cables

For 6G (expected 2030s), this scales capacity 100x beyond 5G fibers. Data centers handling AI (e.g., GPT models need exabytes) benefit from intra-DC links. Submarine cables, spanning oceans, could upgrade to MCF for trans-Pacific routes.NICT press release details future infrastructure

Japan leads with investments in photonics; NICT's IOWN initiative with NTT aims for 125 Pb/s systems.

Japan's Leadership in Photonics Research

Japan dominates optical comm records (over 50% recent highs), thanks to NICT (est. 2004), Sumitomo, Furukawa Electric. Universities like Tokyo Tech, Osaka Univ contribute. Ties to higher ed via joint labs, training photonics PhDs for industry.

Challenges: talent retention amid global competition; solutions include MEXT funding, industry-academia pacts.

Overcoming Challenges and Future Directions

Challenges: inter-core crosstalk, amplifier scalability, DSP complexity (high compute). Overcome via random coupling, custom amps, advanced FEC.

Future: >10 Pb/s with more bands/cores, integration with photonic ICs, field trials. "Major step toward scalable networks," per NICT.

people walking on street during daytime

Photo by Nomadic Julien on Unsplash

Broader Impacts on Society and Economy

This fuels digital economy: telemedicine, VR/AR, autonomous systems. Environmentally, efficient fibers cut energy (optics 100x less power than copper). Japan positions as 6G exporter, boosting jobs in research.Sumitomo's fiber role

For academics: spurs MCF PhD research, collaborations.

Portrait of Dr. Nathan Harlow

Dr. Nathan HarlowView full profile

Contributing Writer

Driving STEM education and research methodologies in academic publications.

Discussion

Sort by:

Be the first to comment on this article!

You

Please keep comments respectful and on-topic.

New0 comments

Join the conversation!

Add your comments now!

Have your say

Engagement level

Frequently Asked Questions

🚀What is the exact speed and distance of NICT's record?

The National Institute of Information and Communications Technology (NICT) transmitted 1.02 petabits per second (Pb/s), equivalent to over 1 million gigabytes per second, across 1,808.1 km using 19-core fiber. NICT press

🔬How does multi-core fiber enable higher speeds?

Multi-core fiber (MCF) packs 19 light paths into one cable, scaling capacity via space-division multiplexing. Random coupling and MIMO-DSP handle interference.

📡What bands and modulation were used?

C-band (1530-1565 nm) and L-band (1565-1625 nm) with 180 wavelengths in polarization-multiplexed 16QAM modulation.

👥Who were the key researchers?

Ruben S. Luis, Georg Rademacher, Benjamin J. Puttnam (NICT), Yuki Goto, and teams from Sumitomo Electric, Eindhoven Univ, etc. Presented at OFC 2025.

📊How does this compare to previous records?

Surpasses 1.7 Pb/s over 63 km and 402 Tb/s in single-mode fiber; highest capacity-distance product at 1.86 Eb/s·km.

🌐What are the implications for 6G networks?

Enables 100x capacity scaling for 6G, supporting AI data floods and ultra-reliable low-latency comm.

🔌Can this fiber integrate with existing infrastructure?

Yes, standard 0.125 mm cladding diameter fits current cables and splicers.

What challenges were overcome?

Inter-core crosstalk via MIMO-DSP, uniform amplification for 19 cores, low-loss design across bands.

💾How does it benefit data centers and submarine cables?

Reduces latency and power for hyperscale DCs; upgrades transoceanic cables for exabyte traffic.

🔮What's next for this technology?

Field trials, more bands/cores for >10 Pb/s, photonic integration per NICT's IOWN vision.

🎓Role of Japanese universities in photonics?

Collaborations with Tokyo Tech, Osaka Univ train experts; MEXT funds advance global leadership.