Physics-informed neural Volterra compensation enabling over 2600× efficiency improvement in 12,057-km ultra-long-haul coherent transmission
Xingchen He, Lianshan Yan, Lin Jiang, Anlin Yi, Wei Pan, Bin Luo, Zhengyu Pu, Alan Pak Tao Lau, Changyuan Yu · 2026
Read Physics-informed neural Volterra compensation enabling over 2600× efficiency improvement in 12,057-km ultra-long-haul coherent transmission by Xingchen He, Lianshan Yan, Lin Jiang, Anlin Yi, Wei Pan, Bin Luo, Zhengyu Pu, Alan Pak Tao Lau, Changyuan Yu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13542127.txt.
Description
Xingchen He and colleagues built a one-step light-signal fix that slashes the cost of long-distance fiber cleanup. Published in 2026, the physics-informed neural method keeps the side-by-side shape of a Volterra-series transfer model that learns its fixes from data. Fiber signal warping caps reach and capacity, and standard digital back-tracking rebuilds signals step by step at high cost.
On a 1,125-km link, the method matched the best back-tracking scores at 19 percent of the work and one twenty-fifth the delay. On a 12,057-km link, costs fell to about 6.6 percent with one one-hundred-seventy-sixth the delay, an overall gain above 2,600-fold. The design suits hardware and keeps performance while cutting the performance-cost-delay trade-off. Readers get a path to usable signal-warp cleanup in next-gen optical networks.
Catalog record
| CREATOR | Xingchen He, Lianshan Yan, Lin Jiang, Anlin Yi, Wei Pan, Bin Luo, Zhengyu Pu, Alan Pak Tao Lau, Changyuan Yu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 53 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |