By the mid-1970s, high-capacity analog submarine telephone cables relied on long strings of sealed repeaters (amplifiers) to boost signals across the ocean. The SG system (including TAT-6, completed in 1976) showed how advances in reliable transistor electronics could greatly increase channel capacity, setting a demanding reliability standard for anything that would replace it.
In the mid-1980s, researchers returned to the idea of using rare-earth elements (like erbium) in glass fibers to amplify light directly. This mattered because long submarine links needed many repeaters, and converting light to electronics and back at each repeater was complex and power-hungry. The renewed research focus helped move repeaters toward all-optical approaches.
Work at the University of Southampton (including Robert J. Mears and colleagues) reported low-noise erbium-doped fiber amplification at around 1.54 μm, a key telecommunications wavelength range. This showed that a short, erbium-doped fiber segment could provide useful optical gain when pumped by a laser. It was a crucial step toward practical optical repeaters for long-haul and undersea systems.
In July 1987, AT&T began laying the first transoceanic telephone cable using fiber-optic technology, aiming for far more capacity than earlier copper systems. This project forced undersea repeater design to handle optical transmission over extreme distances. It also marked a practical transition point from copper-based submarine cables to fiber-based global links.
By 1986–1987, two major groups—one at Southampton and one at AT&T Bell Laboratories—had demonstrated erbium-doped fiber amplification and pushed it toward practical devices. This competitive, parallel progress helped turn EDFA concepts into repeatable engineering designs. The result was a new path for repeaters: amplifying light in the fiber rather than regenerating signals through electronics at every hop.
TAT-8 went into service on December 14, 1988 as the first fiber-optic transatlantic cable. Its repeaters were opto-electric-opto (OEO) regenerators, which converted the optical signal to electronics for reshaping and then back to light for onward transmission. This approach proved fiber could work across an ocean, even before optical amplifiers were ready for wide submarine deployment.
Late-1980s work improved EDFAs by using compact laser diodes to pump erbium-doped fiber, making amplification systems smaller and more suitable for repeater packaging. These improvements mattered for undersea use, where repeaters must run for many years without repair and with strict power limits. Better pumping approaches helped move optical amplification from the lab toward deployable systems.
By the early 1990s, EDFAs were widely recognized as a major solution to a central network problem: how to extend optical links without frequent, expensive electronic regeneration. Because an EDFA can amplify light directly, it can boost multiple channels carried on the same fiber. This capability aligned with the growing push for higher-capacity long-haul and submarine transmission.
In 1994, the Columbus II system entered commercial service, and it (along with the Americas cable) is documented as among the first submarine systems to use EDFAs as optical amplifier repeaters. This marked a major architectural shift: repeaters could now amplify the optical signal without converting it to electronics at each station. The change improved scalability and helped set the direction for future undersea networks.
Once optical amplification was in place, undersea systems were better positioned to use wavelength-division multiplexing (WDM), which sends multiple colors (wavelengths) of light through the same fiber. EDFAs can amplify many wavelengths together over a band, avoiding per-channel electrical regeneration. This made future capacity upgrades more practical as traffic demand rose.
By the mid-1990s, submarine “repeaters” increasingly meant optical amplifier modules (often EDFAs) rather than full OEO regenerators. This reduced the need to rebuild and retime signals electronically at each hop, simplifying the scaling of transoceanic fiber systems. The shift also helped make long submarine links more compatible with rapidly improving terminal equipment on shore.
Between 1975 and 1995, undersea repeaters evolved from high-bandwidth analog transistor amplifiers (for coaxial cables) to optical architectures built around fiber transmission. TAT-8 demonstrated practical transatlantic fiber using OEO regenerators, while EDFAs enabled a new generation of optically amplified submarine systems by 1994–1995. Together, these developments established the core repeater/amplifier approach used in most modern undersea cable networks.
Development of undersea repeaters and erbium-doped fiber amplifiers (1975–1995)