Active Ethernet
A point-to-point fiber access architecture in which each subscriber receives a dedicated fiber strand from an active Ethernet switch, with no splitters in the path.
Active Ethernet, sometimes abbreviated AON (Active Optical Network), is a fiber access architecture in which each subscriber receives a dedicated optical fiber strand from an active Ethernet switch at the central office or remote terminal — with no passive splitters in the optical path. Each subscriber's fiber terminates at an Ethernet-grade ONT or NID/SFP module at the premises, providing a private point-to-point optical link to the operator's switch port.
Active Ethernet is the technical opposite of PON. Where PON uses a passive splitter to share one OLT port across 32 subscribers, Active Ethernet uses a dedicated switch port and dedicated fiber strand for each subscriber. The result is dedicated symmetric bandwidth per subscriber — typically 1 Gbps, 2.5 Gbps, 10 Gbps, or higher depending on the SFP module — with no contention from other subscribers on the same physical path.
Active Ethernet electrical specifications follow IEEE 802.3 Ethernet standards over single-mode fiber: 1000BASE-LX/LH (1 Gbps to 10 km on SMF), 10GBASE-LR (10 Gbps to 10 km), 10GBASE-ER (10 Gbps to 40 km), 25GBASE-LR (25 Gbps to 10 km), and 100GBASE-LR4 (100 Gbps via 4-lane CWDM on a single fiber pair). All operate over G.652D single-mode fiber. Bidirectional (BiDi) SFP modules enable single-fiber operation using WDM (e.g., 1310 nm one direction, 1490 nm other direction) — important for access network economics because halving the fiber count per subscriber significantly reduces the OSP build cost.
Active Ethernet vs PON: PON shares one OLT port across multiple subscribers via passive splitter; Active Ethernet dedicates one switch port per subscriber. The economic trade-off favors PON for residential FTTH (cost per subscriber dominated by the ODN, where 1:32 splitting amortizes feeder cost 32× per port) and favors Active Ethernet for business-grade and high-density data center connectivity (where dedicated symmetric bandwidth justifies the higher capital and operational cost). A typical PON-FTTH deployment achieves ODN cost of $40–$80 per subscriber; Active Ethernet costs $200–$500 per subscriber due to per-subscriber switch ports, dedicated fiber, and field-deployed switch power/cooling.
Active Ethernet is the standard architecture for:
- **Business-grade access**: dedicated symmetric multi-gigabit service with strong SLAs, where shared-bandwidth PON cannot meet bandwidth or jitter guarantees. - **Government and enterprise dark fiber leases**: regulatory and contractual requirements for dedicated physical fiber strands. - **MDU high-density**: in some MDU deployments, an active Ethernet switch in the building serves each unit with a dedicated fiber, providing FTTH-class performance with simpler authentication and provisioning than in-building PON. - **Data center interconnect**: short-haul DCI links between facilities, typically 10G or 100G per fiber pair.
Active Ethernet network operations are more demanding than PON: each field-deployed switch requires AC power, battery backup (typically 8 hours under BEAD rules), HVAC or temperature-controlled enclosure, and remote management. The operating cost of a typical residential Active Ethernet deployment runs 3–5× higher per subscriber than PON over a 20-year asset life. This is why Active Ethernet has not displaced PON for mass-market FTTH, despite its per-subscriber bandwidth advantage.
Accurate Active Ethernet topology in the GIS captures dedicated per-subscriber fiber strand assignments, switch port assignments, SFP module models, and field-switch site locations. MapItRight models Active Ethernet alongside PON in the same network, supporting operators who use Active Ethernet for business customers and PON for residential in the same service area — including shared fiber bundles where some strands serve PON splitters and others serve dedicated business circuits.