Metropolitan Area Networks employ a sophisticated stack of technologies spanning physical layer transmission, data link protocols, and network services. Understanding these technologies is essential for architects, engineers, and operators who design, deploy, and maintain metropolitan infrastructure.

The MAN technology landscape has evolved dramatically over decades—from proprietary carrier systems to standardized, cost-effective solutions that bring LAN simplicity to metropolitan scale. This evolution reflects broader industry trends toward Ethernet ubiquity, optical networking advances, and software-defined control.

This page explores the major technology categories powering modern MANs, their operational characteristics, and guidance for technology selection.

Virtually all modern MANs are built on fiber optic infrastructure. Understanding fiber fundamentals is prerequisite to grasping higher-layer MAN technologies.

Fiber Types in MAN Environments:

Single-Mode Fiber (SMF):

• Core diameter: 8-10 micrometers
• Supports one propagation mode (straight path)
• Lower attenuation: ~0.2 dB/km at 1550nm wavelength
• Maximum reach: 40-100+ km without amplification
• Dominant choice for MAN backbone due to distance requirements
• Supports highest bit rates (100 Gbps, 400 Gbps, and beyond)

Multi-Mode Fiber (MMF):

• Core diameter: 50-62.5 micrometers
• Supports multiple propagation modes (light paths)
• Higher attenuation and modal dispersion
• Practical reach: 550m (OM4 at 10G) to 2km (OM5 at 100G)
• Primarily for building/campus connections within MAN nodes
• Lower cost transceivers and more tolerant of imperfect terminations

Dark Fiber vs. Lit Services:

MANs can leverage fiber infrastructure in two fundamental modes:

Dark Fiber:

• Raw, unlit fiber strands leased or owned
• Organization provides all transmission equipment
• Maximum control over capacity, protocols, and evolution
• Requires significant upfront investment and expertise
• Common for enterprises, research networks, and carriers building their own infrastructure

Lit Services:

• Carrier provides complete transmission service
• Bandwidth delivered at specific capacity tiers
• Lower capital investment; operational expense model
• Limited control; dependent on carrier capabilities
• Faster deployment; simpler operations

Wavelength Services (Lambda Services):

• Hybrid model: carrier provides fiber and DWDM infrastructure
• Customer leases individual wavelengths (lambdas)
• Dedicated capacity without managing optical equipment
• Growing popularity for high-bandwidth MAN links
• Typically sold at 10G, 100G, or 400G per wavelength

Carrier Ethernet represents the dominant paradigm for modern MAN services, extending familiar Ethernet technology to carrier-grade metropolitan networks. The Metro Ethernet Forum (MEF) standards define Carrier Ethernet characteristics and service types.

Defining Characteristics of Carrier Ethernet:

According to MEF, Carrier Ethernet is distinguished by five attributes:

Carrier Ethernet Service Types (MEF Service Definitions):

Carrier Ethernet OAM (Operations, Administration, Maintenance):

Robust OAM capabilities distinguish Carrier Ethernet from simple LAN switching:

IEEE 802.1ag Connectivity Fault Management (CFM):

• Continuity Check Messages (CCM): Periodic 'heartbeat' frames detect connectivity failures
• Loopback Messages (LBM/LBR): Verify reachability to specific endpoints (like 'ping' for Ethernet)
• Linktrace Messages (LTM/LTR): Trace path to destination (like 'traceroute' for Ethernet)
• Maintenance Domains (MD): Hierarchical scopes (operator, provider, customer) for OAM responsibility

ITU-T Y.1731 Performance Monitoring:

• Frame Delay Measurement (DM): Round-trip and one-way latency measurement
• Frame Loss Measurement (LM): Counts frames transmitted vs. received for loss calculation
• Synthetic Loss Measurement (SLM): Injects test frames for loss measurement without affecting customer traffic
• Alarm Indication Signal (AIS): Propagates fault indication to suppress downstream alarms

Synchronous Optical Networking (SONET) and its international counterpart Synchronous Digital Hierarchy (SDH) formed the backbone of telecommunications networks for decades. While largely superseded by Ethernet-based solutions, understanding SONET/SDH remains valuable for interoperating with legacy systems and appreciating modern technology evolution.

Historical Context:

SONET was developed by Bellcore (now Telcordia) in the 1980s, standardized by ANSI. SDH followed as ITU-T's international equivalent. These systems addressed critical limitations of earlier plesiochronous (PDH) systems:

• Standardized bit rates and frame structures across vendors
• Efficient multiplexing of lower-rate signals into higher-rate aggregates
• Built-in operations channels for management
• Sub-50ms automatic protection switching
• Precise timing distribution based on stratum clocking

SONET/SDH Protection Schemes:

SONET's protection mechanisms set the standard for carrier-grade reliability:

UPSR (Unidirectional Path-Switched Ring):

• Traffic transmitted on working fiber; simultaneously copies to protection fiber
• Receiver selects better signal (working or protection)
• Simple signaling; fast switching
• Less bandwidth-efficient (50% protection overhead)

BLSR (Bidirectional Line-Switched Ring):

• Working traffic uses portion of ring capacity
• Protection capacity shared; used only during failures
• More bandwidth-efficient than UPSR
• Two-fiber (BLSR/2) and four-fiber (BLSR/4) variants
• Span switching (single link failure) and ring switching (node failure)

Linear APS (Automatic Protection Switching):

• Point-to-point protection for non-ring topologies
• 1+1 (dedicated) or 1:1/1:N (shared) protection models
• K1/K2 bytes in overhead carry protection switching signaling

Protection Switching Time:

• Standard requires switching complete within 50 milliseconds
• Actual switching often 10-25 ms in well-engineered rings
• This sub-50ms standard influenced all subsequent transport technologies

Ethernet over SONET (EoS):

To preserve SONET investment while supporting Ethernet services, standards enable Ethernet transport over SONET/SDH:

GFP (Generic Framing Procedure):

• ITU-T G.7041 defines encapsulation of various client signals into SONET/SDH
• Frame-mapped GFP: Encapsulates complete Ethernet frames
• Transparent GFP: Maps 8B/10B coded blocks directly

VCAT (Virtual Concatenation):

• Groups multiple lower-rate paths for more granular bandwidth
• Example: Combine 7× VCG-12 = ~700 Mbps (vs. fixed OC-12 = 600 Mbps or OC-48 = 2.4 Gbps)
• Better bandwidth efficiency for intermediate-rate Ethernet services

LCAS (Link Capacity Adjustment Scheme):

• Dynamically adds/removes paths from VCAT groups
• Hitless bandwidth adjustment without service interruption
• Enables pay-as-you-grow capacity models

Current Relevance:

While new MAN deployments favor native Ethernet/OTN, SONET/SDH remains relevant for:

• Legacy system interconnection
• TDM (T1/E1, DS3) circuit transport
• Regions with slower technology refresh cycles
• Precision timing distribution (GPS backup)

Dense Wavelength Division Multiplexing (DWDM) dramatically expands fiber capacity by transmitting multiple independent optical signals on different wavelengths (colors of light) over a single fiber strand. DWDM is the foundational technology enabling petabit-scale fiber capacity.

DWDM Fundamentals:

Operating Principle:

• Each wavelength (lambda, λ) carries an independent data stream
• Wavelengths are closely spaced within the fiber's low-loss windows
• Multiplexers combine wavelengths at transmission; demultiplexers separate at reception
• Each wavelength behaves as an independent 'virtual fiber'

Wavelength Grid (ITU-T G.694.1):

• C-band: 1530-1565 nm (primary DWDM band, ~40-80 wavelengths)
• L-band: 1565-1625 nm (extended capacity, ~40+ additional wavelengths)
• Standard spacing: 100 GHz (~0.8 nm) or 50 GHz (~0.4 nm)
• Ultra-dense systems: 25 GHz or 12.5 GHz spacing (flexible grid)
• Typical metro system: 40-96 wavelengths per fiber

DWDM System Components:

Transponders:

• Convert client signals (Ethernet, OTN, etc.) to DWDM wavelength
• Encode data using advanced modulation (QPSK, 16QAM, etc.)
• Include Forward Error Correction (FEC) for enhanced reach
• May support tunable lasers for flexible wavelength assignment

Multiplexers/Demultiplexers:

• AWG (Arrayed Waveguide Grating): Passive devices using interference patterns
• Thin-Film Filters: Cascaded filters for wavelength separation
• WSS (Wavelength Selective Switch): Reconfigurable optical switching

Optical Amplifiers:

• EDFA (Erbium-Doped Fiber Amplifier): Amplifies all wavelengths simultaneously in C/L bands
• Compensates for fiber attenuation and component losses
• Enables long-distance transmission without electrical regeneration
• Metro systems may require 1-2 amplifier stages; long-haul requires many more

ROADM (Reconfigurable Optical Add-Drop Multiplexer):

• Remotely add/drop individual wavelengths at intermediate nodes
• Enables wavelength routing without manual fiber patching
• Key building block for flexible optical mesh networks
• Colorless (any wavelength to any port), Directionless (any direction), Contentionless (no wavelength blocking) = CDC ROADM

Optical Transport Network (OTN) provides a standardized digital wrapper for transporting multiple client signal types over optical networks. Defined by ITU-T G.709, OTN is often described as 'digital wrapper' technology that brings SONET-like operational benefits to modern high-speed optical networks.

OTN Rationale:

As networks evolved beyond SONET/SDH rates and native Ethernet became dominant, a new transport layer was needed to:

1. Provide structured multiplexing for diverse client signals (10G, 40G, 100G, 400G Ethernet; Fibre Channel; video; etc.)
2. Include robust overhead bytes for OAM similar to SONET
3. Enable end-to-end performance monitoring across multi-domain networks
4. Support tandem connection monitoring for service provider troubleshooting
5. Provide forward error correction to extend optical reach

OTN Frame Structure:

The OTU (Optical Transport Unit) frame contains several overhead and payload regions:

Frame Overhead (OH):

• FA-OH (Frame Alignment): Frame synchronization pattern (FAS bytes)
• OTU-OH: Forward error correction status, general communication channel
• ODU-OH: Path trace, tandem connection monitoring (TCM1-6), maintenance signals
• OPU-OH: Payload structure identifier, client mapping information

Payload Area:

• Carries client data mapped via GMP (Generic Mapping Procedure)
• ODUflex allows precise bandwidth allocation for non-standard rates

Forward Error Correction (FEC):

• Standard Reed-Solomon (255,239) FEC
• Corrects errors up to certain threshold
• Extends optical reach by ~6-10 dB (doubles distance in some cases)

OTN Multiplexing:

OTN supports hierarchical multiplexing (like SONET concatenation but more flexible):

• Multiple ODU0s multiplex into ODU1
• Multiple ODU1s and ODU2s multiplex into ODU3
• ODUflex enables virtually any bandwidth granularity
• Enables sub-wavelength grooming for efficiency

OTN for MAN Applications:

OTN provides specific benefits for metropolitan networks:

Multi-Layer Management:

• Distinct optical channel (OCh), digital path (ODU), and section (OTU) monitoring
• Correlate optical impairments with digital errors
• Enable sophisticated troubleshooting across optical and digital domains

Service Provider Interconnection:

• Defined tandem connection monitoring (TCM) points for multi-operator services
• Each provider monitors their segment independently
• Clear responsibility demarcation for SLA enforcement

Client Flexibility:

• Single OTN infrastructure carries diverse services (Ethernet, FC, video)
• Client-transparent transport without protocol-specific handling
• Future-proof for emerging client signal types

Resilience:

• OTN protection switching (SNCP - Subnetwork Connection Protection)
• Client unaware of underlying protection events
• Consistent protection regardless of client protocol

Multiprotocol Label Switching (MPLS) provides sophisticated traffic engineering, VPN services, and quality of service capabilities in metropolitan networks. While Carrier Ethernet has absorbed many traditional MPLS use cases, MPLS remains essential for complex service provider environments and advanced traffic engineering.

MPLS Fundamentals in MAN Context:

MPLS operates by assigning short, fixed-length labels to packets, enabling fast forwarding based on label values rather than complex IP header lookups:

1. Label Edge Routers (LERs) at network ingress classify packets and apply labels
2. Label Switch Routers (LSRs) in the core forward based solely on label lookups
3. Label Switched Paths (LSPs) provide deterministic paths through the network
4. Labels are often stacked, enabling hierarchical services

MPLS-Based VPN Services:

VPLS (Virtual Private LAN Service):

• Creates multipoint Layer 2 service (like E-LAN using MPLS)
• Full mesh of pseudowires between provider edge (PE) routers
• MAC learning occurs at PE devices
• Scalability challenges with very large number of sites (N² pseudowires)

EVPN (Ethernet VPN):

• Modern successor to VPLS with significant advantages
• BGP distributes MAC/IP reachability information (no flooding)
• Multihoming with active-active forwarding for redundancy
• Integrated with VXLAN for data center/MAN integration
• Preferred for new deployments

L3VPN (Layer 3 VPN):

• Routed service; PE routers participate in customer routing
• BGP carries VPNv4/VPNv6 routes with route distinguishers
• Route targets control route distribution per customer
• Supports any routing protocol between CE and PE
• Common for enterprise WAN and internet access services

Pseudowires:

• Point-to-point Layer 2 transport over MPLS
• Emulates circuits (Ethernet, TDM, ATM) across packet network
• Enables migration from legacy circuits without customer changes

Segment Routing (SR):

Segment Routing represents the evolution of MPLS, simplifying operations while maintaining traffic engineering capabilities:

SR Concepts:

• Network instructions encoded as ordered list of segments
• Prefix segments (node SIDs) identify destination nodes
• Adjacency segments (Adj-SIDs) identify specific links
• Source routing: Entire path encoded at ingress
• No per-flow state in core network (unlike RSVP-TE)

SR Advantages for MAN:

• Simplified protocol stack (no LDP, reduced RSVP complexity)
• Controller-based traffic engineering (PCE integration)
• Better scalability for large networks
• Easier troubleshooting (path visible in packet)
• Smoother transition path from traditional MPLS

SR-MPLS vs. SRv6:

• SR-MPLS: Segments encoded as MPLS labels; interoperates with existing MPLS
• SRv6: Segments encoded in IPv6 extension headers; requires IPv6 infrastructure
• Most MAN deployments favor SR-MPLS for incremental deployment

The MAN technology landscape continues evolving with several emerging trends that will shape metropolitan networking over the coming decade.

Software-Defined Networking (SDN) in MANs:

SDN principles are transforming MAN operations:

Centralized Control:

• SDN controllers maintain global network view
• Path computation and optimization at controller level
• Faster response to traffic changes and failures
• Simplified device configuration (devices follow controller instructions)

Key SDN Technologies for MAN:

• OpenFlow: Direct flow table programming (less common in production MANs)
• NETCONF/YANG: Model-driven configuration management
• Segment Routing with PCE: Controller-based traffic engineering
• Intent-Based Networking: Higher-level abstraction for service definition

Operational Benefits:

• Reduced configuration complexity
• Faster service provisioning (minutes vs. weeks)
• Consistent policy implementation
• Enhanced visibility and analytics

5G Transport (xHaul):

Mobile network evolution creates new MAN requirements:

Fronthaul (Cloud RAN):

• Connects remote radio units to centralized baseband processing
• Requires very low latency (<100 μs) and precisely synchronized timing
• Standards: eCPRI (enhanced Common Public Radio Interface)
• Dense fiber deployment to cell sites

Midhaul:

• Connects distributed units (DU) to centralized units (CU)
• Moderate latency requirements (1-10 ms)
• Transports F1 interface traffic

Backhaul:

• Connects cell sites/CUs to core network
• Traditional MAN connectivity requirements
• Higher bandwidth aggregation

MAN Implications:

• Dramatically increased fiber density in urban areas
• Precision timing distribution (IEEE 1588 PTP, SyncE)
• Latency-optimized routing becomes critical
• Traffic volumes multiply with 5G deployment

Edge Computing Integration:

Edge computing distributes processing closer to users, fundamentally impacting MAN design:

MAN-Hosted Edge Sites:

• Compute capacity located at MAN nodes, aggregation points, or central offices
• Application processing within metropolitan area for minimal latency
• Content caching reduces WAN traffic and improves performance

Design Implications:

• MAN topology optimized for edge site placement
• Power and cooling capacity at network locations
• Low-latency paths between edge sites and end users
• Integration with cloud orchestration systems

Use Cases Driving Edge-MAN Integration:

• Autonomous vehicles requiring ultra-low latency decisions
• Augmented/virtual reality with real-time rendering
• Industrial IoT with local analytics
• Gaming with edge-based streaming

Selecting appropriate MAN technologies requires systematic evaluation of requirements, constraints, and long-term considerations. This framework guides technology decisions.