1. Introduction

A coherent transponder reports post-forward-error-correction bit error ratio (post-FEC BER) as exactly zero for its entire service life, and then reports a traffic-affecting error burst. Nothing in that series told the operations team anything. The same transponder reports pre-FEC BER continuously, and that number moves through three orders of magnitude while the service stays error-free. One of those two counters is in almost every network management system dashboard. The other one is the only one of the pair that predicts anything.

This asymmetry is the subject of the article. Optical transport elements emit far more telemetry than any operations team consumes: per-channel optical power at every reconfigurable optical add-drop multiplexer (ROADM) degree, amplifier gain and pump bias current, laser bias and thermoelectric cooler current in every pluggable, chromatic dispersion and differential group delay estimates from every coherent digital signal processor (DSP), block error counts at every optical data unit (ODU) termination point, plus the alarm stream itself. The collection problem was solved once streaming telemetry over gRPC Network Management Interface (gNMI) displaced polling. The selection problem was not. Collecting a metric is easy; showing that its value at time t shifts the probability distribution of failure at time t + Δ is not.

The distinction that carries operational weight is between a metric that measures the state of a degradation process and a metric that measures the consequence of that process crossing a threshold. Amplified spontaneous emission accumulates, splice loss creeps, a pump laser's slope efficiency falls, a connector oxidises — all of these are slow, monotonic, and observable in the optical domain long before any digital counter increments. The digital counter increments when the accumulated penalty exceeds the receiver's correction capability. By construction, the counter is the last thing to move.

Optical transport networks still run predominantly on reactive fault management, which means service disruption is guaranteed at the onset of a soft failure rather than avoided (Scientific Reports, 2026). The industry response has been machine learning applied to telemetry, and the results are real: a Random Forest regressor with velocity, acceleration and rolling-statistic features predicted time-to-failure with 73.2 ± 0.03 s mean absolute error on a public optical telemetry benchmark of 756 lightpaths across four failure classes (Scientific Reports, 2026; measured on published benchmark data). But the model is downstream of the feature, and the feature is downstream of the metric. A gradient-boosted ensemble trained on link state and alarm counts will not outperform a straight line fitted to pre-FEC BER, because the information is not in the first pair of signals.

This article works through the physics of what degrades and how it becomes observable; the mathematics of margin, trend velocity and the base-rate arithmetic that determines whether an alarm is actionable; the architecture of a telemetry path that preserves the information rather than averaging it away; and the specific, named set of metrics that carry predictive content, alongside the equally specific set that does not. It is written for the engineer deciding what to subscribe to, at what cadence, and what to do with it.

Takeaway: Predictive value comes from metrics that track a continuous physical state variable with an operating margin above a threshold. Metrics that report the crossing of that threshold — post-FEC BER, loss-of-signal alarms, unavailable seconds — record failures rather than anticipate them, regardless of how frequently they are polled.

2. Foundational Concepts

2.1 Leading, Coincident and Lagging Indicators

Borrow the classification from reliability engineering and apply it strictly. A leading indicator changes measurably before the service-affecting event, with a lead time long enough to schedule an intervention. A coincident indicator changes at the same time as the event, which is useful for localisation and root cause but not for prevention. A lagging indicator changes only after the event, and its role is reporting and service-level accounting.

The test for a leading indicator is quantitative, not intuitive. For a candidate metric x and a failure event F at horizon Δ, the metric is leading if the conditional distribution of x given that F occurs within Δ separates from the distribution of x given that it does not — measured by area under the receiver operating characteristic curve, or by the precision achievable at an operationally acceptable recall. Many metrics that feel diagnostic fail this test because the separation exists but is far too small relative to the base rate of failure. Section 4.5 works the arithmetic.

2.2 Hard Failure and Soft Failure Definitions

A hard failure removes the signal: a fiber cut, a card power loss, a pump laser end-of-life open circuit. Transition time is milliseconds or less, the pre-event optical state is often normal, and the only available response is protection switching or restoration. Hard failures are not predictable from the signal that carries them, although they are sometimes predictable from a different observable — mechanical disturbance of a cable precedes a large fraction of aerial and duct fiber cuts, and that disturbance is visible in polarisation, not in power.

Soft failures are gradual degradations that shrink margin without removing service. An EDFA pump losing slope efficiency, a splice degrading under water ingress and freeze cycling, a wavelength-selective switch port with rising insertion loss, a transponder laser drifting in frequency, a connector accumulating contamination. Soft failures dominate the predictable population because they have a state trajectory, and a trajectory can be extrapolated.

The practical consequence is that a predictive monitoring programme should be scoped honestly. It will not predict the backhoe. It will predict the amplifier, the splice, the connector, the pluggable, and — through polarisation transients — a useful subset of the mechanical events that precede the backhoe by minutes to weeks.

2.3 Margin as the Observable State Variable

Every predictable optical failure reduces to one quantity: the difference between the signal quality delivered to the receiver and the signal quality the receiver needs. Everything else — pump current, span loss, tilt, contamination — is a mechanism that moves that difference. Design margin, typically 3–6 dB allocated at planning time for aging, temperature and component drift, is the budget the degradation process consumes.

This gives a single organising principle. The most predictive metric on any lightpath is the one closest to the margin itself, measured continuously, with the operating point held constant or explicitly divided out. Pre-FEC BER and the Q-factor derived from it sit closest, because the coherent DSP computes them from the actual received constellation after all impairments have been applied. Optical signal-to-noise ratio (OSNR) measured by an optical channel monitor sits one step further away, because it captures the linear noise contribution but not nonlinear interference; see the treatment of the OSNR-to-generalised-SNR gap in DWDM channel monitoring with OCM and OSA. Component-level metrics such as pump bias current sit furthest away but have the longest lead time, because they observe the mechanism directly rather than its integrated effect.

Design rule

Build the metric hierarchy from the margin outward. Every metric earns its place by answering one of two questions: how much margin remains on this lightpath, or which component is consuming it. A metric that answers neither is inventory, not telemetry.

2.4 The FEC Waterfall and the Information Cliff

Forward error correction is the reason post-FEC BER carries no predictive information, and the mechanism is worth stating precisely. ITU-T G-series Supplement 39 states the relationship for the Reed-Solomon (255,239) code specified in ITU-T G.709: a bit error ratio of 1.8 × 10-4 at the FEC decoder input corresponds to 10-12 at the decoder output (standard-specified). Modern soft-decision FEC in coherent transponders operates with pre-FEC thresholds in the region of 2 × 10-2, delivering net coding gain in the 11–13 dB range and post-FEC BER below 10-15 (typical values; exact threshold and gain vary by code and vendor implementation).

The transfer function between input and output error ratio is not linear on any scale. Below the code's threshold, the decoder converges and the output error ratio falls to a floor many orders of magnitude below the input. Above it, decoding fails and the output error ratio rises steeply toward the input value. The transition — the waterfall — occupies a narrow band of input error ratio. So a lightpath can lose margin steadily for eighteen months, with pre-FEC BER climbing from 10-4 to 10-2, and post-FEC BER reports zero errors for every second of that period. The first non-zero post-FEC reading arrives when the margin is already gone.

The same argument applies to every performance parameter defined after FEC. ITU-T G.8201 specifies error performance events and objectives for ODUk paths, and where FEC is used, all parameters and events — background block error (BBE), errored second (ES), severely errored second (SES), unavailable second (UAS) — are defined post-FEC (standard-specified). This does not make them useless. It places them on the far side of the waterfall: BBE and BBER carry early information because a single errored block registers long before a second is severely errored, whereas UAS is pure accounting. The relationships and objectives are worked through in BBE, ES, SES and UAS relation in OTN.

Figure 1: Post-FEC Bit Error Ratio Against Pre-FEC Bit Error Ratio. Representative soft-decision FEC waterfall. The output error ratio holds at the decoder floor across two decades of input degradation, then rises by twelve orders of magnitude across a band roughly 20 percent wide in input error ratio. The exact threshold position depends on the code; the shape does not.
Table 1: Post-FEC Bit Error Ratio Against Pre-FEC Bit Error Ratio
Operating pointPre-FEC BERPost-FEC BERReported alarm state
Turn-up, full margin5.0 × 10-3< 1 × 10-15Clear
Mid-life1.0 × 10-2< 1 × 10-15Clear
Margin largely consumed1.8 × 10-21 × 10-14Clear
Waterfall edge2.0 × 10-21 × 10-12Clear
Waterfall2.1 × 10-21 × 10-8Signal degrade
Post-threshold2.4 × 10-21 × 10-3Traffic affected

Takeaway: The FEC waterfall converts a continuous physical degradation into a binary digital outcome. Every metric measured after the decoder inherits that binary character. Every metric measured before it — pre-FEC BER, DSP-estimated SNR, optical power, OSNR — retains the continuous trajectory that prediction requires.

3. Telemetry Architecture

3.1 Source Points in a Coherent Line System

Predictive metrics originate at four points, and each has a distinct sampling characteristic that constrains what the collection path must preserve.

The coherent transponder DSP holds the richest set. Its adaptive equaliser, carrier recovery loop and FEC decoder each expose internal state: pre-FEC BER and the corrected-symbol count behind it, estimated electrical SNR, residual chromatic dispersion, differential group delay, polarisation-dependent loss, carrier frequency offset, and the rotation rate of the received state of polarisation. These registers update at the symbol-block rate internally and are exposed to the management agent at whatever cadence firmware permits — sub-second in current-generation implementations.

The optical amplifier exposes pump bias current, pump temperature, module case temperature, input and output total power, gain, and gain tilt. Pump bias current under a closed gain loop is the single most direct observation of amplifier aging available, because it is the control effort the amplifier is expending to hold its setpoint.

The optical channel monitor and wavelength-selective switch expose per-channel power and, where the monitor's resolution permits, per-channel OSNR, plus WSS port insertion loss and attenuation setting. Both are the instrumented view of the line system between transponders.

The fiber plant is the least instrumented and the most consequential. In-service options are a polarimeter or the coherent receiver's own polarisation tracker for mechanical disturbance, and embedded optical time-domain reflectometry for loss localisation. Neither is universal, and the gap between per-lightpath telemetry density and fiber-plant telemetry density is the largest hole in most operators' predictive coverage.

3.2 Collection Path from DSP Register to Decision Engine

The collection path is where predictive information is most often destroyed, and the loss is almost always accidental. A DSP that computes a polarisation rotation rate every millisecond, feeding an agent that aggregates into 15-minute performance-monitoring bins, delivers a 15-minute mean. The transient is gone. It was not filtered out by a design decision; it was averaged out by a convention inherited from synchronous digital hierarchy.

The modern path replaces the polling convention with subscription. gNMI, defined by OpenConfig over gRPC, carries both configuration and high-rate streaming telemetry, and its push model exists because polling does not scale to optical telemetry density. The protocol layering and its relationship to NETCONF and RESTCONF are covered in gRPC Network Management Interface and streaming telemetry, and the wider control-versus-management split in SDN controller versus NMS in optical networks.

Telemetry Path from DSP Register to Decision Engine Block diagram with four telemetry source blocks at the top (coherent transponder DSP, EDFA and Raman module, OCM and WSS, fiber plant instrumentation) feeding a shared bus into a network element management agent, then a gNMI subscribe and NETCONF transport block, then a distribution bus into three parallel blocks (time-series store, normalisation and residual layer, detector bank), converging into an action layer block. Telemetry Path from DSP Register to Decision Engine Telemetry source blocks Coherent Transponder DSP pre-FEC BER, ESNR, DGD PDL, freq offset, SOP rate cadence: sub-second EDFA and Raman Module pump bias current, gain, tilt input/output power cadence: 1-60 s OCM and WSS per-channel power and OSNR port insertion loss cadence: 1-60 s Fiber Plant Instrumentation polarimeter SOP transients embedded OTDR traces cadence: ms to on-demand Network element management agent Network Element Management Agent YANG-modelled state - OpenConfig / OpenROADM Telemetry transport gNMI Subscribe (STREAM) and NETCONF Get cadence set per path, not per device Analytics blocks Time-Series Store raw samples retained at source cadence no premature averaging Normalisation Layer divide out operating point: gain, channel count, temperature output: residuals Detector Bank EWMA, CUSUM, trend fit, tree-ensemble time-to-failure output: lead time estimate Action layer Action Layer margin report - scheduled work order - preemptive reroute Design note Information loss boundary Aggregation into 15-minute performance-monitoring bins at the management agent removes every event shorter than 15 minutes. A state-of-polarisation transient lasting 10 ms and a pump-current step lasting 2 s both survive a 1 s gNMI subscription and both disappear into the bin mean. Choose the cadence from the event timescale, not from the historical polling convention.
Figure 2: Telemetry Path from DSP Register to Decision Engine. Each aggregation step is a chance to discard information. The design decision is where averaging happens relative to the timescale of the event being detected.
Premium Article — Free 20% Preview

Read the Full Analysis with Premium

The remaining 80% of this article — the design numbers, trade-offs and field guidance — is part of MapYourTech Premium, along with the full premium library, courses and professional tools.

Instant access · Cancel anytime · 48-hour trial available
Sanjay Yadav

Optical Communications & Network Automation Expert | Author of 3 Books for Optical Engineers | Founder, MapYourTech

Optical networking engineer with nearly two decades of experience across DWDM, OTN, coherent optics, submarine systems, and cloud infrastructure. Founder of MapYourTech.

Follow on LinkedIn

Related Articles on MapYourTech