Politecnico di Torino - Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY

+39 011 090 6100 info@tech-share.it

Method for monitoring an optical communications system

Bit Error RateFailure cause identificationInformatica Tsd EnOptical communication systemOptical telemetry


Method for detecting failures in optical communication systems and identifying the source cause of the failure, which exploits optical transmission quality parameters, such as Bit Error Rate (BER) and/or Optical Signal to Noise Ratio (OSNR), directly at the receiver of an optical communication system. Failure detection and source cause identification are based on Machine Learning and data analytics techniques.

Technical features

A continuous monitoring of the BER (and / or the OSNR) of a communication channel is carried out directly at the receiver of the communication system, collecting samples of the BER with a given sampling period (TBER). A BER “window” can thus be defined every W seconds (see figure). The window, therefore, is analyzed in order to collect its features (for example, minimum / maximum / average value of the BER inside the window) and is supplied as input to the classification module based on Machine Learning (ML). Note that this ML module is optimized for the specific communications system under consideration. The ML module classifies each window, deciding whether it presents an anomaly in the BER trend and, possibly, discriminating the cause. Several consecutive sliding windows can be observed and, after a certain number (defined by the system manager) of consecutive “anomaly alarms”, it is possible to start fault repair procedures, which will resume monitoring of the BER.

Possible Applications

  • Optical networks and systems;
  • Software Defined Networks (SDNs);
  • Microwave/wireless links;
  • Predictive maintenance for smart factories.


  • Reduced Mean Time To Repair (MTTR);
  • Specific proactive procedures can be quickly implemented to avoid service disruption;
  • Reduced cost for resources over-provisioning;
  • Provide users with improved quality of service (i.e., without disruption) even in case of equipment malfunctioning.