Reinforcement-Learning based routing for packet-optical networks with hybrid telemetry

TL;DR

This paper introduces a reinforcement learning approach for dynamic routing in packet-optical networks, utilizing physical and link layer measurements to adapt to changing network conditions and optimize performance.

Contribution

It presents a novel RL-based routing algorithm that dynamically adapts to network changes using real-time telemetry data and provides an open-source implementation.

Findings

The RL algorithm effectively adapts to network condition changes.

It improves routing decisions based on real-time measurements.

The approach is validated through simulations or experiments.

Abstract

This article provides a methodology and open-source implementation of Reinforcement Learning algorithms for finding optimal routes in a packet-optical network scenario. The algorithm uses measurements provided by the physical layer (pre-FEC bit error rate and propagation delay) and the link layer (link load) to configure a set of latency-based rewards and penalties based on such measurements. Then, the algorithm executes Q-learning based on this set of rewards for finding the optimal routing strategies. It is further shown that the algorithm dynamically adapts to changing network conditions by re-calculating optimal policies upon either link load changes or link degradation as measured by pre-FEC BER.