Time-delay and estimation uncertainty impact on the heuristic-based power control of optical networks

Abstract

Listen

Translate article

In this work, the performance of distributed power control algorithm based on heuristic optimization and considering the time-delay and estimation uncertainty is investigated in the scenario of optical networks. Time-delay allied to signal-to-noise-plus-interference ratio estimation uncertainty at the receiver side can affect decisively the network stability and the quality solution of the transmitted power optimization. In this work we have deployed a heuristic method based on particle swarm optimization (PSO) approach to allocate suitable transmitted power for the optical network nodes. Our numerical finding has demonstrated a suitable convergence of the PSO algorithm, as well as its limits, when the delay and uncertainty caused by estimation errors arise. Indeed, the rate-of-converge of the PSO algorithm has degraded substantially when delay and uncertainty become superior to 20 %, indicating that a tolerable uncertainty is about 10–15 %.