Predictive Network Control and Throughput Sub-Optimality of MaxWeight

TL;DR

This paper introduces Predictive Network Control (PNC), a novel model predictive control policy for wireless networks that predicts system behavior over extended horizons, leading to improved performance and stability.

Contribution

The paper proposes PNC, a new control policy based on MPC principles, incorporating extended horizon predictions and a sophisticated stochastic system model for wireless networks.

Findings

PNC outperforms standard policies in simulations.

PNC achieves a larger stability region.

Simulation results demonstrate performance gains.

Abstract

We present a novel control policy, called Predictive Network Control (PNC) to control wireless communication networks (on packet level), based on paradigms of Model Predictive Control (MPC). In contrast to common myopic policies, who use one step ahead prediction, PNC predicts the future behavior of the system for an extended horizon, thus facilitating performance gains. We define an advanced system model in which we use a Markov chain in combination with a Bernoulli trial to model the stochastic components of the network. Furthermore we introduce the algorithm and present two detailed simulation examples, which show general improved performance and a gain in stability region compared to the standard policy.