# Event-triggered Online Proactive Network Association to Mobile Edge Computing for IoT

## TL;DR

This paper introduces an online proactive network association method for IoT that minimizes task delay and energy consumption in mobile edge computing environments, using Markov decision processes and Lyapunov optimization.

## Contribution

It proposes a novel event-triggered, mobility-aware proactive network association model with a two-stage online algorithm that does not require prior knowledge of random event distributions.

## Key findings

- The TOPNA algorithm effectively reduces task delay.
- The approach manages energy consumption under system uncertainties.
- Simulation results confirm the algorithm's effectiveness.

## Abstract

Ultra-low latency communication for mobile machines emerges as a critical technology in Internet of Things (IoT). Proactive network association has been suggested to support ultra-low latency communication with the assistance of mobile edge computing. To resolve system dynamics and uncertainty, in this paper, an online proactive network association is proposed to minimize average task delay while considering time-average energy consumption constraints. Under distributed computing and networking environments, we formulate an event-triggered proactive network association model by semi-Markov task states and independent identically distributed (i.i.d.) random events. Then we facilitate the mobility-aware anticipatory network association to predictively consider handover effects caused by the mobility. Based on the Markov decision processes (MDP) and Lyapunov optimization, the two-stage online proactive network association (TOPNA) decision algorithm is proposed without the knowledge nor distribution of random events. Simulation results exhibit the effectiveness of the proposed algorithm.