Smart Handover with Predicted User Behavior using Convolutional Neural Networks for WiGig Systems

TL;DR

This paper tackles the instability of WiGig networks in the 60 GHz band by introducing a CNN‑based proactive handover framework that predicts future user behavior and channel conditions to select APs with superior upcoming performance. The approach collects per‑user, per‑slot data via APs (signal strengths and throughput), trains a CNN to forecast future states, and uses these predictions to drive proactive handovers framed as an AP association problem over future time horizons. A greedy auxiliary algorithm assigns users to APs to maximize predicted future throughput while mitigating unnecessary handovers, with overall complexity $O(NM + NYP)$. Simulation results show significant throughput gains (up to about $1$ Gbps) and reduced handovers compared to reactive or no‑prediction baselines, validating the practical value of learning‑based proactive handovers in WiGig environments. This work advances WiGig network management by enabling real‑time, data‑driven decisions that anticipate user movement and traffic needs, potentially improving QoS in dense, high‑frequency deployments.

Abstract

WiGig networks and 60 GHz frequency communications have a lot of potential for commercial and personal use. They can offer extremely high transmission rates but at the cost of low range and penetration. Due to these issues, WiGig systems are unstable and need to rely on frequent handovers to maintain high-quality connections. However, this solution is problematic as it forces users into bad connections and downtime before they are switched to a better access point. In this work, we use Machine Learning to identify patterns in user behaviors and predict user actions. This prediction is used to do proactive handovers, switching users to access points with better future transmission rates and a more stable environment based on the future state of the user. Results show that not only the proposal is effective at predicting channel data, but the use of such predictions improves system performance and avoids unnecessary handovers.

Figures (5)

• Figure 1: Simplified comparison between WiGig and low frequency communications, and the tradeoff between speed and coverage.
• Figure 2: The main loop of the proposed framework, showing how data is obtained to train the ML model, and how decisions based on the model are sent back to the agents.
• Figure 3: How the model improves in its prediction accuracy as it is allowed to train and see more data from the environment.
• Figure 4:
• Figure 5: The number of handovers that were performed with each method.