FedQML-Edge: Compact Quantum Feature Sketches for Communication-Constrained Roadside Federated Learning

2026-07-30

AI Summary

This work addresses the challenge of enabling efficient federated learning at roadside units under bandwidth constraints and privacy-sensitive conditions in cooperative driving scenarios. The authors propose a quantum feature sketching approach wherein each roadside unit encodes traffic states into inputs for quantum circuits and generates nonlinear feature sketches via Pauli expectation measurements. By coupling these sketches with a logistic classifier and transmitting only model updates—never raw data—the method preserves data privacy. This study represents the first integration of quantum feature mapping into roadside federated learning. Evaluations on SUMO and NGSIM simulations demonstrate its efficacy: on NGSIM, it achieves a 14.4% reduction in test log loss compared to the best classical sketching method; on SUMO, it attains recall performance comparable to large MLPs with 7–28× lower communication overhead, effectively balancing representational capacity, communication efficiency, and privacy preservation.

Abstract

Roadside units (RSUs) supporting connected and autonomous vehicle corridors need compact models to decide when cooperative maneuvers should be rewarded, deferred, or disabled. Raw sensor streams and neural network weight checkpoints are poorly suited to bandwidth-limited, privacy-sensitive roadside learning. This paper presents $\texttt{FedQML-Edge}$, a federated quantum feature-sketching pipeline for traffic-stability gating. Each RSU constructs a traffic-state summary and sends circuit inputs to a quantum computer; Pauli expectations form a nonlinear sketch processed by a logistic classifier. Only classifier updates are shared with an aggregator, whose head supports reward gating. Raw observations, vehicle records, event traces, and quantum sketches remain private. We evaluate the method using NGSIM trajectories, SUMO predictive gating with sensing noise, and IBM Quantum hardware. On NGSIM, the Pauli sketch reduces test log loss by $14.4\%$ relative to the strongest matched classical sketch. On SUMO, it approaches larger MLPs in stable-window recall while using $7-28$ times less communication per round.

Problem

• federated learning
• communication-constrained
• roadside units
• privacy-preserving
• autonomous vehicles

Innovation

• federated quantum machine learning
• quantum feature sketch
• communication-efficient learning
• roadside unit
• Pauli expectation