Joint Vehicle Tracking and RSU Selection for V2I Communications with Extended Kalman Filter

TL;DR

This paper presents joint vehicle tracking and RSU selection algorithms for V2I communications, utilizing an extended Kalman filter framework to improve tracking accuracy and RSU selection efficiency.

Contribution

It introduces a novel analytical framework for vehicle tracking performance evaluation and proposes algorithms for optimal RSU selection and joint tracking to enhance V2I communication reliability.

Findings

Proposed algorithms outperform conventional SNR-based tracking systems.

The analytical metric effectively quantifies vehicle tracking performance.

Joint tracking improves accuracy while reducing sample exchange.

Abstract

We develop joint vehicle tracking and road side unit (RSU) selection algorithms suitable for vehicle-to-infrastructure (V2I) communications. We first design an analytical framework for evaluating vehicle tracking systems based on the extended Kalman filter. A simple, yet effective, metric that quantifies the vehicle tracking performance is derived in terms of the angular derivative of a dominant spatial frequency. Second, an RSU selection algorithm is proposed to select a proper RSU that enhances the vehicle tracking performance. A joint vehicle tracking algorithm is also developed to maximize the tracking performance by considering sounding samples at multiple RSUs while minimizing the amount of sample exchange. The numerical results verify that the proposed vehicle tracking algorithms give better performance than conventional signal-to-noise ratio-based tracking systems.