Integrated Sensing and Communications for V2I Networks: Dynamic Predictive Beamforming for Extended Vehicle Targets

We investigate sensing-assisted beamforming for vehicle-to-infrastructure (V2I) communication by exploiting integrated sensing and communications (ISAC) functionalities at the roadside unit (RSU). The RSU deploys a massive multi-input-multi-output (mMIMO) array mmWave. pencil-sharp mMIMO beams fine range-resolution implicate that point-target assumption is impractical, as vehicle’s geometry becomes essential. Therefore, receiver (CR) may never lie in beam, even when vehicle accurately tracked. To tackle this problem, we consider extended target with two novel schemes. For first scheme, beamwidth adjusted real-time to cover entire vehicle, followed an Kalman filter predict track position of CR according resolved scatterers. An upgraded scheme proposed splitting each transmission block into stages. stage exploited ISAC wide beam. Based on sensed results stage, second dedicated yielding significant improvements. reveal inherent tradeoff between stages terms their durations, develop optimal allocation strategy maximizes average achievable rate. Finally, simulations verify superiorities schemes over state-of-the-art methods.