Abstract
This paper proposes the Sensing Base Station (SBS), a roadside infrastructure that leverages Integrated Sensing and Communication (ISAC) to detect vehicles and pedestrians while simultaneously providing high-speed data services. By repurposing communication signals for radar-like sensing, the SBS eliminates blind spots inherent in vehicle-mounted sensors and significantly improves spectrum efficiency.
Introduction
Modern road safety increasingly relies on Cooperative Intelligent Transportation Systems (C-ITS). However, current implementations face two significant hurdles: sensing blind spots and resource inefficiency. Vehicle-mounted sensors often have limited fields of view, frequently obstructed by buildings or larger vehicles. Furthermore, the separate design of sensing (radar) and communication (V2I) systems leads to hardware redundancy and inefficient use of the radio spectrum.
The proposed SBS addresses these challenges by operating from a higher vantage point, such as a traffic light or utility pole. This "bird's eye" perspective allows the SBS to detect "hidden" vehicles—those obscured from the perspective of other cars—and relay this critical information to prevent collisions.
System Architecture and Methodology
Integrating sensing and communication into a single device necessitates sophisticated interference management and beamforming. The SBS technical framework rests on three pillars:
- Double-Coupling Antenna Array: A circular uniform antenna array equipped with electromagnetic shielding was designed to allow simultaneous transmission and reception of ISAC signals while minimizing internal hardware coupling.
- Joint Beamforming Optimization: The system manages two beam types: Free Detection Beams (FDB) for wide-area scanning and Directional Communication Beams (DCB) for specific users. Using convex optimization, the authors jointly optimize weight vectors to minimize sidelobe leakage and maximize beam gain.
- TSF-DMA Interference Cancellation: A Joint Time-Space-Frequency Domain Division Multiple Access (TSF-DMA) algorithm was developed. This utilizes a Time Division Duplex (TDD) frame structure to stagger sensing echoes and uplink communication, preventing signal drowning.
Performance Analysis and Key Results
The system's performance was validated through rigorous numerical simulations using Rayleigh and Rician fading models to represent urban multipath and line-of-sight environments. Key findings include:
- Communication and Sensing Range: Under standard power limitations (20 dBm in the 24 GHz ISM band), the SBS achieves a communication rate exceeding 100 Mbps and a sensing/communication range of approximately 500 meters.
- The Power Tradeoff: A fundamental tradeoff exists between sensing and communication, governed by a power ratio ($\rho$). Simulations indicate a "sweet spot" at $\rho = 0.25$, where both functions maintain a balanced range of nearly 495 meters.
- Optimization Benefits: Jointly optimizing beams rather than simply superimposing them provides a ~5 dB improvement in beam gain, which is vital for link quality in power-constrained bands.
- Estimation Accuracy: Monte Carlo simulations confirmed that the analytical derivations for Range and Velocity Root Mean Squared Error (RMSE) accurately predict system performance.
Limitations and Future Directions
While the SBS shows great promise, several areas remain for future exploration. The current study is constrained by the 20 dBm power regulation of the 24 GHz band; exploring higher power or different frequencies (e.g., 79 GHz) could yield different performance profiles. Additionally, the current model assumes simplified "point targets" for vehicles; future work will need to address complex, extended targets and high-density pedestrian environments. Finally, real-world deployment will require coordinating multiple SBS units to manage inter-station interference and handovers effectively.
Summary updated