MIMO ISAC for High-Rate Communication and Accurate Situational Awareness in High-Mobility Scenarios

Rutgers University New Brunswick, New Brunswick NJ

Investigators

Abstract

This project addresses the growing demand for high-performance, compact, and energy-efficient wireless systems by advancing Dual Function Radar Communication technology. DFRC systems integrate communication and sensing capabilities into a single platform, reducing hardware redundancy, cost, and power consumption while enhancing spectral efficiency. The research is aimed at next-generation applications — including autonomous vehicles, satellite and UAV systems, and smart manufacturing — where both high-speed communication and high-resolution sensing are critical. The project tackles key challenges in these domains, particularly the efficient allocation of spectral resources and the mitigation of Doppler-induced performance degradation in high-mobility environments where traditional systems fall short. A novel Dual-Function Radar-Communication system is proposed to support high-mobility scenarios while efficiently utilizing bandwidth for both communication and sensing. The system features a monostatic multiple-input multiple-output (MIMO) radar that transmits OTFS waveforms. Bandwidth efficiency is achieved by enabling Doppler-delay domain bins to be shared across transmit antennas. A low-complexity approach is introduced to obtain coarse target estimates within the aperture limits of the receive array. This research further leverages a key strength of MIMO radar: achieving spatial resolution beyond the receive array’s aperture through the use of a virtual array. Forming a virtual array requires orthogonal transmit waveforms. However, ensuring such orthogonality — e.g., through separate, non-overlapping resources for communication and sensing — typically reduces communication rates. Our system addresses this challenge by enabling virtual array-based high-resolution sensing while minimizing its impact on communication throughput. A central innovation is the introduction of Time-Frequency domain private bins, which enable virtual array construction at a collocated receiver. These bins, combined with a discretization of the target space around the coarse estimates, facilitate the formulation of a sparse signal recovery problem, resulting in significant improvement in target parameter estimation. Although using private bins reduces available Doppler-delay domain resources — introducing a trade-off between communication rate and sensing accuracy — only a small number are needed to achieve notable sensing gains with minimal impact on communication performance. The proposed system is implemented on a multi-antenna software-defined radio testbed at Rutgers University’s WINLAB. This prototype will empirically evaluate communication-sensing trade-offs and demonstrate the system’s robustness. The project also offers hands-on research and educational experiences for students. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.