Enabling 4-D Sensing in MU-MIMO-OFDM: A Block-Wise Signal Design With Double-Orthogonal Radar Supplement

Abstract

This paper proposes a novel multi-user multiple-input multiple-output orthogonal frequency-division multiplexing (MU-MIMO-OFDM) based integrated sensing and communication (ISAC) framework. By partitioning the time-frequency resource grid into sub-blocks, the architecture enables 4D sensing (range, velocity, azimuth, elevation) and facilitates radar data cube formation for high-speed standard radar processing. A key innovation is a radar supplement signal featuring double-orthogonality to both the communication channel and signal. Unlike conventional null-space projection (NSP) methods, which transmit radar signals solely through the channel’s null-space, our approach explicitly eliminates the cross-correlation between sensing and communication signals by exploiting the communication signal’s null-space. This minimizes radar estimation error while strictly preserving communication performance. Furthermore, we derive an optimal power allocation strategy and incorporate stabilization techniques, such as block selection, to ensure robust sensing in practical environments. Simulation results demonstrate that the proposed framework achieves superior radar detection performance with an average precision (AP) of 0.90, significantly outperforming both the separate resource allocation (SRA) method, which assigns distinct resources to communication and radar, and schemes that neglect orthogonality with respect to the communication signal. These gains are achieved while maintaining spectral efficiency comparable to a communication-only baseline, effectively validating the efficacy of the stabilized ISAC architecture.