This paper proposes a novel waveform, namely non-uniform OTFS (NU-OTFS), for joint radar and communication applications (Radcom) in multi-user/MIMO scenarios. Based on orthogonal time frequency space (OTFS) modulation, the proposed waveform is realized by using a non-uniform symplectic finite Fourier transform (NU-SFFT) to generate non-overlapping quasi-arbitrary time-frequency representations of OTFS messages. Non-uniform sampling and sparse reconstruction algorithms within the compressed sensing framework are employed to avoid (self-)interference and enhance radar target parameter estimation. The performance of NU-OTFS and its corresponding receivers is evaluated through numerical simulations and measurements, and compared with state-of-the-art MU/MIMO Radcom OTFS system concepts. NU-OTFS allows for increased flexibility in time-frequency resource allocation and larger unambiguous radar parameter estimation while showing comparable performance to state-of-the-art OTFS multi-user communication implementations in realistic high-mobility channel conditions.@en

A generic description for common multi-carrier radar receivers is proposed. Two multi-carrier waveforms - orthogonal frequency division multiplexing (OFDM) and orthogonal time-frequency spacing (OTFS) - which could be used for joint radar and communication (JRC) applications - are considered. Sensing performances of different waveform-receiver pairs are compared theoretically. It is shown that while qualitatively, both waveforms perform similarly under the same receiver, performance differences exist between them.

@en