Joint beat-spectrum averaging in a multi-frequency MIMO radar approach on a slow-fluctuating object range detection

Radar is widely applied in detecting such as aircrafts, ships and motor vehicles, mainly for security and safety purposes. However, a small object is hard to be detected moreover if it is fluctuating. Meanwhile, utilisation of a multiple-input multiple-output (MIMO) in the radar configuration has been acknowledged in many recent works, benefiting from its waveform diversity. In this study, various processing schemes for a MIMO frequency modulated continuous waveform (FMCW) radar were evaluated in detecting a slow-fluctuating object due to water ripples. Employing small and lightweight commercial-off-the-shelf (COTS) modules, a 2×2 co-located MIMO radar configuration was constructed. Beat signals received were post-processed in MATLAB, applying a spectrum averaging (SA), beat averaging (BA) and finally, merging the BA together with SA (BA-SA) schemes. The performance was compared against various averaging methods for MIMO processing, and a single-input single-output (SISO) configuration. Performance was analysed in terms of probability of range error, range error means, root mean square error (RMSE) and scattering index (SI). It was observed that MIMO was performing against SISO, and the combination of BA-SA in MIMO signal processing yielded the best result in all performance indicators compared to other evaluated schemes.