CRB Oriented Transmit Waveform Optimization for One-Bit MIMO Radar

We consider the transmit waveform design for a collocated multiple-input multiple-output (MIMO) radar system, where one-bit analog-to-digital converters (ADCs) are deployed to enable a low-lost and power-efficient hardware implementation. Focusing on improving the target parameter estimation performance, we first derive a novel one-bit Cramér-Rao bound (CRB) metric by exploiting the Bussgang-based linear signal model and the worst-case Gaussian assumption. Then, based on the maximum likelihood principle, we develop a practical one-bit parameter estimation method to approach the derived CRB performance. Next, by minimizing the above one-bit CRB objective subject to a total power constraint, we formulate a transmit waveform optimization problem, which is highly nonconvex due to the one-bit quantization. To solve this problem, a majorization-minimization framework integrated with a projected gradient descent (MMPGD) algorithm is carefully designed whose convergence and complexity analysis are also provided. Finally, numerical results substantiate the tightness of the proposed one-bit CRB and the effectiveness of the MMPGD algorithm, where clear performance improvements can be achieved compared to the existing benchmark schemes.