Transmit Waveform Design for Dual-Function Radar-Communication Systems via Hybrid Linear-Nonlinear Precoding

This article develops a transmit (Tx) waveform design technique for dual-function radar-communication systems that provide both multiple-input multiple-output (MIMO) radar and multi-user multiple-input single-output (MU-MISO) communication functionalities. We propose a hybrid linear-nonlinear precoding (HLNP) signaling scheme, in which the dual-use waveform is the superposition of linearly-precoded communication symbols and a nonlinearly-precoded waveform that improves the radar performance. To attain good radar Tx beampattern and waveform ambiguity properties, we focus on optimizing a weighted sum of the integrated main-lobe-to-sidelobe ratio (IMSR) of the Tx beampatttern and a novel angular waveform similarity metric, while ensuring a predefined signal-to-interference-plus-noise ratio (SINR) for each communication user. Practical constraints are imposed on the Tx waveform, including per-antenna power and peak-to-average-power ratio (PAPR) constraints. We propose an extended feasible point pursuit successive convex approximation (EFPP-SCA) algorithm to solve the resultant nonconvex problem and establish its convergence properties. To reduce the computational cost of designing a long Tx waveform, we further introduce a sub-block design technique. Numerical examples indicate that the proposed HLNP provides a superior performance tradeoff between sensing and communication compared to conventional nonlinear precoding.