Nearfield Beamforming of a Multi-Input Multi-Output Radar System

Abstract

As multi-input multi-output (MIMO) radars become essential for reducing the number of active channels and the overall size, weight, and power (SWaP) of future radars, interest in applying MIMO radar techniques to real-world missions has increased. One challenge that is not fully understood is the near-field effect of MIMO virtual arrays, which is relevant for systems with either a large array or aperture size, relatively long operating wavelengths, or short-range target distances. In this study, we first developed a Python-based, open-source simulation tool that models basic range-dependent effects in MIMO beamforming and supports the incorporation of more realistic antenna patterns. From there, we analyzed the signatures of multiple moving point targets based on simple time-multiplexing transmit MIMO in the array field-of-view, with different speeds and arrival angles, from far field to near field ranges, to derive quantitative relationships that connect the system parameters and antenna performance to the signature sidelobe levels and the angle estimation accuracy. Based on this analysis and data, an algorithm that applies a novel beamforming approach that leverages both range dependence and near-field radiation-pattern compensation to accomplish near-field target detection and location. The effectiveness of the proposed approach is validated through analytical modeling and various simulations using Python-based system simulation tools.